Rethinking Competitiveness

HHassett.indbassett.indb i 111/8/121/8/12 9:279:27 PMPM HHassett.indbassett.indb iiii 111/8/121/8/12 9:279:27 PMPM Rethinking Competitiveness

Kevin A. Hassett, Editor

The AEI Press Publisher for the American Enterprise Institute WASHINGTON, D.C.

HHassett.indbassett.indb iiiiii 111/8/121/8/12 9:279:27 PMPM Distributed by arrangement with the Rowman & Littlefield Publishing Group, 4501 Forbes Boulevard, Suite 200, Lanham, Maryland 20706. To order, call toll free 1-800-462-6420 or 1-717-794-3800. For all other inquiries, please contact AEI Press, 1150 Seventeenth Street, N.W., Washington, D.C. 20036, or call 1-800-862-5801.

Library of Congress Cataloging-in-Publication Data Rethinking competitiveness / Kevin A. Hassett, Editor. pages cm Includes bibliographical references and index. ISBN 978-0-8447-7250-9 (hbk.) — ISBN 0-8447-7250-X (hbk.) — ISBN 978-0-8447-7251-6 (pbk.) — ISBN 0-8447-7251-8 (pbk.) — ISBN (invalid) 978-0-8447-7252-3 (ebook) — ISBN (invalid) 0-8447-7252-6 (ebook) 1. Competition. I. Hassett, Kevin A. HF1414.R48 2012 338.6'048—dc23 2012031415

© 2012 by the American Enterprise Institute for Public Policy Research, Wash- ington, D.C. All rights reserved. No part of this publication may be used or reproduced in any manner whatsoever without permission in writing from the American Enterprise Institute except in the case of brief quotations embodied in news articles, critical articles, or reviews. The views expressed in the publications of the American Enterprise Institute are those of the authors and do not necessarily reflect the views of the staff, advisory panels, officers, or trustees of AEI.

Printed in the United States of America

HHassett.indbassett.indb iviv 111/8/121/8/12 9:279:27 PMPM Contents

LIST OF ILLUSTRATIONS ix

PREFACE xiii

1. TIEBOUT AND COMPETITIVENESS, Kevin A. Hassett, R. Glenn Hubbard, and Matthew H. Jensen 1 Charles Tiebout, Foot Voting, and Competition among Localities 5 Tiebout and a New View of “Competitiveness” 17 Conclusion 23 Notes 23 References 24

2. COMPETITIVE TAX POLICY, Joel Slemrod 32 What Exactly Is Competitiveness? 32 How Can Competitiveness Be Measured? 35 Taxation and Competitiveness 37 A Short Digression on Unpersuasive Economic Arguments 40 Business Taxation 42 International Considerations 49 The Corporate Tax and Prosperity: Facts and Evidence 53 Summary 62 Notes 63 References 65

3. EDUCATION AND GLOBAL COMPETITIVENESS: LESSONS FOR THE UNITED STATES FROM INTERNATIONAL EVIDENCE, Martin West 68 Ranking the U.S. Education System: Quality and Quantity Indicators 72 The Economic Costs of Low-Quality Education 82

HHassett.indbassett.indb v 111/8/121/8/12 9:279:27 PMPM vi RETHINKING COMPETITIVENESS

Policy Lessons 86 Notes 90 References 91

4. IMMIGRATION, PRODUCTIVITY, AND COMPETITIVENESS IN AMERICAN INDUSTRY, Gordon H. Hanson 95 Immigration and Competitiveness in Theory 100 Empirical Evidence on Immigration and Competitiveness 106 U.S. Immigration Policy and American Competitiveness 122 Notes 128 References 129

5. THE ROLE OF INNOVATION AND INTELLECTUAL PROPERTY IN ECONOMIC COMPETITION, Robert J. Shapiro 132 The Economic Value of Innovation 133 The Value of Intellectual Property Rights 135 How Innovations Diffuse across Economies 139 Competition: Efficiency versus Innovation 143 Conclusions 148 Notes 152 References 152

6. AMERICAN COMPETITIVENESS AND THE HEALTH CARE SYSTEM, Michael E. Chernew and Philip I. Levy 156 Productivity versus Competitiveness 159 The Health System’s Effect on the Economy 169 Comparing International Health Approaches 178 Conclusions 187 Notes 188 References 189

7. IS THE UNITED STATES “COMPETITIVE” INTERNATIONALLY IN HEALTH CARE? Benjamin Zycher 196 Some Conventional Wisdom on the U.S. Health Care System 197 Some International Evidence on Prices and Services 202 The Comparative Performance of the U.S. Health Care System 209 Concluding Observations 215 Notes 216 References 220

HHassett.indbassett.indb vivi 111/8/121/8/12 9:279:27 PMPM CONTENTS vii

8. GLOBAL VALUE CHAINS AND THE CONTINUING CASE FOR FREE TRADE: TRADE THEORY AND ILLUSTRATIONS FROM THE UNITED STATES AND EAST ASIA, Claude Barfield and Matthew H. Jensen 225 Two Unbundlings of Trade and Shifting Policy Implications 227 The First Unbundling and the Big Ideas of Trade Theory 228 The Second Great Unbundling 239 East Asian Fragmentation and Supply Chains: The Impact on Trade Data and Analysis 245 The United States and Global Supply Chains 260 Observations and Conclusions for Policy 268 Notes 270 References 273

9. INTERNATIONAL COMPETITIVENESS, Phillip Swagel 278 Uses and Meanings of “Competitiveness” 283 Capital Markets Competitiveness 293 Policy Appeals to Competitiveness 295 International Competitiveness in the Modern Vernacular 299 Notes 300 References 300

INDEX 303

ABOUT THE AUTHORS 325

HHassett.indbassett.indb viivii 111/8/121/8/12 9:279:27 PMPM HHassett.indbassett.indb viiiviii 111/8/121/8/12 9:279:27 PMPM List of Illustrations

Figures 1-1 Consensus Bidding and Sorting 15 1-2 Distribution of Top Statutory Corporate Tax Rates in the OECD 20 1-3 Distribution of Effective Average Corporate Tax Rates in the OECD 21 3-1 Math and Science Achievement in OECD Countries 74 3-2 Math Achievement vs. Educational Expenditure across OECD Countries 75 3-3 Science Achievement vs. Educational Expenditure across OECD Countries 76 3-4 Share of Adults in 2008 with Postsecondary Degree in Select OECD Countries, by Age Range 80 4-1 Share of U.S. Employment by Education Group 107 4-2 Immigrant Share of U.S. Employment by Education and Experience 109 4-3 Immigrant Share of U.S. Employment for Highly Educated Workers 110 4-4 Student Visas Issued by the U.S. Government 115 4-5 Foreign Graduate Students in U.S. Universities 116 4-6 Annual Cap on New H-1B Visas for Skilled Labor 119 4-7 Distribution of Legal Permanent Resident Visas, 1999–2008 121 4-8 Composition of U.S. Foreign-Born Population by Legal Status, 2007 123 4-9 Percent of Foreign Born in the National Population 124 4-10 Percent of Immigrant Population with 13-Plus Years of Education, 2000 125

HHassett.indbassett.indb ixix 111/8/121/8/12 9:279:27 PMPM x RETHINKING COMPETITIVENESS

4-11 Change in Working-Age Population 2005–2020 with Zero Immigration 127 8-1 U.S./UK Export and Productivity Ratios by Industry, 1950–1951 231 8-2 Trade Types, 1989–2002 234 8-3 Components of a Supply Chain 240 8-4 Share of Parts and Components in Total Exports 251 8-5 Complex Network for Hard Disk Drive 252 8-6 Proportion of Processing Trade in China’s Total Trade, 1988–2007 254 8-7 The Dominant Role of Foreign-Invested Enterprises in China’s ATP Surplus 259 8-8 Value Added Percentages for iPhone 4 263 8-9 Apparel Global Supply-Chain Process 264 8-10 Share of U.S. Multinational Parents in Employment, Output, Capital Investment, and R&D 265

Tables 1-1 Empirically Supported Implications of the Tiebout Model 18 2-1 Tax Rates and Growth Rates, Selected OECD Countries 59 7-1 Public Health Expenditures as a Percent of Total Health Expenditures, 2009 200 7-2 Fees for Physician Office Visits, 2008 203 7-3 Fees for Orthopedic Surgery, 2008 204 7-4 Relative Prices for Hospital Services, 2007 205 7-5 Health Care Resources, 2009 205 7-6 Life Expectancies at Age 80, 1987 212 7-7 Annual Compound Growth Rates in Real per Capita GDP, 1970–2010 213 7-8 Five-Year Survival Rates for Cancer of Different Sites, 2000–2002 214 7-9 OECD Five-Year Survival Rates, 2004–2009 215 7-10 CONCORD Study Rankings 216 8-1 Parts and Components (P&C) in Manufacturing Trade 246

HHassett.indbassett.indb x 111/8/121/8/12 9:279:27 PMPM LIST OF ILLUSTRATIONS xi

8-2 Domestic and Foreign Values Added: Normal vs. Processing Exports 254 8-3 Share of China’s Processing Imports by Country of Origin, 2007 255 8-4 Share of China’s Exports by Destination Country, 2007 256 8-5 Chinese ATP Exports to the United States by Trade Regime, 1996–2006 258 8-6 Most Expensive Inputs in the 30GB Third-Generation iPod, 2005 262 8-7 Country or Regional Sources of Value Added in U.S. Imports, Selected Sectors, 2004 266 8-8 Country or Regional Sources of Value Added in U.S. Exports, Selected Sectors, 2004 267

HHassett.indbassett.indb xixi 111/8/121/8/12 9:279:27 PMPM HHassett.indbassett.indb xiixii 111/8/121/8/12 9:279:27 PMPM Preface

Kevin A. Hassett

It is an unwritten law that no politician shall give an economic policy speech without mentioning the word “competitiveness” at least once. Yet, despite the ubiquity of the term, there is little if any agreement concerning its meaning. In particular, academics have been appropriately slow to embrace it, tending to evince a healthy skepticism regarding political utterances that are often reminiscent of the most objectionable assertions of the mercantilists. A politician might hold a treatise on competitiveness aloft, while an economist might prefer to disguise it in a brown paper bag. While scholarship on competitiveness has stagnated, the world has continued to evolve into a flatter, more intertwined marketplace, and vast literatures have emerged that document that competitions between nations, be they over the location of people or machines, have heightened in their ferocity. Against this backdrop, the American Enterprise Institute (AEI) brought together experts from a variety of fields to breathe new life into the competitiveness debate. The experts were asked to think creatively about the extent to which competition between nations may influence outcomes in their area of expertise, with a focus on a simple question: If “competitiveness” were to have a rigorous and relevant meaning in your field, what might that be? The authors presented their findings at a series of three conferences hosted at AEI’s headquarters in Washington, D.C.: “Is Competitiveness Worth Defending?” on September 29, 2011; “Nation vs. Nation: Do Countries Com- pete in Trade and Health Care?” on January 18, 2012; and “Competing for Talent: The United States and High-Skilled Immigration” on January 31, 2012. Nine papers presented at those conferences comprise this volume.

xiii

HHassett.indbassett.indb xiiixiii 111/8/121/8/12 9:279:27 PMPM xiv RETHINKING COMPETITIVENESS

The first chapter of the volume intends to frame the discussion. Glenn Hubbard, Matthew Jensen, and I set out to capture the question underlying this project and summarize the debate surrounding competitiveness. Once we set up the framework for the discussion, we describe a well- known theory of public finance called the “Tiebout model” and extend it to the international stage. Under the Tiebout model, individuals “vote with their feet,” choosing to live in localities with their preferred bundle of public goods and taxes, which creates competition among localities. This model has been well developed and researched; therefore, it is fairly simple to extend its application. We assert that since the world has become “flatter” and more interconnected, the lessons from this model can be applied to the international competitiveness debate. Competitions among nations in this flat world look very much like twentieth-century competitions among municipalities. In his chapter on competitive tax policy, Joel Slemrod of the University of Michigan asserts that global competition does not add a new criterion for judging ideal tax policy, but changes how any policy meets those criteria. He proceeds to point out the inefficiencies of the current U.S. tax system, such as the disincentive to invest. To help focus the debate, he dispels three unhelpful, yet popular, arguments surrounding tax reform. One example is the argument that a sufficient reason to change a policy is that other countries have done so, which is not a convincing argument due to the different economic environments that countries face. Slemrod believes that one should be on the lookout for these arguments when evaluating tax policy. After identifying the inefficiencies in the tax code, Slemrod evaluates them in a global context focusing heavily on corporate taxation, which is most frequently discussed in competitiveness debates. He concludes that the inefficiencies present in the U.S. tax system are exacerbated by the global economy and that competitive U.S. tax policy should consider the international context within which it operates. Another area where competitiveness dominates almost every policy conversation is education. Harvard’s Martin West dissects the familiar argument that U.S. education system is losing its competitive edge, causing the country to fall behind economically. He concludes that there is little evidence to suggest that countries engage in zero-sum competition in education, since productivity abroad can lower the price of U.S. imports

HHassett.indbassett.indb xivxiv 111/8/121/8/12 9:279:27 PMPM PREFACE xv

and provide technological advances that benefit everyone. The competitive rhetoric draws attention to the shortcomings of the American education system and creates a sense of urgency about their improvement. By examining the findings of international performance tests, West finds that U.S. students perform in the middle of the pack of Organisation for Eco- nomic Co-operation and Development (OECD) countries in science, but are falling behind in math. This underwhelming performance of American students is especially striking given U.S. spending on education. These are only some of the challenges faced by the U.S. educational system and West concludes the chapter with policy suggestions, such as paying teachers based on classroom performance, needed to bring the United States back to a competitive level. In the next chapter, Gordon Hanson evaluates the consequences of immigration on U.S. competitiveness by focusing predominantly on legal, high-skilled immigration, which is frequently overshadowed by illegal, low- skilled immigration. This omission is unfortunate in Hanson’s view, since luring high-skilled immigrants is a key focus of many emerging economies, and high-skilled immigrants have a positive impact on productivity and economic growth. Hanson describes the theoretical effects of immigration on production costs, productivity, and trade costs and employs evidence on these internal factors to interpret U.S. competitiveness and theoretical earnings. Questioning the theoretical positive effect of immigration on innovation, Hanson looks to empirical evidence to validate his hypothesis. He concludes the chapter by analyzing current U.S. immigration policies and suggesting improvements. Robert Shapiro, in his chapter on innovation, argues that nations echo the traditional behavior of firms and do in fact engage in competition over the location of intellectual property. Shapiro evaluates the important factors that affect innovation in countries, with a special focus on the United States, and provides several vital findings from the academic literature. He posits that in most cases, a country’s openness to innovation is more important than its natural resources in determining its growth rate. He also finds that more successful economies provide greater protection of intellectual property rights, greater commitments to research and development, more stable political and economic environments, and fewer barriers to new businesses—therefore, they breed the environment necessary for innovation.

HHassett.indbassett.indb xvxv 111/8/121/8/12 9:279:27 PMPM xvi RETHINKING COMPETITIVENESS

Countries that provide fewer intellectual property protections are less successful in the marketplace and tend to grow at a slower rate than those with strong protections. Additionally, Shapiro finds that economies that focus on innovation-based competition have a competitive advantage over economies focused on price-based or efficiency-based competition. He concludes that the competitive advantages offered by innovation-focused economies are available to any nation prepared to actively promote the development of new ideas. The next two chapters explore the health care sector. Michael Chernew and Phil Levy assess in detail the possible meanings of the idea of competitiveness, and go on to analyze the relative efficiency of the U.S. market for health care. They conclude that the link between the U.S. health care system and international competitiveness is rather tenuous. Nevertheless, the authors believe it is important to address inefficiencies in the health care market. Even if these actions do not affect the country’s international competitiveness, they impact the well-being of American citizens. As a result, every policy should be judged by the benefits that it brings. In the following chapter, Benjamin Zycher takes a different look at the health care sector in the United States. Zycher focuses on evaluating the conventional wisdom that the U.S. health care sector is inefficient because it “spends more while getting less.” He evaluates the data available on real spending within the sector along with evidence of real outcomes and concludes that the conventional wisdom is far from correct. The main problem with the literature is the difficulty to accurately compare health outcomes and spending. The rankings and studies that attempt to do this rely on measures that are difficult to define consistently, given the variation in the structure of health care sectors across countries. Zycher systematically reviews the important indicators and concludes that once they are adjusted for the measurement difficulties and definition discrepancies, the U.S. system is not as inefficient as conventionally thought. AEI’s Claude Barfield and Matthew Jensen explore international trade, an area where cooperation, rather than competition, has been traditionally seen as beneficial. They examine the rise of global value chains (GVCs) and the role they have played in influencing and evolving existing theories of trade. The authors analyze developments in international trade to explain how GVCs have undermined the key assumptions of competition.

HHassett.indbassett.indb xvixvi 111/8/121/8/12 9:279:27 PMPM PREFACE xvii

By dividing the production process between many nations, GVCs have increased the importance of free trade and governmental nonfavoritism. Further evaluation of supply chains in East Asia and the United States provides evidence of increasing regional complexity and changes in the nature of global trade between the United States, Japan, and newly industrialized economies. In particular, the content of trade volume has shifted toward increased parts and components of intermediate goods, as opposed to finished products. The authors conclude that the new trade paradigm should encourage the United States to enact policies to increase productivity of U.S. firms, educating high-skilled workers, and implementing regulations and tax policies to encourage competition. Phillip Swagel’s chapter concludes the volume. He analyzes measures of international competitiveness in order to determine their relationship to economic growth and national prosperity. Swagel posits that there are two sets of indicators. First, there are indicators of trade performance, such as real exchange rates, which relate to competitiveness but whose relation to growth and prosperity is complex and often unclear. The second set of indicators, such as output and employment, relates broadly to national well-being, but the link to international competitiveness is more tenuous. Following his thorough analysis of the competitiveness indicators, Swagel touches on financial-sector competitiveness, which is based on measures of initial public offerings, financial resources, and capital market regulation. The chapter concludes that the term “competitiveness” has evolved to mean productivity and well-being and that policymakers should keep this new meaning in mind when comparing policies. Swagel ultimately believes that the objective of the policy—rather than the label—is more important. Taken together, these studies provide a wealth of fresh thoughts and provocative conclusions policymakers would do well to rely upon.

HHassett.indbassett.indb xviixvii 111/8/121/8/12 9:279:27 PMPM HHassett.indbassett.indb xviiixviii 111/8/121/8/12 9:279:27 PMPM 1

Tiebout and Competitiveness Kevin A. Hassett, R. Glenn Hubbard, and Matthew H. Jensen

The concept of “national competitiveness” has been a key focal point of national policy debates at least as far back as Adam Smith, whose notions of specialization and division of labor figure prominently in his early debates with the mercantilists (in particular, see The Wealth of Nations [Smith (1776) 1982]). Later, David Ricardo’s work developing the law of comparative advantage advanced rational economic thinking about competition.1 Yet, while the classical movement refined our understanding and influenced generations of economists, mercantilist arguments that refer to a nation’s “competitiveness” continue to abound even today. Noneconomists regularly appeal to competitiveness when motivating a wide array of policies, while economists protest or look the other way. For the most part, a general consensus has emerged that accepts the analysis of Nobel laureate Paul Krugman, whose 1994 article in Foreign Affairs, bearing the unambiguous title “Competitiveness: A Dangerous Obsession,” disposed of the faulty analysis of the 1990s’ competitiveness mavens as effectively as Smith disposed of the mercantilists. Krugman challenged the idea of competitiveness, arguing that nations usually do not compete with one another in a zero-sum game, even if firms often do. Instead of competing directly with each other, countries benefit from each other’s successes through mutually beneficial trade. In a world with extensive international trade and interconnectedness, competitiveness and productivity are synonymous. When attempting to measure competitiveness according to a nation’s output, you find that the prosperity of one country will often stimulate additional prosperity for others. The

HHassett.indbassett.indb 1 111/8/121/8/12 9:279:27 PMPM 2 RETHINKING COMPETITIVENESS

notion that the success of one comes at the expense of another is most often incorrect. To be sure, Krugman recognizes that competitions may arise in some circumstances but argues that these are, for the most part, not central to debates over macroeconomic policy. This view has been widely accepted among economists, with a few exceptions discussed below. More recently, for example, Cellini and Soci (2002) argued that “the concept of competitiveness is elusive in so far as it neither has a well defined meaning nor is it captured by unambiguous factors” (p. 72). A number of alternative approaches have been attempted in the interim with authors focusing on national well-being as a measure of a nation’s competitiveness and on analysis that relates that well-being to a series of indicators. Michael Porter of the Harvard Business School is a prominent advocate of this view of competitiveness. Porter relates the welfare of a nation to the microeconomic determinants of the competitiveness of its firms, and his view of regional or national competitiveness grows out of this concept. In this context, competitive advantage implies that a firm is more productive than the competition (that is, it can produce more for less). This model is extended to apply to nations because in a competitive and interconnected marketplace, nations often compete for specific competitive advantages. Governments are responsible for creating conditions to foster the success of firms. This output-based view of competition focuses on microeconomic and productivity measures. The interest in competitiveness, and especially in comparing the competitiveness of nations, has led to a growing number of indexes that are consistent with Porter’s view. The two most important are the Global Com- petitiveness Report of the World Economic Forum (WEF) (various years) and the report prepared by the International Institute for Management Development (IMD) (various years) in the World Competitiveness Yearbook. In addition, firms, governments, and other organizations have developed their own ranking and evaluation systems. The Global Competitiveness Report defines competitiveness as “the set of institutions, policies, and factors that determine the level of productivity of a country” (p. 4). As with Porter’s work, the role of the firms is to be productive (competitive), while the role of the government is to create an environment that enables productivity. The Global Competitiveness Report

HHassett.indbassett.indb 2 111/8/121/8/12 9:279:27 PMPM TIEBOUT AND COMPETITIVENESS 3

breaks down the factors to identify twelve pillars of competitiveness. They are institutions, infrastructure, macroeconomic environment, health and primary education, higher education and training, goods market efficiency, labor market efficiency, financial market development, technological readiness, market size, business sophistication, and innovation. The World Competitiveness Yearbook “analyzes and ranks the ability of nations to create and maintain an environment that sustains the competitiveness of enterprises.” The assumption is that wealth creation happens on the firm level, but the national environment can help or hinder the ability of firms to compete in domestic and international markets. The index attempts to analyze the factors that affect the competitive national environment. The 2010 edition analyzed fifty-eight countries according to four primary factors of competitiveness: economic performance, government efficiency, business efficiency, and infrastructure. In all, the World Competitiveness Yearbook includes 327 indicators. Many academics have offered critiques of these indexes, including Sanjaya Lall (2001), who provides a sweeping critique of the Global Com- petitiveness Report. Lall’s main complaint is that the scope of competitiveness embodied in these indexes is too broad. He dismisses the definition of competitiveness that includes productivity and growth and suggests that “all such efforts should be more limited in coverage, focusing on particular sectors rather than pulling in everything the economics, management, strategy, and other disciplines suggest. They should also be more modest in claiming to quantify competitiveness: the phenomenon is too multifaceted and complex to permit easy measurement” (p. 1520). More recent work has focused on the correlation between these country rankings and economic growth. These studies (Berger and Bristow 2009; Ochel and Röhn 2006) find that there is no relationship between competitiveness rankings and economic growth prospects. Berger and Bristow conclude that the indexes lack a common approach and are poor predictors of macroeconomic performance: “The value of such indices is therefore questionable beyond their purpose in reminding us of the continued success of particular nations and the continued paucity of others and thus encouraging policy-makers to indulge in place promotion” (p. 387). Perhaps the state of play is best summarized by Reinert (1995), who analyzed five hundred years of competitiveness theory, arguing that even

HHassett.indbassett.indb 3 111/8/121/8/12 9:279:27 PMPM 4 RETHINKING COMPETITIVENESS

though the term is relatively new, equivalent ideas have prevailed for centuries. Reinert begins with the common definition of competitiveness coined by Bruce Scott: “National competitiveness refers to a nation state’s ability to produce, distribute, and service goods in the international economy in competition with goods and services produced in other countries, and to do so in a way that earns a rising standard of living” (Scott 1985, p. 25). Reinert observes that the criticism levied at competitiveness by Krugman and neoclassical economists can be explained by recognizing that competitiveness does not have any meaning under the assumptions of much of neoclassical theory (representative firms with perfect information and no scale effects). According to this train of thought, the idea of competitiveness, countries increasing their standards of living through competitive activities with other nations, is in opposition to at least a standard neoclassical model wherein common production technologies and competitive markets move the world toward a Pareto-optimal competitive equilibrium. Clearly, existing analysis on competitiveness has produced little that is valuable for policy analysis, even while policymakers continue to appeal to competitiveness to justify practically any policy change. This situation must be observed as a lost opportunity for economists, for if they could agree on an approach that applies rational substance to the notion of competitiveness, then the power of the word in the public debate might well provide a significant impetus for sound policies. The purpose of this chapter is to explore the extent to which a fresh perspective can do a better job of point- ing researchers and policymakers in specific directions. As we explore these analytical gaps, cognizant of the research just discussed, we begin with three observations. First, to be useful as an alternative prism through which one can view questions regarding optimal policies, competitiveness must focus on areas over which nations actually compete. That is, competitiveness will never provide a novel observation on any policy that produces an efficiency or welfare gain that applies to an economy in isolation. If efficiency is higher after a tax reform, for example, then a nation should adopt the reform and do so without ever uttering the word “competitiveness.” Second, as the world becomes “flatter” and it is easier for firms and individuals to choose an international jurisdiction that is optimal for their own preferences, then nations are becoming more and more analogous

HHassett.indbassett.indb 4 111/8/121/8/12 9:279:27 PMPM TIEBOUT AND COMPETITIVENESS 5

to municipalities. This flattening can easily be seen in the complementary areas of language, migration, and trade: English has grown to be the lingua franca for international business, science, and technology, making it easier for people to communicate with each other anywhere, no matter what their native language is.2 Migration is freer due to advances in transportation, particularly air travel, and Europe has paved the road to passport-free zones through the European Union’s Schengen Agreement and the UK-centric Common Travel Area. In addition to free travel zones, there has been an explosion of multilateral and bilateral free trade agreements, including the European Union’s Single Market, the North American Free Trade Agree- ment, the Asian Free Trade Area, and the South Korea–United States Free Trade Agreement. Trade has increased globally at a cumulative rate, on average, of 6.2 percent a year. The integration that is exemplified by the development of a common language, unrestricted travel zones, and free trade provides an opportunity to apply lessons from the local public finance literature. An extensive body of research exists analyzing the “Tiebout” competition among cities and towns, and this research may provide a wealth of insights regarding the likely evolution of Tiebout competition between nations. Indeed, to some extent the literature has begun to move in this direction. Somin (2008), for example, interprets global migration patterns in terms of the Tiebout model. There will obviously be some conclusions from the Tiebout literature that extend to nations and some that do not, but the flattening world makes the extension an interesting focus of research and a promising avenue for adding rigorous analysis to the previously ambiguous notion of competitiveness. Finally, the extent to which a nation’s actions depend on the actions of others should be the focus of competitiveness discussions, even if the competition is not purely zero sum. As we discuss below, Tiebout competition is not always zero sum, and the investigations into global competitiveness should not limit themselves to zero-sum games.

Charles Tiebout, Foot Voting, and Competition among Localities Charles Tiebout was a graduate student at the University of Michigan when he half-jokingly proposed what has become a seminal idea in local public finance and one of the best ways to think about competition among

HHassett.indbassett.indb 5 111/8/121/8/12 9:279:27 PMPM 6 RETHINKING COMPETITIVENESS

localities (see Fischel 2006). It was 1951 or 1952, and Richard Musgrave had just presented work by himself and Paul Samuelson that indicated the impossibility of finding an optimal level of public service through a decentralized pricing system (see Musgrave 1939; Samuelson 1954, 1955, 1958). In markets of noncollective consumption goods, buyers will allocate services for themselves optimally based on decisions of costs and benefits. Likewise, competition among enterprises optimally allocates the factors of production. An optimal allocation could be reached for public services, too, if each user could be polled to ask how much a service is worth to him or her and be counted on to answer truthfully and pay that price as a benefits tax. In aggregate, if the poll indicates that expected tax revenue will meet the cost of providing the service, then it can be provided. Unfortunately, users cannot be counted on to answer truthfully. Instead, they will underre- port their expected benefit from the service, hoping that it will be provided anyway according to others’ demands and, accordingly, that they will then have to pay less than their fair share. Musgrave finished his presentation with the comment that communities, not having a market option, must rely on voting in the political process to determine what levels of public consumption goods to provide. Through the voting process, the government would somehow gain an understanding of the expenditure needs of the typical consumer-voter, and taxation could occur, albeit suboptimally, based on ability to pay rather than benefit. Charles Tiebout at this point made his famous half-joke that an optimal allocation of public goods could also be found if consumer-voters move to communities that satisfy their preferences for public services. Simply put, consumer-voters vote with their feet. In 1956 Tiebout wrote “A Pure Theory of Local Expenditures,” expanding on this concept of “foot voting” and presenting a stylized model in which foot voting reveals consumer-voters’ preference for public services and allows for an efficient allocation of the services at the local level (Tiebout 1956). He was finally taking his idea seriously. The paper has become one of the most cited works in public finance and one of the intellectual bed- rocks of decentralization and federalism (see, for example, Gordon 1983). The driving hypothesis is: “The consumer-voter may be viewed as picking that community which best satisfies his preference pattern for public goods. At the central level the preferences of the consumer-voter are given, and

HHassett.indbassett.indb 6 111/8/121/8/12 9:279:27 PMPM TIEBOUT AND COMPETITIVENESS 7

the government tries to adjust to the pattern of those preferences, whereas at the local level various governments have their revenue and expenditure more or less fixed. Given these revenue and expenditure patterns, the consumer-voter moves to the community whose local government best satisfies his set of preferences” (p. 418). The model of local finance that Tiebout presents relies on a set of strict assumptions and is highly simplified. As we think ahead to our search for the possible lessons that nations can draw from this research, it is fruitful to list these assumptions in detail. The key assumptions are:

1. Consumer-voters are fully mobile and will move to the community that best satisfies their set preference patterns. 2. Consumer-voters have full knowledge of and react to the differences among revenue and expenditure patterns of local governments. 3. There are a large number of communities from which to choose. 4. Consumer-voters live on investment income. In other words, restrictions due to employment opportunities are not considered. 5. Public services exhibit no external economies or diseconomies among communities. 6. There is an optimal community size determined by the fixed resources of land and the pattern of community services demanded by current residents. 7. Communities below optimal size seek to grow by attracting new residents, and those above optimal size seek to shrink.

The first assumption has historically been the key factor limiting the scope of the model to local finance in metropolitan areas, but the flattening world suggests that it is less limiting today. Firms, workers, and the assets of both are highly mobile among nations, and the competition for the location of these may well be comparable in impact to that between municipalities. Aside from that, none of the key assumptions would prohibit thinking about this competition as being between nations. This observation suggests that a thorough understanding of the scope of the local public finance

HHassett.indbassett.indb 7 111/8/121/8/12 9:279:27 PMPM 8 RETHINKING COMPETITIVENESS

research on municipal competition can provide a useful guide to the types of questions that would be raised by a more formal inquiry into the impact of the competition between nations. Tiebout’s model has many implications, and they can be grouped into two broad categories: (1) those that relate to competition between localities for consumer-voters and other entities that could possibly vote with their feet, and (2) those that relate to the optimal distribution of public goods. To the extent that these can be separated, and in the interest of brevity, we focus on the implications of Tiebout’s model that relate to competitiveness. In this first part of this section, we confirm that Tiebout’s model is supported by empirical observation. This step is necessary to begin to motivate the application of Tiebout competition to the international setting. The second part of the section addresses how competition between municipalities (and nations) can lead to more efficient services. The third part describes explicitly how Tiebout’s model provides for an optimal allocation of households among communities based on preferences for the service/tax package. These same mechanisms will apply in the international setting as firms and individuals decide where to locate among nations. Of course, the fourth part then describes the externalities and spillover effects that exist in the real world and contribute to suboptimality in the allocation of households. The fifth part describes how the Tiebout concepts also apply to firms. Finally, the sixth part discusses whether we can consider Tiebout competition zero sum and notes that the welfare benefits from Tiebout competition indicate that the acceleration of global flattening might be a worthy policy goal.

Is the Tiebout Model Supported by Empirical Evidence? Tiebout’s model did not make much of a splash among economists when it was introduced and did not receive any testing until Wallace Oates’s seminal 1969 capitalization paper (Oates 1969). In 2005 Oates described his thought process behind that paper: “It had occurred to me that if mobile households were ‘shopping’ for local public services at the lowest price, we should find that housing prices reflect both the quality of local services and the associated tax bill. I thus (perhaps somewhat naively) proposed, as a test of Tiebout, the capitalization of differentials in local public outputs and local tax liabilities into local property values” (Oates 2005, p. 25). Oates regressed the median value of owner-occupied dwellings in various com-

HHassett.indbassett.indb 8 111/8/121/8/12 9:279:27 PMPM TIEBOUT AND COMPETITIVENESS 9

munities against the characteristics of the dwellings, commuting distance, median family income, public school expenditure per pupil, percentage of low-income families in the community, and effective property tax rates. He found that when property taxes rise, property prices fall by approximately two-thirds as much. The increased cost of public services is capitalized into depressed prices. In his reminiscence, Oates had reason to comment that testing Tiebout with a capitalization study might have been “somewhat [naive].” Soon after Oates’ initial paper, Edel and Sclar (1974) claimed that in the long-run Tiebout equilibrium, there should be no capitalization because Tiebout communities are replicable, and so any fiscally advantaged jurisdiction would soon face competitors. They find empirical evidence that over time the extent of capitalization dissipates. Several theoretical papers following Edel and Sclar argue that the situation is even more complicated and that the extent of capitalization depends on how many jurisdictions are in an area and how expandable they are, how fixed the boundaries are, and how the local decision-making process operates (see Epple, Zelenitz, and Visscher 1978; Yinger 1982; Rubinfeld 1987; Yinger et al. 1988). Fischel (2001), though, points out that there is little ease of entry in the public sector, that boundaries are relatively fixed, and that there is an inelastic supply of jurisdictions, so fiscal advantages can exist for quite some time. On theoretical grounds, Fischel brought the story back to Oates’ original conclusion, but there is also a substantial body of empirical analysis that replicates and builds on Oates’ regression analysis. Dowding, John, and Biggs (1994) provide an excellent technical review of the empirical research surrounding Tiebout’s analysis. They devote a substantial section to capitalization studies and conclude that, although the empirical research has some technical flaws, most papers do find that taxes and public services are capitalized to a significant degree. Studies of capitalization illuminate a secondary effect of the behavioral assumptions from the Tiebout model. According to the capitalization papers, fiscal differentials affect the demand for houses and are ultimately capitalized into house prices. Also, some studies directly measure the behavioral assumptions of the Tiebout model by examining the effect of fiscal differentials on migration flows. This migration research initially focused on welfare and demonstrated that levels of welfare payments

HHassett.indbassett.indb 9 111/8/121/8/12 9:279:27 PMPM 10 RETHINKING COMPETITIVENESS

affect migration flows. Areas with high levels of welfare payment attract potential welfare recipients and discourage the migration of wealthier households (see Aronson and Schwartz 1973; Brehm and Saving 1964; Cebula 1974a; Dye 1990; von Furstenberg and Mueller 1971; Pack 1973). Other studies focus on the nonwelfare expenditures of state and local governments and typically find that higher expenditures are attractive (see Cebula 1974b, 1978; Cebula and Kafoglis 1986, Cebula and Kohn 1975; Day 1992; Ellison 1980; Koven and Shelley 1989; Liu 1977; Mills et al. 1983; Pack 1973; Schneider and Logan 1982). Many of these studies, however, ignore the idea that differences between local tax systems influence location decisions. Cebula (1978) shows that white migrants prefer areas with low property taxes but that nonwhite migrants are insensitive to local tax differentials. This difference reflects the fact at the time that whites tended to have more property ownership than nonwhites. Cebula and Kafoglis (1986) find that migration is significantly responsive to the ratio between per capita state and local tax collections to per capita state and local nonwelfare expenditures. The studies mentioned so far link migration to fiscal differentials using aggregate data. It is possible that they are demonstrating cause but not effect. For example, Cebula (1974a) notes that discrimination in the South during the 1960s was a contributor to the movement of blacks to the North. Microlevel survey data allow researchers to examine the reasons people move by asking them directly. Percy and Hawkins (1992) survey 1,361 recent moves in Milwaukee and find that the top four reasons given for leaving the city were (1) housing values, (2) schools, (3) crime, and (4) taxes. All four of these factors are heavily driven by the expenditure-and- tax package. Percy (1993) recognizes that the Percy and Hawkins paper did not include renters and analyzes newspaper reports of property and deed transactions in Milwaukee. He finds that the tax/service motivation is only of secondary importance for leaving a place, although in regression analysis, schooling, taxes, and local spending all had significant and positive effects. When looking at people who are deciding where to move, he finds that these tax/service motivations are of primary importance. John et al. (1994) similarly find that in London boroughs, tax and service levels are of secondary importance when people decide to leave a place but are of primary importance when they decide where to move to. Understanding

HHassett.indbassett.indb 1010 111/8/121/8/12 9:279:27 PMPM TIEBOUT AND COMPETITIVENESS 11

which factors might affect household location decisions can help identify in which areas localities compete.

Tiebout Competition and the Efficiency of Services. Specific services and taxes, not just an overall assessment of the efficiency of the service-and- tax package, affect consumer-voter behavior. In fact, preferences for particular services could lead some consumer-voters to live in a location with less efficiency than another location that provides less preferred services. Several particular services have been studied and found to affect household location decisions, and for these there is also an indication that Tiebout competition elicits efficiency gains. Voting with feet incites communities to compete not only by providing the services that desirable consumer-voters want but also by improving the efficiency of those services. Public schooling is likely the most significant service for many household location decisions, and the Tiebout research provides clear theoretical grounds for why competition among school districts would raise productivity. While the theory is straightforward, the empirical results have not been entirely consistent. In a widely cited study, Hoxby (2000) concludes that fragmented governance induces competition among school districts by raising productivity. However, her results are disputed by Rothstein (2006). More recent work by Bayer, Ferreira, and McMillan (2004) finds that the competitiveness of a school’s local environment contributes significantly to quality as measured by test scores. Schwager (2007) applies the principles of the Tiebout model to university education in Germany. He concludes that by decentralizing university education and allowing states to choose their tuition level, efficient sorting can occur, with tuition serving as a price signal for mobile students. Tiebout’s ideas have also even been applied to state lotteries. Knight and Schiff (2010) explore the competition among jurisdictions in the context of cross-border shopping for state lottery tickets. The question is whether competition between neighboring states is significant in the sale of lottery tickets. Individuals living near borders respond to prices when deciding between playing the home-state lottery and crossing the border and purchasing tickets in neighboring states. The factors at play are geographic closeness and the size of jackpots. The authors conclude that states face significant competition from neighboring states—the relationship between

HHassett.indbassett.indb 1111 111/8/121/8/12 9:279:27 PMPM 12 RETHINKING COMPETITIVENESS

sales and prices is the strongest in states with small populations and densely populated border regions. States may respond to this competition the same way they might respond to other Tiebout competition: by lowering implicit tax rates (raising the payout of the lottery) or colluding. Regulations can also affect household location decisions. Kahn (2000) argues that regulation limiting ozone levels in the suburbs of Los Angeles has attracted migrants who chose not to live there in the past. This implies that those migrants preferred higher regulatory burdens to higher ozone levels. Households with the opposite preference can move to other cities. These examples develop the essential Tiebout idea that individuals “vote with their feet” according to their preferences for goods and services provided by local government, although such goods and services are by no means an exhaustive list of the determinants that might be important to households.

The Tiebout Model and the Housing Market. Having established that households “vote with their feet,” we can now explore the finer details of the model. Without a market mechanism for distributing more expensive households to jurisdictions with higher local services, there would be a free-rider problem because cities tend to collect the majority of their revenue from property taxes. In other words, there is an incentive to move to a city of McMansions and build a hut. Because of the large tax base from the expensive homes, the city will either provide higher levels of public service, lower tax rates, or both, but the hut dweller will only have to pay property taxes commensurate with the price of his home. Several papers have proposed approaches for either how cities can circumvent this problem or why this problem is resolved by the housing market. In a famous article, Hamilton (1975) proposed that city-initiated zoning could enforce a floor on property prices and by doing so eliminate the efficiency losses from property taxes. Under this approach, in addition to being sorted by preference for public service levels, households are sorted by their preference for levels of housing consumption because, in equilibrium, all households in a community would have identical zoning-specified levels of housing expenditure. Several testable implications of Hamilton’s model, however, have been discredited through empirical analysis.3 Notably, his model indicates that property taxes will not be capitalized in property prices;

HHassett.indbassett.indb 1212 111/8/121/8/12 9:279:27 PMPM TIEBOUT AND COMPETITIVENESS 13

however, empirical studies indicate that capitalization occurs (see Black 1999; Bogart and Cromwell 1997; Carroll and Yinger 1994; Harrison and Rubinfeld 1978; Haurin and Brasington 1996; Jud and Watts 1981; King 1973; McDougall 1976; Noto 1976; Rosen and Fullerton 1977). Ross and Yinger (1999) survey the conceptual literature to find a consensus, and lay out an alternative approach to solving problems: “(1) How does the housing market allocate households to communities when local public services and taxes vary from one community to the next? (2) How do communities select the level of local public services and tax rates? and (3) Under what conditions are solutions to the first two problems compatible, that is, when does an urban equilibrium exist?” (p. 2003). Their approach has two parts: market bidding and sorting. Jurisdictions all have fixed boundaries and vary according to service levels and effective tax rates. Households are utility maximizing and can be grouped by a combination of income and preference for public service levels and taxes. More specifically, the authors assume, on the basis of empirical evidence, that there is a positive correlation between income levels in communities and the public service levels in those communities. In addition, they assume that property taxes and public services are capitalized into house prices. So households can be grouped based on the trade-off they are willing to make between house prices and the level of public service in the community. This is termed the household’s “bid function” and is an upward sloping curve as seen in figure 1-1, where P is the level of house prices, S is the quality of public services, and the three curves each represent different households’ bid functions. Rich households will be more open to paying a larger differential in home prices in exchange for an additional unit of public services than poorer households. Interestingly, the sorted communities will not be completely homogeneous. If the two groups’ bid functions cross at a given level of expenditures, then members from both groups will be willing to live in the community. This result implies a world much like the one where we live: there will be clusters of extremely expensive houses where only the rich live, clusters of extremely cheap houses where only the poor live, and many commingled communities that lie somewhere in between.

Externalities Contribute to the Prize. Taken as a whole, Ross and Yinger’s result implies that communities can attract wealthier households

HHassett.indbassett.indb 1313 111/8/121/8/12 9:279:27 PMPM 14 RETHINKING COMPETITIVENESS

by providing higher levels of nonwelfare public expenditures compared to the tax levels—in other words, more efficient public services. In many ways, there are spillover effects from migration, costs for the region that the migrant leaves, and benefits for the region that a migrant enters. Most important, home prices will rise due to capitalization as the service/tax package improves and demand for housing increases.4 Secondary effects include those positive externalities that might accrue from gaining rich neighbors: Ladd and Yinger (1994) find that a city’s poverty rate affects the cost of police and fire services. Duncombe (1991) also finds that fire protection costs increase with poverty and building age and decrease with the presence of commercial and industrial capital. Bradbury et al. (1984) find that older housing and population density increase the overall cost of local services. In addition, a large body of research has focused on how students garner educational benefits. In particular, more socioeconomically advantaged peers and more intensive parental monitoring of teachers and school administra- tors have been linked to improved educational outcomes (see, for example, Angrist and Lang 2004; Arcidiacono and Nicholson 2005; Betts and Morell 1999; Cooley 2007; Dale and Krueger 2002; Ding and Lehrer 2007; Figlio 2003; Hanushek et al. 2003; Hoxby and Weingarth 2005; Sacerdote 2001; Vigdor and Nechyba 2005; Zimmer and Toma 1999; Zimmerman 2000). Peer effects also influence entrepreneurship. School peers affect students’ entrepreneurial ambitions, according to Falck, Heblich, and Luede- mann (2010). Giannetti and Simonov (2009) and Gompers, Lerner, and Scharfstein (2005) find that peer effects do not just influence entrepreneurship in students; they also influence neighborhood residents to become entrepreneurs and invest more into their own businesses. The wealth segregation implied by the consensus model of Ross and Yinger will cause spillovers across municipal borders. In particular, the secondary “social” effects of living in a richer neighborhood, discussed above, will accrue only to the wealthy municipalities, and the negative externalities of having poor neighbors will fall more often on the poor. Moreover, a large body of literature suggests that the opportunities available to children affect their chances as adults, so the spillover is both cross-jurisdictional and multigenerational.5 Oates (2005) indicates that wealthy cities might even attempt to exaggerate the segregation from Ross-Yinger bidding by instituting zoning rules. Typi- cally, zoning is thought to exist for fiscal reasons. Like in Hamilton’s model,

HHassett.indbassett.indb 1414 111/8/121/8/12 9:279:27 PMPM TIEBOUT AND COMPETITIVENESS 15

FIGURE 1-1 CONSENSUS BIDDING AND SORTING

S S1 S2 S3

SOURCE: Ross and Yinger (1999). This article was published in Handbook of Regional and Urban Econom- ics, Vol. 3, Stepehn Ross and John Yinger, Sorting and Voting: A Review of the Literature on Urban Public Finance, 2001–2060. Copyright Elsevier (1999).

fiscal zoning sets a minimum house price and prevents free riders. Bogart (1993) describes how zoning can also be used to exclude people who will make it more expensive to produce local public services. The Tiebout model assumes no spillover effects, but in reality spillovers do exist and contribute to inefficient sorting outcomes. Atkinson and Stiglitz (1980), for example, note that “where there is a limited number of jurisdictions, or where people act myopically, equilibrium may not exist, and when it does exist it may not be Pareto-efficient” (p. 551). However, the fact that spillovers exist and that sorting outcomes might be inefficient for that reason or others does not mean that communities should not compete for migrants; to the contrary, it means a bigger prize for the winner.6

The Tiebout Model and the Firm. Tiebout originally developed his model to describe the location decisions of households. Given the assumption that consumer-voters are highly mobile, they can “vote with their feet” by moving to the municipality that provides a bundle of public goods most

HHassett.indbassett.indb 1515 111/8/121/8/12 9:279:27 PMPM 16 RETHINKING COMPETITIVENESS

closely matching their preferences. Taxes are taken as the prices consumer- voters pay for the local government services. However, individuals are not the only entity that responds to policy and tax decisions made by local governments. Other research has applied the Tiebout ideas to the location decisions of firms.7 In order to promote economic activity, governments can institute fiscal policies with the intention of attracting businesses,8 but the question remains whether these are important in the decision-making process of firms. The debate centers on whether firms react purely to the tax rates of a municipality or whether they also consider expenditures. Wayslenko (1980) and Fox (1981) find greater business growth in communities with lower taxes but ambiguous firm preferences for expenditures on services. Schneider (1985) conducts an empirical analysis of the effects of fiscal differences on the distribution of firms, using a sample of eight hundred suburbs with data on the number of firms during 1972–1977. His conclusion is that fiscal differences among suburbs significantly affect the distribution of business development; he also points out some differences between retail and manufacturing firms. Manufacturing firms strongly avoid suburbs with high taxes, while retail firms also respond to general market conditions of the communities. Finally, he concludes that local government expenditures do not attract retail or manufacturing firms.

Is Tiebout Competition Zero Sum? The success of the Tiebout model adds an interesting nuance to the observation that analysis of competitiveness focuses either on enhancing productivity, which has merits in its own right, or on zero-sum competitions. To the extent that municipalities (or nations) compete over the location of individuals, individual firms, or some of the assets of those individuals or firms, then the disposition of those individuals, firms, or assets can only occur at the expense of other locations. However, the Tiebout competitions need not be considered purely zero sum. First, if we transition from a world without Tiebout competition to one with competition, then the competition can be expected to benefit all involved, as the characteristics of the competing municipalities (or countries) are driven toward the optimal by the competition itself. The ability of Tiebout competition to benefit all is a key factor that could drive nations to consider adopting policies that “flatten” the world and benefit all.

HHassett.indbassett.indb 1616 111/8/121/8/12 9:279:27 PMPM TIEBOUT AND COMPETITIVENESS 17

Second, empirical research has identified countless spillovers that are not included in the canonical form of the model. These spillovers create a rift between the type of competition that Tiebout predicted and the type of competition that actually exists, and that difference may create circumstances that are not purely zero-sum competitions. For example, consider a competition between Philadelphia and Washington, D.C., for the location of a firm. If Washington wins, then Washington’s neighbor Bethesda might stand to benefit from the competition, while Philadelphia’s neighbor Camden would not benefit if Philadelphia wins. If mechanisms do not exist to allow Washington to capture some of the rents to Bethesda, then the outcome of the competition might not be influenced by these external effects, and the optimality of the outcome would not necessarily follow from the competition.

Tiebout and a New View of “Competitiveness” A rich set of implications of the Tiebout model has been documented empirically. Tax variables and the quality and quantity of public services vary widely, and this variation has a significant impact on foot voting and does not lead to homogenous communities. Competition among jurisdictions improves the efficiency of services, with education being the leading example. People also respond to preferences for specific types of public services, not just overall levels of expenditures/taxes; for example, asthmatics and outdoor enthusiasts, as opposed to muscle-car lovers, would have been more likely to move to Los Angeles after strict ozone standards were put in place. Finally, firms may be even more responsive to these factors than households; firms clearly respond to fiscal differences between communities, and oftentimes firms are most influenced by the level of capital taxes. These results, which are summarized in table 1-1, provide ample fodder for new research in the international arena, but one cannot simply assume that the extension will follow immediately. For example, the international setting challenges the assumption that consumer-voters receive only investment income or, in practical terms, that they can work and live in different regions9 and the assumption that households are completely mobile. As the world flattens, these assumptions will become more and more applicable. In the meantime, the Tiebout studies provide rich motivation to research

HHassett.indbassett.indb 1717 111/8/121/8/12 9:279:27 PMPM 18 RETHINKING COMPETITIVENESS

TABLE 1-1 EMPIRICALLY SUPPORTED IMPLICATIONS OF THE TIEBOUT MODEL

Implication Representative Reference Housing Fiscal differentials, by affecting the demand for Dowding, John, and Biggs houses, are capitalized into housing prices. (1994) Households sort themselves into communities Ross and Yinger (1999) according to their preferences for services and taxation—this does not lead to homogenous communities. “Social” effects accompany rich neighbors. These Ladd and Yinger (1994) can contribute to a suboptimal distribution of housing as municipalities overprovide public goods to attract the rich. Migration The tax/service package is of primary importance Dowding, John, and Biggs when deciding where to move to, but is of only (1994) secondary importance when deciding whether to leave. Regulations, like the tax/service package, influence Kahn (2000) location decisions. Efficiency of Services Competition between localities for mobile con- Bayer, Ferreira, and sumers improves educational outcomes. McMillan (2004) Alternatively, localities can collude. This occurs, Knight and Schiff (2010) for example, with state lotteries. Firm Location Decisions Firms, like households, respond strongly to fiscal Schneider (1985) differences, particularly taxes.

why firms and people move in the international setting and how decision makers can structure optimal policy. Perhaps Ilya Somin (2008) has done the most to begin extending the local urban finance analysis to the migration of people in the international setting. He argues that “overall, freer international migration could stimulate interjurisdictional competition that increases the utility of foot voting

HHassett.indbassett.indb 1818 111/8/121/8/12 9:279:27 PMPM TIEBOUT AND COMPETITIVENESS 19

in much the same way domestic freedom of movement does in a federal system. The potential benefits are potentially far larger because there are so many more options for international migrants than domestic ones and because of the very large policy divergences between countries” (p. 1257). Global interjurisdictional competition is already happening. The Organ- isation for Economic Co-operation and Development (OECD) (2002) describes how support for research, a positive climate for business start-ups and self-employment, the location of multinational corporations, educational opportunities, and political stability attract skilled international immigrants. In other words, skilled immigrants are attracted overwhelmingly by the public service/tax package, just as migrants in a domestic setting are. In the case of international immigration, the free-rider problem can be solved by selectively issuing visas. In the United States, the H-1B visa serves this purpose by requiring that employees sponsor educated workers in specialty occupations. In fact, there has been a shortfall of H1-B visas during the last several years. As Somin (2008) has attested, more freedom of movement would stimulate even more interjurisdictional competition. The extension of the Tiebout research on households’ “voting with their feet” to the international setting is still relatively undeveloped. Much more attention has been paid to the international movement of firms and, in particular, firms’ movement in response to international capital tax competition. Much of the research has focused on the distinction between Tiebout competition and tax competition. But after tax competition plays out, a world more closely matching the Tiebout model will exist. Firms will likely act much more like individuals and make location decisions based on preferences for the tax/service package. The motivation for tax competition is a strong relationship between corporate taxes and foreign direct investment (FDI). Hines (1999) summarizes the empirical research by differentiating between two types of empirical studies and reports a positive correlation between levels of FDI and after-tax rates of return and elasticities of property, plant, and equipment ownership with respect to corporate tax rates ranging from –0.01 to –2.8.10 De Mooij and Ederveen (2006) also analyze the literature of taxation and FDI, and they find that, on average, the literature reports semi-elasticities of around –4. Overall, there is a consensus that differences in corporate taxes between countries have a significant effect on the level of FDI.

HHassett.indbassett.indb 1919 111/8/121/8/12 9:279:27 PMPM 20 RETHINKING COMPETITIVENESS

FIGURE 1-2 DISTRIBUTION OF TOP STATUTORY CORPORATE TAX RATES IN THE OECD

1981 1996 2010 8

4 Kernel Density 2

United States United States United States 0 .1 .2 .3 .4 .5 .6 .1 .2 .3 .4 .5 .6 .1 .2 .3 .4 .5 .6 Tax Rate

SOURCE: Hassett and Mathur (2011).

It is not surprising, then, that international governments, like municipalities, would actively compete to lower tax rates to attract business and capital. OECD countries, for example, began to experience declines in statutory corporate tax rates in the 1980s and 1990s, corresponding with the increased capital interaction within the global market. As international competition increased, average corporate tax rates fell. Devereux, Lockwood, and Redoano (2008) econometrically review competition in the corporate tax system among twenty-one OECD countries from 1983 to 1999 and find evidence that countries compete over both statutory and effective average tax rates. Hassett and Mathur (2007) similarly find evidence of tax competition as countries strategically respond to the tax rates in other countries. In particular, a country is more likely to lower its rates if those rates are higher than the average rates in neighboring countries. Another recent paper by Hassett and Mathur (2011) illustrates the fall of corporate tax rates around the world; results from their papers are displayed in figures 1-2 and 1-3.

HHassett.indbassett.indb 2020 111/8/121/8/12 9:279:27 PMPM TIEBOUT AND COMPETITIVENESS 21

FIGURE 1-3 DISTRIBUTION OF EFFECTIVE AVERAGE CORPORATE TAX RATES IN THE OECD

1981 1996 2010 8

4 Kernel Density 2

United States United States United States 0 .1 .2 .3 .4 .5 .1 .2 .3 .4 .5 .1 .2 .3 .4 .5 Tax Rate

SOURCE: Hassett and Mathur (2011).

Just as researchers do not dispute the existence of international corporate tax competition, a fairly solid consensus has emerged that capital tax competition carries negative effects not associated with the normal Tiebout competition relying on nondistortionary head taxes. In the standard tax competition model, capital is highly mobile, but its overall supply is fixed, so there is a zero-sum game to attract it. Countries are driven to lower their rates competitively with their neighbors, but in the long-run equilibrium, the tax rates will be low, and every country will be on the same footing once again. The lasting effect is that tax competition will lead to an underprovi- sion of public goods.11 The damage of international tax competition is mitigated by several factors in practice. Counterintuitively, tax havens have been identified as a significant release valve. These typically small countries offer favorable tax treatment to foreign investors and have often been maligned in the past as offering corporate and banking secrecy that could be used for criminal purposes and as eroding other countries’ tax bases.

HHassett.indbassett.indb 2121 111/8/121/8/12 9:279:27 PMPM 22 RETHINKING COMPETITIVENESS

Recent research, however, has challenged both of these negative claims. Although the concern about transparency has not yet been entirely neutral- ized, Rose and Spiegel (2006) find that proximity to tax havens can actually improve competition within a domestic banking system as measured by interest rates while additionally spurring banks to extend more credit to the local private sector. More relevant to a discussion of tax competition is the second criticism, that tax havens erode the tax base for other countries. Desai, Foley, and Hines (2006a,b) find that rather than reducing economic activity, the proximity of tax havens actually spurs growth and foreign investment. As Hines (2010) summarizes the mechanism, “Tax-efficient financing structures in tax havens permit taxpayers to avoid costly tax situations in high-tax areas, thereby increasing rates of return and making investment in high-tax places more attractive” (p. 8). Blanco and Rogers (2009) similarly conclude that the investment in a less developed nation is positively influenced by proximity to a large tax haven. By acting as a release valve for high-tax countries, tax havens likely limit tax competition. There have been recent calls from the OECD for tax harmonization that would limit the availability of tax havens in an attempt to bring corporate tax rates in line with each other internationally (see Organisation for Economic Co-operation and Development 1998, 2000). Such harmonization is likely to be counterproductive and would place downward pressure on rates in the large, high-tax countries that currently remain competitive only because their multinationals can receive favorable tax treatment through havens. Tax competition exists, and for now it is probably the most significant form of international competition for firms. Most models indicate that in long-run equilibrium no country will have a lasting advantage. Likely the existence of tax havens will slow, but not halt, the march to zero-capital taxation. The fact that this game ends at a suboptimal state and a level playing field does not mean that the United States, or any other country, can bow out. Instead, the responsiveness of FDI to relative tax rates means that whoever lowers their rates the fastest has much to gain. In order to maintain lost revenues during and after the march to very low taxes, countries should be seeking fundamental tax reforms that leave them with less distortionary taxes to replace their lost revenues. This process

HHassett.indbassett.indb 2222 111/8/121/8/12 9:279:27 PMPM TIEBOUT AND COMPETITIVENESS 23

already seems to be happening throughout the developed world, albeit not in the United States, as countries implement consumption taxes. Once capital taxation has reached zero and countries have shifted to less distortionary taxes, the standard Tiebout model will be more applicable at the international setting for firms. Firms will begin to make location decisions based heavily on their preferences for cost-effective services offered by countries.

Conclusion The conceptual language of the Tiebout model is an apt tool for defining “international competitiveness,” a term most often used without regard for whether nations actually have something over which to compete. The Tiebout analysis tells us that municipalities compete for households and firms, which make their location decisions based largely on preferences for services and taxes. As the world flattens, nations are beginning to compete in the same way. Our chapter has discussed many of nuances of the Tiebout model and how international competitiveness, particularly migration and taxation, can be viewed through its lens. There is still much ground to cover, and that task should be taken up in future research. The rest of this book will discuss many areas of competition, from intellectual property to trade. It is our hope that the lessons from this chapter on Tiebout’s model, the idea of “voting with one’s feet,” and even the questions regarding the allocative efficiency of markets for local public goods will inform our understanding of what is to come.

Notes 1. Ricardo first described comparative advantage in his 1817 book, On the Prin- ciples of Political Economy and Taxation (Ricardo [1817] 2010). 2. HSBC’s “Unlocking the World’s Potential” advertising campaign famously states that there are five times as many people in China learning English as there are people in England (HSBC 2010), and the study titled Key Data on Teaching Languages at School in Europe concluded that more than 90 percent of children in the European Union learn English at some point in their education (Education, Audiovisual and Culture Executive Agency 2008). 3. For example, Hamilton’s model indicates that zoning restrictions should be unique to each community based on the housing demand of that community. Lenon et al. (1996), however, show that zoning, taxing, and spending policies are significantly correlated among neighboring communities.

HHassett.indbassett.indb 2323 111/8/121/8/12 9:279:27 PMPM 24 RETHINKING COMPETITIVENESS

4. Fischel (2001) explores the incentives that “home-voters” have to vote for policies that increase the value of their houses. 5. For more discussion of this point, see Aaronson (1998); Crane (1991); Cutler and Glaeser (1997); Jencks and Mayer (1990); Massey, Gross, and Eggers (1991); Vartanian and Gleason (1999). 6. Another implication is that redistributive policies should be implemented at the federal level rather than the local level. Roy (2009) finds that the Michigan school finance reform of 1994, which evened per pupil school expenditures across districts, increased the value of housing in the lowest spending school districts and, by implication, reduced sorting. However, he argues that continued high demand for residence in the high spending districts implies that peer effects persist. 7. This idea was originally developed in Fischel (1975) and White (1975). 8. Analyzing the effect of direct government action to attract firms, Black and Hoyt (1989) demonstrate that direct payments to attract firms have the potential of enhancing social welfare, assuming there is no private information or strategic behavior. 9. Flatters, Henderson, and Mieszkowski (1974) show that in a model where workers must work and live in the same region, an efficient distribution of labor will only occur in very narrow and unlikely circumstances. This observation does not say that fiscal competition does not occur, and it certainly does not indicate that other sorts of competition between nations for migrants do not occur. 10. One branch of this research uses the time-series estimation of the responsiveness of FDI to annual variation in after-tax rates of return, and the other branch uses cross-sectional data exploiting large differences in corporate tax rates. 11. Wilson (1999) surveys this research.

References Aaronson, Daniel. 1998. “Using Sibling Data to Estimate the Impact of Neighborhoods on Children’s Educational Outcomes.” Journal of Human Resources 33:915–46. Angrist, Joshua D., and Kevin Lang. 2004. “Does School Integration Generate Peer Effects? Evidence from Boston’s Metco Program.” American Economic Review 94:1613–34. Arcidiacono, Peter, and Sean Nicholson. 2005. “Peer Effects in Medical School.” Jour- nal of Public Economics 89:327–50. doi:10.1016/j.jpubeco.2003.10.006. Aronson, Richard J., and Eli Schwartz. 1973. “Financing Public Goods and Their Distribution in a System of Local Government.” National Tax Journal 26:137–60. Atkinson, Anthony B., and Joseph E. Stiglitz. 1980. Lectures on Public Economics. New York: McGraw-Hill. Bayer, Patrick, Fernando Ferreira, and Robert McMillan. 2004. “Tiebout Sorting, Social Multipliers and the Demand for School Quality.” NBER Working Paper. National Bureau of Economic Research, Cambridge, MA.

HHassett.indbassett.indb 2424 111/8/121/8/12 9:279:27 PMPM TIEBOUT AND COMPETITIVENESS 25

Berger, Thomas, and Gillian Bristow. 2009. “Competitiveness and the Benchmark- ing of Nations—A Critical Reflection.” International Advances in Economic Research 15:378–92. doi:10.1007/s11294-009-9231-x. Betts, Julian R., and Darlene Morell. 1999. “The Determinants of Undergradu- ate Grade Point Average: The Relative Importance of Family Background, High School Resources, and Peer Group Effects.” Journal of Human Resources 34:268–93. Black, Dan A., and William Hoyt. 1989. “Bidding for Firms.” American Economic Review 79:1249–56. Black, Sandra E. 1999. “Do Better Schools Matter? Parental Valuation of Elementary Education.” Quarterly Journal of Economics 114:577–99. doi:10.1162/003355399 556070. Blanco, Luisa, and Cynthia Rogers. 2009. Are Tax Havens Good Neighbors? An LDC Perspective. Working Paper, Pepperdine University, Malibu, CA. Bogart, William T. 1993. “‘What Big Teeth You Have!’ Identifying the Motivations for Exclusionary Zoning.” Urban Studies 30:1669–81. doi:10.1080/00420989 320081651. Bogart, William, and Brian Cromwell. 1997. “How Much More Is a Good School Dis- trict Worth?” National Tax Journal 50:215–32. doi:10.1006/juec.1999.2142. Bradbury, Katharine L., Helen F. Ladd, Mark Perrault, Andrew Reschovsky, and John Yinger. 1984. “State Aid to Offset Fiscal Disparities across Communities.” National Tax Journal 37:151–70. Brehm, C. T., and T. R. Saving. 1964. “The Demand for General Assistance Payments.” American Economic Review 54:1002–18. Carroll, Robert J., and John Yinger. 1994. “Is the Property Tax a Benefit Tax? The Case of Rental Housing.” National Tax Journal 47:295–316. Cebula, Richard J. 1974a. “Interstate Migration and the Tiebout Hypothesis: An Anal- ysis according to Race, Sex and Age.” Journal of the American Statistical Association 69:876–79. ———. 1974b. “Local Government Policies and Migration.” Public Choice 19:85–93. doi:10.1007/BF01718548. ———. 1978. “An Empirical Note on the Tiebout-Tullock Hypothesis.” Quarterly Journal of Economics 92:705–11. Cebula, Richard J., and Milton Z. Kafoglis. 1986. “A Note on the Tiebout-Tullock Hypothesis: The Period 1975–80.” Public Choice 48:65–69. Cebula, Richard J., and Robert M. Kohn. 1975. “Public Policies and Migration Pat- terns in the United States.” Public Finance 30:186–96. Cellini, Roberto, and Anna Soci. 2002. “Pop Competitiveness.” Banca Nazionale del Lavoro Quarterly Review 55:71–101. Cooley, LuAnn. 2007. “Transformational Learning and Third-Wave Feminism as Potential Outcomes of Participation in Women’s Enclaves.” Journal of Transforma- tive Education 5:304–16.

HHassett.indbassett.indb 2525 111/8/121/8/12 9:279:27 PMPM 26 RETHINKING COMPETITIVENESS

Crane, Jonathan. 1991. “The Epidemic Theory of Ghettos and Neighborhood Effects on Dropping Out and Teenage Childbearing.” American Journal of Sociology 96:1226–59. Cutler, David M., and Edward L. Glaeser. 1997. “Are Ghettos Good or Bad?” Quarterly Journal of Economics 112:827–72. Dale, Stacy B., and Alan B. Krueger. 2002. “Estimating the Payoff to Attend- ing a More Selective College: An Application of Selection on Observables and Unobservables.” Quarterly Journal of Economics 117:1491–1527. doi:10.1162/ 003355302320935089. Day, Kathleen. 1992. “Interprovincial Migration and Local Public Goods.” Canadian Journal of Economics 25:123–44. De Mooij, Ruud A., and Sjef Ederveen. 2006. What a Difference Does It Make? Under- standing the Empirical Literature on Taxation and International Capital Flows. Euro- pean Economy, Economic Papers 261. European Commission, Brussels. Desai, Mihir A., Fritz Foley, and James R. Hines Jr. 2006a. “Do Tax Havens Divert Economic Activity?” Economics Letters 90:219–24. ———. 2006b. “The Demand for Tax Haven Operations.” Journal of Public Economics 90:513–31. Devereux, Michael P., Ben Lockwood, and Michela Redoano. 2008. “Do Countries Compete over Corporate Tax Rates?” Journal of Public Economics 92:1210–35. doi:10.1016/j.jpubeco.2007.09.005. Ding, Weili, and Steven F. Lehrer. 2007. “Do Peers Affect Student Achievement in China’s Secondary Schools?” Review of Economics and Statistics 89:300–312. Dowding, Keith, Peter John, and Stephen Biggs. 1994. “Tiebout: A Survey of the Empir- ical Literature.” Urban Studies 31:767–97. doi:10.1080/00420989420080671. Duncombe, William D. 1991. “Demand for Local Public Services Revisited: The Case of Fire Protection.” Public Finance Review 19:412–36. doi:10.1177/1091142 19101900403. Dye, Thomas R. 1990. “The Policy Consequences of Intergovernmental Competi- tion.” Cato Journal 10:59–73. Edel, Matthew, and Elliott Sclar. 1974. “Taxes, Spending and Property Values: Sup- ply Adjustment in a Tiebout-Oates Model.” Journal of Political Economy 2:319–28. Education, Audiovisual and Culture Executive Agency. 2008. Key Data on Teaching Languages at School in Europe. Brussels: Education, Audiovisual and Culture Execu- tive Agency. Ellison, Richard. 1980. “Fiscal Impacts on Intrametropolitan Residential Location: Further Insights on the Tiebout Hypothesis.” Public Finance Quarterly 8:189–212. Epple, Dennis, Allan Zelenitz, and Michael Visscher. 1978. “A Search for Testable Implications of the Tiebout Hypothesis.” Journal of Political Economy 86:405–25. Falck, Oliver, Stephen Heblich, and Elke Luedemann. 2010. “Identity and Entrepre- neurship: Do School Peers Shape Entrepreneurial Intentions?” Small Business Eco- nomics 1–21. doi:10.1007/s11187-010-9292-5.

HHassett.indbassett.indb 2626 111/8/121/8/12 9:279:27 PMPM TIEBOUT AND COMPETITIVENESS 27

Figlio, David. 2003. “Fiscal Implications of School Accountability Initiatives.” Tax Policy and the Economy 17:1–36. Fischel, William A. 1975. “Fiscal and Environmental Considerations in the Locations of Firms in Suburban Communities.” In Fiscal Zoning and Land Use Controls, ed. E. Mills and W. Oates. Lexington, MA: Heath-Lexington. ———. 2001. The Homevoter Hypothesis: How Home Values Influence Local Govern- ment Taxation, School Finance, and Land-Use Policy. Cambridge, MA: Harvard University Press. ———. 2006. Tiebout Model at Fifty: Essays in Public Economics in Honor of Wallace Oates. Cambridge, MA: Lincoln Institute of Land Policy. Flatters, Frank, Vernon Henderson, and Peter Mieszkowski. 1974. “Public Goods, Efficiency, and Regional Fiscal Equalization.” Journal of Public Economics 3:99–112. Fox, William. 1981. “An Evaluation of Metropolitan Area Tax Base Sharing: A Com- ment.” National Tax Journal 34:275– 80. Giannetti, Mariassunta, and Andrei Simonov. 2009. “Social Interactions and Entre- preneurial Activity.” Journal of Economics and Management Strategy 18:665–709. Godin, Benoît. 2004. “The Obsession for Competitiveness and Its Impact on Statistics: The Construction of High-Technology Indicators.” Research Policy 33:1217–29. Gompers, Paul, Josh Lerner, and David Scharfstein. 2005. “Entrepreneurial Spawn- ing: Public Corporations and the Genesis of New Ventures, 1986 to 1999.” Journal of Finance 60:577–614. Gordon, Robert H. 1983. “An Optimal Taxation Approach to Fiscal Federalism.” Quarterly Journal of Economics 98:567–86. doi:10.2307/1881778. Hamilton, Bruce W. 1975. “Zoning and Property Taxation in a System of Local Gov- ernments.” Urban Studies 12:205–11. Hanushek, E. A., J. F. Kain, J. M. Markman, and Steven G. Rivkin. 2003. “Does Peer Ability Affect Student Achievement?” Journal of Applied Econometrics 18:527–44. Harrison, David, Jr., and Daniel L. Rubinfeld. 1978. “The Air Pollution and Prop- erty Value Debate: Some Empirical Evidence.” Review of Economics and Statistics 60:635–38. Hassett, Kevin A., and Aparna Mathur. 2007. “Predicting Tax Reform.” AEI Working Paper 138. American Enterprise Institute, Washington, D.C. ———. 2011. “Report Card on Effective Tax Rates: The United States Gets an F.” AEI Tax Policy Outlook 1. American Enterprise Institute, Washington, D.C. Haurin, Donald R., and David Brasington. 1996. “School Quality and Real House Prices: Inter- and Intrametropolitan Effects.” Journal of Housing Economics 5:351–68. Hines, James R., Jr. 1999. “Lessons from Behavioral Responses to International Taxa- tion.” National Tax Journal 52:305–22. Hines, James R. 2010. “Treasure Islands.” Journal of Economic Perspectives 24:103–25. Hoxby, Caroline M. 2000. “Does Competition among Public Schools Benefit Students and Taxpayers?” American Economic Review 90:1209–38.

HHassett.indbassett.indb 2727 111/8/121/8/12 9:279:27 PMPM 28 RETHINKING COMPETITIVENESS

Hoxby, Caroline M., and G. Weingarth. 2005. Taking Race out of the Equation: School Reassignment and the Structure of Peer Effects. Working Paper, Harvard University, Cambridge, MA. HSBC United States. 2010. New Yorkers Discover the World’s Potential in Unexpected Places. http://www.hsbcusa.com/ourcompany/pressroom/2010/news_09302010_ newyork_discover.html. Accessed February 2. International Institute for Management Development (IMD). Various years. World Competitiveness Yearbook. Geneva, Switzerland: IMD. Jencks, Christopher, and Susan E. Mayer. 1990. “The Social Consequences of Growing Up in a Poor Neighborhood.” In Inner-City Poverty in the United States, ed. L. Lynn and M. McGeary, Washington, D.C.: National Academy Press, 111–86. John, Peter, Keith Dowding, and Stephen Biggs. 1994. Residential Mobility in London: A Micro-Level Test of the Tiebout Model. Report to the Economic and Social Research Council. Brunel University, London. Jud, G. Donald, and James M. Watts. 1981. “Schools and Housing Values.” Land Economics 57:459–70. Kahn, Matthew E. 2000. “Smog Reduction’s Impact on California County Growth.” Journal of Regional Science 40:565–82. King, A. Thomas. 1973. Property Taxes, Amenities and Residential Land Values. Cam- bridge, MA: Ballinger. Knight, Brian G., and Nathan Schiff. 2010. Spatial Competition and Cross-Border Shop- ping: Evidence from State Lotteries. NBER Working Paper. National Bureau of Eco- nomic Research, Cambridge, MA. Koven, Steven G., and Mack C. Shelley II. 1989. “Public Policy Effects on Net Urban Migration.” Policy Studies Journal 17:705–18. Krugman, Paul. 1994. “Competitiveness: A Dangerous Obsession.” Foreign Affairs 73:28–44. Ladd, Helen F., and John Yinger. 1994. “The Case for Equalizing Aid.” National Tax Journal 47:211–24. Lall, Sanjaya. 2001. Competitiveness Indices and Developing Countries: An Eco- nomic Evaluation of the Global Competitiveness Report.” World Development, 29:1501–25. Lenon, MaryJane, Sajal K. Chattopadhyay, and Dennis R. Heffley. 1996. “Zoning and Fiscal Interdependencies.” Journal of Real Estate Finance and Economics 12:221–32. Liu, Ben-Chieh. 1977. “Local Government Finance and Metropolitan Employ- ment Growth: A Simultaneous-Equations Model.” Southern Economic Journal 43:1379–85. Massey, Douglas S., Andrew B. Gross, and Mitchell L. Eggers. 1991. “Segregation, the Concentration of Poverty, and the Life Chances of Individuals.” Social Science Research 20:397–420. McDougall, G. A. 1976. “Local Public Goods and Residential Property Insights and Extensions.” National Tax Journal 29:436–47.

HHassett.indbassett.indb 2828 111/8/121/8/12 9:279:27 PMPM TIEBOUT AND COMPETITIVENESS 29

Mills, K. E., M. B. Percy, and L. S. Wilson. 1983. “The Influence of Fiscal Incentives on Interregional Migration: Canada, 1961–1978.” Canadian Journal of Regional Sci- ence 6:207–29. Musgrave, Richard A. 1939. “The Voluntary Exchange Theory of Public Economy.” Quarterly Journal of Economics 53:213–37. Noto, N. A. 1976. “The Impact of the Public Sector on Residential Property Values.” Proceedings of the National Tax Association 192–200. Oates, Wallace E. 1969. “The Effects of Property Taxes and Local Public Spending on Property Values: An Empirical Study of Tax Capitalization and the Tiebout Hypothesis.” Journal of Political Economy 77:957–71. ———. 2005. “The Many Faces of the Tiebout Model.” In The Tiebout Model at 50: Essays in Public Economics in Honor of Wallace Oates, ed. W. A. Fischel. Cambridge, MA: Lincoln Institute of Land Policy. Ochel, Wolfgang, and Oliver Röhn. 2006. “Ranking of Countries: The WEF, IMD, Fraser and Heritage Indices.” Cesifo DICE Report, 48–60. Organisation for Economic Co-operation and Development. 1998. Harmful Tax Com- petition: An Emerging Global Issue. Paris: Organisation for Economic Co-operation and Development. ———. 2000. Towards Global Tax Cooperation. Paris: Organisation for Economic Co- operation and Development. ———. 2002. International Mobility of the Highly Skilled. Paris: Organisation for Eco- nomic Co-operation and Development. Pack, Janet R. 1973. “Determinants of Migration to Central Cities.” Journal of Regional Science 13:249–60. Percy, Stephen L. 1993. Revisiting Tiebout: Moving Rationales, Exiting Behavior, and Gov- ernmental Responses to Metropolitan Mobility. Paper presented at the Midwest Politi- cal Science Association, Chicago. Percy, Stephen L., and Brett W. Hawkins. 1992. “Further Tests of Individual-Level Propositions from the Tiebout Model.” Journal of Politics 54:1149–57. Reinert, Erik S. 1995. “Competitiveness and Its Predecessors: A 500 Year Cross- National Perspective.” Structural Change and Economics Dynamics 6:23–42. Ricardo, David. (1817) 2010. On the Principles of Political Economy and Taxation. Lon- don: John Murray. Reprint, New York: General Books LLC. Rose, Andrew, and Mark Spiegel. 2006. “Offshore Financial Centers: Parasites and Symbionts?” Economic Journal 117:1310–35. Rosen, Harvey S., and David J. Fullerton. 1977. “A Note on Local Tax Rates, Public Benefit Levels and Property Values.” Journal of Political Economy 85:433–40. Ross, Stephen L., and John Yinger. 1999. “Sorting and Voting: A Review of the Lit- erature on Urban Public Finance.” In Handbook of Regional and Urban Economics, ed. P. C. Ceshire and Edward Mills, vol. 3, Applied Urban Economics. Amsterdam: North-Holland.

HHassett.indbassett.indb 2929 111/8/121/8/12 9:279:27 PMPM 30 RETHINKING COMPETITIVENESS

———. 2006. “Uncovering Discrimination: A Comparison of the Methods Used by Scholars and Civil Rights Enforcement Officials.” American Law and Economics Review 8:562–614. Rothstein, Jesse M. 2006. “Good Principals or Good Peers? Parental Valuation of School Characteristics, Tiebout Equilibrium, and the Incentive Effects of Competi- tion among Jurisdictions.” American Economic Review 96:1333–50. Roy, Joydeep. 2009. Effect of a School Finance Reform on Housing and Residential Segrega- tion: Evidence from Proposal A in Michigan. Working Paper, Georgetown University, Washington, D.C. Rubinfeld, Daniel L. 1987. “The Economics of the Local Public Sector.” In Handbook of Public Economics, vol. 2, ed. Alan J. Auerbach and Martin Feldstein. Amsterdam: Elsevier, 571–645. doi:10.1016/S1573-4420(87)80006-X. Sacerdote, Bruce. 2001. “Peer Effects with Random Assignment: Results for Dart- mouth Roommates.” Quarterly Journal of Economics 2:681–704. Salvatore, Dominick. 2010. “Globalization, International Competitiveness and Growth: Advanced and Emerging Markets, Large and Small Countries.” Jour- nal of International Commerce, Economics and Policy 1:21–32. doi:10.1142/ S179399331000007X. Samuelson, Paul. 1954. “The Pure Theory of Public Expenditure.” Review of Economics and Statistics 36:387–89. ———. 1955. “Diagrammatic Exposition of a Theory of Public Expenditure.” Review of Economics and Statistics 38:350–56. ———. 1958. “Aspects of Public Expenditure Theories.” Review of Economics and Sta- tistics 40:332–38. Schneider, Mark. 1985. “Suburban Fiscal Disparities and the Location Decisions of Firms.” American Journal of Political Science 29:587–605. Schneider, Mark, and John Logan. 1982. “The Effects of Local Government Finances on Community Growth Rates.” Urban Affairs Quarterly 18:91–105. Schwager, Robert. 2007. Public Universities, Tuition and Competition: A Tiebout Model. Discussion paper. Zentrum für Europäische Wirtschaftsforschung, Mannheim, Germany. Scott, Bruce R. 1985. “U.S. Competitiveness in the World Economy.” International Executive 27:26. doi:10.1002/tie.5060270112. Smith, Adam. (1776) 1982. An Inquiry into the Nature and Causes of the Wealth of Nations. London: Methuen. Reprint, Penguin Classics. Somin, Ilya. 2008. “Tiebout Goes Global: International Migration as a Tool for Voting with Your Feet.” Missouri Law Review 73:167–76. Tiebout, Charles M. 1956. “A Pure Theory of Local Expenditures.” Journal of Political Economy 64:416–24. Vartanian, Thomas P., and Philip M. Gleason. 1999. “Do Neighborhood Conditions Affect High School Dropout and College Graduation Rates?” Journal of Socio- Economics 28:21–41. doi:10.1016/S1053-5357(99)00011-6.

HHassett.indbassett.indb 3030 111/8/121/8/12 9:279:27 PMPM TIEBOUT AND COMPETITIVENESS 31

Vigdor, Jacob, and Thomas Nechyba. 2005. Peer Effects in North Carolina Public Schools. Working Paper, Duke University, Durham, NC. Von Furstenberg, George M., and Dennis C. Mueller. 1971. “The Pareto Optimal Approach to Income Redistribution: A Fiscal Application.” American Economic Review 61:628–37. Wayslenko, Michael. 1980. “Evidence on Fiscal Differentials and Intrametropolitan Firm Location.” Land Economics 56:339–48. White, M. J. 1975. “Fiscal Zoning in Fragmented Metropolitan Areas.” In Fiscal Zoning and Land Use Controls, ed. E. S. Mills and W. E. Oates. Lexington, MA: Heath-Lexington. Wilson, John D. 1999. “Theories of Tax Competition.” National Tax Journal 52:269–304. World Economic Forum. Various years. Global Competitiveness Report. World Eco- nomic Forum Reports. New York: Oxford University Press. Yinger, John. 1982. “Capitalization and the Theory of Local Public Finance.” Journal of Political Economy 90:917–43. Yinger, John, Howard S., Bloom, Axel Börsch-Supan, and Helen F. Ladd. 1988. Prop- erty Tax and House Values: The Theory and Estimation of Intrajurisdictional Property Tax Capitalization. London: Academic Press. Zimmer, Ron W., and Eugenia F. Toma. 2000. “Peer Effects in Private and Public Schools across Countries.” Journal of Policy Analysis and Management 19:75–92. Zimmerman, B. J. 2000. “Attainment of Self-Regulation: A Social Cognitive Per- spective.” In Handbook of Self-Regulation, ed. M. Boekaerts, P. R. Pintrich, and M. Zeidner, San Diego, CA: Academic Press, 13–39.

HHassett.indbassett.indb 3131 111/8/121/8/12 9:279:27 PMPM 2

Competitive Tax Policy Joel Slemrod1

Calls for more competitive government policy, and in particular a more competitive tax policy, abound in Washington, D.C. For example, maintaining competitiveness is prominent in The Moment of Truth, the recent report of the “deficit commission” (the National Commission on Fiscal Responsibility and Reform). It is mentioned early on in the section on guiding principles and values, where it is associated with investment in education, infrastructure, and high-value research and development. The term appears often in the report’s section on tax reform, where the corporate tax system is singled out as hurting America’s ability to compete, and a lower statutory rate and a territorial system are identified as the keys to improving it. This chapter assesses the role of tax policy in competitiveness and the role of competitiveness in tax policy. While I eventually address policy details and the lessons of economic theory, facts, and evidence, I begin by defining what the term competitiveness means.

What Exactly Is Competitiveness? The terms “compete,” “competition,” and “competitiveness” have a special resonance—and specific meaning—to economists. Because of the discon- nect between the specific meaning of “competitiveness” to economists and the broad use of the term in popular debates, the word has been the source of much miscommunication. We’re Number 1! Because competition is an essential element of most athletic endeavors, sports analogies often crop up. But what is the right analogy? A running race, where each nation is an entry? A race where the

HHassett.indbassett.indb 3232 111/8/121/8/12 9:279:27 PMPM COMPETITIVE TAX POLICY 33

prizes depend on the order of finish, or one where the prizes depend only on how fast one runs? If the runner is a country, and the runner’s time is income per capita, then surely the latter analogy is more apt. The benefits of prosperity are not diminished by another nation’s greater prosperity. To be sure, some people may obtain psychological benefits from living in a country that is number 1. More seriously, there may be military reasons for car- ing about a country’s relative economic prosperity, as economic well-being is what forms the basis for potential military might. But note that military might depends on national income, not on national income per capita. Regardless of their prosperity, small nations must seek strategic military alliances. Even the race-against-time analogy isn’t quite right, because countries have complementary attributes so that by working together and trading with each other the performance of all competitors can be enhanced. This is straight out of Adam Smith—global commerce allows every country to enhance its prosperity by concentrating on doing what it does best and taking full advantage of what other countries do best. In the past the notion of competitiveness has been associated with the relative size of a nation’s export sector or its trade surplus. This is dangerous, because maximizing either of these quantities is not a defensible policy goal.2 A declining trade balance may be a symptom of a problem, but it is not a reliable indicator of economic health. One can easily conceive of a declining economy with a thriving export sector—if, for example, the export sector is propelled by a favorable exchange rate caused by the drying up of attractive domestic opportunities. Conversely, the discovery of natural resources, while a boon to prosperity, often causes a decline in the manufacturing trade balance. In general, a nation’s trade balance reflects its surplus of national saving over domestic investment. It is reasonable and appropriate that a country with favorable investment opportunities will draw on the saving of all nations to finance it, causing a deficit in the current account. This is not a cause for alarm, but rather for celebration. Competitiveness of a country is also sometimes associated with the success of its resident multinational companies, sometimes measured by profitability share, sometimes by market share. But high multinational corporation profits could be achieved simply by a transfer from taxpayers to domestic corporations, a policy that most would agree is not appropriate and would not improve the fundamental soundness of the economy. With

HHassett.indbassett.indb 3333 111/8/121/8/12 9:279:27 PMPM 34 RETHINKING COMPETITIVENESS

government subsidies, domestic firms could gain market share even in cases where the revenue gained is less than the additional cost of production, which is not in the national interest. As this argument comes up in a different form in the debate about corporate tax reform, I return to it below. One striking aspect of the policy debate about competitiveness is that even detailed treatises on the subject often proceed in the complete absence of a definition of the issue at hand. Of course The Economist need not define each term of art every time it uses one. But the absence is troubling in, for example, the 116-plus-page report issued by the U.S. Treasury in 2007 titled Approaches to Improve the Competitiveness of the U.S. Business Tax Sys- tem for the 21st Century. The text refers to the “competitiveness of U.S. companies and workers” (p. i), the “competitiveness of U.S. businesses” (p. iii and many places thereafter), and the “competitiveness of the U.S. economy” (p. 2). Nowhere is there a definition of any of these terms. Maybe it’s just me. Maybe it’s like what Justice Potter Stewart said in the 1964 case Jacobellis v. Ohio about pornography, or what is obscene: “I shall not today attempt further to define the kinds of material I understand to be embraced . . . [b]ut I know it when I see it.” But I don’t think so. Using the term “competitiveness” in so many different contexts begs some important questions: How does the competitiveness of a country depend on the competitiveness of its businesses? How does the competitiveness of a business depend on the competitiveness of its tax system? In what follows I will address these questions, and I will begin immediately by offering a definition of the competitiveness of a country, although admittedly this definition is not a concise or elegant one. A competitive economy would maintain (and preferably expand) the real incomes of citizens, fairly shared, in the face of global international markets and the policies of other countries, which vary in trade policy from a free-trade orientation to aggressive promotion of exports, and vary in tax policy from domestically oriented and open to international cooperation, to aggressive courting of foreign investment, to the beggar-thy-neighbor behavior of tax havens. This definition reveals that I believe that competitiveness, properly interpreted, is not a substitute goal for prosperity. The term “competitiveness” has value to the extent that it reminds us that the conditions under which we strive for prosperity have changed as the integration of national

HHassett.indbassett.indb 3434 111/8/121/8/12 9:279:27 PMPM COMPETITIVE TAX POLICY 35

economies into the world economy has progressed. Just as the training regi- men for a race at high altitude is different, it may be that the way to achieve prosperity is different in a globally integrated economy.

How Can Competitiveness Be Measured? If competitiveness is best thought of as the recipe for prosperity in a global economy, what is that recipe? A broad—though certainly not universal— consensus among economists would stress, among other things, the following aspects of policy:3

1. Fiscal policy discipline; 2. Direction of public spending toward the broad-based provision of key pro-growth, pro-poor services, such as primary education, primary health care, and infrastructure investment, and away from sector-based subsidies; 3. Tax reform—a broadening of the tax base and the adoption of moderate marginal tax rates; and 4. Deregulation—abolition of regulations that impede market entry or restrict competition, and prudent oversight of financial institutions.

I do not have the space here to defend these recommendations or even to qualify them, with or without the lessons to be learned from the 2008–2009 financial crisis and the subsequent world recession. Most of what I have to say does not depend on accepting all of these ingredients to the recipe for prosperity. If one accepts this broad path to prosperity, how does one measure the competitiveness of a country? Arguably the most prominent country measures of competitiveness in this spirit are published by the Davos-based World Economic Forum (WEF). Since 2005 the WEF has based its competitiveness analysis on the Global Competitiveness Index (GCI). To its credit, the WEF provides a succinct definition of competitiveness: “the set of institutions, policies, and factors that determine the level of productivity of a country” (2010, p. 4). Because in the long run, a country’s prosperity

HHassett.indbassett.indb 3535 111/8/121/8/12 9:279:27 PMPM 36 RETHINKING COMPETITIVENESS

depends largely on its productivity, this definition is certainly compatible with the one offered in this chapter. The GCI is a weighted average of more than one hundred indicator variables, many of which come from surveys of executives, each measuring a different aspect of competitiveness. The components are grouped into twelve interdependent “pillars.” The tax system comprises two of the GCI indicators. One, called total tax rate, is an indicator within the sixth pillar, “global market efficiency,” subheading “competition.” The other, called extent and effect of taxation, is based on executives’ answers to survey questions and is placed into two parts of the GCI: half in the competition part of the global market efficiency pillar and half in the seventh pillar, labeled “labor market efficiency,” subheading “flexibility.” In the 2010–2011 Global Competitiveness Index rankings, the United States ranks fourth best of the 139 countries covered, behind only Swit- zerland, Sweden, and Singapore, in that order. However, the United States ranks just seventy-first best on the extent and effect of taxation (Bahrain is first) and eighty-ninth best in the total tax rate indicator (Timor-Leste is first).4 The GCI reports do not provide enough information and documen- tation to perform counterfactual exercises such as what would happen to the U.S. competitiveness score and ranking if, for example, its corporate tax rate went from 35 percent to 25 percent or from 35 percent to 45 percent, but one can infer that the effect would be minor: it is one tax of many that affects two out of more than one hundred indicators of competitiveness. Moreover, one of the twelve pillars (the third) is “macroeconomic environment,” of which three of the six components are government budget balance, government debt, and country credit rating.5 To the extent that a tax cut reduces revenue and increases the deficit, any positive effect on the tax components of the GCI would be to some extent offset by deterioration in the macroeconomic environment pillar of competitiveness. One striking aspect of the World Economic Forum measure of competitiveness is that neither its definition of competiveness nor the construction of the competitiveness index is particularly about the global economy. The definition and explanation of the definition do not mention how globalization affects the recipe for prosperity. Most of the one-hundred-plus indicators do not refer to the fact. In addition, and of note, although the GCI provides a ranking of countries, it does not accept the notion that a

HHassett.indbassett.indb 3636 111/8/121/8/12 9:279:27 PMPM COMPETITIVE TAX POLICY 37

country’s rank makes its residents better off per se. It also suggests that although tax reform may improve global and labor market efficiency, it will enhance a country’s overall competitiveness only if this is not offset by deterioration in its macroeconomic environment. We can quibble about the irresolvable question of whether tax policy comprises 1/50 of the recipe or 1/100 (or even 1/5 or 1/10) of the recipe for prosperity. Of course, such small fractions wound the egos of people in the tax business. Two responses are in order. First of all, just because the World Economic Forum says so does not make it true. Second, and more to the point, even if the tax system is just one of many aspects of the policy environment, this does not mean that we should not strive to get it right. We should. In that spirit I discuss next what kind of tax system is most conducive to prosperity, fairly shared, in a global economy—competitiveness. Before doing so, we must acknowledge an elephant in the room. If competitiveness is prosperity in global clothing, then why not use more direct measures of prosperity, such as real per capita gross domestic product (GDP) or gross national product, to assess it? Indeed, the overall GCI ranking and per capita income are very highly correlated, with a rank-order correlation of 0.85. When assessing the impact of tax policy below, I will indeed use real per capita income, but I readily acknowledge that even real income measures have shortcomings. They do not, for example, account for the value of nonmarket activities, such as child rearing, the quality of the atmosphere, or just plain leisure. Nor do they account for the distribution of prosperity, so they are not measures of prosperity, fairly shared.6

Taxation and Competitiveness In the previous sections I discussed various meanings of competiveness and how it might be measured, and settled on one interpretation. I turn now to what tax policies contribute to competitiveness, so defined.

No New Criteria. Once we accept that a country’s competitiveness is about how to maintain (and preferably expand) the real incomes of citizens, fairly shared, in the face of global international markets and the policies of other countries, we can make use of the economists’ traditional criteria for evaluating policy while recognizing the reality of global competition among

HHassett.indbassett.indb 3737 111/8/121/8/12 9:279:27 PMPM 38 RETHINKING COMPETITIVENESS

firms for profits and among countries for investment, job opportunities, and tax revenues. Thus, recognizing global competition does not add a new criterion by which to judge tax policy, but rather it changes how any given policy meets those criteria. For the most part the policies that best balance these (often competing) objectives are still appropriate in a global economy, and the policies that are misguided in a purely domestic economy still look misguided in a global economy. Thus the required change in policy perspective is not radical, although some important new issues do emerge.

A Low-Rate, Broad-Base Tax System. What kind of tax system achieves this objective? Economists who study taxation have a much more nuanced and sophisticated view than the Global Competitiveness Index of what kind of tax system is most conducive to prosperity, fairly shared, in a global economy. I begin with a bird’s-eye view and then discuss business and corporate taxes more specifically. Most economists, and I, believe that the tax system most conducive to prosperity is a low-rate, broad-base tax system, one that in most cases minimizes the role of the tax system in private decision making and conse- quently allows market forces to determine the allocation of resources. In a narrow-base income tax, resources are attracted to the tax-preferred sectors or forms of income, leading to an inefficient allocation of resources and therefore lower real income. Before proceeding, I note that “base broadening” is another term of art that merits some clarification. Although I advocate a broad-base tax system, not all apparent additions to the tax base are advisable. Consider, for example, the recent move by several states to adopt some type of gross receipts tax. Compared to either a retail sales tax or a value-added tax, a gross receipts tax adds business-to-business sales to the aggregate tax base. Because the base is thereby larger, a gross receipts tax can collect the same amount of revenue with a lower rate than can a retail sales tax, value-added tax, or, for that matter, a corporation income tax. But the broader base alone does not make it a better, read more efficient, tax. On the contrary, nearly all tax experts agree that a gross receipts tax is a less efficient way to raise revenue because it provides an inefficient incentive for vertical integration and generates capriciously varying tax penalties across goods—it is a broader base, but a worse base. Nevertheless, advocates of

HHassett.indbassett.indb 3838 111/8/121/8/12 9:279:27 PMPM COMPETITIVE TAX POLICY 39

gross receipts taxes often trumpet the lower rate as a sufficient argument for its attractiveness. Some base broadeners are not ill advised per se, but are irrelevant. To make my point, consider the following extreme hypothetical example. Imagine that taxable income was replaced by a notion of “really broad taxable income” (RBTI), which is equal to taxable income multiplied by ten. Then the same revenue could be obtained (putting aside the behavioral response of those who don’t figure this all out) as with a taxable-income- based levy, but with a 90 percent reduction in tax rates! Lest you think this is a silly example, recall that the same spirit applies to the Domestic Produc- tion Activities Deduction, passed in 2004, which allows a deduction, now 9 percent, for qualified activities; technicalities aside, this is not essentially different from keeping the tax base as is and applying a tax rate of 31.85 percent rather than the headline statutory rate of 35 percent. Some policies that shrink the base, such as accelerated depreciation, apply in principle to all sectors, although the magnitude of the impact may vary across sectors depending on the mix of capital goods used in a sector.7 Other policies, such as depletion allowances for oil and gas well owners, are targeted to particular sectors. The cost of following a high-rate, narrow-and-bad-base policy is likely to be even higher in an integrated world economy than in a closed economy. Because the global economy expands the menu of available investments, a sector that is relatively disfavored by the tax system will shrink more than otherwise; a sector that is tax favored will attract even more resources than otherwise, resources that could have been put to use more productively elsewhere in the economy. Thus, unless there is a compelling reason for government to overrule the market, the imperatives of the global economy make a neutral tax system even more desirable. Economic globalization also strengthens the argument for a tax system that avoids using the tax system as a means to a hidden industrial policy. On the face of it, this argument may sound backward. After all, in a closed economy it is more apparent that when tax preferences draw resources into one sector, they are drawn away from another domestic sector. But when the competition is German or Japanese firms, isn’t it possible that tax breaks that increase business for domestic firms attract much of that business from foreign corporations? And isn’t that in the national interest?

HHassett.indbassett.indb 3939 111/8/121/8/12 9:279:27 PMPM 40 RETHINKING COMPETITIVENESS

Not necessarily. First of all, if the tax preference is offset by a business tax increase elsewhere, some other sector, also probably facing foreign business competitors, will likely be hurt. Not offsetting the tax preference with a tax increase elsewhere masks its true cost and merely shifts the burden to future generations, who will be shackled with some combination of higher taxes and higher deficits. A streamlined, neutral tax system does not preclude all policies designed to alter market outcomes, nor does it accept that the status quo has achieved such a state of affairs. For example, federal subsidy of research and experimentation can be defended because there are spillover benefits to basic research, and an appropriately designed subsidy will encourage firms to carry out research whose social, but perhaps not private, return justifies its cost. Tax policies to address negative externalities such as pollution may be justified as well. Coming out for a streamlined, neutral tax system does not pin down exactly how to tax business. Later I will go through the economic arguments that are relevant to assessing this. But before doing so, I briefly examine, and dismiss, some unhelpful arguments that are common in tax policy discussions.

A Short Digression on Unpersuasive Economic Arguments In this section I present three economic arguments that I find unpersuasive.

Unpersuasive argument #1 (UA1): Lower taxes on base X are always better because they reduce the disincentive for businesses and individuals to do X. In a narrow sense UA1 is literally true. But, by reductio ad absurdum, it is not a serious argument because it implies that all taxes should be zeroed out to eliminate disincentives. Indeed, it suggests that the government should subsidize everything so that we get more of everything, such as labor supply, investment, saving, and so on. The problem with this argument is that fiscal discipline is also a key part of the recipe for prosperity. Ruling out UA1 is important because it constrains what are acceptable arguments for tax cuts. In the absence of some kind of a revenue constraint, arguments for lower taxes on anything, any time, and in just about any setting are appealing. However, with one exception discussed below, the

HHassett.indbassett.indb 4040 111/8/121/8/12 9:279:27 PMPM COMPETITIVE TAX POLICY 41

only insightful way to evaluate tax systems is to compare systems of equal revenue yield. To counter by saying “We’ll instead cut spending” is evasive and not helpful because governments face a budget constraint, and ensur- ing prosperity, fairly shared, means undertaking certain costly activities. The optimal extent of government spending is a central question that all societies must resolve, and in so doing we should keep in mind that the true cost of government activity must account for the additional disincentives, and therefore inefficiency, that raising the taxes to fund these activities requires. But as a matter of intellectual discipline, debates over tax policy should proceed in a revenue-neutral framework, because this forces the recognition that, while lowering taxes on base X may increase X, making up the lost revenue by raising taxes on base Y and Z will reduce the incentive to do these things. A good tax system needs to balance these disincentives.

Unpersuasive Argument #2 (UA2): Policy X is better than our current policy because most other countries do X. Other countries face different economic environments, have a citizenry with different preferences regarding what government should provide and how they should tax themselves, and often flat out make poor decisions. For these reasons simply mimicking what other countries do is not a compelling way to make policy choices. In addition, it is intellectually disingenuous to advocate policy X (e.g., lower corporate tax rates—see below) because, say, France and Sweden do this, while ignoring the policy implications of the fact that these countries also have policies Y and Z (e.g., national health insurance and much higher tax-to-national-income ratios), policies that the advocates of policy X in the United States would dismiss out of hand. To be sure, in a global economy the United States cannot ignore other countries’ tax policies, because they may affect our prosperity and, more important, may affect what is appropriate U.S. tax policy. In what follows I will take this issue seriously, but not by simply selectively assuming that other countries have figured it all out and that we have to get in step.

Unpersuasive Argument #3 (UA3): Policy X is a good thing because it increases corporate profits. A profitable corporate sector is not an end in itself, although it may be an important instrumental component, or symptom, of a prosper- ous economy. For example, outright unconditional grants of taxpayer funds

HHassett.indbassett.indb 4141 111/8/121/8/12 9:279:27 PMPM 42 RETHINKING COMPETITIVENESS

to corporations would increase profits and, especially if expected to be permanent, increase stock prices. But this would represent merely a transfer mostly to relatively wealthy shareholders from everyone else and would be an unattractive policy. Transfers, or tax breaks, that increase the marginal incentive of businesses to invest or hire workers, as an unconditional transfer would not do, may be appropriate in some cases, though.8

Business Taxation Policy discussions about tax competitiveness tend to focus on the taxation of businesses, and often focus only on corporations.

What Is a Business Tax? Carefully defining the term “business tax” is also a worthwhile exercise. In particular, distinguishing among (1) remitting taxes to the government, (2) having a statutory liability to pay tax, and (3) bearing the burden of a tax is important because of the central role businesses play in the tax remittance and collection process. The impetus behind the central role of business in tax remittance was most elegantly stated by Richard Bird (2002), who wrote: “The key to effective taxation is information, and the key to information in the modern economy is the corporation. The corporation is thus the modern fiscal state’s equivalent of the customs barrier at the border” (p. 199). Collecting taxes from businesses makes use of the economies of scale the tax authority obtains from dealing with a smaller number of larger tax-remitting units, many of which for other purposes have already developed sophisticated systems of record keeping and accounting. One measure of the central role of business in the U.S. tax system is provided by Christensen, Cline, and Neubig (2001), who calculated that in 1999, businesses “paid, collected, and remitted” to all levels of government 83.8 percent of total taxes (p. 498, italics added).9 Of the 83.8 percent, the authors label 31.3 percent as “tax liability of business,” 8.1 percent as the “business as tax collector,” and 44.4 percent as “business as withholding agent” (p. 498 [authors’ calculations]). That a business entity writes a check to the IRS does not indicate anything about who bears the burden of a tax or about its effect on, for example, business investment or job creation. Remitting tax is not the same as bearing

HHassett.indbassett.indb 4242 111/8/121/8/12 9:279:27 PMPM COMPETITIVE TAX POLICY 43

the burden of a tax, both because the tax burden can be shifted through the adjustment of market prices and because, in the case of businesses, it is not informative to say that businesses bear the burden of taxes when we want to know which people—the firm’s stockholders, bondholders, customers, or workers—are worse off because of a tax. Moreover, standard public finance theory holds that who or what entity remits a tax or is legally liable for a tax will have no impact, in the long run, on either who bears the burden or the efficiency cost of raising revenue; only what triggers tax liability matters. For example, abstracting from administration and compliance issues, whether the retailer remits a retail sales tax to the government or the customer remits it, will not, in the long run, affect the price received by the retailer or the price paid by the consumer.10 Thus, focusing on changes in remittance or statutory liability can be seriously misleading with regard to the ultimate impact of a tax reform. For example, replacing corporate income taxes with a value-added tax would not dramatically alter the remittance responsibility of the business sector. But all economists recognize that the impact of a value-added tax would be very different from a tax on business profits, and they understand that the key is what triggers tax liability rather than whether the business must remit the tax in question. Regardless of what triggers tax, to be discussed at length below, there are very good reasons for retaining a business-based tax remittance and collection system.

Should (Business) Income Be Taxed? One of the most important and enduring issues in tax policy is whether tax liability should be based on income or consumption. Taxes on consumption, which take many different forms, including retail sales tax, value-added tax, the Hall-Rabushka flat tax, and personal expenditure tax, all share the quality that they levy no tax on the riskless rate of return to saving or investment and apply tax on the return in excess of the riskless rate. I cannot resolve that argument here, so I will proceed by first evaluating business taxation in the context of a comprehensive income tax and then broadening the context to include tax systems that vary from this framework.

How Should Business Income Be Taxed? What kind of business tax base fits the spirit of a low-rate, broad-base income tax? For one thing, the tax base must be cleaned of provisions carved out solely to benefit specific

HHassett.indbassett.indb 4343 111/8/121/8/12 9:279:27 PMPM 44 RETHINKING COMPETITIVENESS

companies or industries. This objective is behind the call of the National Commission on Fiscal Responsibility and Reform (2010) to eliminate (more than thirty) general business credits and all (more than seventy-five) business tax expenditures, although not all of these fit nicely into this category. Sector-specific tax benefits generally lack a principled economic justification and therefore cause resources to flow to less efficient uses, reducing national income. A principled commitment to a less activist government requires leveling the playing field among businesses and, with only limited exceptions, letting private entrepreneurs and capital owners determine how the economy’s resources are directed. The same argument applies to tax provisions that explicitly or implicitly favor certain kinds of capital assets or financing methods.

The U.S. Corporation Income Tax. Unfortunately the U.S. system of taxing business income fails to meet any of these standards. To see why, first consider how a progressive, comprehensive (i.e., all sources of income are subject to tax) income tax system should work. Business income would be attributed to the owners of the business and taxed at whatever individual tax rate is appropriate, given the total income of the owner. If nonowners have supplied capital to the business, the cost of obtaining the capital would be deducted from business income, and the income paid would be subject to tax at the appropriate tax rate of the supplier of the capital. This is how it works now for all but the biggest public corporations. Any business with one hundred or fewer owners, which accounts for most businesses but a relatively small fraction of business income, can retain the legal advantages of incorporation—principally limited liability and perpetual ownership—while not being subject to corporation income tax. The company’s income is allocated to the owners and added to their individual taxable income. However, an incorporated business (and since 1997, an unincorporated business so electing) is subject to the corporation income tax and its graduated rate structure, which subjects annual taxable income up to $75,000 to first a 15 percent and then a 25 percent rate. Although the graduated rate structure of the corporation income tax mim- ics the graduation of the personal tax, the total income of the owner of a small business may put him or her well into the top (currently 35 percent) individual bracket, so the tax relief afforded by the lower corporate rates

HHassett.indbassett.indb 4444 111/8/121/8/12 9:279:27 PMPM COMPETITIVE TAX POLICY 45

cannot be justified on progressivity grounds. Thus, for most businesses, the corporation tax is not a burdensome double tax, but rather an option for tax reduction. This feature of business taxation could be eliminated either by restricting the ability of companies to opt into the corporation rate structure or by eliminating the low rates of tax on the first $75,000 of income. An entirely different system applies to the big, publicly owned companies that make up only a few thousand of the several million businesses in the country but account for a large fraction of business activity. Publicly owned corporations cannot elect out of the corporation income tax. This produces an odd system in many ways. First, it subjects the corporation’s income to what is effectively a flat rate of 35 percent (the tax benefits of the lower rates in the bottom brackets are trivial compared to the income of public corporations), regardless of what tax bracket the owners of the corporation are in and regardless even of whether the shareholders are tax-exempt entities. The corporate tax system creates a host of well-known distortions. Although interest payments are deductible as a business expense but taxable to the lender, the taxation of equity finance does not follow this pattern. Corporations cannot deduct anything in recognition of the cost of attracting equity financing, but the equity providers (i.e., the shareholders) are taxed to some degree on the income they receive. This system causes inefficient incentives for corporations to raise capital by borrowing and to manage payments from the corporation to the shareholders in tax-efficient, but socially inefficient, ways. Finally, the two levels of tax, corporate and individual, generally cause the tax rate on business income to be higher than it is on other income and the cost of capital for corporate businesses to be higher than it is for other businesses, neither of which is justifiable. Mitigating the double taxation of income flowing through corporations would lower the cost of equity capital to domestic corporations, reduce the tax bias toward debt finance, and at the same time move the tax system in the direction of a comprehensive income tax that taxes all income once and only once. An important function of the corporate income tax is a “backstop” to the individual income tax because in part it applies to the labor income of a corporation’s principals. If the corporation tax were to be abolished, taxpayers could shift what is essentially labor income to the corporate sector and receive it free of tax, while financing their consumption via loans from their companies.11

HHassett.indbassett.indb 4545 111/8/121/8/12 9:279:27 PMPM 46 RETHINKING COMPETITIVENESS

Income Tax Alternatives. One sensible approach to these problems is to allow a personal tax credit to shareholders for some or all of corporation taxes paid. Many countries have a system like this already, although, as discussed in UA2 above, that in itself does not make it good policy. In 2003 the United States adopted a different approach when the personal tax rate on dividends and capital gains was capped at 15 percent, compared to a maximum 35 percent rate on other income. There are two problems with this approach. First, it moves the system toward one where the rate of tax on corporate income is 35 percent regardless of the tax situation of the owner, which is inconsistent with progressivity and certainly inconsistent with the oft-heralded idea of making stock ownership attractive to lower-income people. Second, it cuts the personal tax on corporate-source income while doing nothing to ensure that the corporate-level tax was in fact paid. Nota- bly, the original Bush administration proposal in the 2003 legislation linked the two levels of tax by making dividends tax free (and not just capped at 15 percent) only to the extent that the dividend-paying corporation had actually remitted corporation income tax. Linking the two taxes would ensure that, in the quest for attaining a single level of tax on corporate income, we do not end up collecting no tax at all; this is an important policy issue in light of the apparent but difficult-to-document proliferation of abusive corporate tax avoidance schemes that drain corporate tax collections.

More Radical Income Tax Alternatives. There are, to be sure, other intriguing proposals for rationalizing the taxation of corporate income. For example, under the comprehensive business income tax (CBIT), both corporate dividends and interest payments are tax free at the individual level, but, in parallel, corporate interest payments are no longer a deductible business expense, putting them on a level playing field with dividends and eliminating the distortions to financial behavior the tax system now produces. Under the CBIT, all business income is taxed at 35 percent, as opposed to the current system, under which, for most businesses, the income is taxed at the appropriate tax rate of the owner. As with the Hall-Rabushka flat tax, under a CBIT there is no distinction between corporations and other businesses.

Taxing Business—But Not Business Income. The CBIT eliminates the tax consequences of payments of dividends and interest (to the payer

HHassett.indbassett.indb 4646 111/8/121/8/12 9:279:27 PMPM COMPETITIVE TAX POLICY 47

and the recipient) made by businesses but leaves them in place for other transactions, including mortgage interest payments. One could go further and eliminate the tax consequences of all financial flows, as occurs under a Hall-Rabushka flat tax. This would not cost as much revenue as one might first guess, because it would not only exempt interest receipts from tax but also disallow interest deductions. Because the latter on net are taken by taxpayers in higher tax brackets than those who receive interest payments, attempting to collect revenue on financial flows raises little or no revenue in aggregate. Removing the tax consequences of financial flows would eliminate a highly complex area of the tax law that tries to address the vast array of complicated financial instruments, such as zero-coupon bonds and swaps, but it raises difficult unresolved issues regarding the transition from the current system and how well it would integrate with the tax systems of the rest of the world. Subjecting all businesses to tax at a single rate on a base with no deductions for interest payments and taking dividend and interest receipts out of a completely cleaned-up personal tax base is, with one important exception, exactly the Hall-Rabushka flat tax (or, with a graduated personal tax structure, the X-tax championed by David Bradford). The one exception is that under an income tax or a CBIT, businesses must depreciate the purchase of capital goods, whereas under the flat tax they can immediately write off expenses. In this way, income of all types can be taxed at the business source of income. As noted, taxing (only) at the business source affords considerable simplicity but does not easily accommodate a progressive distribution of the tax burden. Another option is the Allowance for Corporate Equity system, under which companies can deduct from their tax base an imputed normal return on their equity, making the tax treatment of equity essentially parallel to the tax treatment of debt at the business level. If the imputed rate of return is set appropriately, the Allowance for Corporate Equity system provides the equivalent of immediate expensing of investment, so it—like the Hall- Rabushka flat tax—does not cause a disincentive to invest.

Measuring the Tax Disincentive to Investment. As the discussion above suggests, the disincentive to invest provided by taxes depends on both the rate of tax and the definition of the tax base, especially the tax treatment of

HHassett.indbassett.indb 4747 111/8/121/8/12 9:279:27 PMPM 48 RETHINKING COMPETITIVENESS

the cost of doing business.12 An income tax should levy tax on net income (i.e., gross income or revenues, net of costs), so if the costs are mismea- sured, so is net income. For capital investment the key cost of doing business is the depreciation of the asset due to wear and tear and obsolescence. To correctly measure net income, we would need to be sure that depreciation for tax purposes of a capital good equals the decline in the value of the asset over the year. For most assets, this is impossible to measure exactly, so the tax code provides tax depreciation schedules that apply to broad categories of assets. Before 1981, much effort went into trying to approximate true economic depreciation so as to correctly measure net income, but since 1981, tax depreciation schedules have been used as a tool to affect aggregate investment. The more accelerated the tax depreciation allowances are, the greater is their present value to the investing company, so the disincentive to invest declines. The use of accelerated tax depreciation schedules as a fiscal policy tool that began in 1981 was withdrawn in the Tax Reform Act of 1986, but has recently come back with a vengeance with the enactment of 50 percent “bonus depreciation” during the period 2001–2004 and, beginning again in 2008, with 100 percent bonus depreciation (i.e., immediate tax expensing of the cost of capital goods) for 2011. Depending on the type of asset, acceleration of depreciation allowances can offset a substantial amount of the effect of rate changes. In the extreme case where the cost of acquiring capital assets can be deducted immediately on purchase, known as expensing (or more recently as 100 percent bonus depreciation), it offsets all of the disincentive effect of tax on the incentive to invest. In essence, under this system the government contributes a fraction of all investment expenses equal to the tax rate and recovers the same fraction of all gross revenues generated. Any project that was worthwhile in the absence of this “silent partnership” role of the government will still be worthwhile in its presence. The costs of hiring labor are essentially written off for tax purposes as incurred so that, by the same argument invoked above with regard to capital expensing, the corporate tax does not directly affect the incentive to hire labor. It may do so indirectly, though, if it affects the incentive to invest. More capital raises the productivity of labor, thus increasing the attractiveness of labor to firms.

HHassett.indbassett.indb 4848 111/8/121/8/12 9:279:27 PMPM COMPETITIVE TAX POLICY 49

It is important to distinguish between a business as the owner of a set of assets and as an ongoing operation because many tax reforms have differential effects on the value of existing capital assets and the value of newly purchased capital assets. As an example, consider the impact of moving toward a tax system with more accelerated depreciation (in the extreme, toward immediate expensing) for newly purchased assets. That change would tend to increase the after-tax profitability of newly purchased assets and therefore make new investment more attractive. However, it would also make existing assets (that must, in the absence of transition rules, still be depreciated under the preexisting law) less valuable in comparison to comparable assets purchased once the new accelerated depreciation rules are in place. Whether any company’s share price will rise or fall in the wake of this tax change depends on the relative importance of existing (sometimes called “old”) capital and the prospects of profitable future investment (“new” capital) as well as a wide variety of other factors. For this reason there may be a divergence between the kind of tax reform that is good for business, in terms of its share price, and the kind that is good for prosperity. In terms of the latter, the value of new capital matters because a higher value makes new investments more attractive. The shareholders’ interests depend on both the effect on new capital and the effect on the old capital they own, and so this is a case where the interests of the business shareholders and the stewards of the economy are not identical.13 The distinction between the share price impacts and the effect on the incentive to invest is especially important when the type of tax change being considered involves changes in depreciation schedules. Put another way, reducing the corporation tax rate “wastes” some of the revenue cost because it applies to the income of capital already in place; this is essentially a grant of tax relief to businesses that, as discussed in UA3 above, is not good tax policy. In contrast, accelerated depreciation applies only to assets purchased after its date of implementation—that is why it erodes the value of existing capital.14

International Considerations To this point I have for the most part assessed business taxation and the corporation income tax as if the U.S. economy is the only economy in the

HHassett.indbassett.indb 4949 111/8/121/8/12 9:279:27 PMPM 50 RETHINKING COMPETITIVENESS

world. But the idea that the United States is part of a global—and global- izing—world economy is central to the concept of competitiveness. How, if at all, does the global perspective change things?

The Economic Theory of Small Open-Economy Taxation. Economic theory shows that, in stylized circumstances, small open economies should levy no distorting tax on inputs to production that are perfectly mobile loca- tionally. This proscription relies on the observation that if there is at least one perfectly immobile input to production (say, labor supply), this base can be taxed to raise all necessary revenue and can, indeed, be taxed with no efficiency cost resulting from repelling production. If minimizing such efficiency cost is the sole objective of tax policy, then revenue can be raised with no efficiency cost by raising all revenue from taxing the immobile factor. Even if the government has distributional objectives (in the sense that it cares not only about national income but separately about the after-tax wage rate and the after-tax return to capital of its citizens), and even if labor supply is somewhat elastic, economic theory suggests that it is still optimal to levy no tax on the mobile base. This is because in a small open economy all taxes will be borne in the long run by owners of immobile factors, so attempting to tax the mobile factor will be ineffective in shifting the burden from capital to labor. And because taxing these immobile factors directly avoids the efficiency cost of causing mobile factors to leave, this policy dominates taxing the mobile factor even if one has distributional preferences.15

International Income Shifting. The economic theory of taxation in a small open economy rests on the fact that businesses and investors have options outside of any one country to locate real economic activity. In addition, multinational companies have considerable flexibility with respect to where, that is, which countries, their worldwide taxable income appears, and there is substantial evidence that they use this flexibility to reduce their total tax liability by shifting taxable income to low-tax countries. A wide range of income-shifting techniques exists, from transfer pricing, to inter- corporate borrowing, to the use of conduit corporations. In some cases this income shifting is facilitated by the policies of other countries; indeed, some countries, generally referred to as tax havens, are essentially in the business of facilitating tax avoidance of multinational corporations.

HHassett.indbassett.indb 5050 111/8/121/8/12 9:279:27 PMPM COMPETITIVE TAX POLICY 51

From the point of view of the United States, when a multinational corporation chooses to report a dollar of taxable income in another country rather than in the United States, this is a loss of national income. Indeed, the loss exceeds one dollar, because the tax revenue would have to be made up by imposing distorting taxes that reduce the efficiency of the economy. The incentives to shift taxable income across countries are, for the most part, driven by two things: (1) differences in statutory tax rates and (2) the rules that constrain income shifting (and enforcement of those rules) put in place by the high-tax countries, where “high tax” need not be very high relative to the essentially zero rates of tax havens. Generally speaking, these rules are such that it is easier for a multinational company to move taxable income to low-tax countries when it clearly does something there other than putting a plaque on an office building, thus making real activity more attractive in a low-rate country.16 This has been the business-model motivation behind Ireland’s 12.5 percent corporate tax rate—not only to attract taxable income and therefore tax revenue but also to attract real business activity as well. Note that this perspective implies that those who argue for lower corporate taxes as a way to stem outward income shifting and revenue loss should also be for effective enforcement of rules that deter such shifting, but they are generally not. This reasoning applies to policy toward tax havens, jurisdictions that levy no or only nominal taxes and offer themselves as a vehicle for nonresidents to escape tax—legally or not—in their country of residence. A tax haven can offer this service because it has laws and administrative practices that prevent the effective exchange of information on taxpayers benefiting from the low-tax jurisdiction. There is considerable concern that tax havens are “parasitic” on the tax revenues of the nonhaven countries, inducing them to expend real resources in defending their revenue base and in the process reducing the welfare of their residents. A 1998 report from the Organisation for Economic Co- operation and Development (OECD) concluded that “governments cannot stand back while their tax bases are eroded through the actions of countries which offer taxpayers ways to exploit tax havens…to reduce the tax that would otherwise be payable to them” (p. 37). In sharp contrast to this long-standing concern about their deleterious effects, some economists have focused on a potentially beneficial role for tax

HHassett.indbassett.indb 5151 111/8/121/8/12 9:279:27 PMPM 52 RETHINKING COMPETITIVENESS

havens. They note that, as discussed above, under certain conditions small open economies should levy no distorting tax on internationally mobile capital but that countries persist in so doing. In this context, tax havens are a device to save these countries from themselves by providing them with a way to move toward the nondistorting tax regime they should, but for some reason cannot, explicitly enact. But tax havens induce wasteful expenditure of resources, both by firms in their participation in havens and by governments in their attempts to enforce their tax codes. In addition, tax havens worsen tax competition problems by causing countries to further reduce their tax rates below levels that are efficient from the viewpoint of all countries combined. The same economic logic that underlies the call for lower statutory rates to address taxable income shifting suggests that there be substantial attention paid—unilaterally and multilaterally—to defending countries’ revenue bases.

Territorial versus Worldwide Taxation. Many recent proposals for corporation tax reform include moving from our current worldwide system of taxing income to a territorial system. Simply put, under the former system the worldwide income of U.S. individuals and corporations is subject to U.S. taxation, with a limited credit offered for income taxes paid to foreign countries and any residual tax due to the United States deferred until repatriation of the income to the U.S. parent company. Under a territorial system, the United States taxes only the income earned within its borders. In recent years developed countries have been moving toward territorial systems, but, as noted in UA2 above, this observation should not substitute for a reasoned assessment. At first blush, moving the base of taxation from worldwide income to income earned within the United States seems like a narrowing of the tax base rather than a broadening.17 But we should be seeking the best broad base, not any broader base. The principal argument for moving to a territorial system is that it would improve business competitiveness, a concept that is distinct from national competitiveness, of United States–headquartered multinational corporations operating in low-tax foreign countries. For example, if both a U.S. and a German vehicle company are considering setting up a subsidiary in Ireland, both companies would owe the (low) Irish income tax, but only the U.S. company would owe (on repatriation) a residual tax to its home country.

HHassett.indbassett.indb 5252 111/8/121/8/12 9:279:27 PMPM COMPETITIVE TAX POLICY 53

From a global perspective, a worldwide tax system may seem capriciously inefficient: why should multinational companies based in countries with worldwide systems face a tax-related relative disadvantage in low-tax countries? This seems as unjustifiable as having companies face different tax rates depending on the first letter of their company name.18 Granted that this leads to an inefficient allocation of resources from a global perspective, is it also inadvisable for promoting U.S. prosperity? This is a trickier question. Just lowering taxes for some U.S.-based multinational companies fails to be compelling by both UA1 and UA3. As Desai, Foley, and Hines (2009) have suggested, cross-border synergies might redound to make investment of U.S. multinational companies more profitable elsewhere, including in the United States, perhaps leading them to expand their operations here. But nowhere have we argued that prosperity in the United States is better served when U.S.-based multinational companies invest in the United States relative to investment by foreign-based multinationals. U.S. workers employed by Toyota, Royal Dutch Shell, or Nestlé USA have learned that a job is a job. Indeed, relatively speaking, for a U.S. multinational company a territorial tax system improves the relative attractiveness of investing in (low-tax) foreign countries’ investment relative to a U.S. investment, and so the net effect on the incentive to invest and hire workers domestically might be negative. This is not the same criterion as maximizing the prosperity of Americans, but might give some observers pause before supporting a move to a territorial system. Another problem with a territorial system is that, as long as the United States has a high statutory corporate rate relative to many other countries (including tax havens), moving to a territorial system will greatly increase the incentive for outward taxable income shifting. While under a worldwide system shifting taxable income from the United States to a low-tax country in principle only postpones the residual U.S. tax liability until repatriation, under a territorial system it would be lost forever. Defending our revenue base, which is a component of prosperity, would require vigilant enforcement and more onerous rules than now in place.

The Corporate Tax and Prosperity: Facts and Evidence Careful economic reasoning clarifies the costs and benefits of alternative ways of taxing corporate income. Relevant facts and empirical evidence

HHassett.indbassett.indb 5353 111/8/121/8/12 9:279:27 PMPM 54 RETHINKING COMPETITIVENESS

provide context for U.S. policy choices and ideally can provide a quantitative sense for the stakes involved in tax reform. Next I review some key facts and evidence about corporate taxation.

Statutory Corporate Tax Rates. The United States has one of the highest statutory corporate tax rates among developed countries, as shown in the first column of table 2-1.19 The U.S. rate is substantially lower than it was thirty years ago, but corporate tax rates have been falling throughout the world at a faster rate than in the United States.20 Perhaps surprisingly, in most other countries the amount of corporate tax revenue relative to total tax has increased over this period, the main exceptions being Germany, Japan, and the United Kingdom. Both domestic determinants of corporate taxation and increases in international pressures for tax competition seem to be behind this trend.21 There is clear evidence that the corporate tax rate is insulated from a country’s revenue needs: across countries, there is no association of the expenditure–GDP ratio with the corporate statutory rate and only weak evidence of a positive association with the average rate. There is suggestive, but not definitive, evidence that the domestic role of the corporate tax as a backstop to the individual income tax is important: across countries, there is indeed a strong association between the top individual rate and the top statutory corporate rate. Finally, there is intriguing evidence about the role of international competitive pressures on corporate taxation. Measures of openness are negatively associated with statutory corporate rates, although not with revenues collected as a fraction of GDP. Strikingly, larger, more trade-intensive countries do collect more corporate tax revenues, but this may be because these countries are more attractive venues for investment.

Marginal Effective Tax Rates on New Investment. As argued above, the statutory rate is only one component of the tax disincentive to business investment and may be outweighed by the definition of the business tax base, in particular but not limited to the tax depreciation schedules. Calcu- lating the tax disincentive, called by economists the marginal effective tax rate on investment, or METR, is much more difficult than calculating the statutory rate. It is especially difficult to calculate METRs that are comparable across countries.

HHassett.indbassett.indb 5454 111/8/121/8/12 9:279:27 PMPM COMPETITIVE TAX POLICY 55

A recent careful and comprehensive cross-country calculation of marginal effective tax rates was done by Chen and Mintz (2008) for the Foreign Investment Advisory Service of the World Bank. It generally follows a widely accepted methodology but, like all such calculations, must make certain assumptions to make the calculations tractable.22 According to this study, the marginal effective tax rate on capital for the United States in 2008, the latest year the study covers, was 26.5 percent. This is slightly below the 28.7 percent GDP-weighted average METR and puts the United States as the eighteenth highest of the eighty countries covered and seventh highest among OECD countries. Notably, the Chen-Mintz calculation takes account of not only corporate income taxes but also sales taxes on capital purchases, assets taxes, and fees on financial transactions. Indeed, if only corporate income tax provisions are included, the U.S. METR falls to 16.2 percent, which ranks tied for forty-sixth out of the eighty countries in the survey and eighteenth out of the thirty-one OECD countries in the study. Thus, according to Chen and Mintz, noncorporate taxes are what keep the United States from having a lower-than-median METR overall and among developed countries only. The second column of table 2-1 presents the 2008 results, including bonus depreciation, for the corporate tax contribution to METRs. A brief memo updating the 2008 calculations to 2009 by Chen and Mintz (2010) puts the overall U.S. METR at 35.0 percent, sixth highest in the world. The big increase between 2008 and 2009 is not, however, due to a change in U.S. tax law but rather a change in how the U.S. METR is calculated. The 2009 calculation excludes bonus depreciation, which allows half of the cost of capital goods purchased to be deducted in the year of purchase (which takes the METR from 27 percent to 35 percent), on the grounds that bonus depreciation is temporary. Note, though, that the 50 percent bonus depreciation provision applied to investment made in 2008, 2009, and 2010. The Tax Relief, Unemployment Insurance Reauthorization and Job Creation Act of 2010 allows companies to immediately write off 100 percent of their cost of assets (expensing) placed in service after September 8, 2010, and through December 31, 2011. Expensing would, I surmise, sharply lower the METR for the United States, putting it among the lowest of OECD countries. Chen and Mintz (2008) also report the results of some analysis of whether cross-country variations in the METR and the statutory corporate

HHassett.indbassett.indb 5555 111/8/121/8/12 9:279:27 PMPM 56 RETHINKING COMPETITIVENESS

tax can explain cross-country variation in inbound foreign direct investment (FDI). They report two regression specifications estimated for sixty- nine of the eighty countries for which all the necessary data are available. The first specification estimates FDI as a share of GDP in 2006 against the 2006 value of the METR, the growth rate of real GDP lagged one year, and the concurrent inflation rate; the second regression adds as an explanatory variable the statutory corporate tax rate. In both specifications the METR has a negative effect on the FDI ratio that is statistically different from zero and economically nontrivial. In the second regression the statutory tax rate has a separate negative impact, but its estimated value cannot be distinguished from zero with statistical confidence. For at least three important reasons, these regression analyses cannot be taken as compelling evidence of the effect of taxation on business investment. The first is that such a cross-section analysis omits scores of other possible influences on FDI that vary across countries. To the extent that these omitted determinants are correlated with measures of the corporate tax system, as they often certainly are, the estimated coefficients that purport to measure the effect of the corporation tax in fact measure that plus some portion of the effect of all the correlated omitted determining variables. This serious flaw applies to all such cross-sectional analyses. Another problem is that, in a footnote, the authors indicate that they estimated several regression specifications and report in the paper only “the strongest results.” I infer from this caveat that other reasonable specifications did not produce the same statistically significant negative effect of the METR on FDI, thus weakening the inferred causal impact of this measure of the tax disincentive effects. Finally, the regression analyses address only inward foreign direct investment, while the METR would be expected to affect the incentive for all domestic investment, whether it is domestically owned investment or inward foreign direct investment.

The Effect of Corporate Taxation on Real GDP and Real GDP Growth. The Holy Grail of policy analysis is to establish a compelling causal connection between the policy at issue and the ultimate variable of concern—a measure of prosperity such as real GDP per capita or growth in same, fairly shared. For a number of reasons, discussed in Slemrod (1995), establishing such a compelling causal connection is extremely difficult. In part for

HHassett.indbassett.indb 5656 111/8/121/8/12 9:279:27 PMPM COMPETITIVE TAX POLICY 57

the reasons discussed in the preceding section, the most natural kind of analysis, relating cross-country differences in prosperity to differences in policy, is almost never persuasive because it is practically impossible to hold constant, in a statistical sense, all the aspects of a country that affect its real income level. As mentioned earlier, if any of the unobserved influences on real income happen to be correlated with the policy, the estimated relationship will be a biased measure of the causal impact of the policy because it will conflate the true effect of the policy with the effect of the unmea- sured but correlated factors. Moreover, countries choose different policies for a reason, and if the corporate tax system is chosen in part to address perceived inadequacies of the level or growth of real income, purely cross- sectional analysis will conflate the effect of tax on income with the effect of income on the choice of the corporation tax system. A more promising, but still problematic, approach is to relate policy variables to subsequent growth performance; in that way, whatever determines a country’s starting level of performance need not be explained. In the past twenty-five years, a large academic literature has followed this approach to assess what determines countries’ income growth, addressing the effect of such factors as education, trade openness, institutions, corruption, and government spending, with aspects of taxation addressed with varying degrees of attention and with more or less sophistication. Rather than critically evaluate or even catalogue all of these studies, I instead focus on a recent example of this kind of approach by a highly respected team of academic researchers.23 Lee and Gordon (2005) examine cross-country data for seventy countries over the period 1970 to 1997 and focus on the impact of corporate and personal tax rates on subsequent average five-year real growth rates, controlling for other determinants of growth such as openness to trade, inflation rates, the extent of corruption, and educational levels. In one set of regression specifications, they use an instrumental-variables technique to address the possibility that the tax rate choices of countries depend on growth performance, which, as discussed above, would otherwise render a causal interpretation of the results suspect. Lee and Gordon find that a country’s real growth rate is negatively correlated with its statutory corporate tax rate, apparently supporting the notion that high rates are anticompetitive in the sense I have used it here. However, this statistical association obtains only for the subset of

HHassett.indbassett.indb 5757 111/8/121/8/12 9:279:27 PMPM 58 RETHINKING COMPETITIVENESS

non-OECD countries. There is no association between growth rates and statutory corporate tax rates among the developed countries in the OECD. Note, though, that Gordon and Lee investigate only the effect of the statutory corporate tax rate and make no attempt to calculate the tax disincentive to invest as measured by the marginal effective tax rate, which, as they are well aware, is the more appropriate measure of the tax disincentive to invest but is very difficult to measure for many countries over many years using a standardized methodology. The third column of table 2-1 shows the 2000–2009 real growth rates of most OECD countries. There is indeed a noticeable correlation between the statutory tax rates and these decadal real growth rates24—on average, countries with lower rates had higher growth rates. The analysis of Lee and Gordon suggests that other factors explain this correlation, and a causal link cannot be compellingly established. The final column of table 2-1 adds another cautionary note. Many of the low-corporate-tax-rate countries have suffered particularly large downturns during the recent recession, suggesting that the correlation may be eroding.

The Effect of Corporate Taxation on Corporate Revenues. Perhaps in part because of the difficulty of ascertaining the direct impact of corporate tax systems on prosperity and growth, several recent studies attempt to estimate the effect of the statutory rate on corporate revenues. This is of interest because if lowering a tax rate so expands the base that total revenue rises, nothing more need be established to support lowering the tax rate. Ever since Arthur Laffer drew his curve on a napkin in a Washington, D.C., res- taurant, the idea that cutting tax rates raises revenue has been suggested by tax-cut advocates. If it were true, it would be a no-brainer: those whose tax rate was cut would be better off (even though their tax liability increases), and everyone else could be, too, by judiciously sprinkling around the extra tax revenue collected from those whose incentives to produce taxable income increase. There are two problems with using this argument. The first is that most economists think it does not hold in practice, except in very specific circumstances. Second, in the circumstances in which the inverse relationship seems plausible, this is generally because the increased tax revenue is coming at the expense of some other tax base due to income shifting from one

HHassett.indbassett.indb 5858 111/8/121/8/12 9:279:27 PMPM COMPETITIVE TAX POLICY 59

TABLE 2-1 TAX RATES AND GROWTH RATES, SELECTED OECD COUNTRIES

Marginal Corporate Effective Tax Statutory Rate Due to 2000–2009 Real 2009 Real Country Tax Rate Corporate Tax Growth Rate Growth Rate Australia 30.0 (22) 25.9 (24) 0.334 (10) 0.012 (2) Austria 25.0 (12) 23.6 (21) 0.185 (22) –0.038 (16) Belgium 34.0 (27) –3.2 (1) 0.166 (24) –0.027 (13) Canada 29.5 (21) 21.1 (19) 0.232 (16) –0.025 (9) Chile 17.0 (3) 13.9 (7) 0.425 (5) –0.015 (5) Czech Republic 19.0 (4) 18.3 (15) 0.333 (11) –0.041 (17) Denmark 25.0 (12) 18.6 (16) 0.095 (26) –0.047 (18) Finland 26.0 (14) 17.1 (13) 0.207 (17) –0.081 (30) France 34.4 (28) 26.1 (26) 0.156 (25) –0.025 (9) Germany 30.2 (26) 27.3 (27) 0.082 (28) –0.047 (18) Greece 24.0 (10) 15.9 (9) 0.393 (8) –0.020 (8) Hungary 19.0 (4) 13.1 (4) 0.278 (13) –0.063 (26) Iceland 15.0 (2) 13.4 (6) 0.343 (9) –0.068 (28) Ireland 12.5 (1) 13.3 (5) 0.430 (4) –0.076 (29) Italy 27.5 (17) 28.1 (28) 0.056 (30) –0.051 (23) Japan 39.5 (30) 35.0 (30) 0.078 (29) –0.052 (25) Korea, South 24.2 (11) 25.9 (24) 0.530 (2) 0.002 (3) Luxembourg 28.6 (20) 11.7 (3) 0.405 (7) –0.037 (14) Mexico 30.0 (22) 16.5 (12) 0.204 (18) –0.065 (27) New Zealand 30.0 (22) 20.1 (18) 0.267 (14) –0.017 (6) Norway 28.0 (18) 24.5 (22) 0.243 (15) –0.014 (4) Poland 19.0 (4) 14.6 (8) 0.470 (3) 0.017 (1) Portugal 26.5 (16) 19.0 (17) 0.090 (27) –0.026 (11) Slovak Republic 19.0 (4) 11.4 (2) 0.564 (1) –0.047 (18) Spain 30.0 (22) 25.7 (23) 0.292 (12) –0.037 (14) Sweden 26.3 (15) 21.1 (19) 0.168 (23) –0.051 (23) Switzerland 21.1 (9) 17.5 (14) 0.190 (20) –0.019 (7) Turkey 20.0 (8) 15.9 (9) 0.422 (6) –0.047 (18) United Kingdom 28.0 (18) 28.7 (29) 0.186 (21) –0.050 (22) United States 39.2 (29) 16.2 (11) 0.204 (19) –0.026 (11) Average 25.9 19.2 0.268 –0.036

NOTES: Ranks are given in parentheses. For tax rates, a rank of 1 corresponds to the lowest tax rate. For growth rates, a rank of 1 corresponds to the highest growth rate.

SOURCES: Real growth rates for 2000–2008 are from http://stats.oecd.org. The 2009 real growth rate is from the CIA World Factbook 2010 (Washington, DC: GPO). The 2000–2009 real growth rate is calcu-

lated as [(1 + g2000)(1 + g2001)...(1 + g2009)]–1, where gi is the real growth rate in year i. The corporate statutory tax rate is taken from the OECD tax database (www.oecd.org/ctp/taxdatabase). It refers to 2010 and combines central and subcentral tax rates, excludes targeted rates, and uses the top tax rate in graduated tax systems. The marginal effective tax rate due to corporate tax is taken from Chen and Mintz (2008), table 3, column 3, and refers to 2008.

HHassett.indbassett.indb 5959 111/8/121/8/12 9:279:27 PMPM 60 RETHINKING COMPETITIVENESS

base to another or from one time period to another within the same base. When there is income shifting, the fact that the base of a tax goes up when the tax rate falls does not make this tax cut a no-brainer, as the increased revenue is offset by reduced revenue in another base.25 For the most part the Laffer conjecture has been analyzed for the individual income tax, but in recent years there have been some analyses of a corporate tax Laffer curve.26 Several recent studies draw on the experience of OECD countries; as noted earlier, beginning in the early 1980s, statutory corporate tax rates in OECD countries have fallen markedly, while over the same period the ratio of corporate tax revenues to GDP increased in many countries, with the exception of Germany, Japan, and the United Kingdom. After a careful examination of the data, Sørensen (2007) concluded that a major explanation for these trends is that the ratio of corporate profits to total profits increased, reflecting the growing importance of the corporate organizational form for reporting taxable profits. Sørensen concludes that countries “have made up for the drop in statutory tax rates by broadening the corporate tax base, e.g., by eliminating special deductions and moving towards less generous rules for asset depreciation, etc.” (p. 181). Three recent studies directly address the Laffer hypothesis using data from OECD countries.27 Each of these studies begins with a basic regression of corporate tax revenues as a fraction of GDP on the corporate tax rate, the corporate tax rate squared, and year fixed effects to allow for worldwide macroeconomic conditions to affect all countries’ corporate tax system outcomes. These studies point to an implied revenue-maximizing tax rate in the mid-20s to mid-30s, suggesting that the United States might be close to, or over, the revenue-maximizing corporate tax rate.28 However, these estimates are quite sensitive to the set of included controls in the regressions. Each of these studies attempts to account for various aspects of the behavioral response to changes in corporate tax rates. Claus- ing (2007) finds evidence of shifting income between the personal and corporate tax bases, which invalidates the argument that it is never optimal to levy a rate above the corporate-tax-revenue-maximizing rate. Brill and Has- sett (2007) focus on the timing of tax responses and find that the implied revenue-maximizing tax rate has been declining over time. It is critical to note that all three studies omit country-specific fixed effects, despite using

HHassett.indbassett.indb 6060 111/8/121/8/12 9:279:27 PMPM COMPETITIVE TAX POLICY 61

data with a panel structure. This means that these studies exhibit the critical flaw discussed earlier: if there is unobserved heterogeneity across countries (and there surely is) that is related to both the dependent and the explanatory variables, the estimates will suffer from omitted variable bias. Indeed, Gravelle and Hungerford (2007) show that when country-specific fixed effects are added so that the estimates depend only on within-country variations in tax rates and revenues over time within the countries, corporate tax rates are no longer significant predictors of corporate tax revenues, suggesting that unobserved heterogeneity indeed imparts bias to analyses based on cross-sectional data, generating a Laffer-type relationship when none can in fact be confidently established. Omitted variable bias will also occur when relevant aspects of the corporate tax system are not included as explanatory variables. Changes to countries’ statutory corporate tax rate are often accompanied by changes to the corporate tax base, and both of these changes affect corporate tax revenues.29 Without accounting for tax base changes, the joint effect of changing rates and bases is attributed to tax rates alone, which leads to biased estimates of the effect of corporate tax rates on corporate tax revenues. Clausing (2007) and Devereux (2006) include proxies for the tax base as explanatory variables, although their controls are not sufficient measures of all tax base variations.30 Finally, this literature has also not yet carefully considered the likely timing of how changes would affect revenue by influencing the tax base. If lower rates reduce outward income shifting (and perhaps even induce inward income shifting), this would affect the tax base immediately. In contrast, a lower tax rate, by lowering the effective tax disincentive, increases business investment, and would likely reduce the tax base in the short run, because for most investments, costs exceed revenues at first. This is especially true in a regime of accelerated depreciation, where much of the cost of the investment is a tax deduction soon after the investment is made. In summary, the large empirical literature trying to link aspects of the corporate tax system to measures of prosperity has not yielded definitive results, and a recent published study finds no impact of the corporate statutory tax rate on growth among developed countries. Recent attempts to establish a more limited conclusion, how tax rates affect corporate tax revenues, suffer from methodological flaws that prevent drawing clear

HHassett.indbassett.indb 6161 111/8/121/8/12 9:279:27 PMPM 62 RETHINKING COMPETITIVENESS

policy conclusions. For the time being, however, our policy analysis must be mostly based on economic reasoning.

Summary “Competitiveness” is a buzzword that, reasonably interpreted, characterizes policies that support prosperity and growth, fairly shared, in a global economy. Once understood in this way, we can consider what constitutes a competitive education policy, a competitive energy policy, a competitive regulation policy, a competitive research policy, a competitive infrastructure policy, a competitive fiscal policy, a competitive monetary policy, and so on. We should certainly consider what constitutes a competitive tax policy—and in particular a competitive business tax policy—while keeping in mind how tax policy constrains or facilitates the achievement of other aspects of an overall competitive set of national policies, such as the role of budget balance in establishing favorable macroeconomic conditions. The current U.S. system of taxing corporations distorts economic behavior on a number of margins and thus reduces national income, a reasonable but not perfect measure of prosperity. Lowering the statutory corporate tax rate without structural reform reduces the magnitude of some of these distortions, but it introduces other problems, such as the erosion of taxation of labor income. The tax-related disincentive to invest depends not only on the statutory rate but also on the definition of the tax base, most notably the treatment of depreciation allowances. Thus, lowering the tax rate while broadening the base may not encourage investment. Reduc- ing the corporate tax rate also reduces the investment “bang per buck” of revenue loss because some of the tax reduction applies to the income from capital already in place, in contrast to accelerated depreciation. On the other hand, altering the definition of the business tax base can convert an income tax to a consumption tax that causes no disincentive to investment at all, while levying a tax on pure profits and preserving the crucial role of business taxes in tax collection. The fact that the U.S. economy operates within a global economy exac- erbates the cost of a distorting corporation tax system. It also suggests that attempting to tax the income earned in the United States by internationally mobile capital could be counterproductive because it repels capital.

HHassett.indbassett.indb 6262 111/8/121/8/12 9:279:27 PMPM COMPETITIVE TAX POLICY 63

This effect can be minimized either by lowering the corporate tax rate or by moving to a consumption or cash-flow, rather than income, base. One advantage of a lower rate is that it reduces the incentive of multinational corporations to shift taxable income away from the United States to lower- tax-rate countries, including tax havens. Moving to a territorial system, while addressing some capricious penalties to the foreign operations of U.S. multinational companies, would significantly exacerbate outward income shifting unless the statutory tax rate was much lower, and therefore it would have to be accompanied by tax code and enforcement measures that ensure a defense of U.S. revenues.

Notes 1. I am grateful for research assistance from Allison Paciorka and Zachary Rable and for helpful comments from Alan Auerbach, Michael Devereux, Kevin Hassett, and Drew Lyon. 2. Indeed, Paul Krugman called competitiveness a “dangerous obsession” in an influential 1994 article in Foreign Affairs. 3. These are drawn from the recommendations associated with the “Washington consensus” articulated first in the late 1980s by John Williamson (1990). 4. Overall, the rank-order correlation between the GCI measure of overall competitiveness and its measure of the extent and effect of taxation (where rank one goes to the lowest-tax country) is 0.28, but falls to 0.21 if seven Middle Eastern oil countries are dropped from the sample. For the total tax rate, the rank-order correlation is 0.10 and 0.02 for all countries and without the oil countries, respectively. For reasons discussed below, these correlations are not informative about the policy questions of interest. 5. The United States ranks 118th, 122nd, and 11th, respectively, on these three indicators. 6. Nor would it be possible to develop a consensus measure of fairness, although value-free cross-country measures of the inequality in the distribution of income and wealth are available. 7. Accelerated depreciation is of less value for short-lived assets, such as computers, that do not have long useful lives. On the other hand, bonus depreciation as recently enacted in the United States explicitly does not apply to very long-lived assets. Permanent expensing (100 percent bonus depreciation, in current parlance) would be neutral across all assets if applied universally. 8. Some economists have argued that, due to asymmetric information between firms and potential providers of outside financing, a firm’s internal funds are a determinant of investment. Hubbard (1998) reviewed the evidence for this hypothesis, but

HHassett.indbassett.indb 6363 111/8/121/8/12 9:279:27 PMPM 64 RETHINKING COMPETITIVENESS

Erickson and Whited (2000) and subsequent literature have rejected the importance of cash flow in investment as an artifact of measurement error. 9. Shaw, Slemrod, and Whiting (2010) calculate a similar percentage for taxes in the United Kingdom. 10. Slemrod (2008) discusses how administration and compliance cost may alter this invariance property. 11. Of course this argument applies to companies owned by domestic residents, not to foreign-owned companies; however, levying no corporation income tax on the latter would invite domestic individuals to establish corporations that are nominally foreign owned but are really controlled by domestic taxpayers. 12. It also depends on the extent of investment tax credits and other factors. 13. The impact on old capital depends critically on the transition rules that apply to the existing capital. If rules were adopted so that, looking forward, old capital would not be disadvantaged relative to new capital, no capital losses would ensue. This, though, would be a large drain on the revenues collected and would require higher tax rates to keep a reform revenue neutral. 14. Neubig (2006) argues that because public corporations consider the impact of tax legislation on their earnings as reported on financial statements over and above the impact on their “true” after-tax income, they tend to favor corporate rate cuts over investment-incentive-equivalent accelerated depreciation. This is because in the short run, the former will generate bigger increases in after-tax earnings. 15. To be sure, as Sørensen (2007) discusses, there are caveats and exceptions to this theoretical result. For example, if there are “location-specific rents,” that is, extra- normal returns to investment due to some unique aspect of a country, such as natural resources or profit-augmenting infrastructure, revenue can be collected from judiciously constructed tax systems without repelling capital investment. 16. Note that this connection between income shifting and real activity also makes the statutory rate relatively more important in attracting real investment than otherwise. 17. For complex reasons beyond the scope of this chapter, many experts believe that moving to a territorial system with the same rate structure would not actually cost the United States much, if any, revenue. See U.S. Congress (CBO) (2009, p. 245). 18. Indeed, this analogy, posed in Slemrod (1990), raises another issue in the debate about worldwide versus territorial systems. Just as in response to a first-letter- based tax system, one would expect multinational companies to rename themselves, companies with substantial income in low-tax countries might change the country in which their parent company resides for tax purposes, often where it is headquartered, to a country with a territorial system. Why, holding constant where production and jobs are, one should care about where a company is headquartered is controversial. 19. The calculation of the statutory rate on corporate income is not completely straightforward. One has to deal with, for example, subnational corporate levies and regional subsidy programs.

HHassett.indbassett.indb 6464 111/8/121/8/12 9:279:27 PMPM COMPETITIVE TAX POLICY 65

20. The dispersion of corporate tax rates has also been falling during this period. 21. This evidence is analyzed in Slemrod (2004). 22. Among the most important are the following: (1) debt ratios and real interest rates are the same across countries; (2) intangible investments are not considered; (3) property taxes are not considered; (4) tax holidays are not considered; (5) international tax provisions are not considered; and (6) it is assumed that favorable tax provisions are fully used on the grounds that companies are able to shelter income from tax. 23. Other studies come to a different conclusion. For example, in an unpublished study, Arnold (2008) finds that the share of corporate taxes in total taxes is negatively associated with growth for twenty-one OECD countries from 1971 to 2004; no effort is made to reconcile these results with those of Lee and Gordon (2005). 24. The correlation is less strong between the marginal effective tax rate, which measures the tax-related disincentive to investment, and real growth rates. 25. This argument is fleshed out in Saez, Slemrod, and Giertz (2012). 26. Gruber and Rauh (2007) investigate this issue using aggregate U.S. time- series data. 27. Devereux (2006) uses data on twenty OECD countries from 1986 to 2004; Claus- ing (2007) uses data on twenty-nine OECD countries from 1979 to 2002; and Brill and Hassett (2007) use data on twenty-nine OECD countries between 1980 and 2005. 28. Note, however, that in his preferred specification, Devereux’s (2006) estimate of the tax effect is no longer statistically significant. 29. Robinson and Slemrod (2012) investigate this issue in the context of individual income taxation. 30. Clausing (2007) includes the corporate share of GDP, corporate profitability, and the system for taxing worldwide income (i.e., whether a country has a territorial system, a worldwide-with-credit system, or a mixture) to account for the tax base. Devereux (2006) uses the net present value of depreciation deductions on industrial buildings. Brill and Hassett (2007) include no other control variables.

References Arnold, Jens. 2008. Do Tax Structures Affect Aggregate Economic Growth? Empiri- cal Evidence from a Panel of OECD Countries. Working Paper No. 643. OECD Economics Department, Paris. Bird, Richard. 2002. “Why Tax Corporations?” Bulletin of International Fiscal Documen- tation 56(5):194–203. Brill, Alex, and Kevin Hassett. 2007. “Revenue Maximizing Corporate Tax Rates.” Working Paper 137. American Enterprise Institute, Washington, D.C., July 31. Chen, Duanjie, and Jack Mintz. 2008. Taxing Business Investments: A New Ranking of Effective Tax Rates on Capital. Available at http://econ.ucalgary.ca/sites/econ.ucalgary.ca/files/seminars/Mintz.pdf.

HHassett.indbassett.indb 6565 111/8/121/8/12 9:279:27 PMPM 66 RETHINKING COMPETITIVENESS

———. 2010. “U.S. Effective Corporate Tax Rate on New Investments: Highest in the OECD.” Wall Street Journal, May 14. Christensen, Kevin, Robert Cline, and Thomas S. Neubig. 2001. “Total Corporate Taxation: ‘Hidden,’ Above-the-Line, Non-Income Taxes.” National Tax Journal 54(3):495–506. Clausing, Kim. 2007. “Corporate Tax Revenues in OECD Countries.” International Tax and Public Finance 14(2):115–33. Desai, Mihir A., Fritz Foley, and James R. Hines Jr. 2009. “Domestic Effects of the For- eign Activities of U.S. Multinationals.” American Economic Journal: Economic Policy 1(1):181–203. Devereux, Michael M. P. 2006. Developments in the Taxation of Corporate Profit in the OECD since 1965: Rates, Bases, and Revenues. Mimeo. Oxford University Centre for Business Taxation, Oxford, UK, May. Erickson, Timothy, and Toni Whited. 2000. “Measurement Error and the Relation- ship between Investment and q.” Journal of Political Economy 108(5):1027–57. Gravelle, Jane G., and Thomas L. Hungerford. 2007. Corporate Tax Reform: Issues for Congress. CRS Report for Congress. Congressional Research Service, Washington, D.C. October 31. Gruber, Jonathan, and Joshua Rauh. 2007. “How Elastic Is the Corporate Income Tax Base?” In Taxing Corporate Income in the 21st Century, ed. Alan Auerbach, James R. Hines Jr., and Joel Slemrod. Cambridge: Cambridge University Press, 140–63. Hubbard, R. Glenn. 1998. “Capital-Market Imperfections and Investment.” Journal of Economic Literature 36(1):193–225. Krugman, Paul. 1994. “Competitiveness: A Dangerous Obsession.” Foreign Affairs 73(2):28–44. Lee, Young, and Roger H. Gordon. 2005. “Tax Structure and Economic Growth.” Journal of Public Economics 89(5/6):1027–43. National Commission on Fiscal Responsibility and Reform. 2010. The Moment of Truth. Washington, D.C. Neubig, Tom. 2006. “Where’s the Applause? Why Most Corporations Prefer a Lower Rate.” Tax Notes, April 24. Organisation for Economic Co-operation and Development. 1998. Harmful Tax Com- petition: An Emerging Global Issue. Paris, France. Robinson, Leslie, and Joel Slemrod. 2012. “Understanding Multidimensional Tax Sys- tems.” International Tax and Public Finance 19(2):237–67. Saez, Emmanuel, Joel Slemrod, and Seth Giertz. 2012. “The Elasticity of Taxable Income with Respect to Marginal Tax Rates: A Critical Review.” Journal of Economic Literature 50(1):3–50. Shaw, Jonathan, Joel Slemrod, and John Whiting. 2010. “Administration and Com- pliance.” In Dimensions of Tax Design: The Mirrlees Review, ed. S. Adam, T. Besley, R. Blundell, S. Bond, R. Chote, M. Gammie, P. Johnson, G. Myles, and J. Poterba. Oxford: Oxford University Press, 1100–1162.

HHassett.indbassett.indb 6666 111/8/121/8/12 9:279:27 PMPM COMPETITIVE TAX POLICY 67

Slemrod, Joel. 1990. “Tax Havens, Tax Bargains and Tax Addresses: The Effect of Taxation on the Spatial Allocation of Capital.” In Reforming Capital Income Taxation, ed. H. Siebert. Tubingen, Germany: Mohr, 23–42. ———. 1995. What Do Cross-Country Studies Teach about Government Involvement, Prosperity, and Growth? Brookings Papers on Economic Activity 2373–431. ———. 2004. “Are Corporate Tax Rates, or Countries, Converging?” Journal of Public Economics 88(6):1169–86. ———. 2008. “Does It Matter Who Writes the Check to the Government? The Eco- nomics of Tax Remittance.” National Tax Journal 61(2):251–75. Sørensen, Peter Birch. 2007. “Can Capital Income Taxes Survive? And Should They?” CESifo Economic Studies 53(2):172–228. U.S. Congress. Congressional Budget Office. 2009. Budget Options, Volume 2. Wash- ington, D.C.: U.S. Government Printing Office. U.S. Department of Treasury. Office of Tax Policy. 2007. Approaches to Improve the Competitiveness of the U.S. Business Tax System for the 21st Century. Washington, D.C.: U.S. Government Printing Office. December 20. Williamson, John. 1990. “What Washington Means by Policy Reform.” In Latin Amer- ican Readjustment: How Much Has Happened, ed. J. Williamson. Washington, D.C.: Institute for International Economics. World Economic Forum. 2010. The Global Competitiveness Report, 2010–2011. Geneva.

HHassett.indbassett.indb 6767 111/8/121/8/12 9:279:27 PMPM 3

Education and Global Competitiveness: Lessons for the United States from International Evidence Martin West

The 2009 Program for International Student Assessment (PISA), the latest and most comprehensive international test of student achievement, placed the relative standing of the U.S. education system in stark relief. American students’ middling performance on international tests is by now familiar, and PISA 2009 was no exception: our fifteen-year-olds ranked fourteenth out of the thirty-four developed democracies that are members of the Organisation for Economic Co-operation and Development (OECD) in reading, seventeenth in science, and no better than twenty-fifth in math. The new wrinkle in the 2009 study was the participation of China’s Shanghai province, which topped the international league table in all three subjects, besting American students by the equivalent of multiple grade levels in each category. The results from Shanghai, home to the nation’s wealthiest city and a magnet for its most ambitious and talented citizens, are hardly representative of China as a whole. Yet its students’ eye-popping performance seemed to highlight new challenges facing the U.S. economy in an age of unprecedented global trade. U.S. Secretary of Education Arne Duncan deemed the results an “absolute wake-up call” and urged the nation “to get much more serious about investing in education” (Armario 2010). Even before the December 2010 publication of the PISA data, the notion that educational competition threatens America’s future prosperity had been a recurrent theme of the Obama administration’s pronouncements on education policy. In his first State of the Union address, the president

HHassett.indbassett.indb 6868 111/8/121/8/12 9:279:27 PMPM EDUCATION AND GLOBAL COMPETITIVENESS 69

warned that “the countries that out-teach us today will out-compete us tomorrow” (White House 2009). He has repeatedly called attention to America’s declining rank in the share of students completing college degrees and established the goal of reclaiming America’s position as first in the world by this metric by 2020 (U.S. Department of Education 2010). In various speeches, President Obama and Secretary Duncan have emphasized the link between education and national economic competitiveness in making the case for the education funding allocated through the American Recovery and Reinvestment Act, for the state policy changes incentivized by the Race to the Top grant competition, and for increased federal support of early childhood education. More recently, in response to Republican proposals to rein in federal spending, the president has warned that “we cannot cut education. We can’t cut the things that will make America more competitive” (White House 2011). Rhetoric linking education and economic competitiveness is hardly new. Concerns about global competition have long served to motivate federal investment in education and broader efforts to enhance the performance of American schools. A Nation at Risk, the 1983 report that captivated the public and informed a generation of school reform, famously warned that “our once unchallenged preeminence in commerce, industry, science, and technological innovation is being overtaken by competitors throughout the world. . . . What was unimaginable a generation ago has begun to occur— others are matching and surpassing our educational attainments” (National Commission on Excellence in Education 1983). A generation earlier, the Soviets’ launch of Sputnik sparked widespread angst about the quality of American math and science education and the passage of the National Defense Education Act. As far back as 1914, the Commission on National Aid to Vocational Education warned that “the battles of the future between nations will be fought in the markets of the world. That nation will triumph . . . which is able to put the greatest amount of skill and brains into what it produces” (quoted in Grubb and Lazerson 2004, p. 11). The Smith-Hughes Act, which provided the first federal funding for elementary and secondary education programs, was enacted three years later. As Secretary Duncan himself noted in 2010, however, “The relationship between education and international competitiveness is a subject rife with myth and misunderstanding” (Duncan 2010, p. 65). This confusion

HHassett.indbassett.indb 6969 111/8/121/8/12 9:279:27 PMPM 70 RETHINKING COMPETITIVENESS

may stem from the fact that the concept of international competitiveness is notoriously difficult to pin down. Academic economists have long criticized the view that countries in a globalized economy are engaged in a zero-sum game in which only some can emerge as winners and others will inevitably lose out. As Paul Krugman (1996) pointed out in 1994, countries, unlike corporations, “do not go out of business” (p. 6). Moreover, all countries can in theory benefit from international trade by specializing in those activities in which they have a comparative advantage. In what sense, then, does it make sense to talk about national economies competing? These general lessons seem doubly true in the case of education, where the mechanisms by which gains abroad would undermine Americans’ prosperity are altogether unclear. Educational improvements in other countries enhance the productivity of their workforces, which in turn should reduce the costs of imports to the United States to the benefit of all Americans who do not compete directly in producing the same goods. At the top end of the education spectrum, growth in the number of graduate degrees awarded in fields such as science and engineering fosters technological advances from which Americans can benefit regardless of whether the key discoveries were made in the United States. Among developing countries, the ongoing expansion of educational opportunities for women in particular promises to reduce poverty, violence, and political instability. For all these reasons, developments such as Shanghai’s performance on the PISA, although at first glance startling, may in fact represent good news (Duncan 2010). This is not to say, however, that America’s very real educational challenges are irrelevant to its economic performance going forward. On the contrary, the evidence that the quality of a nation’s education system is a key determinant of the future growth of its economy is increasingly strong (Hanushek and Woessmann 2011). If, following Xavier Sala-i-Martin (2010), the economist behind the World Economic Forum’s influential Global Competitiveness Index, competitiveness is redefined as “the set of institutions, policies, and factors that set the sustainable current and medium-term levels of economic prosperity” (p. 1), then the performance of national education systems should be central to its measurement. To the extent that America’s education system stagnates as other countries advance, its relative standing in the world economy and global influence will no doubt suffer.

HHassett.indbassett.indb 7070 111/8/121/8/12 9:279:27 PMPM EDUCATION AND GLOBAL COMPETITIVENESS 71

Assertions of education’s importance for global competitiveness, however misleading, can therefore be useful to the extent that they illustrate the potential for the American education system to perform at much higher levels and foster a sense of urgency about its improvement. Most Americans continue to assign high marks to their own local schools, even as they evaluate the nation’s schools more critically than ever (Bushaw and Lopez 2011). This pattern reflects and works to reinforce a reform agenda centered on increasing the performance of economically disadvantaged students, racial and ethnic minorities, and those at the bottom of the skill distribution. While this task is and must remain an urgent priority, defining the nation’s educational challenge solely in terms of domestic achievement gaps ignores the gains that would result from improved performance and productivity across the American education system as a whole. Greater recognition of the “global achievement gap” may help to alter the perceptions of policymakers and the public and broaden support for politically controversial reforms (Wagner 2008). Discussions of educational competition can also be useful to the extent that they foster systematic investigation of the factors that lead to higher and more equitable levels of performance across countries. Identifying the determinants of the performance of national education systems is challenging, and naïve attempts at “global benchmarking” based on case studies of high-performing countries can yield misleading lessons. However, careful research exploiting the differences in how public education is governed and organized across countries can shed light on which strategies are most promising in the American context. As education researcher Arthur W. Foshay put it in 1962 after conducting the first cross-country study of student achievement, “If custom and law define what is educationally allowable within a nation, the educational systems beyond one’s national borders suggest what is educationally possible” (quoted in Hanushek and Woessmann 2011, p. 90). The remainder of this chapter surveys the emerging scholarly literature on the causes and consequences of differences in the performance of national education systems. I first use international studies of student achievement and attainment to shed light on the scope for improvement in the United States. I then review recent evidence on the link between education and economic growth to highlight the potential economic gains

HHassett.indbassett.indb 7171 111/8/121/8/12 9:279:27 PMPM 72 RETHINKING COMPETITIVENESS

from performance improvements. Finally, I discuss lessons for American policymakers from research examining the policies associated with strong educational performance across countries.

Ranking the U.S. Education System: Quality and Quantity Indicators National education systems can be compared in terms of both the quantity of education citizens receive and its quality. Studies of the former typically focus on the average years of schooling students complete or on the share of students receiving high school or postsecondary degrees. While these indicators are often easy to obtain, they implicitly assume that an additional year of schooling or a given educational credential means the same thing from one country to the next. Quality measures, which are based on common assessments of the academic achievement of a nation’s students at ages when most or all students remain enrolled in school, do not suffer from this drawback. As discussed below, quality measures also appear to be the more reliable barometer of the degree to which the nation is fostering conditions for sustained economic progress.

Student Achievement in Math and Science. Viewed through a quality lens, the evidence is clear that the American education system ranks in the middle of the pack, at best, among industrialized countries when it comes to math and science. The United States regularly participates in two major international testing programs in math and science, the PISA and the longer-running Trends in Mathematics and Science Study (TIMSS), providing multiple perspectives on how the performance of its students compares. TIMSS and PISA use quite different approaches to develop their assessments of student achievement: While TIMSS ensures that the content of its tests is closely aligned to the math and science curricula of participating countries, PISA evaluates students against its own definitions of literacy in math, science, and reading and emphasizes items designed to measure students’ ability to apply knowledge of these subjects in real-world settings. Despite this difference in testing philosophy, however, the scores of countries that participate in both assessments are highly correlated.1 The fact that both assessments paint a similar picture suggests that conclusions concerning

HHassett.indbassett.indb 7272 111/8/121/8/12 9:279:27 PMPM EDUCATION AND GLOBAL COMPETITIVENESS 73

the U.S. performance are not driven by the idiosyncrasies of a particular testing regime. Launched in 2000 as a project of the OECD, the PISA is administered every three years to nationally representative samples of students in each OECD country and a growing number of partner countries (and subnational units such as Shanghai). The seventy-four education systems that participated in the 2009 PISA study represented more than 85 percent of the global economy and included virtually all of America’s major trading partners, making it a particularly useful source of information on our students’ relative standing. Another advantage of the PISA study is its focus on students nearing the completion of secondary schooling. Figure 3-1 displays the performance of the thirty top-performing OECD countries in math and science on the 2009 PISA study.2 Although Chile, Israel, Mexico, and Turkey have been excluded to enhance legibility, these countries do contribute to the average scores plotted for the OECD as a whole. As the figure indicates, U.S. students perform well below the OECD average in math and essentially match the OECD average in science. In math, the United States trails seventeen OECD countries by a statistically significant margin, its performance is indistinguishable from that of eleven countries, and it significantly outperforms only five countries (including the four excluded from the figure). In science, the United States significantly trails twelve countries and outperforms just nine. Countries scoring at similar levels to the United States in both subjects include Austria, the Czech Republic, Hungary, Ireland, Poland, Portugal, and Sweden. The gap in average math and science achievement between the United States and the top-performing national school systems is dramatic. PISA established the scales it uses to evaluate achievement in mathematics in 2003 and in science in 2006 to have a mean of 500 and a (student-level) standard deviation of 100 across the OECD countries. The gap between the United States and the highest-performing OECD country therefore amounts to three-fifths of a standard deviation in math and just over half of a standard deviation in science. By way of comparison, the achievement gap between African American and white high school students within the United States is roughly four-fifths of a standard deviation. The gap between the United States and other countries can also be compared to the amount that students typically learn in one year of schooling, as

HHassett.indbassett.indb 7373 111/8/121/8/12 9:279:27 PMPM 74 RETHINKING COMPETITIVENESS

FIGURE 3-1 MATH AND SCIENCE ACHIEVEMENT IN OECD COUNTRIES (PISA 2009)

OECD Average KOR 540 SWI JPN CAN NLD

520 NZL BEL AUS GER EST ISL DEN SVN 500 AUT FRA SVK NOR POL CZE SWE UK OECD Average LUX HUN PRT IRL Average Math Score SPN ITA USA 480

GRC 460 460 480 500 520 540 560 Average Science Score

NOTES: Countries with dark solid (hollow) markers outperform the United States in two (one) subjects by a statistically significant margin; countries with medium solid (hollow) markers significantly trail the United States in two (one) subjects; countries with light markers are statistically indistinguishable from the United States in both subjects. Slovakia outperformed the United States in math but trailed it in science. The OECD average is the average of the national averages from the OECD member countries, with each country weighted equally. Four OECD countries are excluded from the figure (but included in the calculation of OECD average). Their math (science) scores are as follows: Israel: 447 (455); Turkey: 445 (454); Chile: 421 (447); and Mexico: 419 (416).

SOURCE: Fleischman et al. (2010).

estimated on the basis of age- and demographic-adjusted differences in the performance of students in different grades taking the assessment. In math the average American student by age fifteen is at least a full year behind the average student in six countries, including Canada, Japan, and the Netherlands. Students in six additional countries, including Australia, Belgium, Estonia, and Germany, outperform American students by more than half a year.3

HHassett.indbassett.indb 7474 111/8/121/8/12 9:279:27 PMPM EDUCATION AND GLOBAL COMPETITIVENESS 75

FIGURE 3-2 MATH ACHIEVEMENT VS. EDUCATIONAL EXPENDITURE ACROSS OECD COUNTRIES, 2009

550 KOR FIN JPN SWI CAN

NZL NLD EST GER AUS BEL ISL DEN NOR 500 SVK POL FRA SVN AUT CZE IRL SWE OECD Average PRT UK USA HUN SPN ITA

GRC

Average Math Score 450 TUR ISR

CHL

MEX

400 OECD Average 20,000 40,000 60,000 80,000 100,000

Cumulative Expenditure per Student (USD converted using PPPs)

NOTES: Countries with dark markers outperform the United States by a statistically significant margin; countries with medium markers significantly trail the United States; countries with light markers are statistically indistinguishable from the United States. The OECD averages are averages of the national averages from the OECD member countries, with each country weighted equally. Luxembourg is excluded from the figure (but included in the calculation of OECD averages); it had an average math score of 489 and cumulative expenditure per student of $155,624.

SOURCE: Organisation for Economic Co-operation and Development (2010a, figure I.3.9).

The second-rate math and science performance of American students is particularly striking, given the level of resources the nation devotes to elementary and secondary education. Data on cumulative expenditures per student in public and private schools between ages six and fifteen confirm that the United States spends more than any other OECD country save Lux- embourg (a wealthy city-state that spends nearly 1.5 times as much as the United States). Figures 3-2 and 3-3, which plot average math and science

HHassett.indbassett.indb 7575 111/8/121/8/12 9:279:27 PMPM 76 RETHINKING COMPETITIVENESS

FIGURE 3-3 SCIENCE ACHIEVEMENT VS. EDUCATIONAL EXPENDITURE ACROSS OECD COUNTRIES, 2009

550 FIN

KOR JPN NZL CAN EST AUS NLD GER SWI POL SVN UK OECD Average HUN IRL BEL USA 500 DEN FRA ISL NOR CZE PRT SWE SVK ITA AUT SPN

GRC

450 TUR ISR

Average Science Score CHL

MEX

400 OECD Average 20,000 40,000 60,000 80,000 100,000

Cumulative Expenditure per Student (USD converted using PPPs)

NOTES: Countries with dark markers outperform the United States by a statistically significant margin; countries with medium markers significantly trail the United States; countries with light markers are statistically indistinguishable from the United States. The OECD averages are averages of the national averages from the OECD member countries, with each country weighted equally. Luxembourg is excluded from the figure (but included in the calculation of OECD averages); it had an average science score of 484 and cumulative educational expenditure per student of $155,624.

SOURCE: Organisation for Economic Co-operation and Development (2010a, figure I.3.9).

achievement in 2009 against this spending measure across the OECD, reveal that the relationship between the level of educational investment and student achievement is very weak overall, and altogether nonexistent among those countries spending at least $30,000 per student.4 This pattern suggests that national education systems vary widely in their spending productivity—that is, the achievement outcomes they produce for each dollar spent. This variation does not favor the United States. Most of the countries

HHassett.indbassett.indb 7676 111/8/121/8/12 9:279:27 PMPM EDUCATION AND GLOBAL COMPETITIVENESS 77

that outperform the United States in each subject spend between $60,000 and $80,000 per student, as compared with the nearly $105,000 spent by the United States. The Czech Republic, Hungary, and Poland, all of which perform at essentially the same level as the United States in both subjects, spend less than half as much. These three countries achieve comparable results while spending far less, despite having far lower levels of GDP per capita, fewer than half as many college-educated adults, and, in the case of Hungary and Poland, twice as many students from socioeconomically disadvantaged backgrounds. Some analysts have speculated that, despite the modest performance of its average students, the U.S. education system is characterized by pockets of excellence that can be expected to meet the needs of the knowledge economy. As discussed below, however, there is no clear evidence that educating a subset of students to very high levels is more important than raising average achievement levels for national economic success. Moreover, the United States in fact fares no better in comparisons of the share of students performing at exceptionally high levels. For example, only 9.9 percent of American students taking the 2009 PISA math test achieved at level 5 or 6 (the two top performance categories), a level of accomplishment that PISA claims indicate that students are “capable of complex mathematical tasks requiring broad, well-developed thinking and reasoning skills” (Organisa- tion for Economic Co-operation and Development 2010b, p. 18). As in the case of average math performance, twenty-four of the thirty-four OECD countries outranked the United States by this metric. The share of students achieving level 5 or 6 exceeded 15 percent in ten OECD countries and exceeded 20 percent in another five. In Shanghai, China, fully 50.4 percent of students surpassed this benchmark—more than five times the level in the United States (OECD 2010a). Another common response to our disappointing performance on international assessments has been to emphasize the relative diversity of American students and the wide variation in their socioeconomic status (see, for example, Ladd [2012]). Family background characteristics and other out-of-school factors clearly have a profound influence on students’ academic achievement. The available international assessments, all of which offer only a snapshot of how students have learned at a single point in time rather than evidence on how much progress they are making from one year

HHassett.indbassett.indb 7777 111/8/121/8/12 9:279:27 PMPM 78 RETHINKING COMPETITIVENESS

to the next, are therefore best viewed as measuring the combined effects of differences in school quality and differences in these contextual factors. The latter are poorly measured across countries, making it difficult to pin down their relative import. Even so, it is difficult to attribute the relative standing of U.S. students among OECD countries to out-of-school factors alone. The share of American students with college-educated parents, a key predictor of school success, actually ranks eighth among the thirty-four OECD countries. The typical American student is also well above the OECD average according to PISA’s preferred measure of students’ socioeconomic status (OECD 2011). Cross-country studies based on the PISA and other international assessments do indicate that the United States falls within the top quarter of developed countries in terms of the degree to which various measures of family background predict student achievement (see, for example, Schuetz, Ursprung, and Woessmann 2008; Woessmann 2004). This pattern is consistent with well-known evidence showing large gaps in academic achievement along lines of race, ethnicity, and family income and suggests that educational opportunity is less equitably distributed in the United States than in many other countries. Aggregate differences in family background, however, are unlikely to account for the performance gap between Ameri- can students as a whole and those in top-performing countries.

Educational Attainment. The United States, however, has never fared well in international comparisons of student achievement. The United States ranked eleventh out of twelve countries participating in the first major international study of student achievement, conducted in 1964, and its math and science scores on the 2009 PISA actually reflect modest improvements over the test’s previous administration in 2006.5 America’s traditional reputation as the world’s educational leader stems instead from the fact that the spread of mass secondary education proceeded far earlier here than in most other nations. In the first half of the twentieth century, demand for secondary schooling in the United States surged as technological changes increased the wages available to workers who could follow written instructions, decipher blueprints, and perform basic calculations. America’s highly decentralized school system, in which local communities could vote independently to

HHassett.indbassett.indb 7878 111/8/121/8/12 9:279:27 PMPM EDUCATION AND GLOBAL COMPETITIVENESS 79

support the creation of a high school, provided a uniquely favorable mechanism to drive increased public investment in schooling. As economic his- torian Claudia Goldin (2001) has documented, by 1955 almost 80 percent of American fifteen- to nineteen-year-olds were enrolled full-time in general secondary schooling, more than double the share in any European country. Goldin and Lawrence Katz (2008) provide compelling evidence that this rapid accumulation of human capital played a critical role in America’s twentieth-century economic success. America’s historical advantage in terms of educational attainment has long since eroded. High school graduation rates in the United States peaked in 1970 at roughly 80 percent and have declined slightly since—a trend often masked in official statistics by the growing number of students receiving alternative credentials, such the General Educational Development (GED) certificate (Heckman and LaFontaine 2010). Although the share of students enrolling in college has continued to climb, the share completing a college degree has hardly budged. As this pattern suggests, both the time students are taking to complete college degrees and the dropout rates among students enrolling in college have increased sharply. The apparent stagnation in the accumulation of human capital in the United States in recent decades is especially puzzling in light of the fact that the economic returns to completing a postsecondary degree—and the economic costs of dropping out of high school—have grown substantially over the same period (Heckman and Krueger 2003). Meanwhile, other developed countries have continued to see steady increases in educational attainment and, in many cases, now have postsecondary completion rates that exceed those in the United States. America’s high school graduation rate now trails the average for European Union countries and ranks no better than eighteenth among the twenty-six OECD countries for which comparable data are available (OECD 2010c, table A2.1). Figure 3-4 displays the share of adults of various ages who have completed postsecondary degrees in the United States, OECD countries as a whole, and those OECD countries for whom the share of degree holders among those aged twenty-five to thirty-four exceeds 40 percent. On average across the OECD, postsecondary completion rates have increased steadily from one age cohort to the next: Although only 20 percent of those aged fifty-five to sixty-four have a postsecondary degree, the share among those

HHassett.indbassett.indb 7979 111/8/121/8/12 9:279:27 PMPM 80 RETHINKING COMPETITIVENESS

FIGURE 3-4 SHARE OF ADULTS IN 2008 WITH POSTSECONDARY DEGREE IN SELECT OECD COUNTRIES, BY AGE RANGE

30 Percent

Japan Korea Canada France Ireland Norway Sweden Australia Belgium Denmark United States New Zealand OECD Average 25–34 35–44 45–54 55–64

NOTES: The OECD average is for the thirty-one countries with available data. Countries included are those with completion rates of 40 percent or higher among those aged twenty-five to thirty-four.

SOURCE: Organisation for Economic Co-operation and Development (2010c, table A1.3a).

aged twenty-five to thirty-four is up to 35 percent. The postsecondary completion rate of Americans aged twenty-five to thirty-four remains above the OECD average at 42 percent, but this reflects a decline of 1 percentage point relative to those aged thirty-five to forty-four and is only marginally higher than the rate registered by older cohorts. The comparison between the United States and Canada is particularly striking. Although the two nations have virtually identical postsecondary completion rates among those aged fifty-five to sixty-four, Canada’s rate among the youngest cohort now exceeds America’s by one-third.

HHassett.indbassett.indb 8080 111/8/121/8/12 9:279:27 PMPM EDUCATION AND GLOBAL COMPETITIVENESS 81

To be sure, in many respects the American higher education system remains the envy of the world. Despite recent concerns about rapidly increasing costs, declining degree-completion rates, and the quality of instruction available to undergraduate students, American universities continue to dominate world rankings of research productivity. The 2011 Academic Rankings of World Universities (ARWU), an annual publication of the Shanghai Jiao Tong University, placed 8 American universities within the global top 10, 17 within the top 20, and 151 within the top 500.6 A 2008 Rand study commissioned by the U.S. Department of Defense found that 63 percent of the world’s most highly cited academic papers in science and technology were produced by researchers based in the United States (Galama and Hosek 2008). America remains the top destination for graduate students studying outside of their own country, attracting 19 percent of all foreign students in 2008—9 percentage points more than its closest competitor, the United Kingdom (Wildavsky 2011). Yet surely the most dramatic educational development in recent decades has been the rapid global expansion of higher education (Wildavsky 2008). Harvard economist Richard Freeman (2010) estimates that America’s share of the total number of postsecondary students worldwide fell from 29 percent in 1970 to just 12 percent in 2006, a 60 percent decline. A portion of this decline reflects the progress of developed countries documented above, but the more important factor by far has been the spectacular expansion of higher education in emerging economies such as China and India. In China alone, postsecondary enrollments exploded from fewer than 100,000 students in 1970 to 23.4 million in 2006. The increase over the same period in India was from 2.5 million to 12.9 million students. By way of comparison, just 17.5 million American students were enrolled in postsecondary degree programs in 2006. Although these absolute enrollment numbers reflect China’s and India’s sheer size and say nothing about the quality of instruction students receive, several recent reports have nonetheless concluded that the rapidly shifting landscape of higher education threatens America’s continued dominance in such strategically important fields as science and technology. Perhaps best known is the 2007 National Academy of Sciences committee report titled Rising Above the Gathering Storm, which drew on a variety of indicators of educational and scholarly performance to warn that the “scientific

HHassett.indbassett.indb 8181 111/8/121/8/12 9:279:27 PMPM 82 RETHINKING COMPETITIVENESS

and technological building blocks critical to our economic leadership are eroding at a time when many other nations are gathering strength” (National Academy of Sciences 2007, p. 4). A follow-up report issued by a subset of the committee in 2010 warned that the storm was “rapidly approaching category five” (National Academy of Sciences 2010). While critics claim the committee exaggerated the degree to which the research coming out of emerging economies is comparable to that produced by researchers based in the United States, it seems safe to conclude that in the future America will occupy a much smaller share of a rapidly expanding academic marketplace (Galama and Hosek 2008).

The Economic Costs of Low-Quality Education American students now complete less schooling than those in many other developed countries and, at the secondary level, perform substantially worse in math and science. Moreover, America’s long-standing edge in higher education is fading as developing countries increasingly make investments in higher education a central part of their economic development strategies. How concerned should we be about these developments? And is it the improvement in educational outcomes abroad that should motivate our concern? After all, until very recently the performance of the U.S. economy had far surpassed that of the industrialized world as a whole, despite our students’ mediocre performance on international tests. Some observers have gone so far as to question the existence of a link between available measures of the performance of national education systems and economic success. Education researcher Gerald Bracey in 2002 criticized those asserting that low-quality education threatened our national prosperity, noting that “none of these fine gentlemen has provided any data on the relationship between the economy’s health and the performance of schools. Our long economic boom suggests there isn’t one—or that our schools are better than the critics claim” (Bracey 2002, p. B01).7 Bracey’s evidentiary concern was not entirely misplaced. Economists as far back as Adam Smith have highlighted the theoretical importance of human capital as a source of national economic growth. For technologically advanced countries, highly educated workers represent a source

HHassett.indbassett.indb 8282 111/8/121/8/12 9:279:27 PMPM EDUCATION AND GLOBAL COMPETITIVENESS 83

of innovations needed to further enhance labor productivity (Benhabib and Spiegel 1994). For countries far from the frontier, education is necessary to allow workers to be able to adopt new technologies developed elsewhere (Nelson and Phelps 1966). Because a given country is likely to be both near and far from the technological frontier in various industries at any given point in time, both of these mechanisms are likely to operate simultaneously. Yet rigorous empirical evidence supporting these commonsense propositions has been notoriously difficult to produce. One key limitation of early research examining the relationship between education and economic growth is that it was based on crude measures of school enrollment ratios or the average years of schooling completed by the adult population. Although studies taking this approach tend to find a positive relationship between schooling and economic growth across countries, years of schooling is an incomplete and potentially quite misleading indicator of the performance of national education systems (see, for example, Krueger and Lindahl 2001). As noted above, measures of educational attainment implicitly assume that a year of schooling is equally valuable regardless of where it is completed—despite the clear evidence from international assessments that the skills acquired by students of the same age vary widely across countries. Economists Eric Hanushek and Ludger Woessmann (2008) have addressed this limitation in an important series of papers published since 2008.8 Their key innovation is the use of twelve international assessments of math and science achievement conducted between 1964 and 2003 to construct a comparable measure of the cognitive skills of secondary school students for a large sample of countries. This measure enables them to analyze the relationship between this measure and economic growth rates between 1960 and 2000 across all fifty countries for which cognitive skills and growth data are available, and separately across twenty-four members of the OECD. Hanushek and Woessmann’s work has yielded several notable results. First, after controlling for both a country’s initial GDP per capita and the average years of schooling completed in 1960, a 1 percent standard deviation increase in test scores is associated with an increase in annual growth rates of nearly 2 percent. Taken at face value, this implies that raising the performance of American students in math and science to the level of a top-performing nation such as Finland would increase our growth rate by

HHassett.indbassett.indb 8383 111/8/121/8/12 9:279:27 PMPM 84 RETHINKING COMPETITIVENESS

more than a full percentage point over the long run (i.e., once students educated to that level of academic accomplishment make up the entire national workforce). Second, both the share of a country’s students performing at a very high level and the share performing above a very low level appear to contribute to economic growth in roughly equal amounts, suggesting that there is no clear economic rationale for policymakers to focus exclusively on improving performance at the top or the bottom of the ability distribution (Hanushek and Woessmann 2009). Finally, they show that, after controlling for their test-based measure of students’ cognitive skills, the number of years of schooling completed by the average student is no longer predic- tive of growth rates. This suggests that policies intended to increase the quantity of schooling students receive will bear only economic fruit if they are accompanied by measurable improvements in students’ cognitive skills (Hanushek and Woessmann 2010). Although a clear improvement over previous evidence, skeptics may wonder whether the pattern Hanushek and Woessmann have identified linking education quality and economic growth in fact reflects a causal relationship. It is possible that unidentified factors enhance both the quality of national education systems and economic growth. In addition, the fact that they draw in part on very recent test score measures to predict growth rates between 1960 and 2000 leaves open the possibility that economic growth leads to strong educational performance rather than the other way around. Hanushek and Woessmann (2009) have performed a series of analyses intended to rule out these concerns. For example, they demonstrate that they obtain similar results when they use tests administered prior to 1980 to predict growth rates after that year and when they use changes in a country’s test scores over time to predict variation in growth rates during the time period of their analysis. Although none of these tests of causation is definitive on its own, together they strongly suggest that policies that increase education quality would in fact generate a meaningful economic return. Moreover, the magnitude of the relationship Hanushek and Woess- mann have identified is so large that it would remain important even if a substantial portion of it were driven by other factors. Consider the results of a simulation in which it is assumed that the math achievement of American students improves by 0.25 standard deviations gradually over a twenty-year

HHassett.indbassett.indb 8484 111/8/121/8/12 9:279:27 PMPM EDUCATION AND GLOBAL COMPETITIVENESS 85

period. Such an improvement would raise the United States to the level of countries such as New Zealand and the Netherlands, but not to that of the very top-performing OECD countries. Assuming that the past relationship between test scores and economic growth holds true in the future, the net present value of the resulting increment to GDP over an eighty-year horizon would amount to almost $44 trillion. A parallel simulation of the consequences of bringing American students up to the level of the top- performing countries suggests that doing so would yield benefits with a net present value approaching $112 trillion (Hanushek and Woessmann 2010). Yet despite ubiquitous rhetoric about education’s importance for countries competing in the global marketplace, there is no evidence that these potential gains would come at the expense of other nations. Put differently, there is no reason to suspect that Americans are made worse off in absolute terms by the superior performance of students in places such as Finland, Korea, or even Shanghai. At the higher-education level, American universities clearly face growing competition in recruiting talented international students and faculty and will likely find it difficult to maintain their current dominance of world rankings. Yet, as Richard Freeman (2010) explains, “The globalization of higher education should benefit the United States and the rest of the world by accelerating the rate of technological advance associated with science and engineering and by speeding the adoption of best-practices around the world, which will lower the costs of production and prices of goods” (p. 374). This is not to say that a continued decline in the relative standing of the American education system would leave the U.S. economy entirely unaffected. As Caroline Hoxby (2003) points out, studies of the factor content of American exports and economic growth have long documented their disproportionate reliance on human capital. This pattern suggests that the United States has traditionally had a comparative advantage in the production of goods that depend on skilled labor, which in turns reflects its historical edge in the efficient production of highly educated workers. In recent decades American companies have increasingly addressed labor shortages in technical fields by “importing” human capital in the form of highly educated immigrants, many of whom received their postsecondary training in the United States (Freeman 2010). This strategy cannot be a source of comparative advantage in the long run, however, because other

HHassett.indbassett.indb 8585 111/8/121/8/12 9:279:27 PMPM 86 RETHINKING COMPETITIVENESS

countries are by definition able to import talented immigrants at the same cost. The decline in the relative performance of our educational system may therefore have adverse consequences for the high-tech sectors on which the nation has historically depended to generate overall growth. The ability of the American economy as a whole to adapt in the face of such a disruption is, of course, an open question.

Policy Lessons In short, although there is little indication that education is an area in which countries are engaged in zero-sum global competition for scarce resources, education reform does provide a means to enhance economic growth and, in turn, the nation’s capacity to address its mounting fiscal challenges. Even if that were not the case, the moral argument for addressing the performance of America’s most dysfunctional school systems and the inequalities in social outcomes they produce would be overwhelming. What, then, are the lessons American policymakers should draw from the growing body of evidence examining the performance of national school systems? The first, and most straightforward, is simply that dramatic improvement is possible—and that this is true even of the best-performing state school systems within the United States. Not only do many countries perform at markedly higher levels despite being at lower levels of economic development, but also several of these countries have improved their performance substantially in the relatively short period of time since international tests were first widely administered (Mourshed, Chijioki, and Barber 2010). Nor do the international data suggest that countries face a stark trade-off between excellence and equity when considering strategies to raise student achievement. In fact, the countries with the highest average test scores tend to exhibit less overall inequality in test scores and, in many cases, weaker dependence of achievement on family background characteristics (Freeman, Machin, and Viarengo 2010). A policy agenda centered around closing the global achievement gap between American students and those in other developed countries would provide a complementary, and arguably more encompassing, rationale for education reform than one focused primarily on closing achievement gaps within the United States. The urgency of closing domestic achievement

HHassett.indbassett.indb 8686 111/8/121/8/12 9:279:27 PMPM EDUCATION AND GLOBAL COMPETITIVENESS 87

gaps is without question, but the current emphasis on this goal may well reinforce the perception among middle-class Americans that their schools are performing at acceptable levels. The 2011 Phi Delta Kappa–Gallup Poll shows that more than half of all Americans currently assign the public schools in their local community an “A” or “B” grade, whereas only 17 percent assign one of those grades to public schools in the nation as a whole (Bushaw and Lopez 2011). This gap between local and national evaluations has widened considerably over the past decade, and similar data from the 2011 Education Next–Program on Education Policy and Governance Survey show that well-educated, affluent Americans are particularly likely to rate their local schools favorably (Howell, Peterson, and West 2011). Reporting systems that make it possible to compare the performance of students in specific American school districts to top-performing countries internationally could help to alter these perceptions and broaden support for reform. A second lesson is that reform efforts should aim to improve the quality of education available to American students in elementary and secondary schools rather than merely increase the quantity of education they consume. The large economic return to the completion of college and especially graduate degrees suggests that there is considerable demand for workers who have been educated to those levels, and policymakers would be wise to address issues such as the complexity of financial aid systems that create obstacles to degree completion for academically prepared students. But increasing educational attainment should not be an end in and of itself. Doing so is unlikely to yield economic benefits absent reforms to K-12 schooling that ensure that a growing number of students are equipped for the rigors of postsecondary work. A final general lesson is that additional financial investment is neither necessary nor sufficient to improve the quality of elementary and secondary education. The data presented above clearly show that other developed countries have managed to achieve far greater productivity in their school systems, in many cases spending considerably less than the United States, to achieve superior results. Nor have countries that have increased spending levels in recent decades experienced gains in their performance on international assessments, a pattern that is consistent with the mixed track record of spending increases in producing improved student outcomes within the United States.9

HHassett.indbassett.indb 8787 111/8/121/8/12 9:279:27 PMPM 88 RETHINKING COMPETITIVENESS

If countries with high-performing elementary and secondary education systems have not spent their way to the top, how have they managed to get there? Unfortunately, drawing more specific policy guidance for the United States on the basis of international evidence remains a challenge. While it is straightforward to document correlations between a given policy and performance across countries, it is much harder to rule out the existence of other factors that could explain the relationship. The vast cultural and contextual differences from one country to the next also imply that policies and practices that work well in one setting may not do so in another. Even so, there are three broad areas in which the consistency of findings across studies using different international tests and country samples bears attention.

Exit Exams. Perhaps the best-documented area is that students perform at higher levels in countries (and in regions within countries) with externally administered, curriculum-based exams at the completion of secondary schooling that carry significant consequences for students of all ability levels (Bishop 2006; Hanushek and Woessmann 2011). Although many Ameri- can states now require students to pass an exam in order to receive a high school diploma, these tests are typically designed to assess minimum com- petency in math and reading and are all but irrelevant to students elsewhere in the performance distribution. In contrast, exit exams in many European and Asian countries cover a broader swath of the curriculum, play a central role in determining students’ postsecondary options, and carry significant weight in the labor market. As a result, these systems provide strong incentives for student effort and valuable information to parents and other stakeholders about the relative performance of secondary schools. The most rigorous available evidence indicates that math and science achievement is a full grade-level equivalent higher in countries with such an exam system in the relevant subject (Woessmann 2003).

Private School Competition. Countries vary widely in the extent to which they make use of the private sector to provide public education. In countries such as Belgium, the Netherlands, and (more recently) Sweden, for example, private schools receive government subsidies for each student enrolled equivalent to the level of funding received by state-run schools. Because private schools in these countries are more heavily regulated than

HHassett.indbassett.indb 8888 111/8/121/8/12 9:279:27 PMPM EDUCATION AND GLOBAL COMPETITIVENESS 89

those in the United States, they more closely resemble American charter schools (though they typically have a distinctive religious character). In theory, government funding for private schools provides families of all income levels with a broader range of options and therefore subjects the state-run school system to increased competition from alternative providers. Rigorous studies confirm that students in countries that, for historical reasons, have a larger share of students in private schools perform at higher levels on international assessments while spending less on primary and secondary education, suggesting that competition can spur school productivity (West and Woessmann 2010). In addition, the achievement gap between socioeconomically disadvantaged and advantaged students is reduced in countries in which private schools receive more government funds (Woessmann et al. 2009).

High-Ability Teachers. Much attention has recently been devoted to the fact that several of the highest-performing countries internationally draw their teachers disproportionately from the top one-third of all students completing college degrees (Barber and Mourshed 2007). This contrasts sharply with patterns in the United States and, given the strong evidence that teacher effectiveness is the most important school-based determinant of student achievement, likely plays a decisive role in their success.10 Unfortu- nately, as education economist Dan Goldhaber (2009) points out, the differences in teacher policies across countries that have been documented to date “do not point toward a consensus about the types of policies—or even sets of policies—that might ensure a high-quality teacher workforce” (p. 83). Although increasing average salaries provides one potential mechanism to attract a more capable teaching workforce, there is no clear relationship between teacher salary levels and student performance among developed countries. Especially given the current strains on district and state budgets, any funds devoted to increasing teacher salaries should be targeted at subjects such as math and science (in which qualified candidates have stronger earnings opportunities in other industries) and at teachers who demonstrate themselves to be effective in the classroom. Intriguingly, the only available study on the latter topic shows that countries that allow teacher salaries to be adjusted on the basis of their performance in the classroom perform at higher levels (Woessmann 2010).

HHassett.indbassett.indb 8989 111/8/121/8/12 9:279:27 PMPM 90 RETHINKING COMPETITIVENESS

Over the past two decades, state and federal efforts to improve Ameri- can education have centered on the development of test-based accountability systems that reward and sanction schools on the basis of their students’ performance on state assessments. The evidence is clear that the federal No Child Left Behind Act and its state-level predecessors have improved student achievement, in particular for students at the bottom of the performance distribution (Dee and Jacob 2011; Hanushek and Raymond 2005). Yet the progress made under these policies falls well short of their ambitious goals and, as important, appears to have been limited to a one-time increment in performance rather than launching schools on a trajectory of continuous improvement. International evidence does not provide definitive guidance for closing the global achievement gap between students in the United States and those in the top-performing countries abroad. It does, however, indicate that holding students accountable for their performance, creating competition from alternative providers of schooling, and developing strategies to recruit and retain more capable teachers all have important roles to play in addressing what should be a vital national priority.

Notes 1. The performance of countries participating in the 2003 TIMSS and PISA studies, for example, are correlated at 0.87 in math and 0.97 in science (see Hanushek and Woessmann 2008). Although the relative rank of U.S. students is higher on the TIMSS than on the PISA, this is attributable to the fact that the sample of countries participating in the TIMSS includes fewer industrialized countries and more developing ones. 2. I focus on math and science performance because achievement in these subjects is more reliably measured across countries and because it is in these subjects in which the evidence connecting education quality to economic growth is strongest. As noted above, the United States performs somewhat better on PISA’s reading assessment, with an average score of 500 as compared to the OECD average of 493. Even so, its reading performance significantly trails six OECD countries and is statistically indistinguishable from another fourteen. 3. The reported grade-level equivalents for math and science are 41 and 38 scale score points, respectively. 4. Excluding Chile, Mexico, and Turkey, the correlation between math (science) achievement and cumulative spending per student across OECD countries is 0.08 (–0.06); neither correlation is statistically significant.

HHassett.indbassett.indb 9090 111/8/121/8/12 9:279:27 PMPM EDUCATION AND GLOBAL COMPETITIVENESS 91

5. U.S. performance on the PISA math assessment fell between 2003 and 2006, however, and its 2009 scores are not statistically different from those achieved in 2003. See Loveless (2011). 6. These rankings are available at http://www.shanghairanking.com/. The ARWU ranking is based on the following indicators: number of alumni and staff who have won Nobel Prizes and Fields Medals (10 percent); number of highly cited researchers across a range of fields selected by Thomson Scientific (20 percent); number of articles published in Nature and Science (20 percent); number of articles indexed in Science Citation Index—Expanded and Social Sciences Citation Index (20 percent); and a size-adjusted performance measure that divides each indicator by the number of full- time faculty and academic staff (10 percent). Although the ARWU has been criticized for focusing solely on scholarly productivity and overemphasizing math and science publications, competing rankings yield similar conclusions regarding the relative performance of U.S. institutions. 7. For a more nuanced account questioning the strength of the relationship between education and economic growth, see Grubb and Lazerson (2004). 8. For an earlier attempt to relate measures of educational quality based on student test scores and economic growth, see Hanushek and Kimko (2000). 9. For a review of the international evidence on the relationship between spending and achievement across and within countries, see Hanushek and Woessmann (2011). For a review of evidence from the United States, see Hanushek (2003). 10. For evidence on the importance of teacher effectiveness for student achievement, see Rockoff (2004).

References Armario, Christine. 2010. “Wake-up Call: U.S. Students Trail Global Leaders.” Associ- ated Press Wire, December 7. Barber, Michael, and Mona Mourshed. 2007. How the World’s Best-Performing School Systems Come Out on Top, McKinsey and Company. http://mckinseyonsociety. com/how-the-worlds-best-performing-schools-come-out-on-top/ (accessed July 15, 2012). Benhabib, Jess, and Mark Spiegel. 1994. “The Role of Human Capital in Economic Development: Evidence from Aggregate Cross-National Data.” Journal of Monetary Economics 34:143–74. Bishop, John. 2006. “Drinking from the Fountain of Knowledge: Student Incentive to Study and Learn—Externalities, Information Problems, and Peer Effects.” In Handbook of the Economics of Education, vol. 2, ed. E. A. Hanushek and F. Welch. Amsterdam: Elsevier, 910–44. Bracey, Gerald. 2002. “Why do we Scapegoat the Schools?” Washington Post, May 5, B01.

HHassett.indbassett.indb 9191 111/8/121/8/12 9:279:27 PMPM 92 RETHINKING COMPETITIVENESS

Bushaw, William J., and Shane J. Lopez. 2011. “Betting on Teachers: The 43rd Annual Phi Delta Kappa/Gallup Poll of the Public’s Attitudes toward the Public Schools.” Phi Beta Kappan 93 (1):8–26. Dee, Thomas S., and Brian A. Jacob. 2011. “The Impact of No Child Left Behind on Student Achievement.” Journal of Policy Analysis and Management 30(3):418–46. Duncan, Arne. 2010. “Back to School: Enhancing U.S. Education and Competitive- ness.” Foreign Affairs 89(6):65–74. Fleischman, Howard L., Paul J. Hopstock, Marisa P. Pelczar, and Brooke E. Shelley. 2010. Highlights from Pisa 2009: Performance of U.S. 15-Year-Old Students in Reading, Mathematics, and Science Literacy in an International Context. NCES 2011-004. U.S. Department of Education, National Center for Education Statistics. Washington, D.C.: U.S. Government Printing Office. Freeman, Richard B. 2010. “What Does the Global Expansion of Higher Education Mean for the United States?” In American Universities in a Global Market, ed. C. T. Clotfelter. Chicago: University of Chicago Press, 373–404. Freeman, Richard B., Stephen Machin, and Martina Viarengo. 2010. “Varia- tion in Educational Outcomes and Policies across Countries and of Schools within Countries.” NBER Working Paper 16293. National Bureau of Economic Research, Cambridge, MA. Galama, Titus, and James Hosek. 2008. U.S. Competitiveness in Science and Technology. Santa Monica, CA: Rand Corp. Goldhaber, Dan. 2009. “Lessons from Abroad: Exploring Cross-Country Differences in Teacher Development Systems and What They Mean for U.S. Policy.” In Creat- ing a New Teaching Profession, ed. D. Goldhaber and J. Hannaway. Washington, D.C.: Urban Institute Press, 81–111. Goldin, Claudia. 2001. “The Human Capital Century and American Leadership: Vir- tues of the Past.” Journal of Economic History 61(2):263–92. Goldin, Claudia, and Lawrence Katz. 2008. The Race between Education and Technol- ogy. Cambridge, MA: Belknap Press of Harvard University. Grubb, W. Norton, and Marvin Lazerson. 2004. The Education Gospel: The Economic Power of Schooling. Cambridge, MA: Harvard University Press. Hanushek, Eric A. 2003. “The Failure of Input-based Schooling Policies.” Economic Journal 113(485):F64–98. Hanushek, Eric A., and Dennis D. Kimko. 2000. “Schooling, Labor Force Quality, and the Growth of Nations.” American Economic Review 90(5):1184–1208. Hanushek, Eric A., and Margaret Raymond. 2005. “Does School Accountability Lead to Improved Student Achievement?” Journal of Policy Analysis and Management 24(2):297–327. Hanushek, Eric A., and Ludger Woessmann. 2008. “The Role of Cognitive Skills in Economic Development.” Journal of Economic Literature 46(3):607–68. ———. 2009. Do Better Schools Lead to More Growth? Cognitive Skills, Economic Out- comes, and Causation. NBER Working Paper 14633. National Bureau of Economic Research, Cambridge, MA.

HHassett.indbassett.indb 9292 111/8/121/8/12 9:279:27 PMPM EDUCATION AND GLOBAL COMPETITIVENESS 93

———. 2010. How Much Do Educational Outcomes Matter in OECD Countries? NBER Working Paper 16515. National Bureau of Economic Research, Cambridge, MA. ——— . 2011. “The Economics of International Differences in Educational Achieve- ment.” In Handbook of the Economics of Education, vol. 3, ed. E. Hanushek, S. Machin, and L. Woessmann. Amsterdam: Elsevier, 89–200. Heckman, James J., and Alan B. Krueger. 2003. Inequality in America: What Role for Human Capital Policies? Cambridge, MA: MIT Press. Heckman, James J., and Paul A. LaFontaine. 2010. “The American High School Graduation Rate: Trends and Levels.” Review of Economics and Statistics 92(2):244–62. Howell, William G., Paul E. Peterson, and Martin R. West. 2011. “The Public Weighs In on School Reform.” Education Next 11(4):11–22. Hoxby, Caroline M. 2003. “School Choice and School Productivity: Could School Choice Be a Tide That Lifts All Boats?” In The Economics of School Choice, ed. C. M. Hoxby. Chicago: University of Chicago Press, 287–341. Krueger, Alan B., and Michael Lindahl. 2001. “Education for Growth: Why and For Whom?” Journal of Economic Literature 39(4):1101–36. Krugman, Paul. 1996. Pop Internationalism. Cambridge, MA: MIT Press. Ladd, H. F. 2012. “Education and Poverty: Confronting the Evidence.” Journal of Pol- icy Analysis and Management 31(2):203-227. Loveless, Tom. 2011. The 2010 Brown Center Report on American Education: How Well Are American Students Learning? Washington, D.C.: Brookings Institution. Mourshed, Mona, Chinezi Chijioki, and Michael Barber. 2010. How the Most Improved School Systems in the World Keep Getting Better. McKinsey and Company. http:// mckinseyonsociety.com/how-the-worlds-most-improved-school-systems-keep- getting-better/ (accessed July 15, 2012). National Academy of Sciences. 2007. Rising above the Gathering Storm: Energizing and Employing America for a Brighter Economic Future. Washington, D.C.: The National Academies Press. ———. 2010. Rising above the Gathering Storm, Revisited: Rapidly Approaching Category Five? Washington, D.C.: The National Academies Press. National Commission on Excellence in Education. 1983. A Nation at Risk: The Impera- tive for Educational Reform. Washington, D.C.: U.S. Department of Education. Nelson, Richard F., and Edmund Phelps. 1966. “Investments in Humans, Technology Diffusion, and Economic Growth.” American Economic Review 56(2):69–75. Organisation for Economic Co-operation and Development. 2010a. PISA 2009 Results, vol. 1, What Students Know and Can Do: Student Performance in Reading, Mathematics and Science. Paris: OECD. ———. 2010b. PISA 2009 at a Glance. Paris: OECD. ———. 2010c. Education Indicators at a Glance 2010. Paris: OECD. ———. 2011. Strong Performers and Successful Reformers in Education: Lessons from PISA for the United States. Paris: OECD.

HHassett.indbassett.indb 9393 111/8/121/8/12 9:279:27 PMPM 94 RETHINKING COMPETITIVENESS

Rockoff, Jonah E. 2004. “The Impact of Teachers on Student Achievement: Evidence from Panel Data.” American Economic Review 94(2):247–52. Sala-i-Martin, Xavier. 2010. “The Economics behind the World Economic Forum’s Global Competitiveness Index.” In Dimensions of Competitiveness, ed. P. De Grauwe. Cambridge, MA: MIT Press. Schuetz, Gabriela, Heinrich W. Ursprung, and Ludger Woessmann. 2008. “Educa- tion Policy and Equality of Opportunity.” Kyklos 61(2):279–308. U.S. Department of Education. 2010. A Blueprint for Reform: Reauthorization of the Elementary and Secondary Education Act. Washington, D.C.: U.S. Department of Education, Office of Planning, Evaluation and Policy Development. Wagner, Tony. 2008. The Global Achievement Gap: Why Even Our Best Schools Fail to Teach the New Survival Skills Our Children Need—And What We Can Do about It. New York: Basic Books. West, Martin R., and Ludger Woessmann. 2010. “‘Every Catholic Child in a Catho- lic School’: Historical Resistance to State Schooling, Contemporary Private School Competition, and Student Achievement across Countries.” Economic Journal 120(546):F229–55. White House. 2009. Remarks of President Barack Obama as Prepared for Delivery to Joint Session of Congress, February 24, 2009. http://www.whitehouse.gov/the_press_office/ Remarks-of-President-Barack-Obama-Address-to-Joint-Session-of-Congress/. ———. 2011. Remarks by the President on Education in Arlington, Virginia, March 14, 2011. http://www.whitehouse.gov/the-press-office/2011/03/14/remarks-president- education-arlington-virginia. Wildavsky, Ben. 2008. The Great Brain Race: How Global Universities Are Reshaping the World. Princeton, NJ: Princeton University Press. ———. 2011. “Think Again: Education.” Foreign Policy, March/April. Woessmann, Ludger. 2003. “Central Exit Exams and Student Achievement: Interna- tional Evidence.” In No Child Left Behind? The Politics and Practice of School Account- ability, ed. P. E. Peterson and M. R. West. Washington, D.C.: Brookings Institution Press, 292–323. ———. 2004. How Equal Are Educational Opportunities? Family Background and Stu- dent Achievement in Europe and the United States. CESifo Working Paper No. 1162. Ifo Institute, Center for Economic Studies, Munich, Germany. ———. 2010. “Cross-Country Evidence on Teacher Performance Pay.” Economics of Education Review 30(3):404–18. Woessmann, Ludger, Elke Luedemann, Gabriela Schuetz, and Martin R. West. 2009. School Accountability, Autonom y, and Choice around the World. London: Edward Elgar.

HHassett.indbassett.indb 9494 111/8/121/8/12 9:279:27 PMPM 4

Immigration, Productivity, and Competitiveness in American Industry Gordon H. Hanson

When Google recently decided where to invest in social media-based computer games, it chose San Francisco, California (see Arrington 2010). The primary attraction of the location was the presence of highly skilled labor, including computer scientists, engineers, web designers, and other technical professionals. A substantial fraction of these workers were not born in the United States. They came to the country to study in U.S. universities or to fulfill temporary employment contracts for U.S. companies under H-1B immigration visas and stayed on to become a permanent part of the U.S. labor force. In California, skilled immigrant labor is vital to the state’s ability to attract and to retain businesses in high-tech industries. In 2008, 37 percent of California’s supply of engineers and computer scientists were foreign born.1 If these workers were, for some reason, to leave the state, many of California’s high-tech firms would surely follow. The experience of California raises the broader question of how immigration affects the competitiveness of American industry. Does the presence of immigrant labor enhance the capacity of U.S. companies to compete internationally? By “competitiveness,” I mean the share of global demand that is met by U.S. industries. Immigration may affect the market shares of U.S. firms by making U.S. labor cheaper, by raising the productivity of U.S. factories, or by lowering the cost of trading with or investing in the rest of the world. Whether and how immigration affects U.S. competitiveness affects both the level and the distribution of national income.

HHassett.indbassett.indb 9595 111/8/121/8/12 9:279:27 PMPM 96 RETHINKING COMPETITIVENESS

In this chapter, I delve into ongoing research by economists on the consequences of immigration to examine its importance for U.S. competitiveness. I begin, in the first section, by outlining a theoretical framework that identifies the mechanisms through which immigration affects production costs, productivity, and trade costs. Then, in the second section, I discuss empirical evidence on the quantitative importance of these mechanisms. I proceed, in the third section, to describe how U.S. immigration policy shapes the level and composition of immigration and how future changes in policy may affect U.S. industries. For economists, competitiveness is a slippery concept. It mixes productivity—how much one worker is capable of producing in one hour—with production costs—how much that labor hour costs. Changes in competitiveness do not map cleanly into changes in real earnings for capital or labor, which are the proximate determinants of the national standard of living.2 In the end, competitiveness matters only because it affects U.S. living standards, a point sometimes lost in policy debates. The usefulness of economic theory is that it provides a framework for characterizing the path- ways through which immigration affects the market shares of U.S. industries and how, in turn, these changes in market shares affect factor earnings. Consider two alternative types of immigration shocks, each of which raises global market shares for U.S. industries but which have different effects on the national and international distribution of income. First, suppose that immigration causes aggregate U.S. total factor productivity (TFP) to increase as a consequence of, say, foreign engineers creating new production technologies that make U.S. capital and labor more productive. Higher TFP increases U.S. output in all industries, raising the earnings of all factors of production in the process. The share of global demand met by U.S. producers rises, with the composition of U.S. output remaining unchanged. Because productivity expands across the board, no one industry grows at the expense of another. Furthermore, the earnings of capital and labor rise proportionately because their productivity increase is commensurate. The gain in TFP makes the United States wealthier and makes the rest of the world wealthier as well. Because the productivity shock causes the market shares of U.S. industries to rise globally, aggregate U.S. TFP growth satisfies my definition of an improvement in U.S. competitiveness. But it is one that is good for everybody.

HHassett.indbassett.indb 9696 111/8/121/8/12 9:279:27 PMPM IMMIGRATION, PRODUCTIVITY, AND COMPETITIVENESS 97

Next, consider an immigration-induced change in competitiveness that isn’t good for everybody. Suppose that immigration, by increasing the supply of low-skilled labor, causes the wages of these workers in the United States to fall. U.S. industries intensive in the use of low-skilled labor, such as perishable fruits and vegetables, see their profitability rise, allowing them to hire labor, land, and capital away from other sectors, including perhaps such capital-intensive crops as corn or wheat. As production expands in fruits and vegetables, it necessarily declines in some other sectors. The share of global demand met by U.S. labor-intensive industries rises, but the share met by some other U.S. industries falls. In this example, the changes in competitiveness are sector specific, which entail differential changes in income by factor type. The owners of land suitable for fruit and vegetable production would likely see their real incomes grow, whereas the owners of land suitable for corn or wheat production may see their real incomes shrink. Furthermore, native-born low-skilled workers in the United States may be hurt as a consequence of the increase in labor supply lowering their wages. In the first example, immigration is uniformly good for U.S. industries and factors of production; in the second, it favors some sectors and factors over others. To consider whether either of these stories has relevance for the U.S. economy, we need a model of how immigration affects productivity, wages, and trade costs. I use the global general equilibrium framework developed by Eaton and Kortum (2002) to show how changes in immigration map into changes in outcomes of interest. Their model isn’t about immigration per se, but it offers an elegant and intuitive explanation for the determinants of market shares by industry in the global economy, which makes it appealing for our analysis. One can add on to their framework a model of innovation and economic growth (Eaton and Kortum 2009), which suggests how immigration may affect industry competitiveness dynamically. There is intense debate among economists about the consequences of immigration. Today, immigrants account for 15 percent of the U.S. labor force. Over the last two decades, much of the research on the subject has focused on the U.S. influx of low-skilled labor, mainly from Mexico and Central America (Hanson 2009). Between 1970 and 2009, immigrants as a share of the U.S. population rose from 5 percent to 12 percent, with the newly arriving immigrants concentrated at the extremes of the skill

HHassett.indbassett.indb 9797 111/8/121/8/12 9:279:27 PMPM 98 RETHINKING COMPETITIVENESS

distribution. Currently, 30 percent of working-age immigrants have less than a high school education, compared to only 8 percent of native-born adults. It is natural to imagine that large inflows of low-skilled immigrants put downward pressure on the wages of low-skilled native workers. How- ever, the estimated impact of immigration on wages depends very much on the econometric approach that one takes (Borjas 2003; Card 2001). Immigration may also affect local prices for housing and nontraded services, which may in turn have indirect consequences for U.S. competitiveness. Here, I discuss evidence on the wage and price impacts of immigration and what it means for U.S. industries. The policy debate about immigration has also focused on low-skilled labor. One concern is that immigrants with low earnings potential are a net drain on U.S. fiscal accounts, which increases the tax burden on U.S. individuals and corporations (Hanson 2005). Complicating the discussion is the fact that a large fraction of low-skilled immigrants are in the United States illegally (Passel and Cohn 2010). In 2009, unauthorized immigrants were 29 percent of the U.S. foreign-born population and 5 percent of the U.S. labor force. On the one hand, illegal immigrants lack access to most federally funded entitlements, mitigating their fiscal impact, at least at the federal level (Hanson 2007). On the other hand, many illegal immigrants have children who attend public schools and, if they were born in the United States, have access to subsidized health care (Camarota 2004). The absorption of publicly funded services results in fiscal transfers from native- headed households to immigrant-headed households. Because much of the funding for public education comes from states and localities, the fiscal consequences of immigration vary regionally. Part of immigration’s impact, then, is mediated by tax and spending policies at both the federal and state levels. Industries located in states that tax corporate or individual income, have large immigrant populations, and provide benefits that are generous to immigrant families may face a relatively large tax burden from immigration. As the political debate about illegal immigration has intensified, high- skilled immigration has been lost in the noise. The omission is unfortunate because, arguably, high-skilled immigration matters much more for U.S. productivity growth than low-skilled immigration. Low-skilled immigration may shift income from low-skilled workers to employers or from taxpayers to immigrant-headed households. But these impacts are largely static in

HHassett.indbassett.indb 9898 111/8/121/8/12 9:279:27 PMPM IMMIGRATION, PRODUCTIVITY, AND COMPETITIVENESS 99

nature and have only indirect effects on growth. For low-skilled immigration to have dynamic effects, it must change savings and investment decisions by firms and households. Such dynamic effects are plausible but likely to be small. Because the U.S. tax system is progressive, more low-skilled immigration could increase tax rates disproportionately for high-income earners, reducing their incentive to work or to invest in human or physical capital. But the elasticity of labor supply with respect to income appears to be small (Ashenfelter, Doran, and Schaller 2010), suggesting that such disincentive effects are not large. Furthermore, any such negative consequences may be offset by the fact that low-skilled immigration tends to reduce the price of household services that high-income earners consume (such as child care, house cleaning, and yard care), which enables them to spend more hours at work (Cortes and Tessada 2009). In contrast to low-skilled immigration, high-skilled immigration appears to have direct effects on economic growth. High-skilled immigrants are disproportionately represented in technical fields, in which recent U.S. productivity growth has been most rapid. In 2003, foreign-born students accounted for 51 percent of PhDs in science and engineering fields awarded by U.S. universities, up from 27 percent in 1973 (Bound, Turner, and Walsh 2009). In theory, expanding the supply of scientists and engineers increases the U.S. labor force devoted to research and development (R&D), which contributes to higher rates of innovation and has positive effects on economic growth (Jones 1995).3 Empirically, increases in high-skilled immigration are associated with increases in patenting. While not all new patents find a productive application, a higher level of patenting typically means more innovation and faster productivity growth. Thus, high-skilled immigration affects U.S. competitiveness by raising aggregate TFP and TFP in sectors that are intensive in the use of science and engineering skills.4 A further manner through which high-skilled immigration affects competitiveness is through its impact on trade costs (Rauch 1999). Arriving immigrants bring with them knowledge about market conditions and investment opportunities in foreign countries. Such information flows are likely to lower the cost to U.S. companies of doing business abroad and to foreign companies of doing business in the United States. By hiring immigrant workers, U.S. companies may learn about new foreign inputs they can use in production or new foreign market niches that they can supply. Immigrants

HHassett.indbassett.indb 9999 111/8/121/8/12 9:279:27 PMPM 100 RETHINKING COMPETITIVENESS

who return to their home countries may help U.S. companies invest abroad, as has occurred in China, India, and Taiwan through the return to these countries of individuals who had worked in Silicon Valley (Saxenian 2002). The international transmission of knowledge engendered by immigration may not help all industries or all regions equally. To the extent that information flows are limited to the sectors or the locations in which immigrants are employed, it will enhance U.S. competitiveness in these locations but, through general equilibrium effects, diminish it in others.

Immigration and Competitiveness in Theory I have outlined three mechanisms through which immigration may affect the share of global demand that U.S. industries satisfy. One is by changing production costs. Immigration alters the supply of low-skilled and high- skilled labor in the U.S. economy, which may affect the prices of these factors facing U.S. companies. A second mechanism is by changing productivity. High-skilled immigration adds to the economy workers used intensively in R&D, which may raise the rate of innovation in U.S. industries. And a third mechanism is by changing trade costs. Immigration increases the international transmission of information, which may change the cost of doing business abroad for U.S. firms, as well as the cost to foreign firms of doing business in the United States. In this section, I outline a theoretical framework that incorporates each of these mechanisms into a global general equilibrium setting.

Productivity, Production Costs, and Market Shares. Before contem- plating immigration, we need to characterize the factors that determine the share of global demand in different industries that is satisfied by U.S. firms. The Eaton and Kortum (2002) framework that I employ has three elements: (1) innovation, through which firms TFP is determined, (2) the hiring of labor and intermediate inputs, through which factor prices are determined, and (3) international competition, through which firms with the lowest prices (inclusive of the costs of international trade) win the right to serve particular markets. One can think of the process of innovation as akin to firms playing a lottery, in which some will have good draws and develop processes that

HHassett.indbassett.indb 100100 111/8/121/8/12 9:279:27 PMPM IMMIGRATION, PRODUCTIVITY, AND COMPETITIVENESS 101

give them high TFP (or, equivalently, high product quality), while others will have bad draws, leaving them with low TFP (or low product quality). Notice that innovation lotteries as described here are about the level of TFP and not its growth. I take up sources of TFP growth in the next section. A country’s courts, financial markets, education systems, and other institutions play a role in innovation lotteries. In the United States, well- developed markets for venture capital, strong enforcement of intellectual property rights, and high-quality universities make firms in high-tech industries more likely to have good innovation draws than firms in, say, Vietnam, whose legal, financial, and educational institutions are less developed. However, Vietnam has an abundant supply of low-skilled labor and rich tropical farmland, which may give firms in the country an advantage vis-à-vis rivals in the United States in coming up with cost-effective ways of producing clothing or cultivating rice. Within each U.S. industry, there will be some firms that succeed in innovating and others that do not, with the share of firms that succeed varying across industries according to U.S. potential productivity in the sector, where potential productivity captures the set of factors that determine the likelihood of successful innovation. One easily recognizes this framework as a souped-up version of Ricardian comparative advantage. Potential productivity is roughly the equivalent of absolute advantage in an introductory textbook model. To produce output, firms hire labor and capital and purchase intermediate inputs. The cost of labor is determined by its supply and demand, such that wages are higher where high productivity causes labor demand to be strong and lower where population growth, educational upgrading, or inflows of foreign workers expand labor supply. Strong potential productivity in one industry pushes up the price of labor for other industries, such that firms in industries with weak potential productivity will only survive if they have been particularly lucky in the innovation lottery. The price of intermediate inputs in a country depends on the country’s industrial capacity in upstream industries and the proximity of the country to upstream suppliers in other countries. Immigration has direct effects on input prices through its impact on labor supply. All else equal, U.S. firms will have a cost advantage in industries that rely more heavily on the types of skills that U.S. workers have in abundance (where the astute observer will detect a bit of Heckscher-Ohlin reasoning sneaking into the framework).

HHassett.indbassett.indb 101101 111/8/121/8/12 9:279:27 PMPM 102 RETHINKING COMPETITIVENESS

If immigration increases the supply of these skills, it will enhance existing U.S. cost advantages. The price that a U.S. firm charges in foreign markets will depend in part on its TFP (higher TFP means a firm can get away with charging a lower price) and economy-wide wages (where lower wages are passed through to lower prices) and in part on international trade costs (which represent an ad valorem markup on the foreign price of U.S. goods). Trade costs are determined by importing country tariffs and nontariff barriers, international freight and insurance costs, and foreign marketing and distribution costs. Let’s imagine that competition in industries is Bertrand, such that the lowest-price firm wins the market for a specific product in a particular industry. Each industry consists of a large number of individual product varieties. U.S. firms will capture a larger market share—that is, they will produce a larger share of product varieties—in industries in which U.S. potential productivity is greater (such that a higher fraction of U.S. firms have good innovation draws) or U.S. input costs are lower (due to either a low share of low-skilled labor in production costs or a high share for high- skilled labor in production costs) and in countries in which trade costs vis- à-vis the United States are lower. U.S. export supply capacity in an industry is, then, the product of factors related to U.S. potential productivity, U.S. input costs, and international trade costs. The share of a specific market—for example, semiconductors—that U.S. firms will capture in a particular country—for example, France—will depend on U.S. export supply capacity for semiconductors in France relative to the export supply capacity for semiconductors in France of other countries that also serve the French market. Since I equate competitiveness with market share, U.S. competitiveness rises in semiconductors when U.S. export supply capacity in semiconductors rises relative to other countries. Such an outcome obtains when U.S. potential productivity rises (meaning more firms innovate successfully), U.S. input costs fall (meaning that U.S. cost advantages rise), or trade costs specific to U.S. firms decline (meaning that U.S. firms gain market access relative to foreign rivals). Immigration affects export supply capacity by affecting productivity, production costs, or trade costs, completing the intuition developed in the introduction. I have not yet said anything about how changes in export supply capacity affect the level or the distribution of

HHassett.indbassett.indb 102102 111/8/121/8/12 9:279:27 PMPM IMMIGRATION, PRODUCTIVITY, AND COMPETITIVENESS 103

national income in the United States. I will address these outcomes in the next section, in which I consider how changes in the U.S. economy brought by immigration may have affected U.S. earnings for low-skilled labor, high-skilled labor, and capital.

Innovation and Economic Growth. The framework I’ve outlined so far is static in nature. Innovation lotteries determine the level of TFP among firms in an industry. For there to be economic growth, TFP must rise over time, which requires that potential productivity expand. To understand sources of productivity change, I follow Eaton and Kortum (2009) and draw on models of endogenous economic growth. Imagine that instead of drawing their productivity from a lottery, firms invest in innovation by hiring R&D labor. The more firms invest, the more new ideas they generate and the more likely they are to generate market-winning innovations. The firm with the best innovation captures the market for a specific product, at least until another firm comes up with something better. In this way, production in an industry follows a “quality ladder” or “productivity ladder,” in which new innovations steadily raise quality or TFP relative to the past state of the art (Grossman and Helpman 1991). The rate of innovation, and therefore the rate of economic growth, depends positively on the supply of R&D labor, the productivity of R&D labor (which may vary across countries in a manner similar to variation in potential productivity), and the average magnitude of new innovations (in terms of their impact on raising quality or TFP) and negatively on the rate at which firms discount future profits. How does immigration affect growth in such a framework? If immigration expands the supply of R&D labor in the United States, it raises investment in innovation, which increases the rate at which successful new innovations occur. To the extent that the global supply of R&D labor is constrained—as would be the case if there were only a small number of super- star scientists that were capable of generating significant innovations—the success of the United States in attracting a star scientist would necessarily mean that another country loses her. The international competition for talent, then, may affect which countries generate most new innovations. Country immigration policies may constrain companies in their global talent search. With global trade, innovations developed in the United States

HHassett.indbassett.indb 103103 111/8/121/8/12 9:279:27 PMPM 104 RETHINKING COMPETITIVENESS

would ultimately diffuse to other countries (and vice versa) either through imitation or through foreign direct investment, raising global growth. At least in theory, high-skilled immigration has the potential to affect U.S. growth rates directly by raising industry innovation rates and thereby the rate of growth of industry export supply capacities.

Immigration and Trade Costs. International trade costs impede trade flows and thereby diminish U.S. industrial competitiveness by acting as a tax on cross-border commerce. The commerce-inhibiting effects of trade barriers are obvious when one thinks of import tariffs, import quotas, or the costs of international shipping. However, incomplete information between countries about demand and supply may also inhibit trade. Information barriers may account for why countries with a common language, common colonial history, or common religious traditions tend to trade more, as among these countries cultural barriers to commerce may be small. International migration creates linkages between a country and the rest of the world, which may help reduce information barriers and therefore lower international transaction costs. Casella and Rauch (2002) develop a model in which membership in a group—such as common ancestry or ethnicity—helps individuals in different countries reduce barriers to international trade associated with incomplete information. Relative to purely anonymous trade, the presence of group ties increases the volume of trade and GDP in the trading countries, though individuals lacking group ties are worse off (because they lose access to their more productive potential trading partners). Migration is an obvious mechanism through which cross- national group ties may be established and maintained over time. In empirical analysis of bilateral trade flows using the gravity model, there is a robust positive correlation between bilateral trade and bilateral migration, indicating that a pair of countries trade more today when there have been high levels of migration flows between them in the past. This finding has been interpreted as evidence of a “diaspora externality,” in which previous waves of migration create cross-national networks that facilitate commercial exchange. Gould (1994) finds that bilateral trade involving the United States is larger with countries that have larger U.S. immigrant populations. Head, Ries, and Swenson (1998) find that a 10 percent increase in Canada’s immigrant population from a particular country

HHassett.indbassett.indb 104104 111/8/121/8/12 9:279:27 PMPM IMMIGRATION, PRODUCTIVITY, AND COMPETITIVENESS 105

is associated with a 1 percent increase in bilateral Canadian exports and a 3 percent increase in bilateral Canadian imports. Pushing the analysis a step further, Rauch and Trindade (2002) focus specifically on networks associated with overseas Chinese populations. Successive waves of emigration from southeastern China have created communities of ethnic Chinese throughout Southeast Asia, as well as in South Asia and on the east coast of Africa. Rauch and Trindade find that bilateral trade is positively correlated with the interaction between the two countries’ Chinese populations (expressed as shares of the national population), similar to the findings in Gould and in Head, Ries, and Swenson. More interesting, the correlation between Chinese populations and trade is stronger for differentiated manufacturing products than it is for homogenous goods. To the extent that differentiated products are more subject to informational problems (Rauch 1999), these are the goods one would expect to be most sensitive to the presence of business networks. Rauch and Trindade focus on Chinese business networks owing to their expanse, which makes empirically identifying their impact on trade feasible. Presumably, international business networks are common to many migrant communities. In the United States, large concentrations of immigrants from Latin America, Asia, and Europe, among other regions, may contribute to deepening U.S. trade relations with these regions. One issue about the relationship between trade and migration is whether greater trade is the outcome of increased migration or a reflection of the types of individuals who select into migration. If more skilled and more able individuals are more likely to migrate abroad and more likely to exploit business opportunities, then the correlation between trade and migration may be a by-product of migrant self-selection. U.S. policies to liberalize immigration may not necessarily increase trade with sending countries unless they allow for the admission of individuals with a higher propensity to engage in trade. Consistent with this reasoning, Head, Ries, and Swenson (1998) find that immigrants admitted as refugees or on the basis of family ties with Canadian residents have a smaller effect on trade than immigrants admitted under a point system that values labor-market skills. Other evidence suggests that international migration may increase the flow of investment between countries. In China, India, and Taiwan, the

HHassett.indbassett.indb 105105 111/8/121/8/12 9:279:27 PMPM 106 RETHINKING COMPETITIVENESS

migration of skilled labor to Silicon Valley—where Indian and Chinese immigrants account for one-third of the engineering labor force—has been followed by increased trade with and investment from the United States (Saxenian 2002). By virtue of understanding business conditions in their home countries, these immigrants may be particularly well positioned to exploit investment opportunities abroad. Immigrants attracted today by the lure of engineering jobs in Silicon Valley may contribute tomorrow to investment by Silicon Valley firms in their home countries, which may then lead in the future to expanded trade flows between the United States and these countries.

Empirical Evidence on Immigration and Competitiveness The theoretical framework developed in the last section identifies the mechanisms through which immigration may impact the market shares of U.S. industries in the global economy. In this section, I review evidence on how immigration may have affected production costs and industrial productivity in the United States. In considering these mechanisms, I emphasize what research has to say about the effects of immigration on factor earnings. In the end, what matters is not just whether immigration causes U.S. global market shares to rise but whether individuals in the United States are made better off.

Immigration and Production Costs. Over the last two decades, the U.S. labor market has shifted in favor of utilizing more-skilled labor. Using www. Ipums.org data from U.S. population censuses and American Communities Surveys, I show in figures 4-1a and 4-1b the fraction of hours worked and the total wage bill accounted for by workers at different education levels. Between 1990 and 2008, the share in total employment of workers with a high school education or less fell sharply, whereas the share with a college education or more (bachelor’s degree, master’s or professional degree, or doctorate) rose sharply. Immigration, as we will see next, has attenuated the declining employment share of low-skilled workers and accelerated the rising employment share of high-skilled workers, leading the U.S. supply of labor to be more concentrated at the extremes of the skill distribution than it otherwise would have been.

HHassett.indbassett.indb 106106 111/8/121/8/12 9:279:27 PMPM IMMIGRATION, PRODUCTIVITY, AND COMPETITIVENESS 107

FIGURE 4-1 SHARE OF U.S. EMPLOYMENT BY EDUCATION GROUP

(a) Hours Worked

Less than HS

HS grad

Some college

BA degree

MA or Prof. degree 1990 PhD 2008

0.1.2.3

Share of Hours Worked

(b) Wage Bill

Less than HS

HS grad

Some college

BA degree

MA or Prof. degree 1990 PhD 2008

0.1.2.3

Share of Total Wages

SOURCE: Author’s calculations from 1990 U.S. Population Census and 2008 American Communities Survey, www.Ipums.org.

HHassett.indbassett.indb 107107 111/8/121/8/12 9:279:27 PMPM 108 RETHINKING COMPETITIVENESS

Figures 4-2a and 4-2b indicate how the presence of immigrants in the U.S. labor force has evolved over time. Each graph shows the share of hours worked accounted for by immigrants among individuals with a given level of education, where I plot these shares by year and labor-market experience groups (1 = 0–5 years of experience, 2 = 6–10 years of experience, 3 = 11–15 years of experience, 4 = 16–20 years of experience, 5 = 21–25 years of experience, 6 = 26–30 years of experience, 7 = 31–35 years of experience, and 8 = 36–40 years of experience). In all the graphs, the lines rise over time, with immigrant employment shares being higher in 1980 than 1970, higher in 1990 than 1980, and so on. Clearly, immigrants, both men and women, have become a more important part of the U.S. labor force across all education and experience levels. However, the size of the increase in the immigrant employment share differs sharply across education groups. The graphs for workers with less than a high school education show that between 1970 and 2008 the share of immigrants in hours worked rose from under 10 percent to more than 40 percent for most experience groups, and to more than 50 percent for groups with eleven to thirty years of experience. Among college graduates, there has also been a notable increase in the immigrant share of hours worked, rising from around 5 percent in 1970 to more than 15 percent in 2008. The importance of immigrants for changes in labor supply among high- skilled workers is seen more clearly in figures 4-3a and 4-3b, which breaks the college educated into four schooling groups. In 2008, among younger workers (with less than twenty years of labor-market experience) immigrants accounted for around 20 percent of hours worked among men with a professional degree, 20–30 percent of hours worked among men with a master’s degree, and more than 40 percent of hours worked among men a PhD. Figures for women are similar. Clearly, foreign-born workers are now a significant part of U.S. labor supply among the very highly educated. How does rising immigration change the wages of workers at different skill levels? According to a simple model of labor demand and labor supply, the wages for the most affected groups—the very low-skilled and the very high-skilled—should have been adversely affected by immigration.5 A decline in wages would lower production costs, aiding firms that employ immigrant labor relatively intensively. However, the labor-market consequences of international migration have inspired intense debate among

HHassett.indbassett.indb 108108 111/8/121/8/12 9:279:27 PMPM IMMIGRATION, PRODUCTIVITY, AND COMPETITIVENESS 109

FIGURE 4-2 IMMIGRANT SHARE OF U.S. EMPLOYMENT BY EDUCATION AND EXPERIENCE

(a) Males Less than High School High School .6 .5 .4 .3 .2 .1 0 Some College College .6 .5 .4 .3 .2 Immigrant Share of Employment .1 0 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 Labor Market Experience

1970 1980 1990 2000 2008

(b) Females Less than High School High School .6 .5 .4 .3 .2 .1 0 Some College College .6 .5 .4 .3 .2 Immigrant Share of Employment .1 0 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 Labor Market Experience

SOURCE: Author’s calculations from 1990 and 2000 U.S. Population Census and 2008 American Communities Survey, www.ipums.org.

HHassett.indbassett.indb 109109 111/8/121/8/12 9:279:27 PMPM 110 RETHINKING COMPETITIVENESS

FIGURE 4-3 IMMIGRANT SHARE OF U.S. EMPLOYMENT FOR HIGHLY EDUCATED WORKERS

(a) Males BA Degree MA Degree .5 .4 .3 .2 .1 0 Prof. Degree PhD .5 .4 .3 .2 Immigrant Share of Employment .1 0 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 Labor Market Experience

1990 2000 2008

(b) Females BA Degree MA Degree .4 .3 .2 .1 0 Prof. Degree PhD .4 .3 .2

Immigrant Share of Employment .1 0 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 Labor Market Experience

SOURCE: Author’s calculations from 1990 and 2000 U.S. Population Census and 2008 American Communities Survey, www.ipums.org.

HHassett.indbassett.indb 110110 111/8/121/8/12 9:279:27 PMPM IMMIGRATION, PRODUCTIVITY, AND COMPETITIVENESS 111

scholars, and there is not yet a consensus about the wage impacts of immigration in the United States. Borjas (2003) defines labor markets at the national level according to a worker’s education and labor-market experience, using the same groups as in figure 4-2. Over the period 1960 to 2000, education-experience cells in which immigrant labor supply growth has been larger—such as for young male high school dropouts—had slower wage growth, even after controlling for education or experience-specific wage shocks (i.e., including education- by-year and experience-by-year fixed effects in the estimation). Borjas’s evidence is consistent with immigration’s having depressed wages for low-skilled U.S. workers. A 10 percent increase in the population due to immigration leads to a 3 percent to 4 percent reduction in U.S. wages. The concern about this approach is that it might confound immigration with other labor-market shocks that have hurt low-skilled workers, such as skill-biased technological change. Absent controls for these other shocks, one cannot be sure that the attributed wage changes are really due to immigration. Applying a similar approach to Canada, Aydemir and Borjas (2007) find comparable evidence of the wage effects of immigration. In Canada, where immigration has been dominated by workers toward the top end of the skill distribution—owing to Canada’s use of a skill-based point system to regu- late immigration—immigration is negatively correlated with wages across education-experience cells, with more-educated workers being the ones who have suffered the largest negative wage effects from foreign labor inflows. Since Canada is presumably subject to many of the same technology shocks as the United States, it would not appear that unobserved technology shocks could explain away the wage effects of immigration in both countries. An older and larger literature has searched for immigration’s impact by correlating the change in wages for low-skilled U.S. natives with the change in the immigrant presence in local labor markets, typically at the level of U.S. cities. These area studies tend to find that immigration has little if any impact on U.S. wages. Card (2005) argues that if immigration has affected the U.S. wage structure, one should see larger declines in the wages of native high school dropouts, relative to native high school graduates, in U.S. cities where the relative supply of high school dropouts has expanded by more. In fact, the correlation between the relative wage and the relative supply of U.S. high school dropouts across U.S. cities is close to zero.

HHassett.indbassett.indb 111111 111/8/121/8/12 9:279:27 PMPM 112 RETHINKING COMPETITIVENESS

The area study approach also has its problems. Immigrants may tend to settle in U.S. regions in which job growth is stronger, causing one to underestimate the wage impact of immigration when using city- or state- level data. As a correction, many studies instrument for growth in local immigrant labor supply using lagged immigrant settlement patterns. But this strategy requires strong identifying assumptions. It would be invalid, for instance, if the labor-demand shocks that influence immigrant settlement patterns are persistent over time (Borjas, Freeman, and Katz 1997). One explanation for why low-skilled immigration may not have had larger impacts on U.S. wages is that immigration induces firms to change their investment decisions in a manner that offsets the wage effects. Lewis (2010) finds that in U.S. states experiencing more immigration, firms invested less heavily in automation machinery. Automation machinery replaces low-skilled labor, reducing the demand for their services. Thus, immigration may have induced firms to be less capital intensive in production than they otherwise would have been. Less investment in physical capital that replaces low-skilled labor may have helped mitigate the negative effect of immigration on wages. Lewis’s results apply to U.S. manufacturing over the 1988 to 1993 period. It is unclear how much they matter for other industries or time periods. Despite the conflicting evidence on wages, other results suggest that immigration has had significant economic impacts on prices in U.S. cities that have experienced larger influxes of immigrants. Cortes (2008) finds that in the 1980s and 1990s, U.S. cities with larger inflows of low-skilled immigrants saw larger reductions in prices for housekeeping, gardening, child care, dry cleaning, and other labor-intensive services. She finds that a 10 percent increase in the local population due to immigration is associated with decreases in prices for labor-intensive services of 1.3 percent. As prices for these services fall, highly educated women are likely to increase the number of hours they spend working. Cortes and Tessada (2009) find that as a consequence of U.S. low-skilled immigration over the period 1980–2000, highly educated women increased their weekly labor supply by approximately forty-five minutes, with women working the longest hours having the largest increases. A second impact of immigration is on the price of land. As immigrants are attracted to a location, the demand for housing rises, which may lead

HHassett.indbassett.indb 112112 111/8/121/8/12 9:279:27 PMPM IMMIGRATION, PRODUCTIVITY, AND COMPETITIVENESS 113

to increases in the price of housing, unless the housing supply is relatively elastic. Saiz (2007) examines the correlation between immigration and housing prices and finds that an increase in immigration equivalent to 1 percent of the population of a U.S. city is associated with a 1 percent increase in local housing values. A third impact of immigration is on local or national taxes. Immigration may affect production costs by altering tax rates faced by workers or businesses. Where immigrants pay more in taxes than they receive in government benefits, as is likely the case with high-skilled immigrants, immigration reduces the net tax burden on native taxpayers. Where immigrants pay less in taxes than they receive in government benefits, as is likely the case with low-skilled immigrants, immigration increases the net tax burden on natives. In the United States, immigrant households have historically made greater use of subsidized health care, income support to poor families, food stamps, and other types of public assistance (Borjas and Hilton 1996). Immigrant households tend to be larger than native households, have more children, and have very low incomes, making them eligible for means-tested benefits. In the last decade and a half, however, the difference between immigrant and native use of welfare programs in the United States has fallen or even reversed, largely because of welfare reform in 1996, which restricted noncitizens from having access to many federally funded benefit programs. While immigrant households still make greater use of public health care than native households (largely through their U.S.-born children), they make comparable or less use of other types of public assistance (Borjas 2003; Capps et al. 2005). Because immigrant-headed households tend to have more children of school age than native-headed households, increasing immigrant presence in a locality is likely to increase demand for public education. Thus, even if immigrant use of social services does not differ from that of natives, growth in the immigrant population increases school enrollments, placing a strain on state or local government finances. In 2004, the fraction of school-age children who had foreign-born mothers was 47 percent in California, 30 percent in Nevada, 29 percent in New York, 27 percent in Arizona and Florida, and 26 percent in New Jersey and Texas. Calculating the fiscal consequences of immigration precisely is difficult. One needs to know many details about the income, spending, and

HHassett.indbassett.indb 113113 111/8/121/8/12 9:279:27 PMPM 114 RETHINKING COMPETITIVENESS

employment of immigrants. There are few comprehensive analyses of the fiscal impact of immigration. In one of the few such studies, Smith and Edmonston (1997) estimate that in 1996, immigration imposed a short-run fiscal burden on the average U.S. native household of $200, equivalent to 0.2 percent of U.S. GDP.6 In the United States, the fiscal consequences of immigration appear to matter for individual preferences over immigration policy. Hanson, Scheve, and Slaughter (2007) find that U.S. natives who are more exposed to immigrant fiscal pressures—those living in states that have large immigrant populations and that provide immigrants access to generous public benefits—are more in favor of reducing immigration. This public-finance cleavage on immigration policy is strongest among natives with high earnings potential, who tend to be in high tax brackets. Thus, growth in the local immigrant population may contribute to political conflict over immigration policy, working through immigration’s impact on local public accounts. In sum, there is evidence that in the United States immigration has put downward pressure on the wages of low-skilled labor, high-skilled labor, and the prices of labor-intensive services and has put upward pressure on housing prices and the net tax burden facing native-headed households. Such changes help industries intensive in the use of very low-skilled or very high-skilled labor, but they hurt industries located in regions in which immigrants have caused state or local spending on education or social services to increase. Many U.S. regions are seeing both types of changes occur simultaneously— increases in low-skilled and high-skilled immigrant labor supply and increasing net fiscal burdens on local taxpayers—implying that which industries gain and which lose from immigration depends on specific details regarding industry skill intensity in employment and exposure to tax increases.

Immigration and Innovation. Clearly, immigration matters for the supply of human capital in the United States. Absent the arrival of foreign students and highly skilled foreign labor, the country would have far fewer engineers, scientists, and other technical professionals. The theoretical framework developed in the first section suggests that the presence of such labor affects innovation positively. I turn next to evaluate evidence on whether immigration has altered the scale and composition of innovative activities in the United States.

HHassett.indbassett.indb 114114 111/8/121/8/12 9:279:27 PMPM IMMIGRATION, PRODUCTIVITY, AND COMPETITIVENESS 115

FIGURE 4-4 STUDENT VISAS ISSUED BY THE U.S. GOVERNMENT

400,000

300,000

200,000 F Visas Issued

100,000

0 1970 1980 1990 2000 2010 Year

SOURCE: Institute for International Education, http://www.opendoors.iienetwork.org.

Highly skilled foreign individuals enter the U.S. labor force primarily through three channels: F-1 immigration visas, which permit them to study in the United States and work in U.S. universities; H-1B temporary immigration visas, which permit them to work for a specific U.S. employer for a period of three years (renewable once); or an employer-sponsored permanent immigrant visa, or green card. Other temporary immigration options for high-skilled labor exist but are small by comparison. Figure 4-4 shows the number of new F visas issued by the U.S. Department of State between 1966 and 2010.7 Until 2001, the number of foreign students shows a strong upward trend. Between 1990 and 2000, the number of visas increased by 31 percent. After the events of 9/11, the U.S. government put new restrictions in place that required individuals applying for F visas to undergo an inter- view at a U.S. consulate, and those wishing to study science or engineering to undergo an FBI background check (Alden 2009). The practical impact of these restrictions was a sharp reduction in visas issued, which fell from

HHassett.indbassett.indb 115115 111/8/121/8/12 9:279:27 PMPM 116 RETHINKING COMPETITIVENESS

FIGURE 4-5 FOREIGN GRADUATE STUDENTS IN U.S. UNIVERSITIES

300,000

250,000

200,000

150,000 Enrollment of Foreign Graduate Students 100,000 1985 1990 1995 2000 2005 2010 Year

SOURCE: Institute for International Education, http://www.opendoors.iienetwork.org.

284,000 in 2001 to 216,000 in 2003. The number of F visas has since recov- ered, but in the interim many foreign students likely have decided instead to study in their home countries or in Australia, Canada, France, Germany, or the United Kingdom, which are among the countries that compete with the United States in university education. One consequence of the visa restrictions was, not surprisingly, a reduction in the number of foreign students enrolled in U.S. universities. Figure 4-5 shows data from the Institute for International Education regarding U.S. enrollments for foreign graduate students. After increasing by 78 percent between 1990 and 2003, enrollment of foreign graduate students declined from 280,000 in 2003 to 266,000 in 2007, before recovering slightly. Given the effect of F visas on the flow of new students into the stock of enrollees, one would expect the 9/11 visa restrictions to reduce enrollments with a lag, as figure 4-5 suggests occurred. The 9/11 shock to F visas contributed to a downward trend in the U.S. share of global university education. Between 1970 and 2006, the U.S.

HHassett.indbassett.indb 116116 111/8/121/8/12 9:279:27 PMPM IMMIGRATION, PRODUCTIVITY, AND COMPETITIVENESS 117

share of enrollments in worldwide higher education fell from 29 percent to 12 percent (Freeman 2009). There are many reasons why a decline in U.S. market share for higher education is occurring. As other countries catch up to the United States in terms of income, they are likely to expand local university offerings. Perhaps more of a concern is that the United States is also losing market share in graduate training in science and engineering, fields that are directly related to the country’s capacity for innovation. Free- man shows that in 1975 the United States produced more than four times as many PhDs in science and engineering as did China, Japan, India, and Korea combined; by 2006 these countries produced 23 percent more science and engineering PhDs than the United States. Do foreign students affect U.S. productivity? Stuen, Mobarak, and Maskus (2010) examine the changing number of foreign science and engineering graduate students in U.S. universities over the period 1973–1998, focusing on the top one hundred U.S. schools. Graduate students are key inputs into university research, providing both grunt work as research assis- tants and substantive contributions as collaborators. Stuen, Mobarak, and Maskus report that in a recent survey of Science, a premier scientific journal, 87 percent of papers published had a student or postdoctoral fellow as a coauthor, with 60 percent of these coauthors being foreign born. In their analysis, they explore the fact that shocks in foreign countries, associated with macroeconomic crises, military conflicts, or changes in political regime, affect the ability of students from these countries to travel to the United States for graduate study. These shocks provide a source of exogenous variation in the supply of foreign graduate students in the United States. The authors find that a 10 percent reduction in a university’s share of science and engineering graduate students who are foreign born leads to a 5 percent to 6 percent reduction in research output, measured as publications in peer- reviewed journals or citations of published research. One interpretation of their result is that losing access to foreign students restricts U.S. university access to scientific talent, which reduces the productivity of U.S. researchers. For this interpretation to hold, it must be true that the domestic students who replace foreign students lost to turmoil in their home countries are less able, not as hard working, or less diverse in their intellectual perspectives. The emphasis on science and engineering is warranted by the fact that occupations related to these fields are at the heart of U.S. innovation. Hunt

HHassett.indbassett.indb 117117 111/8/121/8/12 9:279:27 PMPM 118 RETHINKING COMPETITIVENESS

and Gauthier-Loiselle (2008) use the 2003 National Survey of College Graduates to examine patenting activity by native and foreign-born university degree holders. U.S. immigrants patent at twice the rate that natives do, a difference that is explained entirely by the much higher propensity of the foreign born to hold degrees in science and engineering. They find further that for every 1 percent increase in the population made up by immigrant university graduates, patents per capita increase by 6 percent. They offer suggestive evidence that patenting by immigrants does not crowd out patenting by domestic workers (as would occur if immigrants simply captured jobs from native-born workers in which workers are more likely to generate new patents). Instead, there may be positive spillovers from immigrants to natives in innovation. The link of immigrants to university education in the United States turns out to be important for innovative outcomes. Immigrants who did not study in the United States and who arrived in the country at an older age do not exhibit the same positive patenting differential vis-à-vis natives (Hunt 2009). Of course, only a small fraction of foreign students in the United States turn out to be superstars in their chosen fields. But attracting these superstars appears related to regional economic performance. Zucker and Darby (2008) follow 5,400 highly cited scientists from the 1980s to the 2000s. Of this group, 62 percent reside in the United States, the large majority of them having graduated from U.S. universities. Attracting a star scientist to a U.S. region (say, when she is hired by a local university) is associated with a higher likelihood of firm entry in fields related to the scientist’s area of study. It is difficult to make causal inferences from this association, but it appears that innovation in universities may be related to innovation in local industries. After completing university training, foreign students must change their visa status in order to stay and work in the United States. The H-1B immigration visa is a common channel through which foreign students with U.S. university degrees enter the U.S. labor market. (Indeed, Kato and Spulber [2010] suggest that reductions in the supply of H-1B visas lower the academic quality of foreign applicants to U.S. universities.) Workers abroad recruited by employers with U.S. operations are another group that utilizes these visas. As noted earlier, an H-1B visa gives an individual the right to work in the United States for three years. Applications for visas are

HHassett.indbassett.indb 118118 111/8/121/8/12 9:279:27 PMPM IMMIGRATION, PRODUCTIVITY, AND COMPETITIVENESS 119

FIGURE 4-6 ANNUAL CAP ON NEW H-1B VISAS FOR SKILLED LABOR

250,000

200,000

150,000

100,000 Number of H-1B Visas 50,000

19901991199219931994199519961997199819992000200120022003200420052006200720082009 Year

SOURCE: U.S. Department of Homeland Security, Yearbook of Immigration Statistics, http://www.dhs. gov/files/statistics/publications/yearbook.shtm.

filed by a U.S. employer. Renewal for a second three-year period is generally automatic for those complying with visa guidelines. Figure 4-6 shows the number of new H-1B visas made available each year by the U.S. government. The program, which came into existence in 1990, initially made 65,000 visas available each year. The quota was raised to 115,000 in 1999 and to 195,000 in 2001 at the behest of employers in high-tech industries who demanded an increase in visa supplies during the U.S. technology boom of the late 1990s. In 2003 Congress allowed the quota to return to its original 65,000 level, where it remains today. Visas are awarded to applicants on a first-come, first-served basis. When the number of applications exceeds the quota, as occurred on the first day applications were accepted in both 2007 and 2008, the U.S. Department of Homeland Security awards visas by lottery. High-tech companies are the largest users of H-1B visas. Kerr and Lincoln (2010) report that between

HHassett.indbassett.indb 119119 111/8/121/8/12 9:279:27 PMPM 120 RETHINKING COMPETITIVENESS

1995 and 2008, science, engineering, and computer-related occupations accounted for 60 percent of H-1B visas. To hire a worker on an H-1B visa, an employer must apply for and receive a foreign labor certification from the U.S. Department of Labor, verifying that the foreign worker will not displace a U.S. worker. Department of Labor records show the number of foreign labor certifications sought by U.S. companies. In 2008, Cisco Systems applied for 4,747 certifications (for both new and renewing H-1B visas), Hewlett-Packard applied for 4,358, IBM applied for 2,601, Microsoft applied for 3,442, and Sun Microsystems applied for 1,834.8 To examine how H-1B visas affect U.S. innovation, Kerr and Lincoln (2010) explore the fact that the supply of visas has varied over time according to congressional changes in the visa quota. They examine patenting activity by individuals with identifiably Chinese or Indian names (given that patent applications do not list an individual’s immigration status). Over the period 2000–2005, 50 percent of H-1B visas went to individuals from India or China. Since the H-1B program was established, the share of patents awarded to individuals with Indian names increased from 3 percent to 5 percent, and the share going to individuals with Chinese names increased from 4 percent to 9 percent (trend growth in these shares was more modest, before 1990). The patent shares going to these two groups are substantially higher in fields related to computers, electronics, chemicals, or pharmaceuticals. Kerr and Lincoln allocate H-1B visas across U.S. cities according to where the employers who apply for the visas are located. The cities that attract most H-1B visa holders are San Francisco, Miami, Washington, D.C., Raleigh-Durham, Boston, Austin, and New York. A 10 percent increase in the population of H-1B visa holders is associated with a 1 percent to 4 percent increase in city patenting activity (for every standard deviation increase in a city’s share of H-1B visas). The authors also allocate H-1B visas across large employers according to their historic use of the visas. A 10 percent increase in the population of H-1B visa holders is associated with a 4 percent to 5 percent increase in company patenting activity (for every standard deviation increase in a firm’s share of H-1B visas). It thus appears that expanding the number of H-1B visas may increase patenting in companies that specialize in high-tech activities and in U.S. cities where high-tech companies are based. To remain in the United States after the H-1B visa expires, an individual must obtain a U.S. green card. This requires being sponsored either by a

HHassett.indbassett.indb 120120 111/8/121/8/12 9:279:27 PMPM IMMIGRATION, PRODUCTIVITY, AND COMPETITIVENESS 121

FIGURE 4-7 DISTRIBUTION OF LEGAL PERMANENT RESIDENCE VISAS, 1999–2008

Diversity Other 45,000 (4%) 37,000 (4%)

Refugees/asylees 12,000 (11%)

Immediate relatives of U.S. citizens 428,000 (44%)

Family sponsored 210,000 (21%)

Employer sponsored 149,000 (15%)

U.S. family member or by a U.S. employer (or seeking one of the small number of green cards made available through lottery). Though U.S.- educated immigrants who have U.S. work experience under an H-1B visa would seem to be natural candidates for an employer-sponsored green card, the limited supply of these visas means that workers often seek other paths to residency. The Immigration Act of 1990 set an overall annual cap on the number of green cards at 675,000, with specific quotas assigned to immigrants who are family sponsored (480,000) or employment based (140,000) or who are entering under the diversity program (55,000) (U.S. Department of Homeland Security 2009). Immediate relatives of U.S. citizens may obtain visas not subject to a cap. Figure 4-7 shows the allocation of permanent resident visas by category over the period 1999–2008. Two- thirds of visas went to family members of U.S. citizens or legal residents, with 44 percent going to immediate family members and 21 percent going to other family members. Only 15 percent of visas went to individuals

HHassett.indbassett.indb 121121 111/8/121/8/12 9:279:27 PMPM 122 RETHINKING COMPETITIVENESS

sponsored by their employers. Given the relative ease of obtaining visas through family connections, it is perhaps not surprising that among individuals obtaining a green card in 2003 who had first entered the United States on a student visa, 56 percent obtained the green card by being married to a U.S. citizen, whereas only 20 percent were sponsored by a U.S. employer (Rosenzweig 2006). It thus appears that obtaining permanent residence in the United States is enhanced not just by being a good student or managing to land a U.S. job but by succeeding in the dating market as well.

U.S. Immigration Policy and American Competitiveness The lessons from the first section are that U.S. industries capture larger global market shares when their export supply capacity expands. Export supply capacity, in turn, depends on U.S. productivity levels, production costs, and international trade costs. Immigration may lower wages, and therefore production costs, for employers who use intensively the skills that immigrants bring with them. It may also accelerate productivity growth if the arrival of foreign workers expands the supply of R&D labor that is used intensively in innovation. Furthermore, immigration may lower the costs of international trade by reducing informational barriers. Empirical evidence discussed in the first two sections suggests that all these mechanisms are operative. Immigration (probably) lowers wages for native workers who compete with immigrants for jobs and also lowers prices of labor-intensive services. At the same time, immigration tends to bid up housing prices and to increase tax burdens on native taxpayers, at least in regions that have a large population of low-skilled immigrants and that provide relatively generous services to them. Perhaps the most striking and underappreciated evidence about immigration is its link to innovation. Immigrants are far more likely than natives to study science and engineering and as a consequence are more likely to produce innovations in the form of patents. Expanding the supply of high-skilled immigrant workers expands patenting activity in science and engineering fields, particularly in high-tech firms and in U.S. regions that house these firms. There is both a strong theoretical and empirical case to be made that immigration improves U.S. competitiveness in high-tech industries (in the sense that it would expand the share of global demand for high-tech

HHassett.indbassett.indb 122122 111/8/121/8/12 9:279:27 PMPM IMMIGRATION, PRODUCTIVITY, AND COMPETITIVENESS 123

FIGURE 4-8 COMPOSITION OF U.S. FOREIGN-BORN POPULATION BY LEGAL STATUS, 2007

Unauthorized Aliens Legal Permanent 11.9 million (30%) Resident Aliens 12.3 million (31%)

Temporary Legal Residents 1.4 million (4%)

Naturalized Citizens 14.2 million (36%)

products met by U.S. firms). To what extent does U.S. policy incorporate the effects of immigration on competitiveness into decisions about how many visas to allot and to whom to allot them? The short answer is, not much. As we see in figure 4-7, the United States places much more weight on family reunification than it does on employability in allocating permanent residence visas. Figure 4-8 shows the allocation of visas by legal status. Fully 30 percent of U.S. immigrants are unauthorized, with most of these individuals being low skilled. The United States thus appears to favor relatively high levels of low-skilled illegal immigration and immigration of U.S. family members. Furthermore, levels of high-skilled immigration are only weakly responsive to economic conditions. It is clear in figure 4-6 that Congress did not expand H-1B visas until the 1990s technology boom was nearly over, then kept visas high during the 2001–2002 recession (when demand for visas was low), and finally reduced visa supply in 2003 just as the U.S. economy (and visa demand) was beginning to expand once again, yielding binding visa quotas shortly thereafter.

HHassett.indbassett.indb 123123 111/8/121/8/12 9:279:27 PMPM 124 RETHINKING COMPETITIVENESS

FIGURE 4-9 PERCENT OF FOREIGN BORN IN THE NATIONAL POPULATION

30 1999 25 2007

15 Percent

Spain France Austria Ireland Canada Belgium Australia Switzerland Netherlands United States New Zealand United Kingdom

SOURCE: International Migration Outlook, Organisation for Economic Co-operation and Develop- ment, 2011.

It thus appears that the United States tries hard to divert opportunities for immigration away from high-skilled workers. However, things are not as bad as they seem, especially when compared to other high-income countries. In luring high-skilled immigrants, it is the attractiveness of the United States relative to other countries that matters. Relative attractiveness depends, in part, on U.S. wages and the malleability of U.S. admission policies. Compared to other countries, the United States does well on both counts. Figure 4-9 shows immigrants as a share of the population in the United States, Europe, Australia, Canada, and New Zealand. Immigration is on the rise across the board, showing the strongest increases in Ireland and Spain, which are the European countries that until 2008 had both relatively fast income growth and booming construction sectors. Despite U.S. policies, the United States compares favorably in the skill profile of the

HHassett.indbassett.indb 124124 111/8/121/8/12 9:279:27 PMPM IMMIGRATION, PRODUCTIVITY, AND COMPETITIVENESS 125

FIGURE 4-10 PERCENT OF IMMIGRANT POPULATION WITH 13-PLUS YEARS OF EDUCATION, 2000

30 Percent

Italy Japan Spain Canada France Australia Germany United States Netherlands United Kingdom

SOURCE: International Migration Outlook, Organisation for Economic Co-operation and Develop- ment, 2011.

immigrants it attracts. Figure 4-10 shows the share of the immigrant population with thirteen or more years of schooling in high-income countries. The tertiary educated are nearly 60 percent of the foreign-born population in Canada, as a consequence of the country’s point system that gives preference to more skilled individuals in granting immigration visas. The United States has the second most educated immigrant population, with 42 percent of U.S. immigrants having thirteen or more years of schooling. In Europe, education levels among immigrants are far lower, with less than 30 percent of immigrants having tertiary education in all countries. What accounts for the skill profile of U.S. immigrants relative to other nations? Grogger and Hanson (2011) examine the factors that affect the skill composition of bilateral migration flows into OECD economies. They find that countries in which the reward to being a skilled worker is relatively

HHassett.indbassett.indb 125125 111/8/121/8/12 9:279:27 PMPM 126 RETHINKING COMPETITIVENESS

large attract a disproportionate share of more-educated emigrants. In the United States and Canada, the difference in earnings between high-skilled and low-skilled workers is much greater than in continental Europe, with this difference being even larger once we account for Europe’s more progressive tax system and generous welfare benefits. Compared to Europe, the United States has relatively high inequality in pretax earnings, with Europe’s tax and welfare policies making the United States appear even more unequal when earnings are evaluated after taxes. The United States and Canada are home to 51 percent of the OECD’s immigrants but 66 percent of its immigrants with tertiary schooling. Europe, in contrast, receives 38 percent of the OECD’s immigrants but only 24 percent of those who are tertiary educated. Skilled immigrants find their way into the United States—in spite of the country’s immigration policies—because the reward for skill in the country is so large. Canada attracts skilled immigrants by making its policies as favorable toward them as any high-income country. Making U.S. immigration policy more like Canada’s would further enhance the attractiveness of the United States as a destination, given the already large advantage afforded by the U.S. wage structure. Another dimension in which the United States performs relatively well in its immigration policy is in addressing changes in the demographic structure of the labor force. Throughout the rich world, fertility rates have been dropping sharply over the last several decades, such that Korea, Japan, Singapore, and most countries in Europe have population growth rates that are well below replacement. As a consequence, the size of the native-born labor force is beginning to decline. One possible role for immigration policy is to ease adjustment to a future with smaller populations; before long, the population of retired workers will be large relative to the population of active workers (with corresponding pressures on government budget deficits to grow to cover public pension liabilities). In the United States, past high levels of immigration have helped keep birth rates relatively high, allowing the U.S. population to continue to expand. Figure 4-11 shows the change in the working-age population that will result in countries between 2005 and 2020 if they allow no further immigration. Japan’s working-age population will fall by more than 10 percent, and Italy’s and Germany’s will fall by more than 5 percent. In the United States, in contrast, the working- age population will expand by more than 5 percent, even with no additional

HHassett.indbassett.indb 126126 111/8/121/8/12 9:279:27 PMPM IMMIGRATION, PRODUCTIVITY, AND COMPETITIVENESS 127

FIGURE 4-11 CHANGE IN WORKING-AGE POPULATION 2005–2020 WITH ZERO IMMIGRATION

–5

Percent of 2005 Population –10

–15

Japan Italy Spain Greece Canada France Ireland Germany PortugalBelgium Australia Netherlands United NewStates Zealand United Kingdom

SOURCE: International Migration Outlook, Organisation for Economic Co-operation and Develop- ment, 2011.

inflows of foreign workers. The United States faces comparatively minor challenges in adjusting to changes in the age structure of its population. Comparisons to other countries suggest that whatever the United States is getting wrong in its immigration policy, other countries may be doing much worse. Still, were the country to expand the objectives of immigration policy to include promoting U.S. competitiveness, the changes would be profound. Shifting immigration visas toward high-skilled workers would have the potential to expand the U.S. science and engineering labor force, to increase the rate at which U.S. companies create new patents, and to increase the rate of productivity growth in the U.S. economy. In the process, the United States would likely shift resources toward production activities that are intensive in the use of science and engineering talent and away from other activities. Simply reducing visas to family members of U.S. residents

HHassett.indbassett.indb 127127 111/8/121/8/12 9:279:27 PMPM 128 RETHINKING COMPETITIVENESS

by one-quarter (roughly 150,000 visas a year) would allow the number of employer-sponsored green cards to double. Reducing family-based visas by one-half would allow employer-sponsored visas to triple. Relatively modest changes in immigration policy could then have profound effects on U.S. competitiveness, if the redirected (or expanded) immigration visas were targeted toward foreign students with the potential to be stars in science and engineering. The United States does reasonably well in the global competition for talent, but it could do much better.

Notes 1. I calculate labor supply by summing the wage income of workers in engineering and computer science occupations as listed in the 2008 American Communities Survey, www.ipums.org.. 2. To see that competitiveness is not isomorphic to income, consider two alternative shocks to the U.S. commercial aircraft industry. First, imagine that Boeing develops a new aircraft engine that allows it to capture market share from Airbus. The improvement in U.S. competitiveness and the increase in U.S. incomes (from Boeing’s enhanced profitability) would go hand in hand. Next, suppose that a new U.S. firm enters commercial aircraft production, expanding U.S. market share in the industry but driving down product prices, thereby reducing firm profitability in the sector. In this case, U.S. competitiveness (i.e., market share) rises even as U.S. incomes may fall. A further caveat is that in evaluating changes in competitiveness, one wants to hold constant a country’s trade balance. An increase in global market shares associated with rising net exports (implying the deferral of consumption from today until the future) is obviously much different from an increase in market shares with no change in the trade balance (implying no deferral of consumption). The former merely moves exports from the future to today, whereas the latter implies an increase in the steady- state level of exports. 3. For high-skilled immigration to enhance U.S. innovation in this manner, it must be that the gains from innovation are not spread evenly across the globe but captured disproportionately by the United States (as would occur if spillovers from innovation are localized or if innovating firms earn profits at the expense of global consumers). 4. In linking immigration and competitiveness, I am assuming implicitly that immigration does not worsen the U.S. trade balance, an assumption that appears reasonable, given the absence of obvious reasons for which high-skilled immigration would induce U.S. households to reduce savings. 5. The prediction for high-skilled workers depends on highly educated immigrants competing for the same jobs as highly educated native workers. For analysis of occupational choices by highly educated immigrants, see Chiswick and Miller (2011).

HHassett.indbassett.indb 128128 111/8/121/8/12 9:279:27 PMPM IMMIGRATION, PRODUCTIVITY, AND COMPETITIVENESS 129

6. This estimate is based on short-run considerations. Allowing for dynamics could change the results. 7. For analysis of U.S. student visa flows over time, see Lowell and Khadka (2011). 8. Business services companies are the largest users of H-1B visas. These entities in effect subcontract H-1B workers out to U.S. companies that need their services; they appear to attract relatively low-level software programmers. They include multinationals from India (e.g., Wipro, Infosys Technologies, McPhasiS, and Satyam Com- puter Systems) and U.S.-based companies (e.g., Xceltech, Enterprise Business Solu- tions, Business Intelligence Systems).

References Alden, Edward. 2009. The Closing of the American Border: Terrorism, Immigration and Security since 9/11. Washington, D.C.: Harper Perennial. Arrington, Michael. 2010. “Google Secretly Invested $100+ million in Zynga, Pre- paring to Launch Google Games.” TechCrunch, July 10. http://techcrunch. com/2010/07/10. Ashenfelter, Orley C., Kirk B. Doran, and Bruce Schaller. 2010. A Shred of Credible Evi- dence on the Long Run Elasticity of Labor Supply. NBER Working Paper No. 15746. National Bureau of Economic Research, Cambridge, MA. Aydemir, Abdurrahman, and George J. Borjas. 2007. “Cross-Country Variation in the Impact of International Migration: Canada, Mexico, and the U.S.” Journal of the European Economic Association 5(4):663–708. Borjas, George J. 2003. “The Labor Demand Curve Is Downward Sloping: Reexamin- ing the Impact of Immigration on the Labor Market.” Quarterly Journal of Economics 118:1335–74. Borjas, George J., Richard B. Freeman, and Lawrence F. Katz. 1997. “How Much Do Immigration and Trade Affect Labor Market Outcomes?” Brookings Papers on Eco- nomic Activity 1:1–90. Borjas, George J., and Lynette Hilton. 1996. “Immigration and the Welfare State: Immigrant Participation in Means-Tested Entitlement Programs.” Quarterly Journal of Economics 111(2):575–604. Bound, John, Sarah Turner, and Patrick Walsh. 2009. Internationalization of U.S. Doc- torate Education. NBER Working Paper No. 14792. National Bureau of Economic Research, Cambridge, MA. Camarota, Steven A. 2004. The High Cost of Cheap Labor: Illegal Immigration and the Federal Budget. Washington, D.C.: Center for Immigration Studies. Capps, Randall, Michael Fix, Everett Henderson, and Jane Reardon-Anderson. 2005. A Profile of Low-Income Working Immigrant Families. Working Paper B-67. Urban Institute, Washington, D.C. Card, David. 2001. “Immigrant Inflows, Native Outflows, and the Local Labor Market Impacts of Higher Immigration.” Journal of Labor Economics 19(1):22–64.

HHassett.indbassett.indb 129129 111/8/121/8/12 9:279:27 PMPM 130 RETHINKING COMPETITIVENESS

———. 2005. “Is the New Immigration Really So Bad?” Economic Journal 115(507):300–323. Casella, Alessandra, and James E. Rauch. 2002. “Anonymous and Group Ties in Inter- national Trade.” Journal of International Economics 58:19–47. Chiswick, Barry R., and Paul W. Miller. 2011. “Educational Mismatch: Are High Skilled Immigrants Really Working in High Skilled Jobs, and What Price Do They Pay If They Are Not?” In High Skilled Immigration in a Global Labor Market, ed. B. R. Chiswick. Washington, D.C.: AEI Press, 111–54. Cortes, Patricia. 2008. “The Effect of Low-Skilled Immigration on U.S. Prices: Evi- dence from CPI Data.” Journal of Political Economy 116(3):381–422. Cortes, Patricia, and Jose Tessada. 2009. Low-Skilled Immigration and the Labor Supply of Highly Educated Women. Mimeo, Boston University. Eaton, Jonathan, and Samuel Kortum. 2002. “Technology, Geography, and Trade.” Econometrica 70(5):1741–79. ———. 2009. Technology in the Global Economy: A Framework for Quantitative Analysis. Mimeo, University of Chicago. Freeman, Richard B. 2009. What Does Global Expansion of Higher Education Mean for the U.S.? NBER Working Paper No. 14962. National Bureau of Economic Research, Cambridge, MA. Grogger, Jeffrey, and Gordon H. Hanson. 2011. “Income Maximization and the Selection and Sorting of International Migrants.” Journal of Development Economics 95(1):42–57. Grossman, Gene M., and Elhanan Helpman. 1991. “Quality Ladders in the Theory of Growth.” Review of Economic Studies 58:43–61. Gould, David M. 1994. “Immigration Links to the Home Country: Empirical Implica- tions for U.S. Bilateral Trade Flows.” Review of Economics and Statistics 76:302–16. Hanson, Gordon H. 2005. Why Does Immigration Divide America? Washington, D.C.: Institute for International Economics. ———. 2007. The Economic Logic of Illegal Immigration. Council Special Report No. 26. Council on Foreign Relations, New York. ———. 2009. “The Economic Consequences of International Migration.” Annual Review of Economics 1:179–208. Hanson, Gordon H., Kenneth Scheve, and Matthew J. Slaughter. 2007. “Local Public Finance and Individual Preferences over Globalization Strategies.” Economics and Politics 19:1–33. Head, Keith, John Ries, and Deborah Swenson. 1998. “Immigration and Trade Creation: Econometric Evidence from Canada.” Canadian Journal of Economics 31(1):47–62. Hunt, Jennifer. 2009. Which Immigrants Are Most Innovative and Entrepreneurial? Mimeo, McGill University. Hunt, Jennifer, and Marjolaine Gauthier-Loiselle. 2008. How Much Does Immigration Boost Innovation? NBER Working Paper No. 14313. National Bureau of Economic Research, Cambridge, MA.

HHassett.indbassett.indb 130130 111/8/121/8/12 9:279:27 PMPM IMMIGRATION, PRODUCTIVITY, AND COMPETITIVENESS 131

Jones, Charles. 1995. “R&D Based Models of Economic Growth.” Journal of Political Economy 103(4):759–84. Kato, Takao, and Chad Spulber. 2010. Quotas and Quality: The Effect of H-1B Visa Restrictions on the Pool of Prospective Undergraduate Students from Abroad. IZA Dis- cussion Paper No. 4951. IZA, Bonn, Germany. Kerr, William, and William Lincoln. 2010. The Supply Side of Innovation: H-1B Visa Reforms and U.S. Ethnic Invention. NBER Working Paper No. 15768. National Bureau of Economic Research, Cambridge, MA. Lewis, Ethan. 2010. “Immigration, Skill Mix, and Capital-Skill Complementarity.” Quarterly Journal of Economics 126(2):1029–69. Lowell, B. Lindsay, and Pramod Khadka. 2011. “Trends in Foreign Student Admis- sions in the United States: Policy and Competitive Effects.” In High-Skilled Immi- gration in a Global Labor Market, ed. B. R. Chiswick. Washington, D.C.: AEI Press, 83–108. Passel, Jeffrey, and D’Vera Cohn. 2010. U.S. Unauthorized Inflows Are Down Sharply since Mid-Decade. Washington, D.C.: Pew Center for Hispanic Studies. Rauch, James E. 1999. “Networks versus Markets in International Trade.” Journal of International Economics 48(1):7–35. Rauch, James E., and Vitor Trindade. 2002. “Ethnic Chinese Networks in Interna- tional Trade,” Review of Economics and Statistics 84(1):116–30. Rosenzweig, Mark. 2006. “Global Wage Differences and International Student Flows.” Brookings Trade Forum 2006: 57–86. Saiz, Alberto. 2007. “Immigration and Housing Rents in American Cities.” Journal of Urban Economics 61(2):345–71. Saxenian, AnnaLee. 2002. Local and Global Networks of Immigrant Professionals in Sili- con Valley. Monograph, Public Policy Institute of California, San Francisco. Smith, James P., and Barry Edmonston, eds. 1997. The New Americans: Economic, Demographic, and Fiscal Effects of Immigration. Washington, D.C.: National Acad- emy Press. Stuen, Eric T., Ahmed Musfiq Mobarak, and Keith Maskus. 2010. Skilled Immigration and Innovation: Evidence from Enrollment Fluctuations in U.S. Universities. CEPR Dis- cussion Paper No. 7709. Centre for Economic Policy Research, London. Zucker, Lynne G., and Michael R. Darby. 2008. Star Scientists, Innovation, and Regional and National Immigration. NBER Working Paper No. 13547. National Bureau of Economic Research, Cambridge, MA.

HHassett.indbassett.indb 131131 111/8/121/8/12 9:279:27 PMPM 5

The Role of Innovation and Intellectual Property in Economic Competition Robert J. Shapiro1

Among both advanced economies and developing nations, innovations and the intellectual property embodied in them play central roles in economic competition. The development and effective application of innovations can determine the outcome of the competition between domestic companies within an economy and between one nation’s domestic firms and its foreign rivals in both domestic and third-country markets. The pace at which valuable innovations are developed, and how quickly and broadly those innovations are adopted and applied across an economy, also greatly influence how much a country’s GDP, productivity, and incomes increase, both absolutely and compared to other nations. The competition between firms is echoed to some degree by an implicit competition between nations to be the home base for the intellectual property rights associated with the development and application of innovations. Numerous factors affect where firms decide to locate their operations. These factors include, beyond where the major markets for its innovative goods and services are located, the quality of its colleges, universities, and scientific institutions; the strength of its intellectual property protections and their enforcement; the regulatory treatment of the firms and institutions that develop intellectual property; and the tax treatment of the profits derived from intellectual property. A nation can benefit greatly from attracting the intellectual-property-generating activities of foreign-based as well as domestic firms. In this context, one can say that nations around the world may well be engaged in a competition around innovation that policy analysts should monitor closely.

132

HHassett.indbassett.indb 132132 111/8/121/8/12 9:279:27 PMPM THE ROLE OF INNOVATION AND INTELLECTUAL PROPERTY 133

The Economic Value of Innovation

The critical importance of innovation and the underlying intellectual property reflects basic features of markets. Analyses of how markets operate usually begin with labor and capital, both physical and financial. However, the economic value of labor and physical capital also depends ultimately on ideas. Equipment or buildings embody ideas of how to construct them, while the value of most labor depends on the worker’s education and skills. Moreover, the economic importance of innovation lies not only in the stock of economically valuable ideas but also in a continuing process of generating new ideas that can be embodied in new innovations. Innovations generate economic value only when they are adopted and applied. All nations therefore have an interest in promoting openness to innovations, wherever they were originally developed, because most of the economic benefits of innovation are enjoyed by those who use them. There is little doubt that the benefits to workers and corporations from using the Windows operating system far exceed Microsoft’s profits, or that cancer medications provide greater benefits to those who use them and their societies, in productive lives prolonged or saved, than the profits earned by the pharmaceutical makers that develop and patent them. And both the development and application of economically powerful innovations depend on legal and regulatory arrangements that protect the intellectual property rights of those who develop and transfer them, thereby enabling them to profit from doing so. As Paul Romer (1993), a leading theorist of the sources of economic growth, has written, The knowledge needed to provide citizens of the poorest countries with a vastly improved standard of living already exists in the advanced countries. If a poor nation invests in education and does not destroy the incentives for its citizens to acquire ideas from the rest of the world, it can rapidly take advantage of the publicly available part of the worldwide stock of knowledge. If, in addition, it offers incentives for privately held ideas to be put to use within its borders (for example, by protecting foreign patents, copyrights and licenses, and by permitting direct investment by foreign firms), its citizens can soon work in state-of-the-art productive activities. (p. 186)

HHassett.indbassett.indb 133133 111/8/121/8/12 9:279:27 PMPM 134 RETHINKING COMPETITIVENESS

In most cases, therefore, a society’s openness to innovation is more important in determining how fast it develops and grows than its natural resources.2 From 1960 to 2000, for example, economic output and per capita incomes grew more than three times faster in South Korea, with relatively few natural resources, than in Brazil, a country with abundant natural resources (World Bank 2005). Much of the difference can be traced to Korea’s relative openness to technological and other innovations developed elsewhere and imported to Korea through foreign direct investment (FDI) or licensing agreements, as well as to Korea’s greater commitment to broad-based public education that prepared its future workers to adapt to new technologies and ways of doing business (Shapiro and Hassett 2005). Over the last half century, economists have documented the role that innovation and intellectual property play in the economic growth and progress of the United States, the world’s most successful advanced economy. Since the 1950s, researchers starting with Robert Solow (1956, 1957) have established that the development and adoption of economic innovation have been the single most powerful factor determining America’s underlying rate of growth and productivity in the twentieth century. Solow, who was awarded the Nobel Prize for this work, and Edward Denison (1962) and others have estimated that 30 percent to 40 percent of the gains in productivity and growth achieved by the United States from 1900 to the 1980s can be traced to economic innovation in its various forms. These innovations encompass the development of not only new technologies but also new materials and processes; new ways of financing, marketing and distributing goods and services; and new ways of managing a workplace and organizing a business. The dominant role of new ideas in economic life has also been documented in a recent study from the Federal Reserve System by Corrado, Hulten, and Sichel (2004). The authors analyzed the various ways that companies use ideas by examining several categories of business spending on so-called intangibles, including spending on software programs and databases; expenditures for scientific and nonscientific R&D; spending on new- product development by service firms; spending on advertising and market research to create brands; expenditures on developing new business models and corporate cultures; and expenditures on firm-specific training. By clas- sifying these outlays as capital investments intended to increase future earnings, rather than as spending on current output, the authors found that U.S.

HHassett.indbassett.indb 134134 111/8/121/8/12 9:279:27 PMPM THE ROLE OF INNOVATION AND INTELLECTUAL PROPERTY 135

businesses invest as much in these idea-related intangibles—about $1 trillion a year in the early years of this decade—as they do in plant, equipment, and other traditional, tangible forms of investment. The Federal Reserve study further established that in the 1990s, U.S. business spending on long-lasting, knowledge capital grew faster than any other type of business or personal spending. Finally, they traced more than four-fifths of the gains in U.S. productivity in the latter 1990s to the development and application of new ideas, especially involving advances in information technology (IT). Over the years 1995–2001, the development of new information technologies accounted for 28 percent of those gains, capital investment in those technologies accounted for another 34 percent, research and development accounted for 10 percent, and changes in the organization of firms and worker training in response to innovations accounted for another 10 percent (Corrado, Hulten, and Sichel 2004, table 3). By applying this approach to data for 2001–2003, other researchers have estimated that nearly 90 percent of U.S. economic growth during that more recent period can be attributed to increases in the stock of intangible assets (Shapiro and Hassett 2005). Another recent study affirmed these results by reexamining the impact of innovation on U.S. growth during the last quarter of the twentieth century (Van Ark et al. 2009). The authors applied a version of Solow’s growth accounting to two periods—1973 to 1995 and 1995 to 2003—and found that the impact of innovation on growth increased significantly from the first period to the second. They found that the contributions to growth from improvements in labor quality (skills and education) and traditional capital investment in plant and non-IT equipment fell from 30 percent in the first period to 19 percent in the more recent period. The impact on growth of investments in IT capital remained stable—18 percent in the first period and 19 percent in more recent years—as did the contribution of intangibles. However, the contribution of “multifactor productivity,” largely a proxy for the development and application of innovations, increased from 25 percent to 35 percent.

The Value of Intellectual Property Rights Recent research also has explored why most economically powerful innovations originate in advanced economies: innovation thrives in places where

HHassett.indbassett.indb 135135 111/8/121/8/12 9:279:27 PMPM 136 RETHINKING COMPETITIVENESS

commitments to research and development are strong, the political and economic environments are stable, barriers to starting new businesses are relatively low, and intellectual property rights are respected and enforced. Similarly, the most successful developing economies are those that promote the specific economic and political conditions that support the importation of innovations, usually through FDI, and their adoption. In the words of two leading analysts of economic development, “Most of income above sub- sistence is made possible by international diffusion of knowledge” (Klenow and Rodriquez-Clare 2004, p. i). In this regard, the World Bank (2005) reports that since 1980, the world’s greatest economic gains have been achieved by developing nations that aggressively opened their economies to foreign technologies and business methods and then protected the intellectual property rights of their developers. From 1980 to 2002, the developing East Asian economies achieved growth averaging more than 7.4 percent per year, and the developing South Asian economies grew by an average of more than 5.4 percent annually. By contrast, Latin American economies with more restrictive FDI policies and lax intellectual property protections grew by less than 2.5 percent per year over the same period. More generally, another World Bank study of patenting and growth in ninety-two countries over the period 1960 to 2000 found that a 20 percent increase in the annual number of patents granted, wherever the technologies originated, was associated with an increase of 3.8 percent in output (Chen and Dahlman 2004). The role of legal protections for the intellectual property embodied in innovation has also been examined in a recent study of China (Ang et al. 2009). Researchers looked at high-tech firms in China across a number of provinces, controlling for local factors such as government corruption and banking and legal system development. They found that high-tech firms in provinces that enforced intellectual property rights were more successful in securing external debt and equity financing, invested more in R&D, and participated in more joint ventures than their counterparts in provinces that failed to strictly enforce intellectual property rights. It may seem obvious that rates of innovation fall sharply when intellectual property rights are weak or absent, since the innovations developed by one person or company could then be copied and reproduced without compensation by any other company in any country. Yet, some analysts

HHassett.indbassett.indb 136136 111/8/121/8/12 9:279:27 PMPM THE ROLE OF INNOVATION AND INTELLECTUAL PROPERTY 137

have raised questions about the ultimate utility of traditional intellectual property rights for developing countries, including appeals to the World Intellectual Property Organization (WIPO) arguing that those rights actually impair economic progress in many of those nations (Group of Friends of Development 2005). In fact, the relationship between innovation and intellectual property rights is well established in modern economics. To be sure, invention for its own sake does not always depend on those rights. Recent research based on data from the catalogues of inventions displayed at three nineteenth- century world fairs, for example, found that inventors remained active in Denmark and Switzerland during periods in the nineteenth century when those nations lacked intellectual property protections (Moser 2003). How- ever, the researcher also found that the inventions developed in places without strong patent protections mainly involved products in which the inventor could preserve secrecy without a patent, such as new foodstuffs with “secret recipes.” Most inventions that ultimately hastened economic development and lifted living standards—especially new technologies and manufacturing processes—were developed in societies with strong intellectual-property protections in the nineteenth century, most notably the United States and Germany. Part of the explanation lies in how people normally respond to economic incentives. While a few souls are genuine altruists, most innovators expend the effort to develop something that provides economic benefits to others only if doing so also benefits themselves. Moreover, as corporations have come to dominate the development of innovations, the prospect of future gains as an essential incentive to innovate has become conclusive. The development of most modern economic innovations, especially in technology areas, requires highly skilled people and sophisticated equipment and business organizations, and the use of all that labor and capital is costly. The reason that certain businesses are willing to bear those costs and forgo more immediate and certain returns from using their labor and capital in other ways is a prospect of larger returns in the future, protected by patents or copyrights. The rest of the explanation lies in the essential, economic nature of ideas. The ideas that animate economic innovations are what economists call “nonrival goods.” Unlike such “rival goods” as a piece of equipment or

HHassett.indbassett.indb 137137 111/8/121/8/12 9:279:27 PMPM 138 RETHINKING COMPETITIVENESS

real estate, an idea can be used by more than one person at a time and is easily duplicated. Because an idea cannot be physically possessed like land or equipment, its use by those who develop it does not preclude others using it at the same time.3 As a result, the returns from innovations cannot be secure without legal protections for the new ideas that animate them. This dynamic is evident in the results of a survey of American R&D executives in the pharmaceutical sector, who reported that without patent protections, at least 60 percent of the projects that ultimately produced new discoveries would never have been undertaken (Mansfield 1986).4 Nations that are attractive places for innovation benefit from being so, and the actions of other countries can affect how much domestic innovation-related activity occurs in a nation. In this regard, some nations are involved in a complex game of moves and countermoves. For example, the Organisation for Economic Co-operation and Development (OECD) (2007) estimates that counterfeiting and piracy, especially in foreign markets, cost companies as much as $200 billion in 2005 and notes that the methodology used to derive this estimate may have undercounted those costs by hundreds of billions of dollars. These losses undercut the willingness of firms to invest large sums to develop new goods and services, since a substantial share of the foreign market will be lost to intellectual property counterfeiters and pirates. These dynamics also operate in the opposite direction: researchers have found that strong intellectual property protections in developing nations directly stimulate the pace of innovation in advanced economies (Diwan and Rodrik 1991). Innovating firms clearly benefit when they can extract more rents from their intellectual property—that is, claim greater revenues and profits—by extending into new markets. But they also have to weigh the prospect of those gains against the increased risks of counterfeiting and piracy. More- over, there is overwhelming evidence that developing nations benefit from cracking down on piracy and counterfeiting. Researchers have established that technology transfers to developing nations increase as those nations adopt and enforce strict patent protections, especially when the domestic companies in those nations compete with foreign innovators. Studies have also shown that increasing patent protections in a developing nation leads to significant increases in overall U.S. investment in that country (Maskus 1994; Taylor 1994).

HHassett.indbassett.indb 138138 111/8/121/8/12 9:279:27 PMPM THE ROLE OF INNOVATION AND INTELLECTUAL PROPERTY 139

How Innovations Diffuse across Economies

While intellectual property protections promote the development of innovations, their impact on a nation’s overall productivity, and thus its competitiveness, depends on the extent to which a country’s businesses and households adopt them and how effectively they use them. Broad adoption of innovations, however, often takes many years (see, for example, Hall 2004). Jet engine travel, for example, spread slowly because travel on the early jetliners was very expensive and offered travelers only modest advantages compared to prop airplane travel. Moreover, the adoption of an innovation always involves significant costs, which may include new training and organizational changes as well as the price of a new technology. Decades after the development of supersonic jet travel, for example, the high price of the technology continues to block its general adoption for commercial use. However, broad adoption can also occur fairly rapidly, especially in the area of digital technologies. The reason is that these technologies often produce what economists call “network effects” that encourage their diffusion, because the usefulness of a new technology or business method based on it increases as more people or businesses adopt it. The value of a computer operating system such as Windows, for example, increases as more people adopt it and use it to share or exchange information. Such network effects, in turn, tend to take hold as the utility of an innovation increases—for example, as more applications are written for the Windows system—and as its cost declines. Of course, the same basic factors of utility and cost also drive the broad adoption of innovations without network effects. The personal computer achieved broad acceptance only after widely useful applications were developed and prices fell sharply. Some innovations also produce a type of cascade dynamic in which their introduction and adoption are followed by subsequent innovations that build or depend on the initial technology and that may have a comparable or even greater impact on a nation’s productivity and competitiveness than the initial innovation. The most prominent example in recent years is the personal computer and the Internet, although electrification and early twentieth-century mass production also exhibit features of this dynamic. The spread of the Internet depended first on the previous broad adoption

HHassett.indbassett.indb 139139 111/8/121/8/12 9:279:27 PMPM 140 RETHINKING COMPETITIVENESS

of personal computers and later on the innovative development of “killer applications,” starting with e-mail. The Internet, like the personal computer, also produced cascading tiers of additional innovation, especially in the development of Internet-based businesses and the unique services they can provide. A more common type of cascading effect involves incremental improvements or enhancements of an existing innovation that extend its usefulness to more industries or activities. This process may occur through changes in the product or through improvements in the production process that reduce the price and thereby promote its adoption for more purposes and industries. Both processes are apparent in innovations such as cellular telephony and personal computing devices, which gained a broad range of new capacities in a fairly brief time. However, these dynamics and the factors that drive them do not explain why it can take many years for some innovations to achieve broad acceptance and even longer to produce the productivity and consequent competitiveness benefits ultimately associated with them. The most important recent instance is computerization, about which Robert Solow (1987) famously said, “We can see the computer age everywhere but in the productivity statistics” (p. 36). These delays involve lags between the development of an innovation and its broad adoption and lags between its adoption and the time when its benefits are realized (Barras 1986). Many factors affect the rates at which an innovation is adopted and its benefits realized, including price, knowledge, the extent of investment in the existing goods or services displaced by the innovation, and the organizational changes and disruptions associated with its adoption. The adoption of DNA sequencing, for example, requires extensive additional education or training by those using it. Similarly, the adoption of alternative fuels has been slowed by the potential costs associated with writing down or writing off (for tax purposes) the value of plant and equipment that provide current forms of energy and then replacing them. And the adoption of information technologies was delayed by the extensive changes in business organization required for their productive use. As these dynamics suggest, product innovations are generally adopted more quickly than process innovations (Damanpour and Gopalakrishnan 2001). The primary reason is that the successful adoption of process innovations, such

HHassett.indbassett.indb 140140 111/8/121/8/12 9:279:27 PMPM THE ROLE OF INNOVATION AND INTELLECTUAL PROPERTY 141

as mass production or information technologies, depends on major changes in organizational structure and administrative systems. While many forces affect the broad adoption of many innovations in advanced countries, the principal underlying factor is usually the strength of competition. A recent study from the Federal Reserve Bank of New York analyzed a natural experiment to measure the impact of competition on the adoption of new technologies: The researchers compared the pace and extent to which various personal computer manufacturers, competing with each other, adopt new technologies and capacities in their machines, compared to Apple, which maintains an effective monopoly on its form of the personal computer (Copeland and Shapiro 2010). They found that competition drove the makers of personal computers to introduce significantly more new products, with shorter life spans and faster-falling prices, than Apple. Competition has been especially powerful in driving innovation in the personal computer market, because the introduction of a new model sharply drives down the price of earlier models. By contrast, Apple has at least partially shielded itself from this dynamic, which in turn has meant that the company has introduced fewer new models incorporating new capacities, maintains longer product cycles, and offers slower price reductions. Competition also drives the spread of new technologies from businesses in one country to businesses in other places, especially in the case of the advanced economies. The rapid adoption of innovations by companies in developing nations, however, depends on other factors as well, including the nation’s wealth, the educational level of its workforce, its openness to trade, and its history of adopting predecessor innovations (Comin and Hobjin 2003). One broad analysis of how innovations diffuse from one economy to another found that between 49 percent and 87 percent of the differences in the rates of adoption across countries can be explained by five factors, the most important being the relative advantage the innovations confer in competition, their compatibility with existing technologies and organizational arrangements, and their overall complexity (Rogers 2003).5 Researchers have also found that the perception or understanding of an innovation is a critical factor in its adoption both within an economy and across economies. For example, the relative advantage conferred by an innovation depends on how much the innovation is seen to be superior to the existing alternatives. Thus, for example, the use of seatbelts accelerated

HHassett.indbassett.indb 141141 111/8/121/8/12 9:279:27 PMPM 142 RETHINKING COMPETITIVENESS

once it was perceived clearly that the new technology saved lives. The compatibility of an innovation with existing arrangements can also have large effects; for example, farmers in developing nations were slow to adopt techniques of soil conservation because those techniques were initially seen as conflicting with the traditional value of increasing farm production. The importance of complexity is evident in the adoption of computers, which began with small numbers of early adopters and then spread across the American economy and eventually to developing economies, once their apparent complexity was reduced by incremental changes that made computers much more user friendly. Certain factors also can impede the spread of innovations to some countries, especially legal and regulatory arrangements that dampen competition and otherwise impede the normal process by which innovations are widely adopted. For example, barriers to the formation of new businesses can dampen the competitive pressures to innovate. The reason is that new firms are more likely to seize on innovations in order to create market niches for themselves; and if they succeed in claiming a share of the market, their success will produce new competitive pressures for others to adopt the same innovations. These barriers are most evident in instances of state-sanctioned monopolies or oligopolies, where new business formation is precluded by law. However, less drastic barriers to new business creation, such as extensive regulatory requirements or limited access to start-up capital, can have similar effects on the spread of innovations. Product price regulations can also dampen these critical competitive pressures by slowing or stopping the normal scientific and economic process by which the price of many innovations declines over time, which in turn helps to drive their broader adoption. Similarly, trade protections can not only impede the flow of new ideas from one economy to another but also directly shield firms from competition from innovative companies in other countries. Finally, labor regulation can slow the spread of many innovations from other countries by preventing companies from undertaking the organizational changes required to make effective use of certain innovations, especially new technologies. For example, French labor law has broadly prohibited the firing or reassignment of workers without sufficient cause, and reorganizing a business in order to take better advantage of new information technologies has not been considered sufficient cause.

HHassett.indbassett.indb 142142 111/8/121/8/12 9:279:27 PMPM THE ROLE OF INNOVATION AND INTELLECTUAL PROPERTY 143

Competition: Efficiency versus Innovation

The role that innovation plays in the competitiveness of companies and countries also has important implications for the common understanding of how markets operate. Most researchers start with the proposition that companies compete on the basis of price or quality, with innovation often taken to represent the essential element of quality (see, for example, Greenhalgh and Rogers 2010). In practice, this simple, either-or distinction makes sense only when the innovation involves an improvement on an existing product that enables the innovator to gain market share and charge a higher price than its competitors whose products lack the new feature or capacity. This is most common when the innovation is considered especially vital to some consumers and is protected by patents or copyrights, such as a new pharmaceutical or software program which produce a monopoly rent. In many other cases, however, an innovation may enable the innovator to compete on the basis of both price and quality. For example, some innovators will choose to not raise their prices in order to increase market share. In other cases, the innovation involves some aspect of the business process—production, marketing, financing, distribution, or after-purchases services, for example—and produces a price advantage for the product affected by the innovation. In a period of rapid technological advance, such as our current period, competition itself often centers on innovation. Cellular telephony provides an example of innovation-based competition in which U.S. companies such as Apple (iPhone) and Google (Android) have vastly expanded the U.S. share of the global market for a product that European and Japanese firms initially dominated. In such periods, economies such as the United States, which foster the development and adoption of innovations, can secure a significant competitive advantage even when factors such as the cost of labor and capital produce competitive disadvantages. Furthermore, globalization creates opportunities for companies in advanced economies to combine the advantages derived from the conditions that foster innovation with those derived from low factor costs. Ameri- can companies have been particularly successful in this use of globalization, developing new products in the United States while using FDI in low-cost countries for much of the production and assembly. In 2005, 27 percent

HHassett.indbassett.indb 143143 111/8/121/8/12 9:279:27 PMPM 144 RETHINKING COMPETITIVENESS

of the total stock of U.S. FDI was located in low-cost developing nations, compared to less than 10 percent for Germany, France, and Britain (Shapiro 2008). This economically powerful combination helps explain why the global market share of U.S.-based high-tech companies increased from 24 percent in 1990 to 41 percent in 2005 (National Science Board 2006). Over the same years, the market share for high-tech companies based in the big three European economies fell from 26 percent to 18 percent, and from 24 percent to 16 percent for high-tech firms based in Japan (National Science Board 2006). Moreover, the competitive advantages secured through this use of globalization by innovating companies also depend on the effective use of IT and Internet-based innovations to build and manage efficient global networks. The likelihood of innovation-based competition depends on the same factors that drive innovation itself, including the ease or difficulty of form- ing new businesses, the ability of new businesses to secure the talent and financing they need to survive, and the existence or absence of subsidies or other regulatory protections for existing businesses. It also usually depends on the intensity of price-based competition, the condition that often drives existing firms to innovate or adopt the innovations of their rivals. These findings challenge the view of Joseph Schumpeter, who believed that innovation was most likely to occur in large firms in concentrated industries, especially monopolies and oligopolies, because they could use their monopoly or oligopoly rents to finance the innovation process (Schumpeter 1939). But as Nobel laureate Kenneth Arrow and others observed, when intellectual property rights are enforced strictly, producers in competitive industries have greater incentives to innovate than do monopoly or oligopoly firms, which can rely on their special legal status to maintain their market shares (Arrow 1962). One question that remains unsettled, however, is whether large or small firms are more likely to innovate. Large firms are more likely to conduct R&D, but smaller firms that carry out R&D are more R&D intensive on average than big firms (Greenhalgh and Rogers 2010). The answer, at least in the American case, probably lies in the successful combination of large and small firms. In innovation-based industries, such as software and bio- technology, for example, the prevailing model for innovation has become one in which small firms develop something new and then are bought by

HHassett.indbassett.indb 144144 111/8/121/8/12 9:279:27 PMPM THE ROLE OF INNOVATION AND INTELLECTUAL PROPERTY 145

large firms, which complete the development and then commercialize and market the innovation. Michael Porter (Porter et al. 2001) approaches the distinction between innovation-based and price-based competition in a different way. He argues that innovation-based competition generally characterizes economies at a different stage of economic development than price- or efficiency- based competition. Emerging economies just beginning to harness global technologies and business methods for local production, usually relying on FDIs and joint ventures with companies from more advanced economies, generally focus on the manufacture of commodity-like goods and therefore almost always compete on the basis of price or efficiency. By contrast, advanced economies are characterized by innovation-driven economic development. We should note, however, that both stages of development depend on strict intellectual property protections—to enforce the rights and protect the profits of those transferring advanced technologies to developing countries and to protect the incentives to develop those technologies in more advanced economies (see the literature review in Shapiro and Hassett 2005). In practice, as noted earlier, many of the apparent distinctions between competition based on relative prices or efficiency versus competition based on innovation eventually break down. In broad terms, a company’s competitiveness depends on how effectively it adopts best practices in production, technology, and business or management methods; and all of these capacities involve innovation as well as considerations of efficiency (see, for example, Porter 2000). Moreover, many innovations are designed to enhance the efficiency of an existing process and thus drive down costs and prices, rather than provide a new element or capacity to an existing product or process. The question remains of how to assess the impact of innovation on economies invested and engaged in innovation-based competition, and one answer can be derived from measures of the intellectual property embodied in innovations. As expected, the American economy is fairly dominated by intellectual-property-intensive industries. One recent study, for example, focused on “IP [-dependent] industries” in the United States, defined as those industries that depend on copyright and/or patent protections in or for a digital environment (semiconductors, software publishers,

HHassett.indbassett.indb 145145 111/8/121/8/12 9:279:27 PMPM 146 RETHINKING COMPETITIVENESS

motion pictures, advertising, and so on), those industries that depend on patent rights to improve their products and improve their efficiency (the automotive, aerospace, biotech/pharmaceutical, and chemical industries), and those that support the transportation and distribution of intellectual property products (retail and whole merchandisers and transportation industries (Siwek 2005). This research found that in 2003, intellectual- property-dependent industries accounted for 20 percent of all private sector GDP but 40 percent of the growth of all private industry, attesting to the impact of innovations and the intellectual property embodied in them on U.S. productivity. These intellectual-property-dependent industries also accounted for 40 percent of all exportable high-value-added goods and services and nearly 60 percent of the growth of exportable high-value-added goods and services, attesting to the disproportionate impact of innovation on international competitiveness (Siwek 2005). In a similar vein, another recent study found that the announcement of a new product has both a significant positive impact on an innovator’s stock price and a significant negative effect on its rivals’ share price (Akhigbe 2002). All together, these data suggest that U.S. industries organized around the development and adoption of innovation are in critical ways relatively more competitive than other U.S. industries. America’s recent capacity to extract greater productivity gains from IT innovations, compared to Europe or Japan, has spurred a number of studies comparing conditions across the various economies. Research by the U.S. Commerce Department’s Economic and Statistics Administration in the late 1990s points to the benefits of America’s relatively light labor regulation (Department of Commerce 2000). This suggests that the more stringent regulation in the large European economies hindered European businesses from reorganizing their operations to take best advantage of the new technologies and so forced their firms to focus more on price-based competition (Van Ark et al 2008). Similarly, a long line of research points to intense competition as a critical factor in the diffusion of new technologies and associated business methods, suggesting that the relatively stricter overall regulatory environment in Europe and Japan reinforces their focus on price-based rather than innovation-based competition (Ahn 2002, Gru- ber 2000, Hahn 2001, West 2007).

HHassett.indbassett.indb 146146 111/8/121/8/12 9:279:27 PMPM THE ROLE OF INNOVATION AND INTELLECTUAL PROPERTY 147

Another recent study examined more systematically the role of the so- called knowledge economy in explaining why, from 1995 to 2004, productivity gains accelerated in the United States while they were slowing down in the fifteen countries in the European Union (EU-15) (Van Ark et al. 2008). Over this period the EU-15 grew an average of 2.2 percent per year while the United States expanded an average of 3.7 percent annually, producing widening gaps across a number of key economic measures. By 2004 Europe’s GDP per capita was only 74 percent that of the United States, the hours worked per capita by Europeans were 82 percent that of Americans, and GDP and capital input, per hour worked, across the EU-15 were both only 90 percent of the levels in the United States. The researchers applied growth accounting to analyze why these gaps have widened. They found, first, that the traditional factors of economic production—labor and capital investment—could not explain these growing differences. Rather, they traced these differences almost entirely to three factors that together, they argue, comprise the “knowledge economy”—investments in information technologies, the increased use of highly skilled workers, and multifactor productivity encompassing both organizational and technological innovations. The study also disaggregated these growth and productivity differences across sectors. They found that the productivity of European and American non-IT manufacturing firms was generally equivalent over this period, while the contribution to national productivity of U.S. IT producers was nearly twice that of their European competitors. Moreover, U.S. service industries adopted IT more broadly over this period than their European counterparts, and the contribution to productivity by service industries was more than three times greater in America than in the EU-15 (Van Ark et al, pp. 38–41). Other research comparing the recent economic performance in America and Europe points to additional ways in which innovation may have bolstered U.S. competitiveness, such as by generating larger spillover benefits. The authors of that study document how U.S. firms are more likely to involve their suppliers in their new product development process than European firms and how those suppliers can then apply the new inputs to other products (Birou and Fawcett 1994). However, the spillover process can also have adverse effects on innovation when it effectively reduces intellectual property rights. A recent study comparing the United States and Japan, for example, found that Japan’s patent system promotes greater

HHassett.indbassett.indb 147147 111/8/121/8/12 9:279:27 PMPM 148 RETHINKING COMPETITIVENESS

diffusion of information across an industry than seen in the United States but in ways that lower the return to innovators and thus reduce the incidence of innovation (Cohen et al. 2002). The hierarchical and insular character of many Japanese industries may also limit their success in innovation-based global competition, compared to their U.S. counterparts (Dujarric and Hagiu 2009). For example, researchers have traced the relatively small numbers and limited success of Japanese software firms to the strict focus by Japan’s dominant computer and electronics manufacturers on developing hardware and then locking in their customers by supplying the software customized for their hardware.6 In other, more highly innovative sectors of the Japanese economy, such as anime and mobile telephony, industry leaders have focused almost exclusively on domestic sales, limiting the success of their innovations in foreign markets. Much as in the case of the large European economies, Japan’s capacity to grow through innovation and succeed in innovation-based competition is hampered by a number of factors, including weak antitrust enforcement, high barriers to FDI, and a badly underdeveloped venture capital sector. All told, these studies suggest that among advanced economies, those that nurture and promote the conditions for innovation and its broad adoption, and thus become more focused on innovation-based competition, have a competitive advantage over those more focused on price or efficiency-based competition.

Conclusions For a half century, economic research has shown that the single largest factor driving gains in productivity and growth in a modern economy is innovation. That body of research and this analysis approach the subject broadly, encompassing not only the development and application of new technologies, materials, and production processes but also new ways of financing, marketing, and distributing goods and services and new ways of organizing a workplace and managing an enterprise. Especially in a period of rapid technological and organizational changes, like the present one, the competitiveness of firms and nations depends significantly on how effectively they apply and use such economically powerful innovations as information technologies and online business networks.

HHassett.indbassett.indb 148148 111/8/121/8/12 9:279:27 PMPM THE ROLE OF INNOVATION AND INTELLECTUAL PROPERTY 149

These results are apparent not only in the strong productivity and income gains achieved by the United States over the last generation, compared to Japan and the major European economies, and in the growing global market share of American-based high-technology companies, compared to their counterparts in Europe and Japan. The impact of innovation is equally apparent in the enormous economic progress achieved by developing nations such as Taiwan and Korea, and China more recently, which encourage foreign direct investment of advanced technologies and business organizations. Inventive genius and commercial genius are qualities that know no borders, and no society has a monopoly on such talents. But certain nations are homes to disproportionate numbers of companies capable of developing and adopting the powerful and often disruptive innovations that help drive economic progress and competitiveness. These nations share certain economic and political conditions that promote their development and application, starting with strong intellectual property protections for the new ideas that animate innovations. Every advanced economy today has reasonably strict intellectual property rights and protections—a major change from the nineteenth century, for example. Among developing nations—and with the special exception of China—those that have achieved the greatest economic progress respect the intellectual property rights of foreign companies, establishing a critical incentive for foreign direct investments of advanced technologies. The sustained development and broad application of innovations depend on other social, economic, and political conditions as well. An entrepreneurial economic culture and low barriers to the formation of new businesses play significant roles, because new and young businesses are major sources of innovations and are more likely than established firms to quickly adopt innovations from others. Strong government support for basic research and development is important, especially in advanced economies, because the incentives for private firms to undertake basic R&D are notoriously weak. Similarly, sustained public investments in education and training are also critical, for both advanced and developing economies, to ensure a sufficient supply of workers who can operate new technologies or at least perform effectively in workplaces dense with those technologies.

HHassett.indbassett.indb 149149 111/8/121/8/12 9:279:27 PMPM 150 RETHINKING COMPETITIVENESS

Beyond these basic conditions, it is clear that numerous factors have induced many American firms to locate more of their intellectual property abroad, and if this trend persists, it could undermine or reduce some of the historical advantages of the United States. Recent research suggests, for example, that firms shift around, from country to country, large amounts of income; and much of that shifting involves income derived from intellectual property (Grubert 2003). Furthermore, recent data suggest that a firm’s decisions regarding where to locate the company’s intangible assets are quite sensitive to tax rates and tax burdens. One study, for example, estimated that a 1 percent reduction in a nation’s corporate tax rate increases by 1.7 percent the stock of intangible assets held there by multinational companies (Dischinger and Riedel 2011). Moreover, certain other factors that are likely to have comparably powerful effects, such as the enforcement of patent provisions, vary less now across nations. In any case, there is strong evidence of extensive cross-border patenting activity. The World Intellectual Property Organization (2008) reports that nonresident patent filings grew at an average annual rate of 7.3 percent from 1994 to 2006. Moreover, U.S. firms have been especially active in this way. Foreign patent applications by U.S.-based companies increased by approximately 50 percent from 2000 to 2006, growing significantly faster than the growth of U.S. domestic patent applications. A strong competitive environment is the most important economic condition for promoting the broad development and application of innovations capable of enhancing the competitiveness of a nation’s firms. Competitive pressures can provide powerful incentives both to develop new technologies, products, and ways of doing businesses and to adopt the effective technologies and business methods developed by others. These pressures often come from young businesses trying to create their own market niches, and when they succeed, they exert new competitive pressures on more mature businesses to follow suit. Therefore, the various conditions that tend to reduce strong competitive pressures—for example, trade protections, product or industry regulations that insulate dominant businesses from competition, special spending and tax subsidies for established businesses, and many labor regulations that can restrict a firm’s ability to hire, reassign, or discharge employees—tend to slow or reduce the development and especially the adoption of innovations.

HHassett.indbassett.indb 150150 111/8/121/8/12 9:279:27 PMPM THE ROLE OF INNOVATION AND INTELLECTUAL PROPERTY 151

When these conditions are fulfilled, competition itself often centers on innovation or quality, rather than efficiency and price—especially in a time of rapid technological and organizational changes, such as the current one. At such times, firms in countries whose culture and legal and economic arrangements promote such changes may establish significant competitive advantages. These effects have been evident in the generally superior economic performance of the United States over the last generation, compared to Japan and the large European economies. Moreover, they are having transformative effects on the U.S. economy, producing the world’s first, genuine idea-based economy. Industries dominated by their intellectual property account for a growing share of America’s GDP and an even larger share of the growth of that GDP. In addition, for the first time on record, American businesses since the mid-1990s have been investing less in traditional physical assets than in new intangible assets, including not only intellectual property and the R&D that produce it but also software, databases, training, organizational changes, and so on. Moreover, markets now value firms mainly on the basis of those intangible assets: in 1984 the book value of large U.S. public companies—what their physical assets could be sold for on the open market—accounted for three-quarters of their market capitalization (Bryan and Zanini 2005). By 2005 the book value of large U.S, public corporations accounted for just 36 percent of their market caps, so nearly two-thirds of the value of American businesses is now based on their intangible assets (Bryan and Zanini 2005). The competitive advantages of an idea-based economy are available to any society prepared to actively and aggressively promote the development of new ideas and the broad application of the most effective ones, wherever they were developed originally—across the city or across the world. The competitive advantages conferred in recent years on many American companies by our economic, political, and cultural arrangements, moreover, may not be sustained if those arrangements change in basic ways. Some societies will find it difficult to adopt comparable arrangements. For example, while China has been open and quite successful in adopting many technological innovations developed in the United States and other advanced economies, the broad freedom implicit in the arrangements required to spur such development domestically could well exceed the political tolerance of its current leadership. Nevertheless, the competitive advantages conferred

HHassett.indbassett.indb 151151 111/8/121/8/12 9:279:27 PMPM 152 RETHINKING COMPETITIVENESS

by actively promoting an innovation-based economy will produce strong incentives for its spread in China and many other places. The competitive landscape for American companies and the United States, therefore, will be even more challenging in the years ahead.

Notes 1. The author wishes to acknowledge the excellent research support provided by Jiwon Vellucci of Sonecon. 2. This discussion is adapted from Shapiro and Hassett (2005). 3. In economic terms, ideas are also considered to be “partially nonexcludable goods”: an innovator acting as a private agent cannot prevent others from using his idea, as compared to someone who owns a piece of land or a factory and can prevent others from using it by hiring security guards. 4. For a literature review of the connections between innovation and intellectual property protections, see Kanwar and Evenson (2001). 5. The other two factors are “trialability,” or the extent to which the innovation can be experimented with on a limited basis, and “observability,” the degree to which the results of an innovation are visible to others. 6. This approach is similar to that of the U.S. computer firm Apple, which maintains only a modest share of the U.S. personal computer market.

References Ahn, Sanghoon. 2002. Competition, Innovation and Productivity Growth: A Review of Theory and Evidence. OECD Economics Working Paper No. 317. Akhigbe, Aigbe. 2002. “New Product Innovations, Information Signaling and Indus- try Competition.” Applied Financial Economics 15:371–78. Ang, James S., Chaopeng Wu, and Yingmei Cheng. 2009. Does Enforcement of Intellec- tual Property Rights Matter in China? Evidence from Financing and Investment Choices in High Tech Industry. AFA 2011 Denver Meetings Paper. http://papers.ssrn.com/ sol3/papers.cfm?abstract_id=1571392. Arrow, K. J. 1962. “Economic Welfare and the Allocation of Resources for Inven- tions.” In The Rate and Direction of Inventive Activity: Economic and Social Factors, ed. R. R. Nelson. Princeton, NJ: Princeton University Press. Barras, Richard. 1986. “Towards a Theory of Innovation in Services.” Research Policy 15:161–73. Birou, Laura M., and Stanley E. Fawcett. 1994. “Supplier Involvement in Integrated Product Development: A Comparison of US and European Practices.” International Journal of Physical Distribution and Logistics Management 24(5):4–14.

HHassett.indbassett.indb 152152 111/8/121/8/12 9:279:27 PMPM THE ROLE OF INNOVATION AND INTELLECTUAL PROPERTY 153

Bryan, Lowell L., and Michelle Zanini. 2005. “Strategy in an Era of Global Giants.” McKinsey Quarterly 4. https://www.mckinseyquarterly.com/Strategy_in_an_era_ of_global_giants_1689. Chen, Derek H. C., and Carl Dahlman. 2004. Knowledge and Development: A Cross- Section Approach. Policy Research Working Paper 3366. World Bank, Washington, D.C., November. Cohen, Wesley M., Akira Goto, Akiya Nagata, Richard R. Nelson, and John P. Walsh. 2002. “R&D Spillovers, Patents and the Incentives to Innovate in Japan and the United States.” Research Policy 31:1349–67. Comin, Diego, and Bart Hobjin. 2003. Cross Country Technology Adoption: Making the Theories Face the Facts. Staff Report No. 169. Federal Reserve Bank of New York. Copeland, Adam, and Adam H. Shapiro. 2010. The Impact of Competition on Technol- ogy Adoption: An Apples-to-PCs Analysis. Staff Report No. 462. Federal Reserve Bank of New York. Corrado, Carol, Charles Hulten, and Daniel Sichel. 2004. Measuring Capital and Tech- nology: An Expanded Framework. Finance and Economics Discussion Series No. 2004-65. Federal Reserve Board, Washington, D.C., August. Damanpour, Fariborz, and Shanthi Gopalakrishnan. 2001. “The Dynamics of the Adoption of Product and Process Innovations in Organizations.” Journal of Man- agement Studies 38:1. Denison, Edward F. 1962. The Sources of Economic Growth in the United States and the Alternatives Before Us. Supplementary Paper No. 13. Committee for Economic Development, New York. Department of Commerce. 2000. Digital Economy 2000. Economics and Statistics Administration. Dischinger, Matthias, and Nadine Riedel. 2011. “Corporate Taxes and the Loca- tion of Intangible Assets within Multinational Firms.” Journal of Public Economics 95:691–707. Diwan, Ishac, and Dani Rodrik. 1991. “Patents, Appropriate Technology and North- South Trade.” Journal of International Economics 30:1–2, 27–47. Dujarric, Robert, and Andrei Hagiu. 2009. Capitalizing on Innovation: The Case of Japan. Working Paper 09-114. Harvard Business School, Cambridge, MA. Greenhalgh, Christine, and Mark Rogers. 2010. Innovation, Intellectual Property and Economic Growth. Princeton, NJ: Princeton University Press. Group of Friends of Development. 2005. Proposal to Establish a Development Agenda for the World Intellectual Property Organization (WIPO): An Elaboration of Issues Raised in Document Wo/Ga/31/11. World Intellectual Property Organization, April 6. www. ipjustice.org/WIPO/elaborationDA. Gruber, Harold. 2000. Competition and Innovation: The Diffusion of Mobile Telecommu- nications in Central and Eastern Europe. European Investment Bank Working Paper. Grubert, Harry. 2003. “Intangible Income, Intercompany Transactions, Income Shift- ing, and the Choice of Location.” Part 2. National Tax Journal 56(1):221–42.

HHassett.indbassett.indb 153153 111/8/121/8/12 9:279:27 PMPM 154 RETHINKING COMPETITIVENESS

Hahn, Robert W. 2001. A Primer on Competition Policy and the New Economy. AEI– Brookings Joint Center for Regulatory Studies, Working Paper No. 01-3. Hall, Bronwyn H. 2004. Innovation and Diffusion. NBER Working Paper. National Bureau of Economic Research, Cambridge, MA. Kanwar, Sunil, and Robert Evenson. 2001. Does Intellectual Property Protection Spur Technological Change? Discussion Paper No. 831. Yale University, Economic Growth Center, New Haven, CT, June. Klenow, Peter J., and Andres Rodriguez-Clare. 2004. Externalities and Growth. NBER Working Paper 11009. National Bureau of Economic Research, Cambridge, MA, December. Mansfield, Edwin. 1986. “Patents and Innovation: An Empirical Study.” Management Science 32(1):173–81. Maskus, Keith. 1997. The International Regulation of Intellectual Property. Centre for International Economic Studies, University of Adelaide, Seminar Paper 97-11. http://www.adelaide.edu.au/cies/papers/sp9711.pdf. Moser, Petra. 2003. How Do Patent Laws Influence Innovation? Evidence from Nineteenth- Century World Fairs. NBER Working Paper 9909. National Bureau of Economic Research, Cambridge, MA, August. National Science Board. 2006. Science and Engineering Indicators. http://www.nsf.gov/ statistics/seind/. Organisation for Economic Co-operation and Development. 2007. The Economic Impact of Counterfeiting and Piracy. http://www.oecd.org/dataoecd/13/12/38707619.pdf. Porter, Michael E. 2000. “Location, Competition and Economic Development: Local Clusters in a Global Economy.” Economic Development Quarterly 14:15–34. Porter, Michael E, Jeffrey D. Sachs and John W. McArthur. 2001. “Competitiveness and Stages of Economic Development.” Executive Summary of the Global Competi- tiveness Report 2001–2002. World Economic Forum, Geneva, Switzerland. Rogers, Everett. 2003. Diffusion of Innovations. New York: Free Press. Romer, Paul. 1993. “Economic Growth.” In Fortune Encyclopedia of Economics, ed. David R. Henderson. New York: Warner Books. Schumpeter, J. A. 1939. Business Cycles: A Theoretical, Historical and Statistical Analysis of the Capitalist Process. New York: McGraw Hill. Shapiro, Robert J. 2008. Futurecast. New York: St. Martin’s Press. Shapiro, Robert J., and Kevin A. Hassett. 2005. “The Economic Value of Intellec- tual Property.” http://www.sonecon.com/docs/studies/IntellectualPropertyReport- October2005.pdf. Siwek, Stephen E. 2005. Engines of Growth: Economic Contributions of the US Intellectual Property Industries. Economists Incorporated. Solow, Robert M. 1956. “A Contribution to the Theory of Economic Growth.” Quar- terly Journal of Economics 70:65–94. ———. 1957. “Technical Change and the Aggregate Production Function.” Review of Economics and Statistics 39(3):312–20.

HHassett.indbassett.indb 154154 111/8/121/8/12 9:279:27 PMPM THE ROLE OF INNOVATION AND INTELLECTUAL PROPERTY 155

———. 1987. “We’d Better Watch Out,” New York Times Book Review, July 12. Taylor, M. Scott. 1994. “Trips, Trade and Growth.” International Economic Review 35(2):361–81. Van Ark, Bart, Mary O’Mahony, and Marcel P. Timmer. 2008. “The Productivity Gap between Europe and the United States: Trends and Causes.” Journal of Economic Perspectives 22(1):25–44. ———. 2009. Innovation and U.S. Competitiveness: Reevaluating the Contributors to Growth. Research Report 1441. The Conference Board, New York. West, Jeremy K. 2007. Competition, Patents and Innovation. OECD Working Paper No. 72. World Bank. 2005. World Development Indicators. Washington, D.C.: World Bank. World Intellectual Property Organization. 2008. World Patent Report: A Statistical Review. http://www.wipo.int/ipstats/en/statistics/patents/wipo_pub_931.html#b11.

HHassett.indbassett.indb 155155 111/8/121/8/12 9:279:27 PMPM 6

American Competitiveness and the Health Care System Michael E. Chernew and Philip I. Levy

One impetus behind the drive to address the ills of the American health care sector has been a linkage between that sector’s performance and the competitiveness of the American economy. In the early days of the Obama administration, then White House Chief of Staff Rahm Emanuel made the case on CBS’s Face the Nation: “The health care system is a particular example where America’s economic competitiveness, its strength around the world, is sapped because we have a health care system that doesn’t allow American workers and business to compete” (Sweet 2009). While the debate over health care reform that ensued may have been highly par- tisan, the tie between health and competitiveness was not. In a speech in late 2007, Sen. John McCain (R-AZ) said: “You and I both know that rising health care costs are a threat to our global competitiveness, a threat to our families’ budgets, a threat to our government’s solvency, and a threat to the profitability of American business” (McCain 2007). Academic studies have also pointed to the deleterious economic impact of high and growing health spending. For example, Sood et al. (2009) estimated that a 10 percent increase in excess growth in health care spending (defined as outpacing gross domestic product [GDP] growth) would cause a loss of 120,803 jobs, a $28 billion decline in gross output, and $14 billion in lost value added to GDP.1 Baicker and Chandra (2005) estimate that the portion of the 40 percent increase in health insurance premiums between 1996 and 2002 may have caused up to a 6 percent decrease in employment and an 8 percent decrease in full-time work for the employed.

156

HHassett.indbassett.indb 156156 111/8/121/8/12 9:279:27 PMPM AMERICAN COMPETITIVENESS AND HEALTH CARE 157

The importance of the health sector writ large to the broader Ameri- can economy is striking. In 2007 the United States spent 16 percent of its income on health care (Organisation for Economic Co-operation and Devel- opment [OECD] 2010). That is significantly more than the 11 percent share spent by France in 2007, the next most health care–intensive country, and more than triple the 5.9 percent spent by Mexico, the OECD country with the lowest share of GDP devoted to health care. By 2009 the U.S. share had risen to 17.3 percent. These figures highlight one way in which the American system of health care delivery differs dramatically from that in other countries: it consumes substantially more of our national income. Of course, this does not necessarily imply that the American health care system is inefficient. Americans can legitimately choose to spend their income on a different bundle of goods and services than other countries. While spending patterns alone do not prove there is an efficiency problem, for a range of reasons, discussed below, we will argue that in fact the American health care system generates higher spending than the outcomes warrant. This may reflect inefficiency in production or consumption, or it may reflect a transfer from consumers to producers that does not necessarily imply inefficiency in a strict economic sense. It is likely both phenomena are occurring. Yet because even in the latter case we spend more than is justified by output, we will label both as inefficient. This inefficiency does not necessarily imply that the costs of the U.S. health care system adversely affect U.S. competitiveness. Specifically, the U.S. system relies heavily on private provision of health insurance, particularly through employers. Thus the costs of health care appear on employer financial statements, a situation that gives the impression that employers pay for health care. Moreover, other features of our system beyond costs, including the reliance on employer financing as opposed to a tax-financed system, may affect competitiveness. Given the integrated global economy, employer financing of health care has fed the perception that health care spending, and more broadly our health care system, has a significantly adverse impact on America’s ability to compete internationally. However, the arguments linking health policy and American competitiveness are often loose and suggestive, rather than rigorous. Most competitiveness arguments fail even to define the term “competitiveness.” It is used

HHassett.indbassett.indb 157157 111/8/121/8/12 9:279:27 PMPM 158 RETHINKING COMPETITIVENESS

simply as a device to add urgency. A looming problem is more alarming if it portends national failure. Over a decade ago, Paul Krugman (1994) challenged the widespread use of the term “competitiveness” as a motivator for domestic policy reform. He presented a series of arguments for why the analogy of competition was flawed when applied to countries. Krugman also argued that unlike a corporation, which ceases to exist if its market position is uncom- petitive or unsustainable, a country will continue even when it is faring worse than its neighbors. Krugman highlighted a distinction between productivity, which refers to the output a country can produce in absolute terms, given its endowments of resources, and competitiveness, which refers to a country’s ability to win some particular contest. There are many actions a nation may take to enhance its productivity. Because these actions will drive the level of national output, they will necessarily affect a country’s relative standing in international rankings. In this light, countries are only serving as points of comparison, not truly competing. Mindful of this critique, in this chapter we attempt to disentangle the arguments surrounding competitiveness and health care and thereby provide a framework for both future research and more fruitful policy debate. We adhere to a rigorous definition of competitiveness and distinguish between those policies that are truly directed at international competition and those that primarily address national productivity. To do so, we outline the taxonomy of the routes by which health care policy is likely to have a broad economic impact. Although there have been vigorous recent debates about changes to the U.S. health care system, we do not provide an analysis of those changes in this chapter. Rather, here we consider how the distinctive and persistent features of the U.S. health care system affect competitiveness. Our analysis is focused on the “competitiveness” of the nation, not the competitiveness of any industry or segment of the economy. In a general equilibrium framework, a given industry within a country may not be competitive, but that does not necessarily imply the country is not competitive. This focus is important because certain aspects of a country’s system may disadvantage some industries but not others. In fact, the standard theory of comparative advantage in trade economics argues that an open economy (relative to a closed economy) will do just this—boost some sectors and

HHassett.indbassett.indb 158158 111/8/121/8/12 9:279:27 PMPM AMERICAN COMPETITIVENESS AND HEALTH CARE 159

diminish others, all while increasing national welfare. This does not preclude the possibility that there are international competitions and that health policy can affect their outcome, but it raises the bar. Our contention is that even if the American health care system adversely affects the competitiveness of some industries, it may not reduce aggregate American competitiveness. The first section below discusses the distinction between competitiveness and productivity and reviews the efficiency of the U.S. system. The second section considers the dimensions in which we are likely to see global competition and suggests how health policy could factor in “The Health System’s Effect on the Economy.” The third section provides a comparison of international approaches to health care in light of the preceding discussion. We then conclude with some final thoughts.

Productivity versus Competitiveness Rigorous assessment of how the health care sector influences American competitiveness requires a precise definition of competitiveness.

Defining the Concept. The lure of the competitiveness analogy is strong. Whether watching a World Cup soccer match or two businesses vying for market share, we are thoroughly steeped in the schema of competition. Competitors struggle for dominance. Adopting better policies, whether improved training techniques or innovative policies, can make one side more competitive than the other. The more competitive side is likely to win, the less competitive side to lose. A defining feature of such competitions is that they are zero sum. One soccer team emerges as the champion; the rest do not. This domination need not be total, of course. One firm could emerge with 80 percent market share while its less competitive rival emerges with 20 percent, but either firm’s gains must come at the other’s expense. Invoking the specter of international competition can be particularly effective in stirring a nationalistic audience to action. In international economic policy, however, it is also particularly problematic. Trade relations between countries are generally not zero sum; countries can all grow and benefit. There are numerous examples in which the practices one nation adopts to enhance its productivity have a positive effect on the well-being of

HHassett.indbassett.indb 159159 111/8/121/8/12 9:279:27 PMPM 160 RETHINKING COMPETITIVENESS

other nations.2 Even if there are no such beneficial spillovers, it is difficult to find the international equivalent to losing a competition. As Krugman so aptly noted, countries do not go out of business the way corporations do. For the purposes of discussing health care policy and its economic impact, we draw a distinction between those measures that affect productivity and those that affect international competitions. The latter are easiest to define. Definition 1 is as follows: A government policy improves a country’s competitiveness if it boosts one country’s well-being at the expense of other countries.3 In practical terms, this means that competitiveness arguments should be able to identify the prize for which a country is competing. We will defer a more thorough discussion of competitions that meet this criterion to the next section, but legitimate examples of “prizes” could include the location of a factory, rents in an imperfectly competitive industry, or the residency of a particularly desirable group of individuals.4 To make the competitiveness term meaningful, we rule out such prizes as the honorific titles “most productive” or “most improved.” These are frequently the implicit prizes in discussions that center on improving the U.S. standard of living, infrastructure, or employment rate. These we will classify under the broad rubric of “productivity.” This is not to minimize this objective in any way. Improving the domestic standard of living is and should be the principal objective of economic policy. For contrast, we can state Definition 2: A government policy affects a country’s productivity if it boosts one country’s well-being without reducing the well-being of other countries.5 In our context, this would imply that our health care system would harm our competitiveness if it raised the well-being of other countries at the expense of our well-being. It is important to note that while aspects of our health care system may have big effects on productivity and well- being, in the context of trade, the health care system can only affect competitiveness if some industries or jobs are better than others and the distortion created by our health care system reduces our ability to attract desirable industries or jobs. Is this a worthwhile distinction to make? Why should it matter if advocates use the term “competitiveness” as shorthand for “urgent national policy priority”? Krugman (1994) concluded his critique of the concept thus: “Competitiveness is a meaningless word when applied

HHassett.indbassett.indb 160160 111/8/121/8/12 9:279:27 PMPM AMERICAN COMPETITIVENESS AND HEALTH CARE 161

to national economies. And the obsession with competitiveness is both wrong and dangerous” (p. 44). In his argument, the danger comes from introducing an unjustified note of international rivalry, from providing a potential justification for trade protection or international tensions, and from the pernicious knock-on effects of loose and misguided reasoning. For example, if we concern ourselves with competitiveness, we may prefer, relative to the status quo, policies that lead to a reduction in welfare as long as our competitors’ well-being drops more. Similarly, we may reject policies that would improve our welfare but would improve our competitors’ welfare relatively more. Under almost any societal utility function, including the focus on productivity or competitiveness, we would want to maximize the efficiency of every American industry. Yet a blind focus on competitiveness may not lead to an improvement in well-being. Thus even if an inefficient American health care system does not affect American competitiveness, we contend that such inefficiency is important and may potentially diminish the welfare of Americans. In the remainder of this section, we consider prominent arguments for ways in which U.S. health care practices affect the efficiency of the economy and consider whether those arguments deal with productivity or competitiveness. In the next section, we work backward and consider potential prizes in international economic competitions and ask how health policies might influence those outcomes.

The Efficiency of the American Health Care System. Proving the inefficiency of the American health care system is difficult because outputs are multidimensional. They include the quality of clinical care (which itself is multidimensional); nonclinical aspects of quality, such as patient satisfaction; and nonquality attributes of the system itself, such as free choice of physician or treatment. We can readily observe that the American health care system is more expensive than other systems. Yet this multidimension- ality complicates assessment of the benefit obtained from the extra spending. Our argument that the system is inefficient takes several forms. First, we examine the quality measures that are easily observed. On these measures the United States appears to be at the middle or bottom of the pack for developed countries, suggesting we do not get much better quality for the additional money we spend. Second, we look for direct evidence of

HHassett.indbassett.indb 161161 111/8/121/8/12 9:279:27 PMPM 162 RETHINKING COMPETITIVENESS

inefficiency in the American system. This includes review of the literature that examines geographic variation within the United States. This exercise reveals that areas with higher spending do not appear to have systematically better outcomes than areas with lower spending or have better patient satisfaction. Of course, all areas of the United States operate under basically the same rules, so this evidence demonstrates that our system has the capacity to generate both relatively efficient and inefficient outcomes. On balance we think the geographic variation in the United States contributes to the possibility that considerable inefficiency exists in the United States, though we note that our system does not inevitably generate inefficient outcomes. Moreover, we also note that inefficiency within the United States does not necessarily imply inefficiency relative to other countries that may exhibit comparable variation and inefficiency. Third, we examine prices for care in the United States, noting that we pay more for care than other countries do. The evidence suggests higher prices are a significant explanation behind higher American spending. Normally high prices would lead to underconsumption, and thus inefficiency. But other features of health care markets, such as insurance, may counter the tendency for underconsumption and just lead to a transfer from consumers to producers. Inefficiency will arise only if that transfer distorts behavior. Finally, we explore the institutional features of our system that can account for the inefficiency. As the geographic analysis indicates, these features may not affect all areas equally, either because of variation in traits across areas (e.g., competition) or variation in the outcomes that result from a given set of features.

International Comparisons of Outcomes. Before discussing variations in outcomes, it is important to establish that Americans are not significantly sicker (or much healthier) than residents of other countries. In a 2008 study, the McKinsey Global Institute found that the United States had lower disease prevalence in twenty-one of thirty-five medical conditions evaluated (Farrell et al. 2008). Behavioral measures also show mixed results. While obesity is high in the United States, Americans smoke less than the people of other OECD countries, and alcohol consumption falls somewhere in the middle (OECD 2010). Although the United States spent more than $2,400 more per person on health care than the next most expensive OECD nation (Switzerland),6

HHassett.indbassett.indb 162162 111/8/121/8/12 9:279:27 PMPM AMERICAN COMPETITIVENESS AND HEALTH CARE 163

health outcomes were not significantly better and in some cases were actually worse. McKinsey finds that the United States does excel at two measures: adoption of new technologies and cancer survival rates. For instance, new prescription drugs are approved and enter the market one year earlier in the United States than abroad. For other measures, however, the United States has mixed results. Life expectancy at birth and at age sixty-five, albeit a very crude measure of health system effectiveness, was lower in the United States than the average across peer OECD countries (OECD 2010). The United States also lags on many other health system measures. The 2010 annual Commonwealth Fund report on cross-country health systems ranks the United States last among six peer OECD countries on a composite of quality of care, access, efficiency, equity, quality and length of life, and expenditure measures (Commonwealth Fund et al. 2010).7 (Some of these issues, such as access, may be ameliorated by the recent health reform law.) Over the four dimensions of quality measured (effectiveness, safety, coordination, and patient centeredness), the United States falls in the middle of the pack for effectiveness and patient centeredness and lags toward the bottom in safety and care coordination. While the United States excels at some subcomponents of effectiveness, such as preventive care, the overall composite ranks it sixth out of the seven countries. The United States ranks somewhat worse in the 2010 report than the 2008 report, which showed the United States lagging, especially when it came to access and implementation of electronic medical records. So what are Americans buying with the extra spending if the evidence suggests that the quality of care is not better? One possibility is autonomy for patients and physicians. Americans may prefer a system with few restraints and oversights on the care patients can seek and physicians can supply, even if that system leads to higher costs and does not produce better health outcomes. While this is possible, survey evidence indicates otherwise, and reasonable arguments can suggest that political and institutional forces, such as the clout of providers, consumer groups, disease advocacy groups, and organized labor groups, which resist restrictions on care and work to prevent reforms that would lower costs without reducing quality, are at least partially responsible (Blendon et al. 2006). Unfortunately the system is not designed to allow us to assess whether Americans are willing

HHassett.indbassett.indb 163163 111/8/121/8/12 9:279:27 PMPM 164 RETHINKING COMPETITIVENESS

to pay for a system with as few restrictions as ours, or whether their stated and revealed preferences are distorted by subsidies in the system that shield them from the full costs.

Inefficiency in Utilization within the United States: Geographic Variation. Treat- ment, utilization, and spending vary greatly across the United States. Spe- cifically, unadjusted Medicare spending varies around 55 percent between the highest and lowest quintile regions (Fisher et al. 2003 ; MedPAC 2011; Zuckerman et al. 2010). This difference shrinks to around 30 percent when differences in health status and price differences are adjusted for (MedPAC 2011; Zuckerman et al. 2010). While one might believe that patient preferences account for these differences, evidence suggests otherwise. For example, a high concentration of specialists (versus primary care physicians) is associated with higher spending areas (Baicker and Chandra 2004). Yet Anthony et al. (2009) conclude that patient preference for primary versus specialty care does not play a significant role in explaining variation, a finding bolstered by previous work of Fisher et al. (2003), Pritchard et al. (1998), and Anthony et al. (2009). In addition, patient preferences do not appear to play a role in very costly end-of-life care. Related literature illustrates that utilization also varies widely, both between regions and within markets. Wennberg and Gittelsohn’s (1973) research documents substantial differences in practice patterns for similar patients within Vermont. More recent research has confirmed variation for specific procedures, such as rates of cesarean deliveries, antibiotic prescrip- tions, hospitalizations, and types of dialysis access (Baicker et al. 2006; Baker et al. 2008; Hirth et al. 1996; McLaughlin et al. 1989; Wennberg and Gittelsohn 1973; Wennberg et al. 2004). A recent MedPac report estimates that while utilization varies less than spending, there is still a 30 percent difference in utilization between the highest and lowest quintiles. Likewise, differences in prices explain 17 percent of the total 55 percent variation between the ninetieth and tenth percentile areas.

High Prices in the United States. Evidence suggests that part of the reason why spending in the United States exceeds that in other countries is that Ameri- cans pay higher prices. High prices do not necessarily imply inefficiency.

HHassett.indbassett.indb 164164 111/8/121/8/12 9:279:27 PMPM AMERICAN COMPETITIVENESS AND HEALTH CARE 165

They may reflect a transfer from consumers to producers. Yet in standard economic models, distorted prices lead to distorted consumption. In health care this is a bit less clear because insurance offsets the impact of prices on consumer behavior. Whether the insurance effect offsets the high prices to yield efficient consumption is unknown, but programs such as tiered formularies for drugs and utilization review in managed care programs, implemented to reduce spending, may alter consumer behavior and generate distorted consumption. Measuring the differences in prices across countries is complex because market baskets consumed in different countries may differ and product definitions (and quality) may vary. For example, to conclude that prices are systematically higher in the United States, one needs to compare a standard set of services, and the choice and definition of these services may matter. In fact, work by Danzon and Furukawa (2006, 2008) found this to be the case with prescription drugs. Specifically, the gap between American and foreign prices for drugs decreased when the same formulations were compared across countries. Instead, the gap in prescription drug spending between the United States and other countries appeared to be related to access to new formulations. More broadly, typical approaches measure quantity (often crudely) and spending and infer the difference in spending not attributable to quantity is price. For example, much of the relevant literature uses broad product definitions, such as a hospital day. A common approach is to infer prices from measures of spending and crude measures of quantity such as physician visits, hospital days, or inpatient visits. These measures do not indicate the complexity or intensity of work, although substantial differences between the United States and other countries persist over years of data. For example, two studies from the 1990s using this approach document higher prices as the driver behind differences in expenditure between the United States and Canada (Fuchs and Hahn 1990; Welch et al. 1996). More recent analysis of OECD data and research published by the McKinsey Global Institute estimate that the prices in the United States are the primary driver behind spending disparity, based on the finding that spending in the United States is higher but utilization similar or even lower (Anderson et al. 2003; Farrell et al. 2008; Vladeck and Rice 2009). For instance, while Americans spend 40 percent more than Germans on four chronic conditions, they

HHassett.indbassett.indb 165165 111/8/121/8/12 9:279:27 PMPM 166 RETHINKING COMPETITIVENESS

receive 15 percent less services (defined broadly as physician visits or hospital days) (Anderson and Frogner 2008). Nevertheless, a recent report looking at more specific service definitions finds that the cost in the United States is substantially higher than abroad. For example, the average cost for an angioplasty was almost two and a half times more expensive than the next most expensive country (International Federa- tion of Health Plans 2010). For scanning and imaging (angiograms, various CT scans, MRIs), the United States paid the highest fee for every procedure except angiogram, for which only New Zealand paid more. The average hospital and physician spending for a normal delivery was more than 83 percent greater than the spending in the next most expensive country, Australia. The conclusion that spending in the United States exceeds that in other countries largely because of price is consistent with evidence that the United States devotes about the same amount of labor to health care as other countries (Pauly 1993). Results suggest that total employment in the U.S. health sector as a percent of the total workforce is just above the average of OECD countries. Pauly (1993) notes that the United States spends more on that labor. If the labor was equivalent, we would interpret the differences essentially as a price difference. But if labor in the U.S. was more productive at producing health, then we should consider this a quality effect. This is consistent with the notion that the differences in spending between the United States and other countries at the most aggregate level stems from price differences, though inefficiencies in use could exist to a greater or lesser extent at a more detailed level due to how those resources are allocated. Part of the reason why the United States spends more on health care labor is because the United States uses health care labor differently, with more specialists and fewer primary care physicians. One could classify the impact of specialists on spending as working through quantity (or quality), but it seems unlikely that, at a population level, more specialists lead to higher quality. While it is true that in some areas, specialists provide better care, particularly for well-defined conditions and when the match of condition to specialist is good, at a population level, quality appears not to be related to share of specialists (Hirth et al. 1996). For example, Baicker and Chandra (2004) find that state-level utilization of several high-value services is actually inversely related to the concentration of specialists. By comparing quality across all states and ranking them according to a composite

HHassett.indbassett.indb 166166 111/8/121/8/12 9:279:27 PMPM AMERICAN COMPETITIVENESS AND HEALTH CARE 167

score, these researchers find that increasing the number of specialists by 1 per 10,000 physicians is associated with a decrease in overall quality rank by almost nine places. Several other studies employing disparate methods come to a similar conclusion (Shi 1992, 1994; Shi et al. 1999; Starfield et al. 2005; Vogel and Ackermann 1998). So, then, why do we have such a large share of specialists? Some of the explanation may be due to the perception, contrary to the evidence, that specialist care is necessarily better than that provided by general physicians. Though this may be a contributing factor, one must be careful in emphasiz- ing patient preference as a cause because insurance obscures patients’ true willingness to pay for care from a specialist. Perhaps a more salient factor is the large disparity between specialty and primary care pay, a gap wider in the United States than in other countries (Commonwealth Fund et al. 2010). This pay disparity is driven a great deal by the administrative fee schedule in Medicare (which is often used as a template for private fees) that rewards procedure-oriented care. Historically, the Relative Value Unit updates (which define the Medicare fees for different services) have put relatively more value on “technical” procedural tasks and relatively less weight on “cognitive” ones, such as physician visits (Bodenheimer et al. 2007; Sandy et al. 2009). While the longer training and technical skills acquired by specialists (arguably) warrant some increased remuneration, the current gap in the United States is too large to strike a sustainable balance between specialty and primary care. The gap is likely exaggerated by such factors as differences in lifestyle and prestige and the emphasis in medical schools on specialization (Lambert and Holmboe 2005; Sandy et al. 2009). Apart from the existence of more specialists, there are several conceptual reasons to expect prices to be higher in the United States than abroad. Vladeck and Rice (2009) argue that providers constitute a monopoly that drive up per unit spending on services. These authors observe substantial provider market power in the United States generated by barriers to entry and what they describe “Americans’ cultural and political antipathy to certain forms of market power” (p. 3). By this they mean American opposition to strong centralized purchasing power, a prominent feature of the Euro- pean and Canadian systems. Another reason for high prices in the United States may be a lack of competition among providers. American consumers overwhelmingly are

HHassett.indbassett.indb 167167 111/8/121/8/12 9:279:27 PMPM 168 RETHINKING COMPETITIVENESS

insured against the cost of illness. This makes them insensitive to the prices charged by providers. In many cases they are even unaware of the prices charged. The insurance-induced insensitivity to price is exacerbated by information problems, including the inability to observe quality. If consumers take price as a sign of quality, demand elasticity will be dampened. Other models emphasize the nature of competition in the health care sector as a factor leading to high prices. Work by Pauly and Satterthwaite (1981), later augmented by Wong (1996), found empirical evidence to support Satterthwaite’s (1979) general hypothesis that a greater number of sellers complicates consumer search and leads to higher prices in the health care market.

Other Explanations for Inefficiency in the United States. Several other reasons, apart from models that emphasize high prices, may explain why Americans spend more than people in other countries. One model emphasizes the role of physicians in attracting patients to hospitals. Hospitals are assumed to compete for physicians by offering highly specialized (and expensive) services. This feature of competition has been labeled the “medical arms race.” The phenomenon of hospitals competing with nonprice mechanisms was documented in a series of papers in the 1980s that found a correlation between high hospital concentration and a substantially higher medical expenditure (Robinson and Luft 1985, 1987, 1988). While evidence suggests that this trend abated somewhat during the era of managed care (Bamezai et al. 1999; Gift et al. 2002), more recent work shows a resur- gence tied with the increase in specialty service lines (Berenson et al. 2006; Carey et al. 2009; Devers et al. 2003). A frequent rhetorical refrain used in health policy debates is that the United States spends more, and unnecessarily more, on “administrative” costs than its OECD peers. While studies consistently find relatively high administrative costs in the United States, the implications of this depend in large part on how we parse administrative versus nonadministrative costs and how these costs are translated to balance sheets. For instance, in six categories of administrative-related health spending,8 the United States paid around three times as much per person as Canada in 1999 (Woolhandler et al. 2003). Of these six categories, the largest gap was in employers’ cost to manage health benefits (approximately 600 percent), and the smallest

HHassett.indbassett.indb 168168 111/8/121/8/12 9:279:27 PMPM AMERICAN COMPETITIVENESS AND HEALTH CARE 169

gap was for nursing home administration (approximately 50 percent). In addition, nondelivery personnel make up a greater portion (27.3 percent) of our health sector employment, not including insurance industry personnel. Woolhandler et al. estimate that more than 31 percent of American health spending goes toward administrative costs. In addition, they estimate that physicians’ administrative work, which includes time spent dictating, sub- mitting claims, hiring staff, paying rent, and paying accounting and legal fees, comes to around 27 percent of physicians’ gross income. Another aspect of comparison is the relative cost of the United States’ predominantly private provision of health insurance and the more national- ized European models. In 2007 the United States spent an estimated $156 billion on insurance administration, not including the time costs to hospitals and physicians in negotiating fees (Commonwealth Fund 2009). Greater administrative spending in the United States is not in itself an inefficient arrangement, but given evidence of comparable quality across systems, there is a good case for substantial inefficiency and opportunity for a shift from administrative activity toward more productive, “health-generating” work. In summary, the United States spends a higher share of its GDP on health care than any other country. Evidence that we get more for that spending is scant. For example, common quality indicators do not suggest that the quality of care or patient satisfaction is better in the United States than in other countries. Thus, justification of our higher spending would need to rely on assertions that there are other features of our system, such as patient and provider autonomy, that we value. Furthermore, existing evidence suggests that our greater spending reflects, in part, higher prices, which may stem from imperfect competition among health care providers. This may represent a transfer as opposed to inefficiency, but responses to the high prices likely generate inefficiencies. In any case, there is no denying that Americans spend more on health care and that the concerns over such spending extend beyond the health care system to the economy as a whole.

The Health System’s Effect on the Economy The high cost of the American health care system does not necessarily imply that our health care system hampers our competitiveness. To the extent that the high costs are driven by high prices, consumers pay more, but providers

HHassett.indbassett.indb 169169 111/8/121/8/12 9:279:27 PMPM 170 RETHINKING COMPETITIVENESS

earn more. This is a transfer. Yet many have argued that the high cost does have meaningful broader effects. In this section, we consider four broad avenues by which the U.S. approach to health care provision could affect the economy: its effect on labor costs, costs for retirees, health effects, and broader macroeconomic effects due to the system’s overall cost. For each avenue, we consider the findings of the literature about the likely magnitude of impact and whether these should properly be classified as determining U.S. competitiveness.

Labor Costs. Not only does the health sector account for roughly one-sixth of U.S. economic activity; the U.S. approach to health care provision has substantially broader effects on the functioning of U.S. business through its two-pronged impact on the U.S. labor market. The first prong is the important role that employer provision of health insurance plays in the compensation of employees (compensation effects). The second prong is the potential for the health system to help determine the physical well-being and productivity of workers (physical effects). The compensation effects have received substantially more attention in the literature. One popular argument is that employer provision of health insurance disadvantages U.S. firms relative to counterparts based in countries that do not rely on employer-provided health insurance. The popular form of this argument was captured in an article in the Washington Post some years back: “For each mid-size car DaimlerChrysler AG builds at one of its U.S. plants, the company pays about $1,300 to cover employee health care costs—more than twice the cost of the sheet metal in the vehicle. When it builds an identical car across the border in Canada, the health care cost is negligible” (Downey 2004, p. E01). The same article goes on to offer a corporate take on the implications for U.S. competitiveness: “High health care costs have ‘created a competitive gap that’s driving investment decisions away from the U.S.,’ Ford Vice Chairman Allan Gilmour said in a speech at a recent auto industry conference. ‘If we cannot get our arms around this issue as a nation, our manufacturing base and many of our other businesses are in danger,’ he said, according to a transcript of the speech.” This line of reasoning raises several conceptual issues. First, in a standard model of labor demand, a firm will pay a worker the worker’s marginal

HHassett.indbassett.indb 170170 111/8/121/8/12 9:279:27 PMPM AMERICAN COMPETITIVENESS AND HEALTH CARE 171

revenue product. This is a theory of total compensation, not of cash wages. In the abstract, from the firm’s perspective, it makes little difference in what form the compensation is delivered: direct deposit to a bank account or through health insurance premiums. It may well matter to the employee, but these preferences presumably shape the firm’s balance of wages and benefits. An employee may prefer to receive health insurance through an employer because of the favorable tax treatment offered to such benefits in the United States. Even if there were an individual market for health insurance comparable to the group insurance market, the tax deductibility of health insurance benefits means that for a given level of total compensation, an employee could afford more generous insurance through an employer than that employee could purchase as an individual. The presumption, then, is that the cost of labor to business—the compensation bill for a firm’s employees—will be unaffected by the costliness of health benefits in the long run. Health cost increases could be offset by slower wage growth or even a cut in wages (see, for example, Gruber 1994; Gruber and Krueger 1991; Pauly 2003). There are a number of reasons why such offsets could be difficult. For example, in the short run, wages may be set through extended contracts, which might delay the offset. Other features of the market may impede a full wage benefit offset even in the long run. Institutional barriers, such as the minimum wage, represent one such barrier. However, market features could also yield the result that higher health care costs increase total compensation. For example, in equilibrium, total compensation reflects both labor supply and demand. While demand is based on marginal productivity, supply depends on the utility of any wage benefit package. If high health care costs reflect an inefficient system, it is possible that workers would demand higher compensation for any given amount of labor supplied. An upward shift in labor supply associated with inefficiency in the health care sector would yield higher wages. Firms would still be paying workers their marginal product, but because workers’ marginal product depends on the number of workers hired, the upward shift in the labor supply curve will result in fewer workers being hired at higher total compensation. We should note that high (or inefficient) health care costs are analogous to high (or inefficient) costs for any commodity. The institutional feature that ties health care premiums to employment is largely just an accounting

HHassett.indbassett.indb 171171 111/8/121/8/12 9:279:27 PMPM 172 RETHINKING COMPETITIVENESS

convention. Employers pay for employee food, housing, clothing, and every other consumption expenditure via the wage. We would expect wages (and hence total compensation) to reflect the costs of all these commodities. For example, we would expect to see higher wages in high-cost housing markets than low-cost housing markets because of a labor supply effect. This will affect total compensation and employment just as high health care costs might. Ultimately the impact of high health care costs on wages is an empirical question. While it is difficult to isolate the effects of increased insurance cost, most existing work suggests the trade-off is close to dollar for dollar (Gruber and Krueger 1991; Miller 2004; Olson 1994; Ryan 1997; Sheiner 1999). Moreover, there is some evidence that the wage offset is targeted to classes of workers experiencing high or increasing costs. For example, Gruber (1994) found that the wages of workers of childbearing age were disproportionately reduced by mandated coverage for maternity benefits. Likewise, Sheiner (1999) found that groups with higher baseline costs face a greater reduction in wages when overall insurance costs rise. Yet that literature is largely based on studies where variation in premium reflects benefits workers may value. Studies whose variation in costs may reflect inefficiency tend to find evidence consistent with the hypothesis that high health care costs in specific industries adversely affect employment in those industries and thus may affect equilibrium wages. A key distinction is that in the case of nonhealth goods, consumers have more say over the bundle they wish to purchase, whereas in health care the employer decides based on collective preferences and labor market forces. This may contribute to the “value gap” between the value employees place on health insurance (and indirectly the care it facilitates access to) and the cost of health insurance. Nevertheless, the key question is what wage benefit package employers can offer to attract employees and how that is influenced by health care costs. We expect that to the extent that workers value the health benefit and will trade wages for coverage dollar for dollar, high health care costs will not affect total compensation. Yet if workers in high-cost areas do not value health benefits sufficiently, they will demand higher compensation, and we may expect some equilibrium labor market effects. The extent to which workers will be able to maintain higher wages will depend on the extent to which the output is tradable across countries.

HHassett.indbassett.indb 172172 111/8/121/8/12 9:279:27 PMPM AMERICAN COMPETITIVENESS AND HEALTH CARE 173

Greater tradability will lead to similar total compensation, and a shift in the supply curve will result in a smaller domestic industry. Yet this does not imply a drop in aggregate national competitiveness. Workers may shift to other industries in the long run, as the theory of comparative advantage would suggest. A drop in competitiveness requires some industries to be better than others and for workers to shift away from those industries. The Washington Post story mentioned above regarding an auto investment that might take place in either Detroit or Windsor would seem to be an example of the effects of health care on U.S. competitiveness. There is an identifiable prize—the investors’ capital—and the configuration of the U.S. health care system arguably alters the likelihood of winning this prize. Yet the case also illustrates some of the pitfalls of this line of argument. Let us assume that U.S. and the Canadian auto workers are equally productive. There is no obvious reason why the total compensation of the U.S. worker would need to be higher than the total compensation for the Canadian counterpart. A discrepancy in total compensation could certainly occur if wage levels were fixed through a collective bargaining agreement, for example. If the mandated wages were sufficiently high that the addition of employer-provided health benefits pushed the total U.S. compensation package above the Canadian level, the U.S. worker could look relatively expensive, and the investment could head north. But that is generally true of situations in which compensation is held above market-clearing levels and would be at least as attributable to the fixed wage levels as to the role of health benefits in the compensation package. If workers in the United States demand higher wages because health care (or housing) is more expensive, we could also see labor-intensive industries avoiding these areas. The fundamental question this raises is whether this shift will be offset by general equilibrium labor market adjustments. More generally, industry location decisions (and, in turn, the composition of national output) will be based on multiple determinants, including compensation levels, productivity, quality of living (nonpecuni- ary benefits), taxation, regulatory environment, and infrastructure. Is one configuration inherently better than another? In theory, some configurations are better than others. This brings us back to the question of prizes from our initial discussion of competitions. In the example of attracting the capital investment for a new factory, for

HHassett.indbassett.indb 173173 111/8/121/8/12 9:279:27 PMPM 174 RETHINKING COMPETITIVENESS

example, the question translates into whether the community gets more than it pays. There is a global cost of capital, and the owners of that capital will generally demand that return. If a particular industry, however, gives better than it gets, then it may qualify as a prize. This could be because it trains its workers, stimulates other local industries through its innovation, or captures unusual rents because it is in an imperfectly competitive industry. In each of these cases, the industry is a prize because it generates positive externalities. An economy with higher health care costs could be particularly undesir- able if it served to disproportionately discourage industries with such positive externalities or spillovers. To our knowledge, the literature has not yet established such an argument. To do so would first require identification of industries with positive local spillovers, a contentious exercise.9 Further- more, to the extent that positive-spillover industries depend on high skills and therefore high wages, they would have greater capacity for wage offsets when faced with high health care costs. Thus, there may be a presumption that they would be relatively less susceptible to health cost discouragement than other industries. It is, in fact, this relative effect on industries that matters. It is tempt- ing to argue that high U.S. health care costs discourage employment in all industries, but the U.S. system has coexisted with periods of both high and low unemployment. Furthermore, other OECD nations with lower health care costs have certainly experienced periods of high unemployment. In fact, some analyses of the deleterious employment effects associated with high health care costs are partial equilibrium analyses. For example, Sood et al. (2009) examine the association between high health care costs across industries. Reductions in employment in industries that provide generous health insurance to workers (and thus are more affected by high health care costs) may be offset by gains in industries less likely to provide workers with coverage. However, Baicker and Chandra (2005) employ an alternative method by using malpractice premiums as an instrumental variable at the state level. They find that a 10 percent increase in health insurance premiums is associated with a 1.6 percent reduction in probability of employment and a 1.9 percent increase in the probability that a worker is employed only part-time (Baicker and Chandra 2005). This raises the possibility that health care costs do affect aggregate employment and potentially competitiveness.

HHassett.indbassett.indb 174174 111/8/121/8/12 9:279:27 PMPM AMERICAN COMPETITIVENESS AND HEALTH CARE 175

Retiree Costs. A related argument about high costs of care in the United States relates to the burden employers face from retiree coverage. This burden has been felt in the last twenty years especially, as the number of firms offering retiree coverage has more than halved since the late 1980s (Kaiser Family Foundation and Health Research and Educational Trust 2009). One of the most prominent instances in which retiree health costs figured prominently in corporate viability was the bankruptcy of General Motors. In 2009 the company, which had once been a paragon of American corporate might, was compelled to reorganize (Ingrassia 2010). One driving force was the legacy of generous health benefits offered to employees and retirees. The open-ended nature of this commitment meant that GM was more exposed to rising health costs than many of its competitors. In this sense, it could be argued that rising health costs dramatically affected the competitiveness of an American industry. Claims to this effect were cited above. Multiple factors led to GM’s bankruptcy. If we oversimplify and focus in on the burgeoning cost of retiree health benefits, we find several reasons to question the linkage between health costs and national competiveness. First, as described above, there is an important distinction between industry competitiveness and national competitiveness. Second, even in this instance, what was bad for GM was not equally bad for the entire auto industry in America. Not all auto manufacturers had made equivalent promises to their retirees. Finally, even for GM, retiree health demands were ultimately separate from the operations of the firm, since they were formally separated in the bankruptcy process. In general, retiree health benefit expenses are a fixed cost to firms and are unlikely to affect existing behavior, though they may affect firm viability and induce bankruptcy.10 It is certainly true that long-term commitments to retirees for health coverage posed a significant risk to firms that offered them, given the variability of health costs. But such long-term bets have ample precedents among firms that have long investment lead times or make other irrevocable decisions well in advance. In this particular respect, there is nothing unique about the effects of health costs. In essence, the problem was both the generous promises about retiree health benefits made to workers in the auto industry and rising health care costs per se. Even if the rising costs reflected rising value, GM would have faced a higher-than- anticipated fiscal burden.

HHassett.indbassett.indb 175175 111/8/121/8/12 9:279:27 PMPM 176 RETHINKING COMPETITIVENESS

Health Effects. Another pathway through which our health care sector could influence competitiveness is by influencing worker health and thus productivity. In cases of developing countries, or even in past decades in the United States, it was reasonable to believe that health care affected productivity. For example, at the beginning of the twentieth century, John. D. Rockefeller made a $1 million donation to form the Rockefeller Sanitary Commission, whose purpose was to eradicate hookworm, then prevalent in the American South. Bleakley (2007) recently evaluated the economic effects of hookworm eradication and found that school enrollment, atten- dance, literacy, and longer-term financial improvements can be linked to the eradication of the parasite. Looking more directly at agricultural productivity during this period, Brinkley (1995) finds that half of the 16.33 percent increase in southern agricultural income between 1910 and 1920 can be attributed to the eradication of hookworm. Even if health has an important impact on productivity, it is not clear that the American health care system generates a beneficial effect through this pathway. For example, many of the interventions that affect health in ways that would impact productivity are likely to be interventions related to worker behavior, including diet and exercise. Medical interventions, with the exception of care for chronic diseases, may not be as salient. When it comes to managing chronic disease, the U.S. health care system (and general U.S. population) does a relatively poor job. For example, McGlynn et al. (2003) find that Americans receive only about 50 percent of recommended care, much of which relates to chronic disease management. Moreover, American adherence to medications for treating chronic disease is far from perfect (Gibson et al. 2005, 2006; Goldman et al. 2004). Finally, to the extent that the American health care system devotes considerable resources to those over the age of sixty-five or otherwise out of the labor force, it is unlikely that our system would boost competitiveness. Of course, relative performance matters, but at least compared to other developed countries, our performance on treating chronic disease is not exceptional (Commonwealth Fund et al. 2010).

Broader Macroeconomic Effects. Apart from the impact of the health care system on the economy via cost or health effects, the institutional features of the American health care system may have an impact. Specifi-

HHassett.indbassett.indb 176176 111/8/121/8/12 9:279:27 PMPM AMERICAN COMPETITIVENESS AND HEALTH CARE 177

cally, our system is employer based and, prior to the Affordable Care Act, voluntary. Thus some firms offered coverage, and others did not. Even the mandates in the Affordable Care Act do not force all firms to offer coverage, so heterogeneity in firm offerings will likely continue. This employer-based system may influence the labor market in several ways and thus influence competitiveness.

Job Lock. The employer-based system of insurance may influence job mobility. Specifically, because our system is tied to employment, Americans may be less willing to switch jobs for fear of losing coverage. This phenomenon is termed “job lock.” The empirical evidence on job lock suggests a large effect between 20–30 percent, although this value is somewhat questionable due to differences in study design. Several studies that find large coefficients (22–32 percent) do not have statistically significant results due to lack of power (Buchmueller and Valletta 1996; Mitchell 1982), and two studies find a small and insignifi- cant effect (Holtz-Eakin 1994; Kapur and House 1998). In addition, Gruber and Madrian (1996, 1997) find a significant but much smaller (12–15 percent) effect when COBRA was implemented, a law that ensured some continued coverage. Of those analysts who find large effects, Anderson (1997) evaluated only men with pregnant spouses, and Buchmeuller and Valletta (1996) included only men and women with dependent spouses. As with health care costs, inefficiencies in the system may not affect competitiveness. An impact on competitiveness requires that some industries or jobs be more desirable than others and that job lock adversely affects our ability to attract those industries or jobs.

Taxes. To this point, our discussion has largely addressed the potential disadvantages of costly private financing and provision of health care, mostly channeled through employers. It is worth noting, however, that the proper comparison is not to free provision of health care but to viable alternative systems. The dominant alternative system in the world is public provision of insurance (and, in some cases, care) financed by taxes and generally more heavily regulated. These systems generally include some form of mandated (or automatic) coverage. Of course, in the United States, there is a large (and soon to be larger) public system, including Medicare and Medicaid.

HHassett.indbassett.indb 177177 111/8/121/8/12 9:279:27 PMPM 178 RETHINKING COMPETITIVENESS

There are important costs associated with a tax-financed system that can parallel those of employer-provided insurance. The extent of those costs will depend on the efficiency of tax systems, which lies beyond the scope of this chapter. However, we note one potential inefficiency associated with taxes that might affect competitiveness: the crowding out of other public investments. Specifically, if voters have a limit regarding the absolute rate of taxation they will tolerate, health care spending will divert resources away from other activities, such as building infrastructure or education (Rein- hardt 1989). Such a diversion can impact the development of human capital or infrastructure that promoted economic activity. Clearly that can affect well-being. It will affect competitiveness if it drives away desirable industries or jobs. Overall, two broad observations related to tax-financed coverage and provision of care apply: (1) the less efficient the system of taxation, the costlier public provision will be; and (2) the more U.S. costs reflect inefficiencies as opposed to greater quality or quantity, the more desirable an efficient system of public provision of care (or more heavily regulated private provision) will look. Yet, while the relative desirability of the American system will depend on the distortions associated with taxes and the inefficiencies through the public and private health care sectors, the impact of those features on competiveness reflects their impact on desirable jobs and industries.

Comparing International Health Approaches The outcomes of the American health care system, relative to those in others systems, will reflect, in part, the structural differences.

The American Health Care System. In analyzing any health care system, we need to distinguish between health care financing and health care delivery. In both areas, the American health care system is a mixed, public/ private system. In health care financing the government pays for about 45 percent of care (spending) through government programs (OECD 2010). The largest of these is Medicare, which is financed by a payroll tax. Medi- care provides coverage for hospital, physician, prescription drug, and some

HHassett.indbassett.indb 178178 111/8/121/8/12 9:279:27 PMPM AMERICAN COMPETITIVENESS AND HEALTH CARE 179

limited long-term care services for retired and disabled beneficiaries. Dental services and vision services are excluded. Most Medicare beneficiaries (about 80 percent) are enrolled in the traditional Medicare program, which imposes a government-set fee schedule for all services except prescription drugs. The traditional Medicare benefit package does not pay for the full cost of services. Beneficiaries face a number of cost-sharing requirements, including inpatient deductibles and 20 percent coinsurance for physician and outpatient visits. Almost all providers choose to participate in Medicare, and thus enrollees in the traditional Medicare program face few limits on choice of provider. The Medicare program relies on markets in a number of ways. First, most Medicare beneficiaries have supplemental coverage for the gaps in Medicare coverage (89 percent in 2007) (Kaiser Family Foundation 2009). Coverage is provided by private insurers and purchased by former employers (34 percent of Medicare beneficiaries, including those without supplemental coverage) or by individuals in a regulated market for such coverage (39 percent) (Kaiser Family Foundation 2009). Second, prescription drugs for Medicare beneficiaries are covered in a separate program, which relies on private insurers to provide coverage, subject to some constraints on the benefit packages. These private insurers negotiate with drug companies over the price of drugs and determine the formulary and cost-sharing provisions faced by enrollees. Third, about 20 percent of Medicare beneficiaries choose to be enrolled in the Medicare Advan- tage program. This program provides premium support for private health plans that agree to provide beneficiaries with coverage at least as generous as the benefit package in traditional Medicare, though most plans provide more generous coverage. The Medicare Advantage plan uses a wide range of techniques, including restrictions on provider networks, to control the cost of care. In addition to the Medicare program, low-income Americans may qualify for the Medicaid program. Most beneficiaries are low-income individuals with families, but most expenditures are for long-term care services for individuals with low income and assets (perhaps because they spent down to Medicaid eligibility thresholds). Medicaid is a joint federal/state program with details such as program eligibility, fees, and reliance on private plans decided at the state level but subject to some federal restraints.

HHassett.indbassett.indb 179179 111/8/121/8/12 9:279:27 PMPM 180 RETHINKING COMPETITIVENESS

For individuals who do not qualify for Medicare or Medicaid (or a few other government programs), insurance is most commonly provided by employers. Small- and medium-size employers typically purchase coverage from a private health plan. Large employers often self-insure and use private plans to administer benefits. Private insurers typically pay providers 25 percent more than Medicare for services, while Medicaid fees are generally even lower. In addition, there is a small market for individual coverage, which is generally expensive. Still, around 50.7 (16.7 percent) million Americans were uninsured as of 2009. While these individuals had some access to care through community health systems and other safety net providers, evidence suggests they received fewer services and had more difficulty accessing care. In the future, provisions in the Affordable Care Act will require all citizens to buy coverage and will provide substantial subsidies starting in 2014, which is expected to substantially reduce the number of insured Americans. The health care delivery system in the United States is dominated by private firms, including hospitals and medical practices. Most hospitals are nonprofit firms, commonly affiliated with larger hospital systems. Only 18 percent of hospital beds are in for-profit entities (Kaiser Family Foundation 2010). Physician distribution in the United States is heavily skewed away from primary care physicians. For example, 35 percent of practicing physicians in the United States are considered to practice primary care (family medicine, general medicine, general internal medicine, and pediatricians) (Dodoo et al. 2005; Macinko et al. 2007). In addition to the private provider system, there are some publicly run facilities. About 22 percent of hospitals are in government-run facilities (including county hospitals and Veterans Administration hospitals) (Health Forum LLC and American Hospital Association 2010). There is also a system of safety-net providers, including federally qualified health centers and public hospitals. Public hospitals service around ten million people per year, and more than sixteen million use federally qualified health centers as their usual source of care (Commonwealth Fund 2006; Regenstein and Huang 2005). The typical patient receiving care at one of these centers earns below 100 percent of the federal poverty level and is either uninsured (38 percent) or covered by Medicaid (35 percent) (U.S. Department of Health and Human Services 2008).

HHassett.indbassett.indb 180180 111/8/121/8/12 9:279:27 PMPM AMERICAN COMPETITIVENESS AND HEALTH CARE 181

Other Countries. The health care systems of most other countries are much more heavily dominated by government involvement. Typically the government takes a larger role in financing care and is more stringent in regulating provider markets, particularly prices.

United Kingdom. One extreme is the United Kingdom, where both the delivery and financing of health care are largely government dominated. Legal residents of the United Kingdom’s four constituent countries (England, Scotland, Wales, and Northern Ireland) are provided with universal coverage and access via the National Health Service (NHS). The NHS is funded via direct taxes, value-added taxes, income taxes and local taxation, national insurance contributions, and user contributions at point of service. Every three years, the government sets budgets for the NHS as a whole and for primary care trusts, which account for 85 percent of the NHS budget, and receive funding based on a risk-adjusted capitation formula. Although the program is funded centrally, the tasks of purchasing services and making specific policy decisions lie with the individual countries as well as with local primary care trusts, and so characteristics of coverage may vary slightly from country to country. Approximately 80 percent of national health expenditures in the United Kingdom are publicly financed. Most UK residents rely on the National Health Service for the receipt of medical care. Coverage under the NHS system includes physician services, inpatient and outpatient hospital care, preventive services, mental health care, rehabilitation, learning disabilities treatment, and inpatient and outpatient drugs. Unlike Medicare recipients, NHS beneficiaries do not face any cost-sharing requirements for the receipt of these basic medical services, although a few cost-sharing requirements exist for optical and dental services and some prescription drugs. These point-of-service costs account for slightly more than 10 percent of total health expenditures (WHO 2007). The system has in place several safety nets for those with low income and other special groups. Prescription drug copayments are waived for those with low income, children under the age of sixteen, full-time students between the ages of sixteen and eighteen, individuals over the age of sixty, women who are pregnant or have had a child in the last year, and individuals who have disabilities or certain medical conditions. Those who require large quantities of prescription drugs are likewise eligible for discounts.

HHassett.indbassett.indb 181181 111/8/121/8/12 9:279:27 PMPM 182 RETHINKING COMPETITIVENESS

A private health insurance industry, including both for-profit and nonprofit insurers, offers supplementary coverage alongside the NHS. Pri- vate health insurance offers an alternative to some of the inconveniences of the NHS system, including greater comfort and privacy, choice of specialists, and no-wait lines for elective procedures. Care can be obtained at both private and public (NHS) hospitals. Private coverage plays a minimal role in the UK health care system, accounting for only 1 percent of total expenditures in 2009, with only 12 percent of UK residents obtaining private coverage (OECD 2011). Out-of-pocket expenditure represents around 90 percent of spending in the relatively small private insurance market (Boyle 2008). Local commissioners of health services control much of the NHS budget and contract with providers to organize the infrastructure for service delivery. While technically self-employed, most general practitioners are effectively employees of the NHS, receiving reimbursement directly from primary care trusts via a mix of salary, capitation, fee for service, and, more recently, pay-for-performance arrangements. Specialists perform most of their work in NHS hospitals and cannot be consulted without a referral from a general practitioner. While most providers work for the NHS, some may choose to have a private practice or to supplement their income by providing services to private patients. Private providers set their own fee- for-service rates and are not reimbursed by the NHS. The system is made up of both public and private hospitals. Public hospitals include those organized as NHS trusts that answer to the Department of Health, as well as self-governing, semiautonomous foundation trusts. Private hospitals that are owned and staffed by private providers also exist, and in some cases the NHS purchases care from these hospitals.

Canada. The Canadian model shares some similarities with the United Kingdom’s model (budgets), with a few differences. Universal coverage is provided to all Canadian residents under the system known as Medicare. Care is for the most part publicly funded via federal (income), provincial, and territorial taxes, while care is provided via the private sector. Federal funds are transferred to the individual provincial and territorial governments, which in turn administer their own health plans and determine local physician salaries and service fees, while abiding by the restrictions set by

HHassett.indbassett.indb 182182 111/8/121/8/12 9:279:27 PMPM AMERICAN COMPETITIVENESS AND HEALTH CARE 183

the Canada Health Act. In 2006, 65.7 percent of Canada’s health expenditures were publicly funded. As part of the Medicare system, Canadian residents receive coverage for physician, hospital, and diagnostic services that are deemed medically necessary; no copayments are charged for the receipt of these services. Prescription drugs, medical equipment, and optical and dental services are not covered under Medicare, and as such, residents have the option of paying out of pocket for the full cost of care or obtaining supplementary private coverage. A majority of Canadians (65 percent) enroll in private supplementary coverage for coverage of the aforementioned services that are not covered or only partially covered under Medicare. Private plans do not offer coverage for core medical services because, unlike in the UK, provincial laws to ensure equity in access to and quality of care prevent private plans from offering coverage of services that are available via Medicare. Most hospitals in Canada operate as nonprofit municipalities, voluntary organizations, or community boards of trustees. While relatively autono- mous in daily operations, they must adhere to annual global operating budgets set by provincial or territorial governments. Approximately 40 percent of hospitals are private, while only 1 percent are private and for profit (OECD 2011). Physicians working within the Canadian system are privately employed and contract with provincial or territorial plans for compensation. They are compensated on a fee-for-service basis, with payment reflective of fee schedules negotiated locally. Most primary care physicians work in small-group practices and are responsible for providing primary medical services (prevention, injury/disease treatment, and emergency care) and coordinating referrals to specialists. Incentives have been placed in the system to discourage self-referral, although a referral is not required for specialist visits. Approximately 47 percent of Canadian providers are general practitioners; the remaining 53 percent provide specialist care (OECD 2011).

Japan. Japan has a universal health insurance system and requires that all individuals obtain coverage via one of three insurance schemes: employer- based insurance, national (public) insurance, or insurance for the disabled

HHassett.indbassett.indb 183183 111/8/121/8/12 9:279:27 PMPM 184 RETHINKING COMPETITIVENESS

or elderly. Japan has an employment-based financing system, with a combination of federal funds, employer and employee payroll taxes, and income-based coinsurance payments from the self-employed funding the health care system (Commonwealth Fund 2011). Though subject to strong government regulation, the provision of health care within the Japanese system is mostly privatized, with privately operated health insurance plans, hospitals, and clinic systems. In 2008, 81 percent of total health expenditures were publicly financed (OECD 2011), with around 60 percent of the population covered by employers. All three insurance schemes offer similar health benefits, that are mandated under a statutory benefit package, although there are small differences in individual contribution to medical costs. Premiums vary according to income, and monthly coinsurance amounts depend on the type of program in which the individual is enrolled. Physician services, hospital care, most dental care, prescription drugs, and mental health care are covered in all plans. Coinsurance rates for services are generally around 30 percent, with monthly limits that vary by income. Currently, more than 3,500 health insurers participate in the Japanese health system (Common- wealth Fund 2011). Japan’s national health program offers coverage for individuals who do not receive employer-sponsored coverage or are not eligible for the country’s plan for the elderly or disabled. Beneficiaries are also charged a fixed premium (split by employee and employer, or government if self-employed) as well as the 30 percent coinsurance fee for all health care costs. Households below the poverty line are eligible for welfare support and receive medical services devoid of any cost sharing. Ambulatory providers in the Japanese delivery system work from private, nonprofit office settings. Reimbursement is allocated on a fee-for- service basis and must adhere to a uniform national fee schedule set by the Ministry of Health and Welfare (in other words, providers are not allowed to charge rates above the nationally recommended price for a certain procedure, and all insurers must pay the same fees to providers). Rates for the national fee schedule, known as the point-fee system, are updated every two years and are based simply on the complexity of the procedure, with no authorized variations on the basis of provider geographic location or

HHassett.indbassett.indb 184184 111/8/121/8/12 9:279:27 PMPM AMERICAN COMPETITIVENESS AND HEALTH CARE 185

setting, qualification, or the actual cost of services rendered (Common- wealth Fund 2011). Most hospitals are nonprofit and privately owned and operated (around 55 percent of beds) (Commonwealth Fund 2011). The remainder of hospitals are operated by national, municipal, or prefectural governments or quasi-public agencies. Hospitals may elect to receive either per diem rates that are case-mix adjusted, or to be paid fee-for-service (FFS) under the nationally set fee schedule. Around half of the hospitals choose the per-diem rate (which covers physician fees as well as hospital fees), whereas the other half choose FFS. In addition, the Western paradigm of specialist versus primary care is not pervasive in Japan, as many specialists can be accessed without referral and provide primary care services (Com- monwealth Fund 2011).

France. France’s national system of health insurance, Couverture Maladie Universelle (CMU), provides coverage to all citizens and foreign residents of France. The public system is financed by a series of federal and state taxes, including a national income tax, employer and employee payroll taxes, taxes on the revenue of private health insurers, and taxes on tobacco and alcohol sales. In 2007, 79 percent of France’s heath expenditures were publicly financed. The CMU offers full coverage for ambulatory physician services, inpatient and outpatient hospital care, and prescription drugs. Coverage for outpatient optical and dental services is limited. An intricate system of cost sharing applies for all services and drugs that are publicly financed. Coinsurance rates ranging from 0 percent to 50 percent of the full service cost are charged and vary based on the type of care (doctor visit vs. hospital stay), patient class (chronic or poor vs. average), therapeutic value of the service (prescription drug) received, and compliance with the extant gatekeeping system. Many of these coinsurance fees can be reimbursed by supplementary private health insurance, but users are also responsible for nonreimbursable copayments for expensive treatments and ambulance journeys and for a small per-prescription drug and per-doctor visit fee. A majority of the population in France (90 percent) has supplementary private health insurance, which covers most cost-sharing requirements for

HHassett.indbassett.indb 185185 111/8/121/8/12 9:279:27 PMPM 186 RETHINKING COMPETITIVENESS

publicly financed benefits (Commonwealth Fund and Durand-Zaleski 2010). Most private health insurance plans are nonprofit, employment- based mutual associations, with coverage limited to the same services offered via public health insurance. These complementary private plans are also available to low-income individuals with substantial government subsidies. In addition, some commercial insurance plans (both for profit and nonprofit) offer services not defined in the public benefits package. These commercial plans represented around 20 percent of supplementary options throughout the 1990s and early 2000s (Buchmueller and Couffinhal 2004). The system has an extensive safety net so as to ensure that those with chronic illnesses, disabilities, and low income receive adequate care. In addition to free optical and dental care, the system offers vouchers for supplementary private health insurance (at little or no cost) for individuals and families with low income. Low-income patients are exempted from coinsurance for most services and nonreimbursable copayments, and doctors are required to charge reference prices for all services rendered to low-income patients. Coinsurance exemptions also apply to all patients with specific chronic illnesses and individuals receiving invalidity and work injury benefits. The CMU exempts all patients, regardless of income or health status, from cost sharing for certain surgical services as well as hospital fees beyond the first thirty-one days of hospitalization. About one-third of French hospitals are private, for-profit organizations; the remaining two-thirds of hospitals are nonprofit, government-owned institutions. While private hospitals only serve patients with supplementary private insurance, public hospitals are accessible to both private and public- only patients. Under the CMU system, physicians are self-employed and have the prerogative to set their own fee-for-service rates, which typically vary as a function of professional experience. The government, public insurance scheme, and medical unions negotiate reference prices, which doctors can refer to as they set their own rates. Government reimbursement for services is limited to the reference prices that are set; any difference between the reference price and charged price must be covered by the patient or supplementary private health insurance plan. Physicians working in public, nonprofit hospitals are salaried.

HHassett.indbassett.indb 186186 111/8/121/8/12 9:279:27 PMPM AMERICAN COMPETITIVENESS AND HEALTH CARE 187

Conclusions

In lieu of rendering a verdict on whether the U.S. health care system undermines national competitiveness, this chapter has sought to clarify the issue. The U.S. system differs from those in other major economies in important ways, ranging from the level of per-capita spending to the reliance on employer provision. One central question in the literature has been whether the higher U.S. levels of spending reflect inefficiencies or whether Americans are getting more for their substantial expenditures. The literature has struggled with this question for several reasons. It is difficult to separate price effects from quality effects in studies, and the obvious comparisons in health outcomes can be muddied by factors that lie outside the provision of medical care (such as violent deaths). Even were we able to present a clear answer on the relative efficiency of the U.S. health care system, the implications for competitiveness are not straightforward. The competitiveness of a nation is quite distinct from the competitiveness of an individual firm or even an industry. We set a high standard for national competitiveness—a country had to be engaged in a zero-sum competition with other countries. While it is not difficult to imagine such a competition, it is much harder to establish the case empirically, demonstrating the existence of worthwhile prizes, such as industries with positive local spillovers. Even if we were to stipulate that certain industries qualified as prizes, it would remain to be demonstrated that the particular features of the U.S. health care system made the country more or less likely to snatch those prizes. There are unresolved arguments in the literature about whether high health care costs even have an appreciable effect on total compensation within industries, much less whether total compensation effects demonstra- bly alter the outcomes of these contexts. This all describes a relatively tenuous link between the U.S. health care system and international competitiveness. In no way does this imply, however, that the inefficiencies and failings of the U.S. health care system are unimportant. Even if they do not alter the results of particular international competitions, they can have a very significant impact on the well-being of Americans. Although warnings about international competition may serve

HHassett.indbassett.indb 187187 111/8/121/8/12 9:279:27 PMPM 188 RETHINKING COMPETITIVENESS

as an easy way to stir up sentiment about health care change, potential improvements ought to be judged on their own merits.

Notes 1. We must note that these studies are, as the authors acknowledge, based on partial equilibrium analysis. The authors are careful not to make claims about the aggregate impact of health insurance costs on the overall economy. 2. As one example, see Coe, Helpman, and Hoffmaister (1997). They find that research and development activities in developed countries enhance productivity in developing countries. 3. More generally, we can think of changing the probability distribution over potential outcomes. 4. In some ways, the narrower definition of competitiveness that we adopt may seem paradoxical. If we imagine, for a second, that every industry in a country produces internationally traded goods and services, then how can it be that each industry is engaged in an international competition while the economy as a whole—the sum of all those parts—may not be? The answer depends on linkages between the sectors that drive general equilibrium effects. Before all sectors could lose competitions, the losses of some would drive down costs for the others. Conversely, before every sector could win at its international competition, the incipient expansion of output would tax limited resources and drive up costs, causing some to contract. Thus, cutthroat industry-level competition can coexist with ensured national survival. 5. We are using national productivity loosely to describe gains that do not stem from international contests. There are more precise definitions common in the study of national accounts. Note that this definition describes international Pareto improvements. 6. Amount in 2000, adjusted for purchasing power parity. 7. Australia, Canada, Germany, Netherlands, New Zealand, United Kingdom, and the United States. 8. The categories are insurance overhead, employers’ cost to manage health benefits, hospital administration, nursing home administration, administrative costs of practitioners, and home care administration. 9. For one seminal example of assessing the spillover effects of an industry, see Irwin and Klenow (1994). Note that the spillovers must be local. If the industry throws off benefits that spread rapidly around the world, then there is no particular benefit to having that industry nearby. 10. It has been argued that the financial squeeze imposed on GM by retiree costs pushed the automaker to compromise on car quality, to the detriment of its performance. Whether or not that occurred in GM’s case, a firm with a large fixed cost should attempt to maximize profits through its production and design choices, just as an unencumbered firm would. There may be other explanations for unfortunate management decisions.

HHassett.indbassett.indb 188188 111/8/121/8/12 9:279:27 PMPM AMERICAN COMPETITIVENESS AND HEALTH CARE 189

References Anderson, Gerard F., and Bianca K. Frogner. 2008. “Health Spending in OECD Countries: Obtaining Value per Dollar.” Health Affairs (Millwood) 27(6):1718–27. Anderson, Gerard F., Uwe E. Reinhardt, Peter S. Hussey, and Varduhi Petrosyan. 2003. “It’s the Prices, Stupid: Why the United States Is So Different from Other Countries.” Health Affairs (Millwood) 22(3):89–105. Anderson, Patricia M. 1997. The Effect of Employer-Provided Health Insurance on Job Mobility: Job-Lock or Job-Push? Unpublished paper, Dartmouth. Anthony, Denise L., M. Brooke Herndon, Patricia M. Gallagher, Amber E. Barnato, Julie P. Bynum, Daniel J. Gottlieb, Elliott S. Fisher, and Jonathan S. Skinner. 2009. “How Much Do Patients’ Preferences Contribute to Resource Use?” Health Affairs (Millwood) 28(3):864–73. Baicker, Katherine, Kasey S. Buckles, and Amitabh Chandra. 2006. “Geographic Variation in the Appropriate Use of Cesarean Delivery.” Health Affairs (Millwood) 25(5):w355–67. Baicker, Katherine, and Amitabh Chandra. 2004. “Medicare Spending, the Physician Workforce, and Beneficiaries’ Quality of Care.” Health Affairs (Millwood) (Supple- ment, Web exclusive):w184–97. ———. 2005. The Labor Market Effects of Rising Health Insurance Premiums. NBER Working Paper 11160. National Bureau of Economic Research, Cambridge, MA. Baicker, Katherine, Amitabh Chandra, and Jonathan S. Skinner. 2005. “Geographic Variation in Health Care and the Problem of Measuring Racial Disparities.” Supple- ment, Perspectives in Biology and Medicine 48(1 Suppl):S42–53. Baker, Laurence C., Elliott S. Fisher, and John E. Wennberg. 2008. “Variations in Hos- pital Resource Use for Medicare and Privately Insured Populations in California.” Health Affairs (Millwood) 27(2):w123–34. Bamezai, Anil J., Jack Zwangziger, Glenn A. Melnick, and Joyce M. Mann. 1999. “Price Competition and Hospital Cost Growth in the United States (1984–1994).” Health Economics 8(3):233–43. Berenson, Robert A., Thomas Bodenheimer, and Hoangmai H. Pham. 2006. “Specialty- Service Lines: Salvos in the New Medical Arms Race.” Health Affairs (Millwood) 25(5):w337–43. Bleakley, Hoyt. 2007. “Disease and Development: Evidence from Hookworm Eradica- tion in the American South.” Quarterly Journal of Economics 122(1):73–117. Blendon, Robert J., Mollyann Brodie, John M. Benson, Drew E. Altman, and Tami Buhr. 2006. “Americans’ Views of Health Care Costs, Access, and Quality.” Mil- bank Quarterly 84(4):623–57. Bodenheimer, Thomas, Robert A. Berenson, and Paul Rudolf. 2007. “The Pri- mary Care-Specialty Income Gap: Why It Matters.” Annals of Internal Medicine 146(4):301–6. Boyle, Sean. 2008. The UK Health Care System. Commonwealth Fund. http://www. commonwealthfund.org/usr_doc/UK_Country_Profile_2008.pdf?section=4061.

HHassett.indbassett.indb 189189 111/8/121/8/12 9:279:27 PMPM 190 RETHINKING COMPETITIVENESS

Brinkley, Garland L. 1995. “The Economic Impact of Disease in the American South, 1860–1940.” Journal of Economic History 55(2):371–73. Buchmueller, Thomas, and Agnes Couffinhal. 2004. Private Health Insurance in France. OECD Health Working Papers. http://www.oecd.org/dataoecd/35/ 11/30455292.pdf. Buchmueller, Thomas, and Robert Valletta. 1996. “The Effects of Employer-Pro- vided Health Insurance on Worker Mobility.” Industrial and Labor Relations Review 49:439–45. Carey, Kathleen, James F. Burgess and Gary J. Young. 2009. “Single Specialty Hospi- tals and Service Competition.” Inquiry 46(2):162–71. Coe, David T., Elhanan Helpman, and Alexander W. Hoffmaister. 1997. “North-South R&D Spillovers.” Royal Economic Society. Economic Journal 107(440):134–49. Commonwealth Fund. 2006. The Commonwealth Fund 2006 Health Care Quality Survey. http://www.commonwealthfund.org/Surveys/2006/The-Commonwealth- Fund-2006-Health-Care-Quality-Survey.aspx. ———. 2009. How Health Care Reform Can Lower the Costs of Insurance Adminis- tration. http://www.commonwealthfund.org/~/media/Files/Publications/Issue%20 Brief/2009/Jul/Admin%20Costs/1299_Collins_how_hlt_care_reform_can_lower_ costs_ins_admin_v2.p df. ———. 2011. International Profiles of Health Care Systems, 2011. http://www.commonwealthfund.org/~/media/Files/Publications/Fund%20Report/2011/Nov/1562_ Squires_Intl_Profiles_2011_11_10.pdf. Commonwealth Fund, and Sean Boyle. 2010. International Profiles of Health Care Systems: The English Health Care System, 2009. http://www.commonwealthfund. org/Topics/International-Health-Policy/Countries/~/media/Files/Publications/ Other/2010/Jun/International%20Profiles/1417_Squires_Intl_Profiles_ England.pdf. Commonwealth Fund, K. Davis, C. Schoen, and K. Stremikis. 2010. Mirror, Mirror on the Wall: How the Performance of the U.S. Health Care System Compares Interna- tionally, 2010 Update. http://www.commonwealthfund.org/Content/Publications/ Fund-Reports/2010/Jun/Mirror-Mirror-Update.aspx?page=all. Commonwealth Fund, and I. Durand-Zaleski. 2010. International Profiles of Health Care Systems: The French Health Care System, 2009. http://www.commonwealthfund.org/ Topics/International-Health-Policy/Countries/~/media/Files/Publications/Other/ 2010/Jun/International%20Profiles/1417_Squires_Intl_Profiles_France.pdf/. Danzon, Patricia M., and Michael F. Furukawa. 2006. “Prices and Availability of Bio- pharmaceuticals: An International Comparision.” Health Affairs 25(5):1353–62. ———. 2008. “International Prices and Availability of Pharmaceuticals in 2005.” Health Affairs (Millwood) 27(1):221–33. Devers, Kelly J., Linda R. Brewster, and Lawrence P. Casalino. 2003. “Changes in Hospital Competitive Strategy: A New Medical Arms Race?” Part 2. Health Services Research 38(1):447–69.

HHassett.indbassett.indb 190190 111/8/121/8/12 9:279:27 PMPM AMERICAN COMPETITIVENESS AND HEALTH CARE 191

Dodoo, Marty N., George E. Fryer, Larry A. Green, Robert L. Phillips, Ginger R. Ruddy, and Jessica L. McCann. 2005. Patterns of Visits to Physicians’ Offices, 1980 to 2003. http://www.graham-center.org/online/graham/home/publications/one pagers/2005/op35-patterns-visits.html. Downey, Kristin. 2004. “A Heftier Dose to Swallow: Rising Cost of Health Care in U.S. Gives Other Developed Countries an Edge in Keeping Jobs.” Washington Post, E01. Farrell, Diana, Eric Jensen, Bob Kocher, Nick Lovegrove, Fareed Melhem, Lenny Mendonca, Beth Parish, and McKinsey Global Institute. 2008. Accounting for the Cost of U.S. Health Care: A New Look at Why Americans Spend More. McKinsey & Company, December. Fisher, Elliott S., David E. Wennberg, Therese A. Stukel, Daniel J. Gottlieb, F. L. Lucas, and Etoile L. Pinder. 2003. “The Implications of Regional Variations in Medicare Spending, Part 1: The Content, Quality, and Accessibility of Care.” Annals of Inter- nal Medicine 138(4):273–87. Fuchs, Victor R., and James S. Hahn. 1990. “How Does Canada Do It?” New England Journal of Medicine 232(13):884–90. Gibson, Teresa B., Tami L. Mark, Kimberly A. McGuigan, Kristen Axelsen, and Shao- hung Wang. 2006. “The Effects of Prescription Drug Copayments on Statin Adher- ence.” American Journal of Managed Care 12(9):509–17. Gibson, Teresa B., Catherine G. McLaughlin, and Dean G. Smith. 2005. “A Copay- ment Increase for Prescription Drugs: The Long-Term and Short-Term Effects on Use and Expenditures.” Inquiry 42(3):293–310. Gift, Thomas L., Richard Arnould, and Larry DeBrock. 2002. “Is Healthy Compe- tition Healthy? New Evidence of the Impact of Hospital Competition.” Inquiry 39(1):45–55. Goldman, Dana P., Geoffrey F. Joyce, Jose J. Escarce, Jennifer E. Pace, Matthew D. Solomon, Marianne Laouri, Pamela B. Landsman, and Steven M. Teutsch. 2004. “Pharmacy Benefits and the Use of Drugs by the Chronically Ill.” Journal of the American Medical Association 291(19):2344–50. Gruber, Jonathan. 1994. “The Incidence of Mandated Maternity Benefits.” American Economic Review 84(3):622–41. Gruber, Jonathan, and Alan B. Krueger. 1991. “The Incidence of Mandated Employer- Provided Insurance: Lessons from Workers’ Compensation Insurance.” Tax Policy and the Economy 5:111–43. Gruber, Jonathan, and Brigitte Madrian. 1996. “Health Insurance and Early Retire- ment: Evidence from the Availability of Continuation Coverage.” In Advances in the Economics of Aging, ed. D. Wise. Chicago: University of Chicago Press. ———. 1997. “Employment Separation and Health Insurance Coverage.” Journal of Public Economics 66(3):349–82. Health Forum LLC and American Hospital Association. 2010. Fast Facts on US Hospitals. http://www.aha.org/aha/resource-center/Statistics-and-Studies/fast- facts.html.

HHassett.indbassett.indb 191191 111/8/121/8/12 9:279:27 PMPM 192 RETHINKING COMPETITIVENESS

Hirth, Richard A., A. Mark Fendrick, and Michael E. Chernew. 1996. “Specialist and Generalist Physicians’ Adoption of Antibiotic Therapy to Eradicate Helicobacter pylori Infection.” Medical Care 34(12):1199–204. Hirth, Richard A., Marc N. Turenne, John D. Woods, Eric W. Young, Friedrich K. Port, Mark V. Pauly, and Philip J. Held. 1996. “Predictors of Type of Vascu- lar Access in Hemodialysis Patients.” Journal of the American Medical Association 276(16):1303–8. Holtz-Eakin, Douglas. 1994. “Health Insurance Provision and Labor Market Efficiency in the United States and Germany.” In Social Protection versus Economic Flexibility: Is There a Tradeoff? ed. R. Blank and R. Freeman. Chicago: University of Chicago Press. Ingrassia, Paul. 2010. “The Lessons of the GM Bankruptcy.” Wall Street Journal, June 1. http://online.wsj.com/article/SB100014240527487041135045752646411452 27612.html. International Federation of Health Plans. 2010. 2010 Comparative Price Report Medi- cal and Hospital Fees by Country. http://www.ifhp.com/documents/IFHPPrice reportfinal.pdf. Irwin, Douglas A., and Pete J. Klenow. 1994. “Learning-by-Doing Spillovers in the Semiconductor Industry.” Journal of Political Economy 102(6):1200–227. Kaiser Family Foundation. 2009. Examining Sources of Supplemental Insurance and Pre- scription Drug Coverage among Medicare Beneficiaries: Findings from the Medicare Cur- rent Beneficiary Survey, 2007. http://www.kff.org/medicare/upload/7801-02.pdf. ———. 2010. Hospital Beds per 1,000 Population by Ownership Type, 2008. http://state- healthfacts.kff.org/comparebar.jsp?ind=385&cat=8. Kaiser Family Foundation and Health Research and Educational Trust. 2009. Employer Health Benefits. http://ahier.blogspot.com/2009/09/kaiser-employer-health-benefits-2009.html. Kapur, Navneet, and Allan House. 1998. “Improving ‘New Deal’ Shifts for Junior House Officers.” Journal of Hospital Medicine 59(12):960–66. Krugman, Paul. 1994. “Competitiveness: A Dangerous Obsession.” Foreign Affairs 73(2):28–44. Lambert, Emily M., and Eric S. Holmboe. 2005. “The Relationship between Specialty Choice and Gender of U.S. Medical Students, 1990–2003.” Academic Medicine 80(9):797–802. Macinko, James, Barbara Starfield, and Leiyu Shi. 2007. “Quantifying the Health Ben- efits of Primary Care Physician Supply in the United States.” International Journal of Health Services 37(1):111–26. McCain, John. 2007. Speech at Detroit Economic Club, Detroit, Michigan. Octo- ber 9. http://www.cfr.org/publication/14460/john_mccains_speech_in_detroit_ michigan.html. McGlynn, Elizabeth A., Steven M. Asch, John Adams, Joan Keesey, Jennifer Hicks, Alison DeCristofaro, and Eve A. Kerr. 2003. “The Quality of Health Care Delivered to Adults in the United States.” New England Journal of Medicine 348:2635–45.

HHassett.indbassett.indb 192192 111/8/121/8/12 9:279:27 PMPM AMERICAN COMPETITIVENESS AND HEALTH CARE 193

McLaughlin, Catherine G., Daniel P. Normolle, Robert A. Wolfe, Laurence F. McMa- hon, Jr., and John R. Griffith. 1989. “Small-Area Variation in Hospital Discharge Rates: Do Socioeconomic Variables Matter?” Medical Care 27(5):507–21. MedPAC. 2011. Medicare Payment Advisory Commission Report to the Congress: Regional Variation in Medicare Service Use. http://www.medpac.gov/documents/Jan11_ RegionalVariation_report.pdf. Miller, Richard D. 2004. “Estimating the Compensating Differential for Employer- Provided Health Insurance.” International Journal of Health Care Finance and Eco- nomics 4(1):27–41. Mitchell, Olivia. 1982. “Fringe Benefits and Labor Mobility.” Journal of Human Resources 17:286–98. Olson, Craig. 1994. Part Time Work, Health Insurance Coverage and the Wages of Mar- ried Women. Mimeo, University of Wisconsin-Madison. Organisation for Economic Co-operation and Development. 2010. Health Data 2010. http://stats.oecd.org/Index.aspx?DataSetCode=HEALTH. ———. 2011. OECD Health Data: Health Care Resources. OECD Health Statistics Data- base. http://www.oecd.org/document/30/0,3746,en_2649_34631_12968734_ 1_1_1_1,00.html. Pauly, Mark V. 1993. “U.S. Health Care Costs: The Untold True Story.” Health Affairs 12(3):152–59. ———. 2003. “Should We Be Worried about High Real Medical Spending Growth in the United States?” Health Affairs (Web exclusive):w3/15–13/27. Pauly, Mark V., and Mark A. Satterthwaite. 1981. “The Pricing of Primary Care Physi- cians Services: A Test of the Role of Consumer Information.” Bell Journal of Econom- ics 12(2):488–506. Pritchard, Robert S., Elliott S. Fisher, Joan M. Teno, Sandra M. Sharp, Douglas J. Reding, William A. Knaus, John E. Wennberg, and Joanne Lynn. 1998. “Influence of patient preferences and local health system characteristics on the place of death. SUPPORT Investigators. Study to Understand Prognoses and Preferences for Risks and Out- comes of Treatment.” Journal of the American Geriatrics Society 46(10):1242–50. Regenstein, Marsha, and Jennifer Huang. 2005. Stresses to the Safety Net: The Public Hospital Perspective. Kaiser Commission on Medicaid and the Uninsured. http:// www.kff.org/medicaid/upload/Stresses-to-the-Safety-Net.pdf. Reinhardt, Uwe E. 1989. “Health Care Spending and American Competitiveness.” Health Affairs (Millwood) 8(4):5–21. Robinson, James C., and Harold S. Luft. 1985. “The Impact of Hospital Market Struc- ture on Patient Volume, Average Length of Stay, and the Cost of Care.” Journal of Health Economics 4(4):333–56. ———. 1987. “Competition and the Cost of Hospital Care, 1972 to 1982.” Journal of the American Medical Association 257(23):3241–45. ———. 1988. “Competition, Regulation, and Hospital Costs, 1982 to 1986.” Journal of the American Medical Association 260(18):2676–81.

HHassett.indbassett.indb 193193 111/8/121/8/12 9:279:27 PMPM 194 RETHINKING COMPETITIVENESS

Ryan, S. 1997. Employer-Provided Health Insurance and Compensating Wage Differen- tials: Evidence from the Survey of Income and Program Participation. Mimeo, Univer- sity of Missouri-Columbia. Sandy, Lewis G., Thomas Bodenheimer, L. Gregory Pawlson, and Barbara Starfield. 2009. “The Political Economy of U.S. Primary Care.” Health Affairs 28(4):1136–44. Satterthwaite, Mark A. 1979. “Consumer Information, Equilibrium Industry Price, and the Number of Sellers.” Bell Journal of Economics 10(2):483–502. Sheiner, Louise. 1999. Health Care Costs, Wages, and Aging. Federal Reserve Board of Governors. Stop 83. Washington, D.C. Shi, Leiyu. 1992. “The Relationship between Primary Care and Life Chances.” Journal of Health Care for the Poor and Underserved 3(2):321–35. ———. 1994. “Primary Care, Specialty Care, and Life Chances.” International Journal of Health Services 24(3):431–58. Shi, Leiyu, Barbara Starfield, B. Kennedy, and I. Kawachi. 1999. “Income Inequality, Primary Care, and Health Indicators.” Journal of Family Practice 48(4):275–84. Sood, Neeraj, Arkadipta Ghosh, and Jose J. Escarce. 2009. “Employer-Sponsored Insurance, Health Care Cost Growth, and the Economic Performance of U.S. Industries.” Part 1. Health Services Research 44(5):1449–64. Starfield, Barbara, Leiyu Shi, and James Macinko. 2005. “Contribution of Primary Care to Health Systems and Health.” Milbank Quarterly 83(3):457–502. Sweet, Lynn. 2009. “Rahm Emanuel on CBS Face the Nation.” http://blogs.suntimes. com/sweet/2009/03/rahm_emanuel_on_cbs_face_the_n.html. U.S. Department of Health and Human Services. Health Resources and Services Administration. 2008. Health Center Data 2008. http://bphc.hrsa.gov/healthcenter datastatistics/NationalData/2008/2008datasnapshot.html. Vladeck, Bruce C., and Thomas Rice. 2009. “Market Failure and the Failure of Dis- course: Facing Up to the Power of Sellers.” Health Affairs (Millwood) 28(5):1305–15. Vogel, Robert L., and Richard J. Ackermann. 1998. “Is Primary Care Physician Sup- ply Correlated with Health Outcomes?” International Journal of Health Services 28(1):183–96. Welch, W. Pete, Diana Verrilli, Steven J. Katz, and Eric Latimer. 1996. “A Detailed Comparison of Physician Services for the Elderly in the United States and Canada.” Journal of the American Medical Association 275(18):1410–16. Wennberg, John E., and Alan Gittelsohn. 1973. “Small Area Variations in Health Care Delivery.” Science 182(117):1102–8. Wennberg, John E., Elliott S. Fisher, Therese A. Stukel, and Sandra M. Sharp. 2004. “Use of Medicare Claims Data to Monitor Provider-Specific Performance among Patients with Severe Chronic Illness.” Health Affairs (Millwood) (Supplement, Web exclusive):VAR5–18. Wong, Herbert S. 1996. “Market Structure and the Role of Consumer Information in the Physician Services Industry: An Empirical Test.” Journal of Health Economics 15(2):139–60.

HHassett.indbassett.indb 194194 111/8/121/8/12 9:279:27 PMPM AMERICAN COMPETITIVENESS AND HEALTH CARE 195

Woolhandler, Steffie, Terry Campbell, and David U. Himmelstein. 2003. “Costs of Health Care Administration in the United States and Canada.” New England Journal of Medicine 349(8):768–75. World Health Organization. 2007. Table of key indicators, sources and methods by country and indicators. http://apps.who.int/nha/database/StandardReport.aspx?ID=REP_ WEB_MINI_TEMPLATE_WEB_VERSION&COUNTRYKEY=84042. Zuckerman, Stephen, Timothy Waidmann, Robert Berenson, and Jack Hadley. 2010. “Clarifying Sources of Geographic Differences in Medicare Spending.” New England Journal of Medicine 363(1):54–62.

HHassett.indbassett.indb 195195 111/8/121/8/12 9:279:27 PMPM 7

Is the United States “Competitive” Internationally in Health Care? Benjamin Zycher1

International “competitiveness” is difficult to define, particularly in the context of goods and services that for the most part are not traded in global markets.2 This may be true in particular for the U.S. health care sector, with respect to which the term “competitiveness” has been subjected to a number of differing definitions and approaches. But one obvious question is whether the U.S. health care system achieves results that systematically are better or worse than those observed in other nations, in terms of both costs and health outcomes.3 Accordingly, the issue addressed here is whether the conventional wisdom on the U.S. health care system—that we “spend more and get less”—is supported by the evidence, that is, whether the U.S. health care system would be competitive internationally if health care services were traded internationally (see, for example, World Health Organization [WHO] 2000; Organisation for Economic Co-operation and Development [OECD] 2011a; Davis, Schoen, and Stremikis 2010; Aaron and Ginsburg 2009; Garber and Skinner 2008). In economic terms, that question can be restated a bit more rigorously as follows: Does the U.S. health care system consume more real resources than other systems in the delivery of given health care outputs?4 Or, in a comparison of the United States with other nations, which among them produces given health outcomes actually attributable to the health care system at a lower cost in terms of other goods and services forgone?5 This chapter reviews the available evidence on real resource consumption by the U.S. health care sector in comparison with those of other

196

HHassett.indbassett.indb 196196 111/8/121/8/12 9:279:27 PMPM IS THE U.S. “COMPETITIVE” IN HEALTH CARE? 197

advanced economies, and the available evidence as well on health care outcomes that reasonably can be attributed to the health care system, as opposed to lifestyle choices and other such parameters that health care per se is poorly positioned to influence. The first section summarizes some of the prominent literature on U.S. health care spending and outcomes, and then addresses several problems with that body of analysis that have not received nearly as much attention as the literature itself. Some important effects of governmental financing of health care delivery and the resulting implications for biases in reported spending are discussed as well. The second section discusses the available data on prices, service delivery, and other parameters, as well as the increased use of nonprice rationing and other predictable effects for several advanced economies. The third section discusses the available data on comparative outcomes for life expectancy, infant mortality, and cancer treatment. The last section offers some concluding observations.

Some Conventional Wisdom on the U.S. Health Care System Several large and influential statistical analyses have influenced the public debate on international comparisons of health systems heavily. One was published some years ago by the World Health Organization, presenting a comprehensive assessment and ranking of the health care systems of 191 nations (see World Health Organization 2000, annex table 1). The central conclusion is that the United States ranks first in health care expenditures per capita, thirty-seventh in “overall health system performance,” and seventy-second in terms of the “level of health.” The WHO analysis suffers from several methodological problems that make the national rankings generally and the findings for the United States in particular deeply problematic. One loud hint about that reality is offered by the identities of the highest-ranked nations in terms of “overall health system performance”: France, Italy, San Marino, Andorra, and Malta, a group of nations not commonly perceived collectively as a model of efficiency in the delivery of health care services.6 The basic problem is that the WHO analysis in its estimation of health system performance defined five criteria (each of which comprised several underlying criteria), of which three are measures of (purported) distributional equality rather than actual

HHassett.indbassett.indb 197197 111/8/121/8/12 9:279:27 PMPM 198 RETHINKING COMPETITIVENESS

quality of health care services.7 Under the WHO weighting system, these three sum to 62.5 percent—almost two-thirds—of a given nation’s ranking; moreover, the subjectivity of the underlying criteria cannot be overstated. “Responsiveness” obviously is difficult to define, and half of that criterion was based on a WHO assessment of the respective nations’ “respect for per- sons.” The other half was the assessed “client orientation” of the given health care system, itself a function of such parameters as the “quality of amenities” and “prompt attention.” The Commonwealth Fund study finds that among seven Western economies, the United States spends far more per capita but for years has ranked sixth or seventh in terms of five measures of health system performance (Davis et al. 2010, exhibit ES-1).8 These five measures are the quality of care (comprising “effective care,” “safe care,” “coordinated care,” and “patient-centered care”), access (comprising “cost-related access problems” and “timeliness of care”), efficiency, equity, and “long, healthy, productive lives.” Among these five criteria, the U.S. rankings are, respectively, sixth, 6.5 (tie for last with Australia), seventh, seventh, and seventh. All of these criteria introduce serious conceptual problems as measures of the performance of the health care system in terms of treating health problems. But the “equity” measure is particularly poor, as it divides the national populations into below- and above-median incomes for each country without consideration of the (complex) factors that influence income disparities, some of which affect the consumption of health care services and health outcomes in substantial ways. The Commonwealth Fund study justifies this approach by quoting the Institute of Medicine definition of “equity”: “providing care that does not vary in quality because of personal characteristics such as gender, ethnicity, geographic location, and socioeconomic status” (see Davis et al. 2010, pp. 13–15). Socioeconomic characteristics as a group are an example of powerful drivers of “access” that the health care system itself can affect only marginally; moreover, the “below- and above-median income” dichotomy is a poor proxy for such factors. Thus, to a significant degree, the “equity” criterion is little more than a measure of income disparities, themselves driven substantially by the magnitudes of income transfer (“welfare”) programs; nations characterized by larger relative entitlement systems will tend to drive down such income disparities, while reducing overall living

HHassett.indbassett.indb 198198 111/8/121/8/12 9:279:27 PMPM IS THE U.S. “COMPETITIVE” IN HEALTH CARE? 199

standards (per capita GDP).9 It is obvious that the health care system, again, is poorly positioned to influence income disparities in the aggregate, and to the extent that “equity” drives the relative rankings, it is, accordingly, a poor indicator of comparative health system performance as opposed to the magnitude of (measured) income disparities. Moreover, the empirical literature uniformly finds that the demand for health care services rises with income (or wealth); such services clearly are a “normal” good and may be a “luxury” good.10 In other words, the poor simply have smaller demands for health care services than the nonpoor and thus consume fewer such services, ceteris paribus. Accordingly, the U.S. “equity” ranking may stem largely from the income status of below-median respondents in the United States relative to that in the other countries, particularly in terms of the bottom quintiles of the income distribution. Again, this is a parameter not particularly revealing about the performance of the health care system per se. The OECD (2011a) analysis asserts that “the United States spends two-and-a-half times more than the OECD average health expenditure per person. . . . It even spends twice as much as France, for example, a country which is generally accepted as having very good health services. At 17.4% of GDP in 2009, US health spending is half as much again as any other country, and nearly twice the average” (p. 1). One central problem with such spending comparisons is the distortion introduced by formal and informal price suppression—explicit or implicit price controls—on reported spending.11 Price suppression has the effect of biasing downward the reported value (or total amount) of the resources used in the delivery of health care services, an impact discussed more fully below. For now it is important to see why price suppression for health care services is likely to be more important in the rest of the OECD than in the United States. The reason is straightforward: government health care spending represents a far larger proportion of the market in the other members of the OECD. Table 7-1 below presents these data, which show that government expenditures for health care services are substantially smaller as a share of the market in the United States than elsewhere in the OECD. As a crude generalization, that share is less than half in the United States but approximately three-quarters for the other major members of the OECD. In the context of direct and indirect price suppression, this is important

HHassett.indbassett.indb 199199 111/8/121/8/12 9:279:27 PMPM 200 RETHINKING COMPETITIVENESS

TABLE 7-1 PUBLIC HEALTH EXPENDITURES AS A PERCENT OF TOTAL HEALTH EXPENDITURES, 2009

Nation Percent Australia 68.0a Canada 70.6 Finland 74.7 France 77.9 Germany 76.9 Italy 77.9 Japan 80.8a New Zealand 80.5 Spain 73.6 Sweden 81.5 United Kingdom 84.1 United States 47.7

aData are for 2008.

SOURCE: OECD (2011), http://www.oecd.org/document/16/0,3746,en_2649_33929_2085200_1_1_ 1_1,00.html (accessed May 17, 2012).

because of the differing incentives confronting government decision makers as distinct from those faced by private ones. Government has interest groups rather than patients (or customers), a crucial distinction from the conditions that confront health care providers; for policymakers, dollars not spent on a given constituency can be spent on others. Accordingly, government as a buyer of medical goods and services has powerful incentives to suppress prices in various ways both explicit and implicit and to opt for lower-priced medical goods and services over higher-priced ones in ways that do not reflect the incremental medical benefits of the latter, even if the value of those additional benefits exceeds the additional pecuniary costs.12 In short, government policymakers in the context of a given health coverage program have incentives (at the margin) to favor current budget savings at the expense of benefits enjoyed by the beneficiaries of that program. Such choices might not reduce costs, defined properly as including the marginal benefits of alternative medical treatments more effective but also

HHassett.indbassett.indb 200200 111/8/121/8/12 9:279:27 PMPM IS THE U.S. “COMPETITIVE” IN HEALTH CARE? 201

more expensive, forgone in pursuit of budget savings that can be spent on other interest groups.13 This bias is likely to weigh against the higher medical effectiveness (if any) of the more expensive alternatives as perceived by the participants in the government programs, but perhaps perceived only weakly if at all by government decision makers, an effect likely to be even stronger in the context of future treatments aimed at future beneficiaries, that is, those who do not vote in current elections.14 This powerful incentive on the part of government policymakers to view “costs” narrowly as budget outlays for which constituencies compete, rather than as a broader concept of benefits for constituents forgone in the pursuit of budget savings, is exacerbated by the short time horizons of public officials.15 Outlay savings are available today (or during a given policymaker’s term in office), and those savings can be spent on current constituencies. To some substantial degree, the potential adverse effects of policy decisions on coverage and other such matters—reduced health care outlays—will be felt in the future. There is no particular reason to believe that government as an institution has incentives to adopt a time horizon longer than that relevant for the private sector; indeed, one assumption, wholly plausible at a minimum, is that the time horizon for many public officials is the next election. Of course, to say that a given official views the next election as the “long run” is different from arguing that government acting collectively would display the same behavior; but the profit motive provides incentives for the market to consider the long-run effects of current decisions, while no similar constraint operates in the public sector, except perhaps crudely through democratic processes.16 In addition, such policies as campaign finance restrictions (in the U.S. context) may weaken the constraints that the political parties can impose on officeholders: As the parties are long-lived institutions with some incentives to adopt time horizons longer than those of particular officeholders, the net effect may be a tendency to discount the future effects of policies more heavily. Private-sector producers of health care services (or health coverage) in principle are entities with infinite lives, constrained by consumer preferences and by the market rate of interest to consider the future effects of current decision making. In this sense, the profit motive leads the providers implicitly to represent the interests of future patients, an incentive very different from that driving policymakers with short time horizons. Because

HHassett.indbassett.indb 201201 111/8/121/8/12 9:279:27 PMPM 202 RETHINKING COMPETITIVENESS

policymakers have important incentives to constrain explicit outlays on health care, nations with greater public-sector shares in terms of financing are likely to display greater degrees of explicit and implicit price suppression. Accordingly, reported health care spending will tend to understate the true resource (or opportunity) cost of health care programs. Because resources are limited, always and everywhere, government cannot finance all health care services demanded when the demanders (patients) pay substantially less than marginal cost, or little, for services. Some services or demanders, or both, must be denied, whether explicitly or implicitly. Such nonprice rationing can take a number of forms: waiting lists, denial of given health care goods and services for all or part of a population, underinvestment in particular technologies, restrictions on the choice of physician or hospital, and the like. As with price suppression, denial or delay in the provision of services may reduce reported outlays, but the value of the services not delivered or consumed is a real economic cost not included in reported budgets. To the extent that those services are worth more to patients than their true economic cost, the forgone net value is a real economic cost of a system of health care financed by government. Moreover, the use of nonprice rationing, by reducing the production of health care services below levels that would be observed otherwise, has the effect of increasing the forgone value of the services not produced and consumed. The reduced availability of services, whether through price suppression or nonprice rationing used to allocate the services available, increases the marginal value of the services that are available.17 The next section discusses these biases in reported spending and the increased use of nonprice rationing in greater detail.

Some International Evidence on Prices and Services The discussion in the previous section suggests that reported (or official) prices for health care services can be predicted to be lower in nations with greater government-financing shares of the health care market because of various forms of price suppression, and that nonprice rationing as well would be observed in those nations to a degree greater than the case for nations with lower government-financing shares, although “nonprice rationing” presents some difficulties in terms of definition and measure-

HHassett.indbassett.indb 202202 111/8/121/8/12 9:279:27 PMPM IS THE U.S. “COMPETITIVE” IN HEALTH CARE? 203

TABLE 7-2 FEES FOR PHYSICIAN OFFICE VISITS, 2008 (DOLLARS)

Government Private Government Country Payer Payer Finance Share (%) Australia 34 45 68.0 Canada 59 NA 70.6 France 32 34 77.9 Germany 46 104 76.9 United Kingdom 66 129 84.1 United States 60 133 47.7 NA, not applicable.

SOURCES: Table 7-1; Laugesen and Glied (2011); author’s computations.

ment. These effects obtain because of powerful incentives confronting public decision makers to constrain explicit health care outlays in order to shift spending toward other constituencies. Table 7-2 presents data on physician fees for office visits for six OECD countries in 2008.18 The U.S. fees uniformly are the highest for both government and private payers, with the sole exception of government payers in the United Kingdom. Moreover, the fees paid by government in every nation uniformly are lower than those paid by private purchasers of office visits with physicians. These data are broadly consistent with the financing shares summarized in table 7-1 and with the qualitative summary model of the incentives of public officials discussed in the previous section. Table 7-3 presents similar data for 2008 for orthopedic surgery.19 U.S. fees for both government and private purchasers are higher than those for all the other countries for which data are available. As with physician fees for office visits, the fees paid by government are lower than those paid by private payers. In short, government uniformly pays less than private payers, and fees in the United States are almost always the highest among the six nations. Table 7-4 presents data on relative prices for seven common hospital services (or procedures) for eleven members of the OECD in 2007.20 In every case the U.S. price was greater than those of the other nations. The more detailed OECD survey of prices for twenty-nine hospital services (or procedures) for thirteen OECD members finds that the U.S. price was

HHassett.indbassett.indb 203203 111/8/121/8/12 9:279:27 PMPM 204 RETHINKING COMPETITIVENESS

TABLE 7-3 FEES FOR ORTHOPEDIC SURGERY, 2008 (DOLLARS)

Government Government Private Finance Share Country Payer Payer (percent) Australia 1,046 1,943 68.0 Canada 652 NA 70.6 France 674 1,340 77.9 Germany 1,251 NA 76.9 United Kingdom 1,181 2,160 84.1 United States 1,634 3,996 47.7 NA, not applicable.

SOURCES: Table 7-1; Laugesen and Glied (2011).

greater than the average for the other twelve nations for twenty-eight of the twenty-nine medical services (Koechlin, Lorenzoni, and Schreyer 2010). For all services taken as a (weighted) whole, the U.S. price was more than 60 percent higher than the average for the other twelve nations. The U.S. price was greater than those of every one of the other nations for twenty of the twenty-nine procedures. In short, the international spending comparisons can be misleading in terms of the value of the real resources consumed in the delivery of health care services, in that the greater government financing role overseas has had the effect of increasing the importance of price suppression. Table 7-5 presents some data on the use of real resources in health care delivery for several OECD nations per unit of population. For physicians and hospital beds, the United States is lowest among these large OECD economies. For the sophisticated-equipment categories in table 7-5—magnetic resonance imaging (MRI), computed tomography (CT), and radiotherapy units—the United States ranks at or near the top. More generally, among the thirty- four member nations in the OECD, the U.S. rankings per unit of population are twenty-sixth in physicians, twenty-ninth in hospital beds, and twenty- ninth in physician consultations, but it ranks first in radiation equipment; second in MRI units, MRI exams, and CT exams; and fifth in CT scanners. The United States ranks twenty-ninth in the average length of hospital stays,

HHassett.indbassett.indb 204204 111/8/121/8/12 9:279:27 PMPM IS THE U.S. “COMPETITIVE” IN HEALTH CARE? 205

TABLE 7-4 RELATIVE PRICES FOR HOSPITAL SERVICES, 2007 (U.S. = 100)

Nation Ger- Service/Procedure Aus. Canada many Finland France Sweden U.S. Appendectomy 63.4 62.8 37.0 47.0 57.2 62.3 100 Normal birth 67.0 62.9 40.2 34.2 65.0 58.2 100 Caesarean birth 95.2 64.7 50.1 64.5 78.1 85.6 100 Coronary angioplasty 49.6 64.5 23.2 38.8 48.9 64.7 100 Coronary bypass 63.2 66.1 40.9 68.3 67.3 61.8 100 Hip replacement 91.5 68.8 51.1 62.2 64.1 66.5 100 Knee replacement 97.7 66.3 67.0 66.4 83.1 69.2 100

SOURCES: Koechlin, Lorenzoni, and Schreyer (2010, tables A.3.5 and A.3.7); author’s computations.

TABLE 7-5 HEALTH CARE RESOURCES, 2009

Hospital MRI CT Radiation Nation Physicians Beds Units Scanners Equipment Australia 3.0a 3.8a 5.9 38.7 8.8 Canada 2.4 3.3a 8.0 13.9 NA Finland 2.7a 6.2 16.9 20.4 8.8 France 3.3 6.6 6.4 11.1 NA Germany 3.6 8.2 NA NA NA Italy 3.4 3.7 21.6 31.7 6.5 Japan 2.2a 13.7 43.1a 97.3a NA Spain 3.5 3.2 NA NA NA UK 2.7 3.3 5.6a 11.6 4.9a U.S. 2.4 3.1 25.9b 34.3b 11.3c

NOTES: Physicians and hospital beds per 1,000 population; MRI units, CT scanners, and radiation equipment per million population. aData are for 2008. bData are for 2007. cDatum is for 2010. NA, not available or not applicable.

SOURCE: OECD (2011b).

HHassett.indbassett.indb 205205 111/8/121/8/12 9:279:27 PMPM 206 RETHINKING COMPETITIVENESS

but first in knee replacements, second in tonsillectomies, and third in coronary angioplasty procedures (OECD 2011b).21 Accordingly, it is far from obvious that the higher reported spending and the lower physician/population ratio imply that the United States “spends more and gets less.” The crude data suggest that the United States in several dimensions uses fewer resources and delivers more services than most of the other members of the OECD. In particular, the data suggest that the U.S. health care system may be less labor intensive and more capital intensive than those of the other OECD economies and may deliver more services. One reason for that last outcome may be the relatively greater use of capital equipment (e.g., MRI, CT, and radiation units): Such capital intensiveness may result in lower marginal costs than alternative ways of delivering health care services that are more labor intensive.22 Since reported price and spending data are likely to reflect average costs (in an accounting sense) rather than marginal costs, those data may not represent the more relevant costs of delivering additional services. Accordingly, there may exist a systematic bias in the data. As discussed in the previous section, government officials have powerful incentives to reduce (the growth of) explicit outlays on health care services so as to increase the availability of budget dollars for other constituencies. Price suppression is a tool with which to effect that outcome so that systems with greater proportions of government financing can be predicted to display lower reported prices. Another such tool is nonprice rationing in its many forms. The measurement of price differentials entails some complexities but is relatively straightforward; however, a description of “nonprice rationing” for purposes of comparison is more difficult, as there is no simple definition of the phenomenon. But a review of the literature describing the health care systems of the United States and other nations is instructive.23 As noted above, even (or especially) a policy of unrestricted universal health coverage cannot be achieved even in principle because resources always are limited. Consider the case of France, often described as a model for the United States.24 Many health care services are not covered by the government system, and some providers do not accept the official fee schedules; accordingly, most French residents purchase supplementary private insurance. More generally, the government system classifies various technologies as

HHassett.indbassett.indb 206206 111/8/121/8/12 9:279:27 PMPM IS THE U.S. “COMPETITIVE” IN HEALTH CARE? 207

providing “insufficient medical service rendered,” which essentially is a centralized benefit/cost decision (Tanner 2008, p. 9). It is, therefore, a form of rationing, as those services are excluded from payment under the government system. The number of practicing physicians is limited by sharp constraints on medical students admitted to the second year of medical school; and the government more recently has imposed restrictions on access to physicians, particularly specialists. Aggregate (or “global”) bud- geting combined with fee restrictions (price controls) on hospitals have yielded reduced capital investment and a relative shortage of advanced medical technology, and there has been recently a process of delisting for drugs from the national formulary (see Petkantchin 2007). Tanner (2008, p. 12) notes that the French system has avoided the worst rationing outcomes observed in other systems because there is substantial cost sharing by patients, because the private insurance market is relatively unregulated, and because consumers have the option of paying privately (out of pocket) for better or additional care. The Italian system is plagued with lengthy waiting lists for diagnostic services, strict formulary restrictions for drugs, poor hospital services (in particular in the southern parts of the country), and suppressed capital investment in the face of rapid cost growth and efforts by the regional governments (which control most health care outlays) to achieve budget savings. Physician choice by patients is limited geographically. Waiting lists are common for diagnostic services in particular as a result of suppressed capital investment and a resulting shortage of advanced equipment. Many of the newest and most innovative drugs have not been introduced widely into the Italian system because of price controls and formulary restrictions (see Mingardi 2006). The Spanish national health care system is operated by the regional governments and covers essentially the entire population. Individual patients are assigned primary care physicians on a geographic basis; in order to change physicians or find a shorter waiting list, Spaniards must change residences. Referrals are required for specialized care, which is plagued with lengthy waiting lists of at least two months on average. Other specialized services—examples are rehabilitation and convalescence—are essentially unavailable except through private insurance or out-of-pocket spending (see Bosch 2002). A private insurance system has arisen, substantially as a

HHassett.indbassett.indb 207207 111/8/121/8/12 9:279:27 PMPM 208 RETHINKING COMPETITIVENESS

result of the lengthening waiting lists and quality problems inherent in the government system, and now provides supplementary coverage for about 14 percent of the population. Interestingly, private insurance payments are more than 20 percent of total health care expenditures, and out-of-pocket spending by patients is about one-quarter of all health spending. Shortages of modern medical technology are pronounced (Tanner 2008, p. 15). The Greek system nominally is employer based, but in practice is essentially a single-payer system subject to extensive and highly detailed centralized finance and regulation (for a useful description, see Mossialos, Allin, and Davaki 2005). In principle, health care services are “free”: patients receive services with no deductibles and small cost sharing for only a few services. Unsurprisingly, therefore, demands exceed the ability of the system to provide services—again, resources always are limited—so that nonprice rationing and cost suppression are unavoidable (see Kresge 2012). Capital investment has been suppressed in pursuit of budget savings, so that technology levels are poor.25 Low wages have yielded sharp personnel shortages, and “most observers agree that waiting lists are a severe problem at almost every level of care, and [are] particularly bad at both NHS and public hospitals” (Tanner 2008, p. 21). Informal payment—bribery—is a routine tool with which patients can see a particular doctor not formally available to that given patient, avoid or reduce waiting times, obtain treatments not authorized by government guidelines, switch from the government system to care from a private practice, and so forth (see Forelle 2011). The Greek system illustrates starkly the problems inherent in the use of official “spending” figures for systems in which price suppression is an important tool used to reduce reported costs, that is, budget outlays. The problems afflicting the British National Health Service have received widespread attention, so only a few summary observations are offered here. Patients pay little or nothing for services, so government policies have focused heavily on the control of outlays; nonetheless, budget pressures plague the system, an outcome wholly predictable in that “needs” in the absence of substantial cost-sharing by patients are certain to exceed available resources.26 Waiting lists are long for both physician and hospital care and have affected the quality of care substantially (see, for example, Lewis and Appleby 2006, pp. 10–13). For example, 20 percent of NHS patients diagnosed with colon cancer considered curable at the time of diagnosis

HHassett.indbassett.indb 208208 111/8/121/8/12 9:279:27 PMPM IS THE U.S. “COMPETITIVE” IN HEALTH CARE? 209

are deemed incurable when treatment finally is obtained (Browne 2001). Formal rationing has been implemented for certain expensive treatments, with those deemed too ill or too old denied treatment altogether (Tanner 2008, pp. 24–25). As with the Greek and other systems, such institutional responses to the problem of resource scarcity make the official “spending” figures biased downward in terms of the true economic costs of the system and thus highly unreliable. The experiences of other nations with substantial government finance for health care services are broadly similar: greater governmental shares in terms of finance and/or the allocation of health care resources are correlated with the enhanced use of price suppression and nonprice rationing.27 In particular, price suppression and nonprice rationing hide some of the real resource costs of health care, and informal payment—again, bribery—used as a tool with which to compete for services is a parameter unreported by definition and so may add even more to the aggregate cost of health care services.28 Accordingly, reported spending is a poor measure of real resource use, and the quality of the services delivered is affected adversely by several predictable responses that are somewhat difficult to quantify. There are good reasons to believe that the value of real resource use—a parameter related to but distinct from reported spending—is greater than indicated by the official data for other advanced economies relative to that for the United States because of the adverse effects of price suppression and the resulting downward distortion in reported resource use. Accordingly, it is far from clear that the U.S. health care system “spends more” than those of other advanced economies. The next section discusses the available data on health care outcomes, that is, whether patients in the United States systematically “get less.”

The Comparative Performance of the U.S. Health Care System As noted in the introduction, the prominent studies comparing health care systems internationally use a wide range of criteria in addition to spending, either per capita or as a percentage of GDP. Many of these are highly qualitative: “access,” “coverage,” “fairness,” “responsiveness,” and others, the definition, measurement, and weighting of which are afflicted with considerable subjectivity (see, for example, the WHO [2000] and Commonwealth

HHassett.indbassett.indb 209209 111/8/121/8/12 9:279:27 PMPM 210 RETHINKING COMPETITIVENESS

Fund [Davis, Schoen, and Stremikis 2010] studies).29 In terms of the relative overall contribution of the health care systems themselves to “health,” many analyses have focused on narrower comparisons of life expectancies at birth and infant mortality rates. For life expectancy at birth, the United States ranks twenty-seventh out of thirty-four OECD nations (despite ostensibly ranking first in spending), with a life expectancy at birth 1.3 years below the OECD average (OECD 2011b). An obvious problem with the life expectancy rankings used as measures of health care quality is straightforward: life expectancies are affected by important factors relatively independent of the health care system. Some of these factors are obvious: accident rates, lifestyle choices, rates of violent crime, suicides, and the like.30 Ohsfeldt and Schneider (2006, tables 1-4 and 1-5) noted that for 1980–1999, U.S. life expectancy at birth ranked nineteenth out of twenty-nine OECD nations, but first after controlling for deaths due to fatal injuries.31 Ohsfeldt and Schneider (2006, p. 19) also note: According to data from the World Health Organization, the death rate from transport accidents (motor vehicle or common carrier) in the United States is about three times higher than the rate in Sweden or the United Kingdom, and one and a half to two times higher than in Australia, Canada, Denmark, Germany, or Japan. . . . In the United States, unintentional injury was the fifth leading cause of death in the year 2000 overall, but was the leading cause of death among individuals between the ages of one and thirty-four, and was the second and third most common cause of death among individuals between the ages of thirty-five and forty-four and forty- five and fifty-four, respectively. By contrast, among individuals sixty-five years of age and over, unintentional injury was the ninth leading cause of death. The unusually high death rates from unintentional injury among young Americans reduce the estimated life expectancy at birth for the United States, but they do not necessarily signal a deficiency in the U.S. health care system. Fatalities from injuries (whether accidental or intentional) certainly are not wholly independent of the health care system. The literature suggests that for the United States, about 80–85 percent of trauma deaths are

HHassett.indbassett.indb 210210 111/8/121/8/12 9:279:27 PMPM IS THE U.S. “COMPETITIVE” IN HEALTH CARE? 211

inevitable, given the severity of the injuries; that is, death either is immediate or occurs within hours and cannot be avoided through timely trauma care (see, for example, Acosta et al. 1998; personal correspondence between the author and Robert Ohsfeldt, January 13, 2012). Accordingly, the quality of the health care system is largely irrelevant for those cases. But for the remaining 15–20 percent, death may occur weeks or months later, and so health care delivery may be relevant for the life expectancy data. One way to adjust for this factor is to examine life expectancies for a subpopulation for which deaths from injuries, whether accidental or intentional, are relatively unimportant. Data for the United States show that more than 74 percent of those succumbing to accidents are younger than seventy-five years of age; for deaths due to intentional self-harm and assaults, the respective figures are 92 percent and almost 98 percent (Centers for Disease Control and Prevention 2009, table 10). Accordingly, life expectancy at an advanced age (rather than at birth) is a crude way to evaluate the comparative performance of health care systems without the complications introduced by injuries, although such other factors as lifestyle choices also relatively independent of the health care systems clearly complicate this analysis. Manton and Vaupel (1995) have reported the life expectancies for both men and women at age eighty for the United States, Japan, France, Sweden, and England for 1987, as summarized in table 7-6. The findings in table 7-6 suggest, crudely, that the U.S. health care system does relatively well in terms of extending life expectancies when at least one factor largely unrelated to health care—deaths from accidental and intentional injuries—are excluded from the analysis for the most part. At a minimum these data suggest that the simple life-expectancy comparisons are problematic, and they call into question the usual assumption that relatively lower life expectancies in the United States suggest poorer comparative performance by the U.S. health care system. The OECD reports additional data showing that in terms of years of life expectancy gained during 1960–2009, the United States (+8.3) ranked twenty-sixth out of thirty-four OECD members and almost 3 years below the OECD average of +11.2 (OECD 2011a, 7). Note that for the 1960–2009 period, the outliers are the Republic of Korea (+27.9), Turkey (+25.5), Chile (+21.4), Mexico (+17.8), Portugal (+15.6), and Japan (+15.2). To the extent that the OECD is suggesting that this is an indictment of the U.S.

HHassett.indbassett.indb 211211 111/8/121/8/12 9:279:27 PMPM 212 RETHINKING COMPETITIVENESS

TABLE 7-6 LIFE EXPECTANCIES AT AGE 80, 1987 (YEARS)

Nation Men Women United States 7.0 9.1 England 6.2 8.1 France 6.7 8.6 Japan 6.9 8.5 Sweden 6.5 8.3

SOURCE: Manton and Vaupel (1995, table 1).

health care system, that argument would be exceedingly weak. The obvious explanation of this pattern of relative increases in life expectancies is wealth: as nations’ respective per capita GDPs rise, health and life expectancies increase as well. Table 7-7 presents data on the growth of real per capita GDP for several OECD nations for the period 1970 through 2010.32 The outliers for increased life expectancy in table 7-7—Korea, Turkey, Mexico, Portugal, and Japan (with the exception of Mexico)—also had the highest growth rates in real per capita GDP. Greater wealth yields increases in the consumption of health care and a vast array of other goods and services offering improved life expectancies (Liu and Chollet 2006). Unless one is willing to ignore the preferences of others, it is not inefficient for individuals to consume more health— health care, safety, improved diets, greater sanitation, ad infinitum—when incomes or wealth increase. With respect to the infant mortality data—again used widely as a measure of the relative effectiveness of health care systems—OECD data for 2008 rank the United States thirty-first out of thirty-four OECD member states (OECD 2011b). The OECD average for that year was 4.6 infant deaths per 1,000 live births; the U.S. figure was 6.5, and only Chile, Mexico, and Turkey had higher figures. The U.S. infant mortality rate was higher than those of all the other high-income OECD economies and higher than those of Estonia, Hungary, Poland, Slovakia, and Slove- nia. The infant mortality comparisons, however, are problematic in that they are affected by issues of definition, the incidence of preterm and

HHassett.indbassett.indb 212212 111/8/121/8/12 9:279:27 PMPM IS THE U.S. “COMPETITIVE” IN HEALTH CARE? 213

Table 7-7 ANNUAL COMPOUND GROWTH RATES IN REAL PER CAPITA GDP, 1970–2010 (PERCENT)

Nation Compound Rate Australia 1.7 Canada 1.8 France 1.7 Germany 1.9 Italy 1.7 Japan 2.2 Korea 6.2 Mexico 1.3 Portugal 2.4 Turkey 2.3 United Kingdom 2.0 United States 1.8 OECD Total 1.9 Note: Earliest OECD data for Chile are for 1986. Sources: OECD (2010); author’s computations.

low-weight births, and other factors not substantially related to the quality of the health care system.33 Some nations do not count as “live” very-low-weight premature births (e.g., less than five hundred grams in weight) and/or very small births (e.g., less than thirty centimeters in length). Other nations classify infants surviving less than twenty-four hours as “stillborn.” In parts of the Euro- pean Union, babies born before twenty-six weeks of gestation who then die within one year are not counted in infant mortality data. In the United States, all live births are counted as such. Moreover, the incidence of preterm and low-weight births is correlated strongly with the teenage fertility rate; the United States has the fourth highest teenage birth rate among the thirty-four OECD members. Interestingly, the three members with higher teenage fertility rates in 2008—Mexico, Chile, and Turkey—also had the highest infant mortality rates (OECD 2011d). As with the comparative life expectancy data, the infant mortality comparisons are problematic in terms of drawing inferences about the effectiveness of the health care system.34

HHassett.indbassett.indb 213213 111/8/121/8/12 9:279:27 PMPM 214 RETHINKING COMPETITIVENESS

TABLE 7-8 FIVE-YEAR SURVIVAL RATES FOR CANCER OF DIFFERENT SITES, 2000–2002 (PERCENT)

Tumor Type United States Europe Breast 90.1 79.0 Colorectal 65.5 56.2 Corpus uteri 82.3 78.0 Lung 15.7 10.9 Melanoma 92.3 86.1 Non-Hodgkin lymphoma 62.0 54.6 Prostate 99.3 77.5 Stomach 25.0 24.9 All malignancies (men) 66.3 47.3 All malignancies (women) 62.9 55.8

SOURCE: Verdecchia et al. (2007).

It is far more useful to compare health outcomes (or “outputs”) that health care systems can affect more or less directly, controlling, however crudely, for important and independent influences. One important class of such comparative outcomes is cancer treatment, which at least plausibly is a useful proxy for health care effectiveness overall. Several comparisons of five-year survival rates for various cancers have been reported.35 Table 7-8 shows data reported by Verdecchia et al. (2007) for 2000–2002. Five-year survival in the United States is greater for every type of cancer reported by Verdecchia et al. (2007). Table 7-9 presents comparative data reported by the OECD for 2004–2009. For cervical cancer, the U.S. figure is 2 percentage points below the average for the OECD 16 and is higher only than the figures for Germany and the United Kingdom among the nations shown.36 For breast cancer, the U.S. rank is first, 5.8 percentage points above the figure for the OECD 16. For colorectal cancer, the U.S. figure is 4 percentage points above the OECD 16 figure, ranking below only Iceland and Japan (and Korea, not shown in the table). It is difficult to conclude from these data that the U.S. cancer treatment yields outcomes systematically less favorable than those observed in other advanced economies.

HHassett.indbassett.indb 214214 111/8/121/8/12 9:279:27 PMPM IS THE U.S. “COMPETITIVE” IN HEALTH CARE? 215

TABLE 7-9 OECD FIVE-YEAR SURVIVAL RATES, 2004–2009 (PERCENT)

Tumor Type Nation Cervical Breast Colorectala Canada 64.9 86.6 63.5 Finland 66.3 86.3 61.9 France NA NA NA Germany 62.9 63.3 60.5 Iceland 67.3 86.3 68.0 Japan 70.2 87.3 68.1 Netherlands 67.0 84.4 61.1 Norway 78.2 86.5 63.2 Sweden 68.1 86.0 60.8 United Kingdom 58.8 81.3 53.4 United States 64.4 89.3 64.6 OECD 16 66.4 83.5 60.6 NA, not available. aAverage for males and females.

SOURCES: OECD (2011a, charts 5.8.2, 5.9.2, 5.10.2); author’s computations.

The CONCORD study ranked a large number of nations with respect to five-year survival rates for four tumor types for the period 1990–1999 (Coleman et al. 2008). Table 7-10 shows the highest five rankings among thirty-one nations. The United States ranks first or second for each of the tumor types in terms of five-year survival rates. Again, it is difficult to conclude from these data that the U.S. health care system systematically offers outcomes poorer than those for other nations.

Concluding Observations New pharmaceuticals, as a crude generalization, are introduced one to two years earlier in the United States than elsewhere (Farrell et al. 2008, p. 89). The United States has faster adoption of new surgical techniques and

HHassett.indbassett.indb 215215 111/8/121/8/12 9:279:27 PMPM 216 RETHINKING COMPETITIVENESS

TABLE 7-10 CONCORD STUDY RANKINGS

Tumor Type Colorectal Breast Prostate (women) Colorectal (men) United States United States France Japan Canada Austria United States United States Sweden Canada Canada Australia Japan Australia Australia France Australia Germany Japan Canada

SOURCE: Coleman et al. (2008, fig. 1).

advances in anesthesia (Farrell et al. 2008, p. 89). Internationally, the substantial majority of medical “tourists” in pursuit of higher-quality care choose to come to the United States (Tilman, Guevara, and Mango 2008, p. 4). Five major U.S. hospitals conduct far more clinical trials than any other OECD nation (Farrell et al. 2008, p. 90). The United States has shorter waiting times for visiting a specialist and for elective surgery than the other advanced economies, with the possible exception of Germany (Farrell et al. 2008, p. 91). Does the United States “spend more and get less”? That conventional wisdom is far from obviously correct: proper adjustment of the spending data for the distortions introduced by explicit and implicit price controls and for nonprice rationing, combined with the available outcomes data, suggests that the United States indeed may spend less and get more. Assertions that greater value will be obtained from a U.S. health care system subjected to the greater centralization observed in many advanced economies in Europe and elsewhere cannot be accepted on the basis of the available evidence.

Notes 1. The views expressed are those of the author alone. Sincere thanks are due Joseph Antos, Michael F. Cannon, Laurence A. Dougharty, H. E. Frech III, Kevin A. Hassett, Robert B. Helms, John R. Lott, and Thomas P. Miller for useful suggestions; any remaining errors are the responsibility of the author.

HHassett.indbassett.indb 216216 111/8/121/8/12 9:279:27 PMPM IS THE U.S. “COMPETITIVE” IN HEALTH CARE? 217

2. For a brief and nontechnical discussion of the meaning of “competitiveness,” see Lawrence (2002). 3. For an approach differing from that discussed here—the effect of the U.S. health care system on the competitiveness of particular U.S. economic sectors and of the economy as a whole—see Chernew and Levy (2012). 4. As discussed more fully below, “health care outputs” differ from “health” or “health status” in that the latter two conditions result from numerous parameters, only some of which are related directly to the quality and cost of health care services. 5. That is, in terms of the value of the resources consumed in the production of health care services. This sort of comparison of “comparative advantage” is the standard analytic approach used by economists for examination of international trade flows and, more generally, for the analysis of economic specialization. It is therefore applicable to the analysis of competition in both domestic and international markets and to the comparison of the competitiveness of given sectors in an international context. 6. For a detailed critique of the WHO methodology, see Whitman (2008). See also Atlas (2011). 7. The five are “health level,” “health distribution,” “responsiveness,” “responsiveness distribution,” and “financial fairness.” 8. The seven nations are Australia, Canada, Germany, Netherlands, New Zealand, the United Kingdom, and the United States. 9. See the World Bank (2012) comparisons of per capita GDP. 10. For a useful review of the literature, see Liu and Chollet (2006). A normal good has an income elasticity of demand greater than zero; for a luxury good, it is greater than one. 11. I use the term “price suppression” here broadly to include both formal ceilings on the prices of given health care services and/or coverage and the price effect of buying (“monopsony”) power exercised by government purchasers of such services. Ser- vice restrictions are an additional tool with which to suppress prices, as is the explicit or implicit use of such other forms of nonprice rationing as waiting lists. This distortion introduces a gap (or “wedge”) between prices or spending and the true economic value of the resources used in the production of health care services. Note that price suppression has the effect of reducing reported spending if the demand elasticity for the given health care service is less than one (in absolute value), that is, if demand is inelastic. Except for long-term care services, that is what the empirical literature finds; see Liu and Chollet (2006, table IV.A). 12. At the same time, concentrated interests may pressure government officials to opt for alternatives that are more costly. A good example outside the health-care context is the Davis-Bacon requirement that “prevailing wages” be paid on federally financed projects, a policy that creates an important subsidy for unionized workers and perhaps their employers. In the context of health care, the recent decision by the Obama administration to require that private employers and/or insurers pay for

HHassett.indbassett.indb 217217 111/8/121/8/12 9:279:27 PMPM 218 RETHINKING COMPETITIVENESS

contraceptives and abortifacients without cost sharing is likely to yield an increase in the demand for branded versions of such drugs over generic versions, thus yielding a gain for the producers of the former. Whether this shift would yield medical benefits for patients is an issue outside the scope of the discussion here. 13. Note that a “better” alternative may not offer benefits sufficiently great to justify the greater cost. In addition, if government has elastic demands, then a lower-cost alternative might yield higher spending, but most government programs are struc- tured in ways that yield effective demand elasticities lower than one (in absolute value), largely because users of the service (e.g., Medicare beneficiaries) do not pay prices reflecting full marginal cost or because Congress structures government programs so that a lump-sum budget is exchanged for a lump-sum basket of goods and services. More broadly, “cost” comprises both real resource consumption and the net value of services not consumed (forgone “consumer and producer surplus”) because of regulatory restrictions imposed on government programs. “Cost” is not merely government outlays. 14. The new comparative effectiveness review (CER) process implemented by the Patient Protection and Affordable Care Act of 2010 is very likely to be used by policymakers in a comparative cost context, even though CER is constrained legally to compare only the relative effectiveness of alternative treatments. It is not, therefore, benefit/cost analysis, strictly speaking. But the point here is that policymakers have incentives to use CER findings, whatever their shortcomings, to drive cost/benefit evaluations that give outlay savings heavier weight than clinical improvements, relative to the case in which patients were to make such decisions even when confronted with the full costs of their choices; see Zycher (2011). 15. For a useful discussion of the incentives of public officials to discount the future too highly, see Buchanan and Lee (1982a,b). 16. For example, note that government does make such long-lived investments as highways and other infrastructure and education. This may be the result of interest- group demands from the users and providers of such capital assets. 17. Consider the simple case of effective price controls on gasoline. One obvious effect is the substitution of nonprice rationing—queuing—in place of price rationing, the upshot of which is lines at gasoline stations, as observed in 1973 and 1979. The other effect is more subtle: the production of gasoline is reduced, raising its market value. Because the marginal consumer is willing to “spend” an amount of dollars and time for gasoline summing to its marginal value, and because consumers must compete for gasoline using nonprice forms of competition (queuing), the sum of the legal maximum price and the value of time spent in line is greater than the market- clearing price in the absence of price ceilings. For a discussion of the distinction between prices and the value of real resource use in medical care, see Pauly (1993). See also Frech (2000). 18. The dollar conversions in the Laugesen-Glied (2011) study raise some issues outside the scope of the discussion here, but an alternative methodology would be unlikely to change the relative rankings, particularly for the United States.

HHassett.indbassett.indb 218218 111/8/121/8/12 9:279:27 PMPM IS THE U.S. “COMPETITIVE” IN HEALTH CARE? 219

19. The definition of “orthopedic surgery” is a bit unclear in the Laugesen-Glied study and so may be subject to some variation across nations. 20. The eleven nations are Australia, Canada, Finland, France, Italy, Israel, the Republic of Korea, Portugal, Slovenia, Sweden, and the United States. 21. These comparisons obviously are somewhat crude. The number of “physician consultations” per capita, for example, yields little information about the length of such consultations or about other dimensions of service quality. Hospital stays may be shorter because they are more productive in terms of improved health. Some procedures may be more common in the United States simply because of differences in usual and customary clinical practices, and so forth. 22. The cost of using capital equipment may be particularly low if such equipment is long-lived; that cost may be only some depreciation, some electric power, and the economic value of the time spent evaluating the findings by a technician or a physician. 23. A useful discussion of the features of several national health care systems is offered by Tanner (2008). 24. See, for example, the OECD description of France as “a country which is generally accepted as having very good health services” (OECD 2011c, p. 1). 25. For example, the per capita availability of radiation therapy equipment is less than half that of the United States (see OECD 2011b). 26. This outcome is certain because additional resources would reduce the length of waiting lists or otherwise increase the availability of services, which has the effect of inducing more patients (or demanders) to enter the system and/or increasing demands by existing program beneficiaries. As long as the market is not satiated with medical services—as long as the marginal value of medical services is greater than zero—the addition of more resources without additional cost sharing must induce an increase in demands placed on the system. 27. See the Tanner (2008) summaries for several nations not discussed here. 28. Such informal payment, however, would reduce such other hidden costs as the adverse effects of waiting lists, lower-quality care, and so forth. 29. The underlying composition of such criteria, and the weights assigned to them and to the underlying components, are essentially arbitrary. See the discussion above. 30. Some may be more subtle: does a lower geographic population density yield longer ambulance response delays? 31. The OECD argues that GDP should not have been included in the Ohsfeldt- Schneider econometric model and that exclusion would yield a U.S. ranking of seventeenth. I do not agree with the OECD argument, as GDP is a useful proxy variable for the demand for health care, but that is an issue outside the scope of the discussion here (see OECD 2008). 32. The earliest year for which the OECD reports data for most member nations is 1970. 33. Two useful references are National Center for Health Statistics (2008) and Rich- ardus et al. (1998). Another is Sanghavi (2007).

HHassett.indbassett.indb 219219 111/8/121/8/12 9:279:27 PMPM 220 RETHINKING COMPETITIVENESS

34. Another major problem, outside the scope of the discussion here, is the effect of the health care system on behavior itself. Consider the favorable impacts that modern medicine can have on the adverse effects of obesity. Various classes of drugs—statins, ACE inhibitors, and so forth—can ameliorate the serum cholesterol and hypertension effects of obesity. Knee and hip replacements can reduce the adverse long-term effects of excessive weight. Accordingly, it is reasonable to hypothesize that the availability of such medical care and technology can reduce the degree to which individuals choose to discipline their behavior. For the classic analysis of such effects in the context of automobile safety, see Peltzman 1975. 35. Strictly speaking, the five-year survival rate for a given cancer can be misleading. Consider a cancer that invariably kills its victims seven years after the tumor begins to grow, regardless of treatment. Nation A screens for this particular cancer on a widespread basis and uniformly detects the tumors early. Its five-year survival rate will be 100 percent, even though treatment is utterly useless. Nation B screens much less intensively and detects these tumors only after, say, four years. Its five-year survival rate will be 0 percent. But in this (extreme) case, the difference in the five-year survival rates between nations A and B is meaningless; all patients die at the end of the seven-year period. This example, again, is extreme; it simply is implausible that comparative five-year survival rates for cancers do not have some significant relationship to the effectiveness of health care, although earlier screening may obscure the effects of treatment to some degree. For example, the higher five-year survival rate in the United States for prostate cancer may reflect earlier screening to a significant degree. At the same time, it is not plausible that earlier screening is worthless in terms of medical treatment and favorable outcomes. But the data must be interpreted carefully. 36. The United States ranks higher than Denmark, Germany, the Czech Republic, Austria, the United Kingdom, and Ireland.

References Aaron, Henry J., and Paul B. Ginsburg. 2009. “Is Health Spending Excessive? If So, What Can We Do about It?” Health Affairs 28(5):1260–75. Acosta, Jose A., Jack C. Yang, Robert J. Winchell, Richard K. Simons, Dale A. Fort- lage, Peggy Hollingsworth-Fridlund, and David B. Hoyt. 1998. “Lethal Injuries and Time to Death in a Level I Trauma Center.” Journal of the American College of Surgeons 186(5):528–33. Atlas, Scott W. 2011. “The Worst Study Ever?” Commentary, April. Bosch, Xavier. 2002. “Spain: The Old Frequently Live with Their Families.” British Med- ical Journal 324(7353). http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1172217/. Buchanan, James M., and Dwight R. Lee. 1982a. “Tax Rates and Tax Revenues in Political Equilibrium: Some Simple Analytics.” Economic Inquiry 20(3):344–54. ———. 1982b. “Politics, Time, and the Laffer Curve.” Journal of Political Economy 90(4):816–19.

HHassett.indbassett.indb 220220 111/8/121/8/12 9:279:27 PMPM IS THE U.S. “COMPETITIVE” IN HEALTH CARE? 221

Centers for Disease Control and Prevention. 2009. Deaths: Final Data for 2009. http:// www.cdc.gov/nchs/data/dvs/deaths_2009_release.pdf. Chernew, Michael E., and Phillip I. Levy. 2012. “American Competitiveness and the Health Care System.” In Rethinking Competitiveness, ed. Kevin A. Hassett. Washing- ton, D.C.: AEI Press. Coleman, Michael P., Manuela Quaresma, Franco Berrino, Jean-Michel Lutz, Roberta De Angelis, Riccardo Capocaccia, Paola Baili, Bernard Rachet, Gemma Gatta, Timo Hakulinen, Andrea Micheli, Milena Sant, Hannah K. Weir, J. Mark elwood, Hideaki Tsukuma, Sergio Koifman, Gulnar Azevedo e Silva, Silvia Francisi, Mari- ano Santaquilani, Arduino Verdecchia, Hans H. Storm, John L. Young, and the CONCORD Working Group. 2008. Cancer Survival in Five Continents: A Worldwide Population-Based Study (CONCORD). July 17. http://www.thelancet.com/oncology. http://healthcare.procon.org/sourcefiles/CONCORDCancerSurvivalStudy.pdf. Comanor, William S., H. E. Frech III, and Richard D. Miller, Jr. 2006. “Is the United States an Outlier in Health Care and Health Outcomes? A Preliminary Analysis.” International Journal of Health Care Finance and Economics 6(1):3–23. Davis, Karen, Cathy Schoen, and Kristof Stremikis. 2010. Mirror, Mirror on the Wall: How the Performance of the U.S. Health Care System Compares Internationally, 2010 Update. Commonwealth Fund, June. http://www.commonwealthfund.org/~/ media/Files/Publications/Fund%20Report/2010/Jun/1400_Davis_Mirror_Mir- ror_on_the_wall_2010.pdf. Ehrbeck, Tilman, Ceani Guevara, and Paul D. Mango. 2008. “Mapping the Market for Medical Travel.” McKinsey Quarterly, May. http://www.cosmashealth.com/work- space/assets/uploads/McKinsey-MedicalTravel.pdf. Farrell, Diana, Eric Jensen, Bob Kocher, Nick Lovegrove, Fareed Melhem, Lenny Mendonca, and Beth Parish. 2008. Accounting for the Cost of US Health Care: A New Look at Why Americans Spend More. McKinsey Global Institute, December. http:// www.mckinsey.com/Insights/MGI/Research/Americas/Accounting_for_the_cost_ of US_health_care. Frech, H. E., III. 2000. “Physician Fees and Price Controls.” In American Health Care: Government, Market Processes, and the Public Interest, ed. Roger D. Feldman. New Brunswick, NJ: Transaction Publishers, pp. 346–62. Garber, Alan M., and Jonathan Skinner. 2008. “Is American Health Care Uniquely Inefficient?” Journal of Economic Perspectives 22(4):27–50. Goodman, John. 2011. Do We Really Spend More and Get Less? November 30. http:// healthblog.ncpa.org/do-we-really-spend-more-and-get-less/. Goodman, John. 2009. Health Care Reform: Do Other Countries Have the Answers? Statement submitted to the Energy and Commerce Subcommit- tee on Health, U.S. House of Representatives, June 24. http://www.ncpa.org/ pdfs/062409ECHearingNCPAGoodman.pdf. Koechlin, Francette, Luca Lorenzoni, and Paul Schreyer. 2010. Comparing Price Levels of Hospital Services across Countries: Results of Pilot Study. OECD Health Working

HHassett.indbassett.indb 221221 111/8/121/8/12 9:279:27 PMPM 222 RETHINKING COMPETITIVENESS

Papers 53. OECD Publishing. http://www.oecd-ilibrary.org/docserver/download/ fulltext/5km4k7mrnnjb.pdf?expires=1337291145&id=id&accname=guest& checksum=BE3145D6FE0254CB1E079E1BDA6AAE68 . Kresge, Naomi. 2012. Greek Crisis Has Pharmacists Pleading for Aspirin as Drug Supply Dries Up. January 10. http://www.bloomberg.com/news/2012-01-10/greek-crisis- has-pharmacists-pleading-for-aspirin-as-drug-supply-dries-up.html. Laugesen, Miriam J., and Sherry Glied. 2011. “Higher Fees Paid to US Physicians Drive Higher Spending for Physician Services Compared to Other Countries.” Health Affairs 30(9):1647–56. Lawrence, Robert Z. 2002. “Competitiveness.” In Concise Encyclopedia of Economics, Liberty Fund Library of Economics and Liberty. http://www.econlib.org/library/ Enc1/Competitiveness.html. Liu, Su, and Deborah Chollet. 2006. Price and Income Elasticity of the Demand for Health Insurance and Health Care Services: A Critical Review of the Literature. Mathematica Policy Research, March 24. http://www.mathematica-mpr.com/publications/pdfs/ priceincome.pdf. Manton, Kenneth G., and James W. Vaupel. 1995. “Survival after the Age of 80 in the United States, Sweden, France, England, and Japan.” New England Journal of Medicine 333(18):1232–35. Mingardi, Alberto. 2006. “A Drug Price Path to Avoid.” Washington Post, November 12. http://www.washingtonpost.com/wp-dyn/content/article/2006/11/10/AR2006 111001489.html. Mossialos, Elias, Sara Allin, and Konstantina Davaki. 2005. “Analyzing the Greek Health System: A Tale of Fragmentation and Inertia.” Supplement, Health Econom- ics 14(S1):S151–68. National Center for Health Statistics. 2008. Recent Trends in Infant Mortality in the United States. NCHS Data Brief No. 9, October. http://www.cdc.gov/nchs/data/ databriefs/db09.htm. Ohsfeldt, Robert L., and John E. Schneider. 2006. The Business of Health: The Role of Competition, Markets, and Regulation. Washington, D.C.: AEI Press. O’Neill, June E., and Dave M. O’Neill. 2007. “Health Status, Health Care and Inequal- ity: Canada vs. the U.S.” Forum for Health Economics and Policy 10(1). Organisation for Economic Co-operation and Development (OECD). Infant Mortality. Family Database. http://www.oecd.org/dataoecd/4/36/46796773.pdf. ———. 2008. OECD Economic Surveys: United States. Volume 16. http://www.oecd. org/document/32/0,3746,en_2649_34111_41803296_1_1_1_1,00.html. ———. 2010. Stat Extracts. http://stats.oecd.org/Index.aspx?DataSetCode=SNA_ TABLE1. ———. 2011a. Health at a Glance 2011: OECD Indicators. http://www.oecd.org/dataoecd/6/28/49105858.pdf. ———. 2011b. Health Data 2011. http://www.oecd.org/document/16/0,3746,en_ 2649_37407_2085200_1_1_1_37407,00.html.

HHassett.indbassett.indb 222222 111/8/121/8/12 9:279:27 PMPM IS THE U.S. “COMPETITIVE” IN HEALTH CARE? 223

———. 2011c. “Why Is Health Spending in the United States So High?” In Health at a Glance 2011: OECD Indicators. http://www.oecd.org/dataoecd/12/16/49084355. pdf. ———. 2011d. Share of Births out of Wedlock and Teenage Births. OECD Family Data- base. http://www.oecd.org/document/4/0,3746,en_2649_37419_37836996_ 1_1_1_37419,00.html. Pauly, Mark V. 1993. “U.S. Health Care Costs: The Untold True Story.” Health Affairs 12(3):152–59. http://content.healthaffairs.org/content/12/3/152.full.pdf. Peltzman, Sam. 1975. “The Effects of Automobile Safety Regulation.” Journal of Politi- cal Economy 83(4):677–726. Petkantchin, Valentin. 2007. “Bureaucratic Drug Delisting: The French Example.” EU Reporter, January-February. http://www.institutmolinari.org/bureaucratic-drug- delisting-the,321.html. Richardus, Jan H., Wilco C. Graafmans, S. Pauline Verloove-Vanhorick, and Johan P. Mackenbach. 1998. “The Perinatal Mortality Rate as an Indicator of Quality of Care in International Comparisons.” Medical Care 36(1):54–66. Roy, Avik. 2011. “The Myth of Americans’ Poor Life Expectancy.” Forbes, Novem- ber 23. http://www.forbes.com/sites/aroy/2011/11/23/the-myth-of-americans- poor-life-expectancy/. Sanghavi, Darshak. 2007. “Baby Gap: The Surprising Truth about America’s Infant- Mortality Gap.” Slate, March 16. http://www.slate.com/articles/health_and_science/ medical_examiner/2007/03/baby_gap.html. Tanner, Michael. 2008. The Grass Is Not Always Greener: A Look at National Health Care Systems around the World. Policy Analysis No. 613. Cato Institute, Washington, D.C., March 18. http://www.cato.org/pubs/pas/pa-613.pdf. Verdecchia, Arduino, Silvia Francisci, Hermann Brenner, Gemma Gatta, Andrea Micheli, Lucia Mangone, Ian Kunkler, and the EUROCARE-4 Working Group. 2007. “Recent Cancer Survival in Europe: A 2000–2002 Period Analysis of EURO CARE-4 Data.” Lancet Oncology 8(9):784–96. Whitman, Glen. 2008. WHO’s Fooling Who? The World Health Organization’s Problem- atic Ranking of Health Care Systems. Briefing Paper 101. Cato Institute, Washington, D.C., February 28. http://www.cato.org/pubs/bp/bp101.pdf. World Bank. 2012. “GDP per Capita.” http://data.worldbank.org/indicator/NY.GDP. PCAP.CD. World Health Organization. 2000. World Health Report 2000. http://www.who.int/ whr/2000/en/whr00_en.pdf. Zycher, Benjamin. 2007. Comparing Public and Private Health Insurance: Would a Single- Payer System Save Enough to Cover the Uninsured? Center for Medical Progress Report No. 5, Manhattan Institute for Policy Research, New York, October. http:// www.manhattan-institute.org/html/mpr_05.htm. ———. 2011. Comparative Effectiveness Reviews: Quantitative Analysis of Research and Development Investment Effects. Monograph, Pacific Research Institute, July.

HHassett.indbassett.indb 223223 111/8/121/8/12 9:279:27 PMPM 224 RETHINKING COMPETITIVENESS

http://www.pacificresearch.org/publications/comparative-effectiveness-reviews- quantitative-analysis-of-research-and-development-investment-effects. Zycher, Benjamin, Joseph A. DiMasi, and Christopher-Paul Milne. 2008. The Truth about Drug Innovation: Thirty-Five Summary Case Histories on Private Sector Contribu- tions to Pharmaceutical Science. Center for Medical Progress Report No. 6, Manhat- tan Institute for Policy Research, New York, June. http://www.manhattan-institute. org/html/mpr_06.htm.

HHassett.indbassett.indb 224224 111/8/121/8/12 9:279:27 PMPM 8

Global Value Chains and the Continuing Case for Free Trade: Trade Theory and Illustrations from the United States and East Asia Claude Barfield and Matthew H. Jensen

To borrow a well-known GM slogan, “It’s not your father’s trading system anymore.” Over the past several decades, a new trade paradigm has arisen, one that deemphasizes domestic, vertically integrated firms competing in end products with similarly integrated firms from other nations. Instead, from automobiles to electronics, chemicals, and clothing, the production process has dispersed. Parts and components crisscross borders multiple times, and lead firms interact with many foreign partners through a variety of contractual arrangements. Underpinning the new trade paradigm have been dramatic reductions in transportation costs and a technological revolution in communications and information. Today, the most successful firms have divided the supply chain into stages, or tasks and fragments, in order to take advantage of global relocation possibilities that “slice through” national boundaries. In the oft-cited example of the iPad, Apple has its headquarters, R&D, and design facilities in the United States, but the iPad’s hard drive is designed in Japan and actually manufactured in factories in China and the Philippines. The controller chip is designed by a U.S. firm, but production is split between firms in Taiwan and the United States. There are more than four hundred smaller parts and components of the iPad that are manufactured by firms scattered around East Asia. And finally, the final product

225

HHassett.indbassett.indb 225225 111/8/121/8/12 9:279:27 PMPM 226 RETHINKING COMPETITIVENESS

is assembled by Taiwanese corporations located in China before being shipped back to the United States and distributed by Apple. In this chapter, we attempt to analyze the implications of the rise of global value chains (GVCs), both from a theoretical perspective and from an empirical perspective, while drawing heavily from an existing, burgeoning body of literature. The first half of the chapter explains how the new trade paradigm fits within existing theories of trade. New developments in theory during the 1970s and 1980s implied a concrete definition of international competitiveness in trade, where countries could compete for key industries through strategic industrial policy. Although this competition is defensible as an intellectual construct, our chapter will explain why these developments came too late and why the ability of countries to compete on this basis has waned, if it ever existed in practice. The rise of GVCs has undermined key assumptions that lent the theory credibility and has undermined the ability of nations to gain by enacting industrial policy. The bottom line is that the advent of GVCs and the “slicing” of the production process have rendered free trade and government nonfavoritism in industry even more important. The second half of the chapter will explore in some depth the evolution of supply chains in two key regions: East Asia and the United States. The goal is to provide a clear understanding of the increasingly complex nature of regional and global value chains and of the varying roles of the United States, Japan, the newly industrialized economies (Korea, Taiwan, and Singapore), and developing East Asia (Thailand, Malaysia, the Phil- ippines, etc.). The China phenomenon—and myths attached to it—will receive separate treatment. More generally, the chapter will chart how supply chains produce changes in the volume and nature of global trade, first by their very nature increasing the volume of trade as parts and components pass back and forth across borders and, second, by changing the content of what is traded, with the steep rise in intermediate goods as opposed to final products. Within these fact-driven sections, the chapter will also point to anomalies in the measurement of trade flows and misleading conclusions concerning trade balances that stem from supply chain dynamics. These misperceptions are particularly important in dealing with the realities of China’s trade with the United States and East Asia. The chapter will explain

HHassett.indbassett.indb 226226 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 227

how traditional instruments of trade policy—anti-dumping actions, countervailing duties—often not only miss the target but also redound the economic interests of the nation wielding these weapons. Finally, we will lay out a set of observations and policy conclusions that stem from the challenges posed by the new trade paradigm. Whether surprising or not, the theme is an old one for trade economists: trade policy per se is not the first or second key to global competitiveness, and activist trade policy is generally damaging. It’s the basics that count: policies to broadly enhance the productivity of U.S. firms, including education for a high-skilled workforce, regulatory and tax policies that do not deter competition, balance in macroeconomic policy, and strong intellectual property laws and enforcement, among others.

Two Unbundlings of Trade and Shifting Policy Implications Richard Baldwin has described globalization as occurring in two great unbundlings (see, for example, Baldwin 2006, 2011; Baldwin and Martin 1999). The first was driven by falling transportation costs and led to an explosion of trade from 1850 to 1914 and again after the world wars from 1960 through today. The second unbundling was driven by high-speed transportation and lower communication and coordination costs and is characterized by trade in tasks and services in GVCs.1 It began around 1980 and is still occurring today. The international trade of final goods that stemmed from the first unbundling has been thoroughly described by traditional trade theory. Two broad areas of consensus emerged between the early 1800s and the late 1900s. First, trade is about mutually beneficial exchange, not competition. Even those nations that do not have an absolute advantage in producing a single good benefit from trade; that is, if a nation participates in trade, the overall purchasing power of its residents will be higher than if it did not. Second, there are a few areas where policy can improve the “competitiveness” of a nation, as defined by its ability to gain comparative advantage in key industries where it otherwise wouldn’t be able to. The ability of nations to gain from this competition was limited in practice by the skill and integrity of policymakers, but it was theoretically possible.

HHassett.indbassett.indb 227227 111/8/121/8/12 9:279:27 PMPM 228 RETHINKING COMPETITIVENESS

The next section of this chapter will briefly discuss the history of the first unbundling of trade and then will explain each of the areas of theoretical consensus in some depth. The advent of GVCs in the second unbundling broadly supports the first consensus of trade theory, that trade is a mutually beneficial activity. Their rise, though, undermines some of the theoretical support for nations to compete for key industries, and it implies even greater practical challenges to determining which industries are worth competing for. The third section of this chapter will provide some historical perspective of the second unbundling of trade and then build on the theoretical underpinnings of the second section to explain how GVCs make trade more beneficial, competition through industrial policy less beneficial, and strengthen the case for free trade. Paul Krugman (1999) writes that “the Big Ideas in trade theory probably can be summarized in not much more than 5 minutes.” We will attempt that in the next section.2

The First Unbundling and the Big Ideas of Trade Theory Before the transportation revolution of the nineteenth century, self- sufficiency was forced on communities. Without the widespread use of rail and steamship that developed from the 1820s to the 1870s, communities were largely limited to consuming what they were able to produce. As shipping costs fell in the mid-nineteenth century,3 factories could be moved away from consumers, allowing nations and regions to specialize in products based on comparative advantage. The world wars and the Great Depression halted the progress in the mid-twentieth century, but then transportation improvements and drastic reductions of tariffs and other trade barriers led to a second wave of trade growth after the Second World War.4 The disaggregation of production and consumption is the hallmark of the first unbundling. Nearly two centuries ago, David Ricardo described the benefits of this disaggregation and why the international trade that allows it is a mutually beneficial behavior rather than a competitive one (Ricardo 1817).

HHassett.indbassett.indb 228228 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 229

Ricardo, Comparative Advantage, and Gains from Trade. Today, a two-good, two-country Ricardian model is still the classic tool for making the point that trade is mutually beneficial. Although this exercise has been conducted in countless other papers and textbooks, we feel compelled to include a numerical example, given our mandate to define competitiveness in trade. Later, once we have dispelled the broad notion that countries compete to export more goods to each other, we will defend an alternative definition of competitiveness. Suppose Home and Foreign are two countries of equal size that both produce wine and cloth. Following Ricardo for simplicity, we will assume that the productive capacity of each country relies solely on the productivity of its workforce, which in turn can be thought of as resulting from technological differences. Home labor can produce one gallon of wine in one hour or two yards of cloth in one hour. Foreign labor can produce eight gallons of wine in one hour or four yards of cloth. This means that the relative price of cloth in terms of wine for the home country is two yards of cloth per gallon of wine. The relative price of cloth in terms of wine for the foreign country is 0.5 yards of cloth per gallon of wine. The world equilibrium price for cloth in terms of wine will have to be between these two values, 0.5 and 2. Let’s assume it is 1.5 Before proceeding we should note that Home labor is less productive in both viniculture and textiles. In other words, Foreign has an absolute advantage in both industries. However, Home has a comparative advantage in textiles, and Foreign has a comparative advantage in viniculture because the opportunity cost of producing cloth is lower for Home than it is for Foreign, and the opportunity cost of producing wine is lower for Foreign than it is for Home. So what happens if we allow the two countries to trade? Each country will specialize in its comparative advantage. Foreign will have “discovered” an amazing new way to “produce” cloth: rather than spending an hour producing four yards of cloth, it will spend that hour making eight gallons of wine and then trade that wine for eight yards of cloth, assuming Home has the excess labor capacity to produce the extra cloth. Home will have discovered a new way to “produce” wine: it will spend an hour making two yards of cloth and then trade that for two gallons of wine. Both countries will have improved their ability to consume the product for which their labor is less productive. If we assume that both labor forces are at full capacity after one hour of work, then the trade patterns will look like what is in this example:

HHassett.indbassett.indb 229229 111/8/121/8/12 9:279:27 PMPM 230 RETHINKING COMPETITIVENESS

Production and consumption per hour of labor, before and after trade Before Before Trade Pro- Trade Pro- duction and After Trade After Trade duction and After Trade After Trade Consump- Production, Consump- Consump- Production, Consump- tion, Wine Wine tion, Wine tion, Cloth Cloth tion, Cloth Home 1 gallon 0 gallons 2 gallons 2 yards 4 yards 2 yards Foreign 8 gallons 10 gallons 8 gallons 4 yards 3 yards 5 yards

When popular commentators or politicians bemoan the ability of the United States, or any other country, to “compete in a global marketplace,” this is the counterargument that should come to mind.6 By specializing in products where a comparative advantage exists, countries improve the amount of goods that their labor can purchase. In other words, trade is almost always positive sum.7 Before moving on to a more sophisticated model of trade, it should be mentioned that, simple as it is, the Ricardian model has been relatively successful at predicting patterns of trade. For example, the classic test is of British and U.S. trade in the period following the Second World War. For twenty-eight industries, Balassa (1963) found that in 1950–51 U.S. labor was more productive than UK labor in every single industry, from a low of 11 percent more productive in shipbuilding and repairing to a high of 466 percent more productive in automobiles, trucks, and tractors.8 The United States had an absolute advantage in every industry; however, as is displayed in figure 8-1, the United States was only a net exporter to the United King- dom in those industries where American productivity exceeded British productivity by more than around 250 percent. In other words, the United Kingdom had a comparative advantage in industries where its productivity exceeded 40 percent of the U.S. level.

Heckscher-Ohlin and the Causes of Trade. The Ricardian model employed here is extremely useful for illustrating the concept of comparative advantage. It makes the point that in most cases trade liberalization is beneficial not only for the country that undertakes it but also for the rest of the world. This is not to say, though, that it comprehensively tells the story of international trade. Indeed, the Heckscher-Ohlin (H-O) model,

HHassett.indbassett.indb 230230 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 231

FIGURE 8-1 U.S./UK EXPORT AND PRODUCTIVITY RATIOS BY INDUSTRY, 1950–1951

400

350

300

250

200 Exports 150

100

0 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5 Labor Productivity (in Hundreds)

which relies on differences of factor endowments to determine comparative advantage, has become a workhorse of international trade theory.10 Under the H-O model, countries will export those goods for which production requires their most abundant resources and import those goods for which production requires their scarcest resources. The preferential use of the H-O model rests heavily on the fact that it makes stronger predictions about the distributional effects of trade than the Ricardian model does, and it allows for factor endowments to influence trade, not just technological differences. The H-O model, however, has come under significant empirical scru- tiny, with the most famous example offered up by Nobel laureate Wassily Leontief in 1953. He found that that the United States imported more capital-intensive activities than it exported, a result quite contrary to the H-O model’s prediction, given that the United States is the most capital- abundant country in the world. This has been termed the Leontief Paradox.

HHassett.indbassett.indb 231231 111/8/121/8/12 9:279:27 PMPM 232 RETHINKING COMPETITIVENESS

Several authors have critiqued Leontief’s methodology, and the debate surrounding the H-O model is still lively.10 For example, Adrian Wood (1994), in his aptly titled article, “Give Heckscher and Ohlin a Chance!” argues that the criticisms of the H-O theory have been exaggerated based on a misspecification of the model: since capital is largely mobile, it should not be treated as a factor similar to land or even the labor force. If capital is excluded but high-skilled labor and low-skilled labor are treated as two separate factors, then the H-O model is quite adept at explaining the trade in manufactures between the developed North and developing South.11 It can also explain, for example, why the United States might import cars and export advanced computer chips, even though cars are much more capital intensive to manufacture. The reason is that cars are mass-produced goods that require little skilled labor, whereas computer chips rely heavily on a skilled workforce.12 It turns out that factor endowments sometimes provide an explanation of trade, and when they do not, the H-O model still allows technological differences to be invoked. The important point to take away from this section is that Heckscher- Ohlin is not a departure from the comparative advantage framework that we examined using a Ricardian model; it merely allows comparative advantage to be determined by factor endowments, not just technological differences. In the next subsection, we will describe how some trade does not depend on comparative advantage at all.

New Trade Theory and Internal Increasing Returns to Scale. One aspect of trade confounds both Ricardian and Heckscher-Ohlin models: the two-way trade in similar goods between countries. One example of this type of trade is the flow of cars and car parts between Canada and the United States. Between 1992 and 1999, for example, about 55 percent of this exchange was in similar products differentiated only by variety, not price or quality, according to Montout, Mucchielli, and Zignago (2002). Two- way trade in similar goods, often termed “horizontal intra-industry trade,” is inexplicable under a comparative advantage framework: if trade is based on technological or factor endowment differences between countries, why would two countries ever trade the same good back and forth? In the late 1970s and early 1980s, a few economists began to offer an answer to this question, and their work has become known as New Trade

HHassett.indbassett.indb 232232 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 233

Theory.13 Widely accepted today, their explanation hinges on increasing returns to scale (IRS), or, in other words, when the per-unit cost of a good depends on the size of the individual firm or industry. If IRS exist, then specialization is advantageous, and countries will trade in order to allow more specialization. In this case, the gains from trade will be twofold. First, just like under trade motivated by comparative advantage, each country’s purchasing power will increase. An additional gain from trade that is not present under comparative-advantage-motivated trade is that IRS will also allow for a greater variety of goods. Given these effects, gains from trade should be largest when IRS are strong and when product differentiation is valuable. Throughout history, increasing returns to scale have occasionally been mentioned as a reason for trade,14 but until an industrial organization paper by Dixit and Stiglitz in 1977, there was little capacity to model them. It has even been claimed that the difficulty of modeling prompted them to be ignored. The problem was that internal IRS, where the per-unit cost of a good depends on the size of the individual firm, implied imperfect competition. The existing trade models, however, relied on perfectly competitive markets where constant or decreasing returns are assumed.15 Basically, the effects of IRS were assumed away. So it was really the easy way to model imperfect competition provided by Dixit and Stiglitz’s 1977 paper that allowed for the formation of New Trade Theory.16 As long as increasing returns to scale are internal to a firm, they will not supplant comparative advantage as a motivation for trade. If two countries are substantially different, trade between those two countries will stem from those differences. As can be seen in figure 8-2, horizontal intra-industry trade is a small part of overall world trade. Internal IRS can be considered a supplemental motivation for trade. For example, two countries with similar factor endowments and labor forces might not have much motivation to trade based on comparative advantage, but would display a great deal of horizontal intra-industry trade in an industry characterized by internal IRS and diversified products. Now we have an explanation for the automobile trade between the United States and Canada. What is not clear from New Trade Theory is why Toyota assembles Corollas and some Lexus in Canada but Camrys in the United States.17 Generally, patterns of horizontal intra- industry trade are left unexplained by New Trade Theory.

HHassett.indbassett.indb 233233 111/8/121/8/12 9:279:27 PMPM 234 RETHINKING COMPETITIVENESS

FIGURE 8-2 TRADE TYPES, 1989–2002

70.0

Inter-industry trade 60.0

50.0

40.0

30.0 Intra-industry trade (vertical)

20.0 Intra-industry trade (horizontal)

10.0 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002

SOURCE: Fontagné, Freudenberg, and Gaulier (2006).

An Outdated Definition of Competitiveness. New Trade Theory is not, in itself, central to our mission of discussing a definition of competitiveness in trade, but it has had a major effect. By establishing a highly plausible framework where internal increasing returns to scale motivate trade, it sparked a newfound interest in external increasing returns to scale, where the per-unit cost of a good depends on the size of the industry and not necessarily the firm. These external IRS became central to any discussion of “competitiveness” in trade.18 While internal IRS supplement but do not supplant comparative advantage as a determinant of trade, it is feasible that external IRS can create comparative advantage where none would otherwise exist. The formalization of this theoretical point led to a lively debate in the trade policy world about the possibility of enacting policy— essentially industrial policy—to protect or develop external IRS and create a “competitive” advantage.

HHassett.indbassett.indb 234234 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 235

Here we will focus on both the theoretical case for industrial policy that was developed during the 1980s and 1990s and the difficulties of enacting it in the real world. In the next section, we will explain how the advent of GVCs undermines many of the theoretical arguments that were developed in support of industrial policy and has made a difficult task in the real world all but impossible.

External Increasing Returns to Scale. The first undertaking is to explain what causes external IRS and how they arise. There are two main types of external IRS, pecuniary and technological. Pecuniary IRS rely on the size of the market, and the most representative of these are labor market pooling and input sharing. The main technological increasing return to scale is knowledge spillover. Labor market pooling describes how suitable workers will concentrate in an area around an industrial cluster. This gives workers better job opportunities and allows firms a better pool of workers to hire from.19 Just as is the case with labor market pooling, service providers and providers of other inputs will tend to pool around industry clusters. This is termed “input sharing.”20 Technological IRS rely on “knowledge spillovers.” These exist when firms learn from each other through firm-level interaction or often through interactions among employees. The idea that knowledge spillovers exist has been around for more than a century. In 1879 Alfred Mar- shall and Mary Paley Marshall wrote, “Where large masses of people are working at the same kind of trade, they educate one another. The skill and the taste required for their work are in the air, and children breathe them as they grow up…each man profits by the ideas of his neighbors” (Marshall and Paley Marshall 1879, p. 197). This is still how knowledge spillovers are thought to occur today, from knowledge floating around “in the air.” A broad empirical body of literature has confirmed that the pecuniary and technological external IRS described here both exist and contribute to industry clustering.21 However, there is substantial difficulty involved in measuring external IRS; in particular, it is difficult to know how far pecuniary and technological spillovers extend across industrial specialization, geography, and time.22

HHassett.indbassett.indb 235235 111/8/121/8/12 9:279:27 PMPM 236 RETHINKING COMPETITIVENESS

The Theory-Driven Argument for Industrial Policy. When the forces that pull firms together, agglomeration forces, are stronger than the forces that push firms apart, dispersion forces, industries will tend to cluster, external IRS can be exploited, and international specialization will follow. What’s more, the specialization will have an arbitrary quality: the Swiss watchmaking industry likely developed because John Calvin banned the wearing of jewels in 1541, so Swiss jewelers and goldsmiths began making watches instead.23 Much of the lasting success of these centers can be attributed to a self-reinforcing cycle where agglomeration strengthens external IRS, and external IRS cause agglomeration. This international specialization due to external IRS has the same effect on patterns of trade as does specialization due to technological differences or differences in factor endowments. Simply put, external IRS can create comparative advantage. This should be contrasted with the way that internal IRS influence trade. Internal IRS provide an additional motivation for trade, outside of comparative advantage, that manifests itself in horizontal intra-industry trade. External IRS actually create comparative advantage by supplanting or exaggerating the advantages gained from factor endowments or technological differences. This will manifest itself in vertical intra- industry or inter-industry trade. Moreover, historically it was difficult for private markets to establish an industry where external IRS might be present. This was true even in a country that might have a comparative advantage in that industry if it were established. While it is typically assumed that financial markets function well enough to finance a firm while it develops internal economies of scale,24 economists found it harder to imagine entire industries being financed as they grew large enough for external economies of scale to be realized.25 In fact, the canonical models of trade that show external IRS contributing to comparative advantage rely on the assumption that “firms contemplate entry only at such small scale that they believe they can have negligible impact on national output and their own productivity” (Gross- man and Rossi-Hansberg 2010, p. 832).26 Economists in the 1980s and 1990s identified several implications that would follow if firms did not consider entering an industry at a large enough scale to “internalize” the external economies of scale and if strong agglomerative forces pulled together external IRS industries into industrial

HHassett.indbassett.indb 236236 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 237

clusters. The first and most important was that countries could theoretically compete for comparative advantage by supporting industries where strong external IRS exist or could exist.27 During an industry’s formative years, or during transitional periods when technology shifts were changing patterns of trade, import barriers or export subsidies could allow a country to establish comparative advantage. The motivation would be that after having developed external IRS, the industry could be taxed and the gains distributed to the rest of the population. The proper definition of competitiveness in trade was thus thought to be the ability of a country to exhibit a comparative advantage in external IRS industries.28 The next three implications were loosely related to the first. The second was that comparative advantage could be established in suboptimal locations if accident or policy allowed external IRS to develop in a location that would not otherwise offer comparative advantage. Perhaps productivity in the watch industry, for example, would be higher if it had been established in France rather than Switzerland. Third, there could be global losses from trade if international competition caused external IRS industry clusters to form that prevented a country from developing an industry in which it might otherwise have obtained a comparative advantage. And fourth, the competition that arises as countries attempted to gain external IRS–based comparative advantage could have significant costs. It could cause countries to erect costly trade barriers in the hopes of developing an industry, and it could cause more industrial clusters to form than would be optimal.29 The assumption that firms could not “internalize” external economies of scale has been undermined by the advent of GVCs and is a point we will focus on in the next section, when we dismantle the case for industrial policy based on developments from the second unbundling. We will also discuss how GVCs oftentimes reverse the self-reinforcing cycle of agglomeration and external IRS and are reducing the extent to which external IRS are a determinant of trade.

New Trade Theory’s Insights for the Application of Industrial Policy. The fact that countries theoretically could compete for comparative advantage from external IRS led many commentators and politicians to say that the United States should.30 The modeling capacity that New Trade Theory adopted allowed for the separation of internal IRS from external IRS and allowed

HHassett.indbassett.indb 237237 111/8/121/8/12 9:279:27 PMPM 238 RETHINKING COMPETITIVENESS

economists to identify external IRS as the most intellectually sound reasons to support industrial policy in developed nations. Having a theoretical case to target industries with external IRS was not enough in order for industrial policy to be useful, though. Policymakers also needed a way to identify those industries that display external IRS. Michael Porter (1990), in The Competitive Advantage of Nations, and Paul Krugman, in a 1993 book chapter titled “The Current Case for Industrial Policy,” offered a method based on the agglomerative nature of external IRS industries. Firms in industries that exhibit labor pooling, input sharing, and knowledge spillovers tended to cluster geographically even when that clustering led to dispersion forces such as higher prices for office space and congestion. High external IRS industries, then, could be identified by the amount of clustering that occurred. The thinking was that policymakers need not rely on blind guesswork to decide which industries might offer external IRS; they could look at the real world and see the effects of external IRS, just as doctors often identify diseases by their symptoms. This thinking did not recognize the fact that GVCs were weakening the forces that led to clustering and would make external IRS industries harder to identify. The recommendations for policy, then, of this theorizing from the 1980s and 1990s were to subsidize and protect industries that exhibited external IRS and were complementary to a nation’s natural comparative advantage. Industries that had not formed yet in a country should be subsidized if they clustered in other countries where they had formed, or if similar industries tended to exhibit large amounts of external IRS. Indus- trial clusters that had formed should be subsidized when they were being threatened by other countries’ industrial policy or when major technological shifts were occurring. In reality, it was very difficult for nations to stick to this prescription; policymakers misunderstood the theory, understood the theory but exercised hubris, or pandered to special interests and used industrial policy as an excuse to seek votes or money. It is unclear whether the method of conducting industrial policy described above has or hasn’t been successful in the past,31 but its usefulness in the present is severely limited. In the next section, we describe how the second unbundling of trade and the rise of GVCs undermine both the theoretical case for industrial policy and the practicality of applying it.

HHassett.indbassett.indb 238238 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 239

The Second Great Unbundling It is important to understand what is causing the second unbundling of trade and how it has manifested itself, before discussing its implications for industrial policy. In this section, we will explain, generally, the causes and trade consequences of the second unbundling as well as how the changing nature of trade has undermined the theoretical case for industrial policy and the practical ability to conduct it. In the next section, we will offer case studies of the United States, East Asia, and China and describe how each of them fits into the “global value chains” that have arisen from the second unbundling of trade. These two contrasting examples, the United States and East Asia, portray the increasing complexity of trade in a globalized world and offer real-world examples of the concepts discussed in this section.

The Changing Nature of Trade. The first unbundling was characterized by the growth of inter-industry trade and horizontal intra-industry trade that could be analyzed largely at the industry level motivated by declining shipping costs and trade liberalization. The second unbundling has led to an explosion of vertical intra-industry trade and needs to be analyzed at the firm level. This new type of trade goes hand in hand with the development of global value chains.32 Industries no longer conduct every stage of production in the same country and have begun to “offshore”33 production activities, or “tasks,” to affiliates and foreign firms. These GVCs are networks of geographically dispersed firms that supply unique steps of production and interact to produce a final good. Figure 8-3, produced by the United States International Trade Commission (2011), illustrates some steps of production that might be distributed to separate countries. The commission’s report also provides a short story:

A domestic firm might provide the R&D, design of a product, and then produce the initial intermediate inputs using local raw materials [the head]. Then these intermediate inputs would be exported to a second country, where a firm would use them to produce a semifinished product. That firm would then export the semifinished good to a third country, where the final good is assembled and packaged. The third country would then export the good back to the domestic firm, which would oversee the

HHassett.indbassett.indb 239239 111/8/121/8/12 9:279:27 PMPM 240 RETHINKING COMPETITIVENESS

FIGURE 8-3 COMPONENTS OF A SUPPLY CHAIN

R&D (Design)

Financial Raw Materials Services

Quality Intermediate Inputs Control Services Semi-finished Good

Final Good Assembly and Packaging

Marketing Logistics Services Warehouse Services

Retail/Delivery

Customer Service

SOURCE: United States International Trade Commission (2011).

marketing, retailing, and delivery of the product domestically and abroad [the tail]. (pp. 32–33)

The ability of firms to conduct this complex and geographically dispersed production has hinged on several recent developments, most notably drastic improvements in communication and coordination technologies, a steep fall in rapid-transportation costs, and new management practices that have arisen to take advantage of these new capabilities. Although the cost of most shipping has not changed much since the 1980s, the cost of air transport began to fall rapidly after jet engines started to become widespread in the late 1950s. One measure of technological change, revenue to the airline for each ton-kilometer shipped, fell

HHassett.indbassett.indb 240240 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 241

92 percent from 1955 to 2004. Quality-adjusted air freight price indices showed a much smaller but still substantial drop-off.34 During this period, the U.S. air share of trade value increased from 8.1 percent of imports and 11.9 percent of exports in 1965 to 31.5 percent of imports and 52.8 percent of exports in 2004. Comparatively, ocean shipping contributes 99 percent of world trade by weight, but drastic price changes have not been present since the 1950s (Hummels 2007). During this same time period, although more loaded toward the present, there have been major advances in Internet and communication technologies. The use of faxes, cell phones, and, most important, the Internet has exploded in recent years. According to Baldwin (2011), there were 1,024 Internet hosts in the world in 1984, 6.6 million in 1995, and 106.8 million in 2000. This technology allowed for cheap exchange of data and many forms of instant communication. The advances in computing power during the same time period allowed for the adoption of data-management software and geographically separated group work. All these advancements made more complicated logistical operations possible. In particular, they allowed for the coordination of production facilities across long distances and for the adoption of more efficient production methods, such as “just in time” manufacturing, where inputs arrive at the next stage of production just in time for that production to take place. This contributes to a reduction of in-process inventory and, as a result, smaller firms can actively participate in GVCs. GVCs are a major factor in the growth of vertical intra-industry trade. Figure 8-2 shows how vertical intra-industry trade increased significantly as a percentage of overall trade between 1988 and 2002. An influential paper by Kei-Mu Yi (2003) at the Federal Reserve Bank of New York found that the growth in vertical intra-industry trade contributed to more than 50 percent of the growth in trade between 1962 and 1999. In other words, the second unbundling of globalization and the rise of GVCs have played as big a role or bigger than recent continuation of the first unbundling. In order to provide a better understanding of the way GVCs affect trade flows, in the next section we will provide an overview of the United States’ role in GVCs. We will then provide an overview of GVCs in East Asia, with a particular focus on China. Before moving to those case studies, though, we will discuss the policy implications of global value chains.

HHassett.indbassett.indb 241241 111/8/121/8/12 9:279:27 PMPM 242 RETHINKING COMPETITIVENESS

Gains from Trade in Tasks and the Motivations of Trade. In order to understand the policy implications of trade in tasks, we need to know what will happen in the simplest case, where there are no increasing returns to scale, and trade is based on comparative advantage deriving from factor endowments or technological differences. As it turns out, trade in tasks can be thought of in the same Ricardian framework that we used to model inter- industry trade, only with one trivial modification. In a previous section, we showed that countries were always able to consume more wine and cloth when trading allowed them to specialize in their comparative advantage industry; trading always increases the purchasing power of labor, so there were always gains from trade. To better model trade in tasks, we need to make just one modification, changing “cloth” to “oranges.” When trade is available, both countries will still be able to obtain more oranges and wine than they could otherwise. But now, instead of consuming the wine and oranges separately, the thirsty workers are going to cut up the oranges and mix them with the wine to make sangria. In other words, consumption in a two-good, two-country model is no different from production of a final good in a two-country, two-intermediate good, one-final good model. More sangria can be made and drunk with trade than without it. Trading tasks has the same effect on both countries as would technological progress, and gains from trading tasks exist in a fundamental setting!35 It is important to note the other major similarities between trade in tasks and trade in final goods. Largely, trade in tasks is motivated by technological or factor endowment differences within a comparative advantage framework, just as trade in final goods is. Trade in tasks when firms exhibit internal IRS will cause horizontal intra-industry trade where similar, but specialized, tasks are traded back and forth between countries. This is exactly the same framework as New Trade Theory presented for trade in final goods when internal IRS exist. The area where the rise of GVCs breaks down the old frameworks is in their treatment of external IRS industries. In fact, the existence of trade in tasks even undermines the way policymakers should think about trade in final goods when external IRS exist.

Trade in Tasks and External IRS. The increasing ability of firms to offshore tasks has several important implications for external IRS industries.

HHassett.indbassett.indb 242242 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 243

None of these completely eliminates a case for industrial policy, but each undermines it. The overarching story is that the value of clustering in external IRS industries will begin to decline relative to the gains that come from geographic dispersion. Essentially, dispersal forces will begin to outweigh agglomerative forces. First, the agglomerative forces are weaker for intermediate tasks than they are for the industries as a whole, and this will cause industrial clusters to get smaller and more polarized, with only the head and tail of the global value chains remaining. Moreover, the weaker agglomerative forces for intermediate tasks will have the knock-on effect of generally decreasing agglomeration in the industry. Likely, the first section of the industry to relocate will be the parts “subindustry,” which typically has few inputs to contribute to input sharing, almost never contributes to or takes advantage of knowledge spillovers, and does not depend on the home market for sales. Moreover, the parts subindustry likely exhibits internal IRS, which it would not have been able to fully exploit when it was geographically bundled with the other “subindustries,” but will be able to exhibit when it is moved to another location. Once the parts subindustry has moved, the intermediate components industry might follow it, because for it input pooling does matter, and parts are those inputs. Final assembly will likely stay in its original location due to the large home market, and R&D will likely stay due to the existing knowledge spillovers and labor pooling of high-skilled labor.36 Second, the industry tasks that do exhibit agglomerative forces will form separate subindustry clusters that are much smaller than external IRS industries have been in the past. As a result, it will be easier for a new firm to contemplate entering one of these subindustries at a large enough scale to have a non-negligible impact on its own productivity; in other words, it will be able to “internalize” the external economies of scale. This directly breaks one of the assumptions that are required for trade models to predict losses from trade when external IRS are present, as we discussed in the last section. The reason is that if a firm can realize external economies of scale by growing to be large enough, financing will be available for that firm from a well-developed financial market. If financial markets can finance firms or small industries while they develop external economies of scale, then there is no role for the government to do so. Obviously some subindustries will

HHassett.indbassett.indb 243243 111/8/121/8/12 9:279:27 PMPM 244 RETHINKING COMPETITIVENESS

still be too large for external economies of scale to be internalized by individual firms, but not all.37 Third, since agglomerative forces are weaker, and industrial clusters are less likely to form, it will be harder to tell when a new-to-the-world industry might exhibit external IRS if it grew large enough. The external IRS industries of the future may not exhibit strong enough external IRS to form in a cluster. This prevents policymakers from using the rule of thumb that we identified for actually putting industrial policy into practice: there won’t be as many clusters to find and to identify external IRS industries by. Fourth, nations may be tempted to enact industrial policy to keep industrial clusters from dispersing, but this would be a mistake. The point of a government’s subsidizing external IRS industries is to be able to gain future rents from them after they’ve grown large enough to exhibit hysteresis. If the natural tendency of a large cluster is to deterio- rate, government will be unable to ever recoup the cost of the subsidy or protection. Fifth, it may still be possible to identify some industries where strong external IRS continue to cause agglomeration. Given the experience of history, policymakers should be more wary about the potential for new innovations in coordination technology to break up these clusters. The value of gaining or keeping a cluster must be weighed against the probability that it might break up in the future. Sixth, given the first three points, it is unlikely that a target could be identified for favoritism that will cluster and exhibit strong external IRS in maturity. Even if such an industry is identified, protection will likely have to include limits to all offshoring during the industry’s development, making it even less competitive versus other established industries in other countries and thus even more expensive to protect. Each of these six points undermines the case for industrial policy, but most rely on the fact that global value chains are an important part of global trade. In particular, they rely on the empirical assertion that trade in parts and intermediate components will have increased due to the second unbundling of trade. The next sections support these assertions with data from the United States and East Asia. They also discuss several related issues for policy and the measurement of trade flows.

HHassett.indbassett.indb 244244 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 245

East Asian Fragmentation and Supply Chains: The Impact on Trade Data and Analysis

The world trade in parts and components increased in value from US$400 billion in 1992 to US$1,000 billion in 2003. Though developed economies account for much of this trade, the share of developing countries has increased rapidly in recent decades—with developing East Asia leading the way. The share of East Asia in total exports of components rose from 31 percent in 1992 to 43 percent in 2003 (Athukorala and Yamashita 2006). The rise masked several substantial intraregional changes. Japan steadily lost ground as a leader in components, dipping from 15 percent of total world trade down to 11 percent in 2003. Meanwhile, the share of developing East Asia (minus Japan) increased from 16 percent to 31 percent during these same years. A further factor has been the growing importance of China and Hong Kong (“Greater China”; for more details on the impact of China, see below). The share of Greater China in total world component exports increased from 6 percent in 1992 to 10 percent in 2003 (the imports increase was even greater, from 7 percent to 16 percent). Contrary to popular belief, this increase did not result in a “crowding out” of other East Asian economies. Indeed, a major driver in the expansion of trade in East Asia over the past decade has been international processing, with China functioning as a central manufacturing base relying on inputs imported from other Asian countries. Production sharing has thus provided the means for developing countries to reap the benefits of differences in comparative advantage and specialization. Networks initially linked Japanese companies vertically with Korea and Taiwan in low-skill assembly activities, which more recently have been transferred to lower-wage countries, such as Malaysia, the Philippines, and Thailand (Indonesia is a laggard in this regard). These networks are now also being transferred to Vietnam and the less-developed regions of China. These various production networks overlap and supply one another, thus creating an expansion of two-way trade in components (Haddad 2007; Ng and Yeats 1999, 2003). Table 8-1 presents a comparison of the share of components in total manufacturing exports and imports in 1992, 1996, and 2003. Within East Asia, those nations that belong to AFTA (ASEAN Free Trade Agreement),

HHassett.indbassett.indb 245245 111/8/121/8/12 9:279:27 PMPM 246 RETHINKING COMPETITIVENESS Growth of of P&C to Mfg. Exports Contribution Growth of P&C Exports RADE T Growth of Mfg. Exports ANUFACTURING M IN 8-1 ABLE T (P&C) (A) Exports Share of P&C in Mfg. Exports OMPONENTS C AND

ARTS P ($billion) Value of P&C 1992 1996 2003 1992 1996 2003 1992–2003 1992–2003 1992–2003 Country or Region East Asia Japan Developing East Asia China 70.4 169.7 136.8 Hong Kong SAR 330.1 275.8 66.5 Rep. of Korea Taiwan 17.5 447.5 106.1 14.1AFTA 28.5 117.4 26.7 19.2 27.9 3.6 Indonesia 11.4 82.9 21.2 27.9 28.0 62.4 Malaysia 9.8 131.2 26.7 30.2 27.9 12.8 Philippines 12.9 24.7 59.5 43.4 27.9 Singapore 22.4 4.4 26.7 0.6 35.0 Thailand 17.1 3.2 5.5 10.0 29.3 33.6 40.6 Vietnam 0.8 1.7 25.2 1.2 23.5 28.3 9.8 13.0 25.5 8.8 28.8 4.3 2.7 15.2 33.9 39.4 4.1 6.3 4.1 39.5 20.7 38.7 3.7 4.8 56.5 3.8 19.8 9.5 42.6 7.3 2.3 27.0 7.4 52.5 2.5 42.7 15.8 33.2 39.7 6.1 13.9 6.2 63.8 19.1 34.9 46.7 5.4 23.4 4.5 47.1 11.7 2.7 8.6 26.7 3.7 46.4 3.9 49.5 30.9 4.1 17.1 4.9 8.3 13.7 52.4 6.0 5.2 44.6 5.4 24.5 70.0 59.8 31.0

HHassett.indbassett.indb 246246 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 247 South Asia CER 0.3NAFTA 113.5 4.0 182.4 USA 225.5 Mexico 3.8 0.6MERCOSUR 2.5 25.3 13.2Andean Pact 1.0 27.2 3.8 18.9Europe 202.4 25.6 277.5 91.0EU 192.1 17.3 2.6 6.7 4.9 146.2 258.5 379.5Eastern Europe 13.1 2.6 13.1 170.6 337.5Rest of Europe 2.7 0.2 15.5 8.9 13.5 26.8World 446.7 28.0 15.9 4.9 16.2 0.2 729.4Developed countries 12.2 30.5 1.5 20.7 17.7 16.6 1047.8 4.0 375.9Developing countries 29.2 8.8 19.4 16.7 0.4 552.4 17.9 6.9 70.8 2.7 10.0 3.0 12.2 21.1 677.5 177.0 20.3 2.2 5.0 25.4 4.0 2.2 370.3 18.6 2.1 16.7 21.1 4.1 9.1 5.8 15.0 20.4 11.1 26.0 13.5 17.8 5.6 19.8 20.2 12.3 2.8 2.6 20.2 2.5 23.1 5.8 11.7 2.5 2.2 2.9 2.0 5.8 8.4 4.9 2.3 11.3 18.0 3.4 4.9 32.5 17.9 21.3 3.4 2.4 11.9 6.7 2.5 24.4 22.5 6.2 21.9 26.5 12.4

HHassett.indbassett.indb 247247 111/8/121/8/12 9:279:27 PMPM 248 RETHINKING COMPETITIVENESS Growth of of P&C to Contribution Mfg. Imports Imports Growth of P&C RADE T Imports Growth of Mfg. ) ANUFACTURING M IN continued 8-1 ( (P&C) (B) Imports Share of P&C ABLE in Mfg. Imports T OMPONENTS C AND

ARTS P ($billion) Value of P&C 1992 1996 2003 1992 1996 2003 1992–2003 1992–2003 1992–2003 Country or Region East Asia Japan Developing East Asia 90.1 China 218.0 104.1 Hong Kong SAR 387.4 249.4 14.1 Rep. of Korea 21.1 433.6 Taiwan 15.6 31.3AFTA+5 39.2 10.6 30.2 19.8 31.2 46.2 12.2 102.4 Indonesia 37.4 27.9 20.8 65.6 125.5 Malaysia 14.2 22.6 105.3 34.6 12.4 14.7 Philippines 28.2 19.3 35.5 17.6 3.6 Singapore 27.5 20.4 39.3 3.6 21.5 25.2 21.1 3.5 Thailand 31.4 55.5 11.0 47.1 27.4 1.9 34.3 6.7 16.9 27.1 3.1 16.0 33.6 10.9 2.7 35.0 3.1 2.6 5.9 36.5 50.3 6.6 19.1 6.8 37.3 18.5 35.2 5.8 3.1 49.6 24.8 20.9 23.8 47.5 30.0 43.6 4.8 17.2 2.8 18.5 55.7 48.8 5.8 42.8 4.7 63.1 24.7 9.5 45.4 49.2 –0.5 32.9 4.3 3.0 5.5 27.7 32.5 6.1 2.6 48.5 67.8 38.4 –0.5 2.6 40.7 4.9 9.4 57.0 4.6 18.5 3.7 74.4 76.5 70.8 41.0

HHassett.indbassett.indb 248248 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 249 : Athukorala and Yamashita (2006); data compiled from UN Comtrade database. OURCES South Asia CER 1.1NAFTA 104.2 9.0 183.9 USA 232.7 10.0 Mexico 3.0MERCOSUR 4.7 18.9 14.6Andean Pact 3.8 23.6 11.0 15.2Europe 190.3 17.7 71.9 10.5 12.1 6.8 267.0EU 177.3 8.0 126.0 17.9 363.4 242.0 14.1Eastern Europe 150.3 20.7 10.1 2.2 315.9 3.5Rest of Europe 17.2 15.0 14.6 17.5 41.1World 408.2 15.3 21.5 16.6 12.6 2.0 21.7 18.7 705.8 1.9Developed countries 18.9 1044.3 17.4 299.0 11.1Developing countries 15.4 30.6 2.1 10.1 109.2 17.6 462.4 2.5 16.8 3.8 9.3 3.2 14.9 28.7 243.4S 12.6 582.5 28.8 2.0 19.8 461.8 18.7 3.5 16.1 1.7 9.7 20.7 9.2 19.5 12.6 4.9 18.4 14.3 9.1 11.9 23.0 12.5 3.2 16.9 3.3 17.0 2.9 20.3 28.6 12.3 2.6 3.0 1.1 2.3 2.4 6.7 8.0 4.3 2.2 25.6 11.6 3.8 21.1 13.8 21.8 -0.2 2.7 33.0 11.4 5.9 2.1 24.4 -1.9 18.2 22.2 33.5 11.9

HHassett.indbassett.indb 249249 111/8/121/8/12 9:279:27 PMPM 250 RETHINKING COMPETITIVENESS

particularly Malaysia, the Philippines, Singapore, and Thailand, show a heavy dependence on production sharing for trade expansion. By 2003, parts and components made up 40 percent of total manufacturing exports in AFTA, up from 24 percent in 1992. Even for the more advanced East Asian developing economies such as Taiwan and Korea, parts and components exports increased during this same time period, belying the belief that these nations had shifted in large measure from component- to final- goods production. It is also demonstrable from table 8-1 that developing East Asia is more heavily dependent on component trade for trade expansion. Much of the steep rise in manufacturing exports has been driven by vertical specialization. In 2003, components accounted for 28 percent of total manufacturing exports in developing East Asia, compared with 17 percent for the European Union and 25 percent for North America. Figure 8-4 provides a picture of vertical specialization by country in 2003. A large share of exports from Malaysia, the Philippines, Thailand, and Singapore are accounted for by vertical specialization, with only Indonesia lagging behind in intra- industry export growth (China was just at the outset of the steep rise in intra-industry trade). Recent research by Koopman et al. 2010 has provided the conceptual framework to produce a more comprehensive picture of the domestic and foreign value-added content of exports and imports in given countries. These researchers show that countries in developing Asian economies perform different roles in global supply chains than developing countries in other regions. Most particularly, their exports pass through more borders than those from developing countries in other regions. Most of these countries, including China, send their exports to countries that perform either additional tasks or final assembly for consumers in other countries. Thus, in most cases, Asian supply chains have “relatively dispersed sourcing of imported intermediate inputs” (p. 1; see also Hummels, Ishii, and Yi 2001) The World Bank’s Mona Haddad (2007) describes the evolving Asian production networks:

The traditional production network is changing. It is no longer the simple model whereby Japan and the NIEs [newly industrializing economies] supply high-quality components and capital goods to

HHassett.indbassett.indb 250250 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 251

FIGURE 8-4 SHARE OF PARTS AND COMPONENTS IN TOTAL EXPORTS

World Eastern Europe EU MERCOSUR Mexico USA NAFTA South Asia

Thailand Singapore Philippines Malaysia Indonesia AFTA Taiwan Korea Hong Kong China Japan

0 1020304050607080

SOURCE: Haddad (2007); Athukorala and Yamashita (2006).

emerging East Asia, including China, which assembles them for the final markets in the European Union and the United States. Another sort of production network involving the transshipment of components is also appearing in the region. In this more sophisticated and complex network, Japan and the NIEs provide high-quality materials, including design, to their FDI [foreign direct investment] affiliates in developing East Asia, which uses them to produce components. The components are then sent back to the originators for further processing. The originators perform quality control, organize the components, and send them back to developing East Asia as kits for final Assembly. (p. 13).

See figure 8-5 for an illustration of this network.

HHassett.indbassett.indb 251251 111/8/121/8/12 9:279:27 PMPM 252 RETHINKING COMPETITIVENESS

FIGURE 8-5 COMPLEX NETWORK FOR HARD DISK DRIVE

USA Disk, Head, Suspension Mexico Head

China PCBA, Carriage HGA, Base, Head Suspension Japan Cover, Disk, Screw, Seal, Ramp, Top Hong Thailand Clamp, Latch, Plate Kong Taiwan Case, Label, Filter Spindle Motor, Base, Filter Cap Top PCBA, Suspension Carriage, Flex Cable, Clamp Pivot, Seal, VCM, Top Cover, PCBA, HGA, HAS

Philippines Damping Plate, Coil Support, Malaysia PCBA Base, Pivot, Spacer, VCM, Base Card, Top Clamp, Disk

Singapore Cover, Screw, Pivot, PC ADP, Indonesia Disc Suspension, VCM, PCBA

NOTE: The figure shows the sourcing of the parts for hard drives assembled in Thailand; after assembly, the drives are shipped to various markets to be used in electronic products. It should also be noted that the hard disk drive itself is an intermediate part, which will be shipped somewhere else and assembled—together with an equally complex chain of sourced parts—into a final electrical product.

SOURCES: Haddad (2007); Baldwin (2006).

China and Regional and Global Supply Chains: Myth and Reality. China’s rise to the top global position has been the result of a combination of technological and economic factors—abetted strongly by national policies. As this chapter has set forth above, the dramatic decrease in transportation costs and the equally dramatic impact of the communications revolution acted as spurs to global trade and investment flows. And in China’s case, geographical location was an important asset. China’s relative

HHassett.indbassett.indb 252252 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 253

proximity to East Asian parts and components firms enhanced its attractiveness as an export platform (Ma, Van Assche, and Hong 2009). Finally, though often condemned in political circles for pursuing mercantilist trade policies, in many ways China does not fit a traditional mercantilist model. The most important difference stems from the estab- lishment from the outset of the reform period in the early 1980s of a policy of open investment—and indeed a policy that often favored foreign firms over domestic firms. As part of this policy to attract foreign investments and companies, China granted duty exemptions for imported raw materials and other inputs (parts and components) as long as they were used solely for exports. It also gave tax breaks to foreign firms locating in its territory. The upshot, as this chapter will detail below, is that China was able to take advantage of increasing vertical specialization led by multinational corporations from the United States, Japan, the European Union, and increasingly from the newly emerging economies (Korea, Taiwan, and Singapore) to emerge as the world’s top trading nation, with annual export growth for the past twenty years averaging 19 percent, twice the rate of growth of world exports.

Dual Trade Regime. Within these gross numbers, however, there is the fact of a dual export regime: exports as a result of processing trade and far fewer exports unrelated to processing trade. The rise of China’s processing trade regime has been chronicled and analyzed by economists for some years (see, for instance, Amiti and Freund 2008; Branstetter and Lardy 2006; Dean and Lovely 2008). Recently, new data sets and models have allowed a more detailed look at the internal dynamics of Chinese processing and nonprocessing trade (Koopman, Wang, and Wei 2008; Ma, Van Assche, and Hong 2009). As shown in figure 8-6, the share of processing exports in China’s total trade has risen from 30 percent in 1988 to 51 percent in 2007. For imports, the numbers are 27 percent and 38 percent, respectively. Further research by Koopman, Wang, and Wei (2008) has demonstrated the great differences in the content, source, and technological differentiation of processing versus nonprocessing imports and exports. As depicted in table 8-2, they found that in 2006 the domestic content of processing exports was about 18 percent, leaving by implication about

HHassett.indbassett.indb 253253 111/8/121/8/12 9:279:27 PMPM 254 RETHINKING COMPETITIVENESS

FIGURE 8-6 PROPORTION OF PROCESSING TRADE IN CHINA’S TOTAL TRADE, 1988–2007

50 Export

40 Import

30 Percentage 20

0 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006

SOURCE: Ma, Van Assche, and Hong (2009) using China’s Customs Statistics data.

TABLE 8-2 DOMESTIC AND FOREIGN VALUES ADDED: NORMAL VS. PROCESSING EXPORTS (IN PERCENT OF TOTAL EXPORTS)

Normal Exports Processing Exports 1997 2002 2006 1997 2002 2006 All Merchandise Total Foreign value-added 5.3 10.8 11.3 81.9 74.3 81.9 Direct foreign value-added 1.9 4.5 4.6 81.7 72.5 80.9 Total Domestic Value-added 94.7 89.2 88.7 18.1 25.7 18.1 Direct domestic value-added 34.4 31.0 29.3 15.0 11.4 10.5 Manufacturing Goods Only Total Foreign value-added 5.7 11.6 11.7 82.3 74.9 82.3 Direct foreign value-added 2.1 4.9 4.8 82.2 73.0 81.4 Total Domestic Value-added 94.3 88.4 88.3 17.7 25.1 17.7 Direct domestic value-added 30.9 28.5 28.3 15.0 9.5 10.4

NOTE: The estimates for 2006 are preliminary, as they use the trade statistics in 2006, but the I/O table is for 2002, which is the latest available. The next benchmark I/O table—the 2007 table—is scheduled to be released in 2010.

SOURCES: Koopman et al. (2008); authors’ estimates.

HHassett.indbassett.indb 254254 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 255

TABLE 8-3 SHARE OF CHINA’S PROCESSING IMPORTS BY COUNTRY OF ORIGIN, 2007 (%)

Adjusted for Hong Kong Transshipments Unadjusted 1997 2002 2007 2007 East Asia 68.8 73.3 76.1 86.6 Hong Kong — — — 47.1 Japan 26.9 26.5 23.7 10.6 South Korea 15.02 14.1 15.7 10.8 Singapore 3.2 3.4 4.3 2.9 Taiwan 16.9 19.0 20.3 9.6 Malaysia 2.2 3.9 4.5 1.5 Thailand 2.0 2.8 2.8 1.3 Philippines 0.2 1.7 3.5 2.1 Vietnam 0.2 0.1 0.2 0.1 Indonesia 1.8 1.3 0.9 0.4 Macau 0.4 0.6 0.3 0.2 Non-Asian OECD 23.8 21.8 18.1 9.3 United States 10.4 9.1 7.7 3.9 EU-19 9.0 9.8 7.9 4.1 Canada 0.7 0.5 0.8 0.5 Australia 2.7 1.3 0.8 0.4 Other OECD 1.0 1.1 1.0 0.4 Rest of the world 7.3 4.9 5.8 4.1

SOURCE: Ma, Van Assche, and Hong (2009) using China’s Customs Statistics data.

82 percent of the value added to foreign content. In contrast, the domestic content of nonprocessing exports was about 89 percent, thus leaving only 11 percent accounted for by foreign inputs. In addition, foreign firms pre- dominate in the processing sector: the share of foreign-invested enterprises (FIEs) has evolved over the past two decades, ranging from a high of about 90 percent in 1995 to a low of 75 percent in 2007 (compared with FIEs’ share of nonprocessing exports remaining below 30 percent). Greater detail about the structure of the global production networks into which China is integrated can be obtained after adjusting for transshipments through Hong Kong. Table 8-3 shows that on the import side China

HHassett.indbassett.indb 255255 111/8/121/8/12 9:279:27 PMPM 256 RETHINKING COMPETITIVENESS

TABLE 8-4 SHARE OF CHINA’S EXPORTS BY DESTINATION COUNTRY, 2007 (%)

Adjusted for Hong Kong Transshipments Unadjusted 1997 2002 2007 2007 East Asia 36.0 33.4 29.2 51.4 Hong Kong — — — 32.8 Japan 18.6 15.9 11.4 7.9 South Korea 5.0 4.8 5.0 3.7 Singapore 3.6 3.6 3.7 2.3 Taiwan 2.4 2.3 2.6 1.5 Malaysia 1.7 2.1 2.0 1.4 Thailand 1.3 1.5 1.3 0.6 Philippines 1.3 1.3 1.1 0.4 Vietnam 0.5 0.5 0.6 0.3 Indonesia 0.9 0.7 0.7 0.4 Macau 0.7 0.6 0.7 0.2 Non-Asian OECD 54.7 59.9 61.8 42.0 United States 28.9 32.4 28.8 20.1 EU-19 20.1 22.1 27.2 18.1 Canada 1.8 1.8 1.8 1.1 Australia 1.7 1.6 1.7 1.1 Other OECD 2.1 2.0 2.4 1.6 Rest of the world 9.4 6.7 9.0 6.6

SOURCE: Ma, Van Assche, and Hong (2009); China’s Customs Statistics data.

heavily sourced inputs for processing from East Asian countries. In 2007, these economies accounted for about 76 percent of processing imports (in contrast, the European Union and United States accounted for just under 16 percent of such imports). On the exports side (table 8-4) we see a very different pattern: in 2007, about 62 percent of processing exports went to non-Asian developed countries, while the share of such exports to Asian economies stood at only 29 percent (Ma, Van Assche, and Hong 2009). Taking a different cut with earlier data, Gaulier, Lemoine, and Unal-Kesenci (2005, 2011) found that in 2002, 60 percent of Chinese imports from the newly emerging economies (in this case adding Hong Kong to the data from

HHassett.indbassett.indb 256256 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 257

Korea, Taiwan, and Singapore) were inputs for processing industries, and for Japan 40 percent of imports were for processing.

China, the United States, and High-Technology Trade. In the political arena, China’s steeply rising trade surplus with the United States after 2001, and, most disturbingly for many, China’s increasing surplus in the category of advanced-technology products (ATP), has been taken as a sign of U.S. decline and China’s emergence as a technological giant (see Ferrantino et al. 2010). A closer look at the underlying evidence from supply chain realities, however, reveals a much different—and much less alarming—picture. Certainly, the apparent deterioration of the U.S. ATP trade balance with the world—and with the People’s Republic of China—has been evident over the past decade. The total U.S. advanced-technology trade deficit with the world stood at about $65 billion in 2010, while the deficit with China reached some $75 billion. The apparent steep increase in Chinese advanced-technology exports to the United States and to the world initially produced tremendous anxiety in developed countries. Some analysts noted that China’s export specialization, as revealed by its export composition, had become that of an advanced technological power (Preeg 2004; Schott 2008). Schott, for instance, describes how China’s export bundle increasingly overlaps with that of more developed countries, rendering it more sophisticated than countries with similar endowments. Others argued that Chinese government–sponsored industrial policies had allowed Chinese companies to leapfrog ahead technologically in a way that posed major challenges to U.S. and other developed country high-tech competitiveness in global trade (Rodrik 2006). Recent and more detailed analysis of the internal composition (and ultimate origin) of Chinese advanced-technology exports reveals a very different picture, one much more in line with traditional economic theory. In the first place, as noted above, foreign firms dominate in the Chinese export sectors. This dominance becomes even more complete for ATP products. As table 8-5 shows, in 2006, 95.8 percent of Chinese ATP exports to the United States were from export processing sectors. The bilateral Chinese ATP surplus took off after 2002, but at the same time, the processing surplus was behind this increase—and indeed is large enough to account for almost the entire Chinese ATP surplus with the United States. Furthermore,

HHassett.indbassett.indb 257257 111/8/121/8/12 9:279:27 PMPM 258 RETHINKING COMPETITIVENESS

TABLE 8-5 CHINESE ATP EXPORTS TO THE UNITED STATES BY TRADE REGIME, 1996–2006

Exports (%) Year Processing exports Normal exports Other exports 1996 92.9 3.5 3.6 1997 93.2 3.5 3.4 1998 92.7 3.4 3.9 1999 92.1 4.4 3.5 2000 93.4 5.0 1.6 2001 94.6 3.8 1.7 2002 95.8 2.5 1.8 2003 96.5 1.9 1.6 2004 96.4 1.7 1.9 2005 96.6 1.7 1.7 2006 95.8 2.1 2.1

SOURCE: Ferrantino et al. (2010) using China Customs data.

as figure 8-7 illustrates, foreign firms dominate ATP trade. About 85 percent of the ATP trade surplus stems from FIEs, with much of the remaining share going to joint ventures between Chinese and foreign firms. There are two important, linked factors behind these numbers: first, as described throughout this chapter, the technological and economic imperatives behind the fragmentation of production and, second, Chinese policies that have systematically (at least until recently) favored foreign direct investment. In concluding this section, we should note that China may well be rethinking the fundamentals of its open investment policy, with active incentives to integrate its economy into the global value chains. Recent research and solid Chinese experience raise questions as to whether one central goal of Chinese investment and trade policies has been achieved: that is, a process of “catching up” with developed economies and a more rapid ascent up the technological ladder. For instance, using detailed Chi- nese export data, Blonigen and Ma (2007) conclude, “The general pattern over our time period, 1997–2005, run exactly counter to what one would expect if Chinese firms were catching up—foreign firms’ share of exports

HHassett.indbassett.indb 258258 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 259

FIGURE 8-7 THE DOMINANT ROLE OF FOREIGN-INVESTED ENTERPRISES IN CHINA’S ATP SURPLUS

10 Joint Venture 5 Wholly Foreign Collective Billion U.S. Dollars Private 0

–5 State-owned

–10 1996 1990 1992 1994 1996 1998 2000 2002 2004 2006

SOURCES: Ferrantino et al. (2010); China Customs Statistics; U.S. Census ATP definition.

by product category and foreign values relative to Chinese unit values are increasing over time, not decreasing” (p. 31). And writing in 2010, Ferran- tino et al. express doubt that policies to enhance and give favors to foreign firms in the processing trade “have accelerated China’s economic growth.” They continue: “At the very least, the large role of FIEs in China’s ATP exports suggests that relatively little progress has been made so far toward China’s goal of promoting growth through indigenous innovation by Chi- nese firms. One would expect to observe the results of successful innovations in the form of ATP exports by domestic enterprises, either private or state-owned, but we do not observe this.” All of this suggests, they state, that “Chinese authorities have begun to appreciate the limits of an economic growth path driven by policy-induced high-tech exports” (p. 223). For the past several years, Western firms and governments have become increasingly concerned that Beijing was turning away from the open- investment, relatively free trade policies that had characterized Chinese

HHassett.indbassett.indb 259259 111/8/121/8/12 9:279:27 PMPM 260 RETHINKING COMPETITIVENESS

international economic policy since the early 1990s. If that is the case—and the stepped-up efforts to foster “indigenous innovation,” combined with determined efforts to level the playing field between native Chinese firms and FIEs (e.g., equalization of tax policies, subsidies for native product development, pushing technology transfer and intellectual property sharing), render this change quite plausible—then it may well be that a kind of “golden age” for foreign corporations and free trade and investment polices is coming to an end. The technological and business model imperatives that have created global value chains and supply fragmentation over the past several decades ultimately cannot be reversed. But Beijing is certainly capable of throwing spanners into the spinning wheel of global competition, even at the cost of greatly increased trade and investment friction.

The United States and Global Supply Chains As an advanced industrial nation with highly competitive service sectors, the United States or, more precisely, U.S. firms participate on many levels in global supply chains. First, there are U.S. firms that lead production supply chains in such sectors as electronics, semiconductors, computers, chemicals, and motor vehicles and parts (see U.S. International Trade Commission 2011). Second, particularly in the apparel and textiles sectors, as well as in consumer goods and durables, large retailers and branded marketers lead what have been labeled “buyer-driven” chains (Gereffi and Frederick 2010). These latter chains consist of standardized, generally low-tech goods that do not require either highly sophisticated capital equipment or skilled labor. In both global production and buyer-driven chains, U.S. multinationals have increasingly performed the role of coordinating networks. As one recent paper concluded: “The advent of global supply chains has required major U.S. companies to revise their business models to become network facilitators, with successful firms acting as coordinators of capabilities among multiple strategic allies” (U.S. International Trade Commission 2011, p. 324). Finally, the increasing complexity of global supply chains has generated subtasks in the provision of business services—accounting, financial, legal, communications, and information technology—that today provide a separate set of roles for U.S. service firms. While there has been a focus on the potential for offshoring service tasks—payroll processing and call

HHassett.indbassett.indb 260260 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 261

centers, for example—recent research has also pointed out that many services require a skilled, professional workforce and that the United States retains a comparative advantage in these technical services. A recent report stated: “The United States is a net exporter of services and has a comparative advantage in many services sectors. U.S. workers in many tradable services sectors have more education and higher skills that in the lower-paying nontradable services sectors. These facts imply that the United States may gain good jobs in tradable services as services exports grow and become more integrated into global supply chains” (U.S. International Trade Commission 2011, p. 314). In 2009, the United States had a trade surplus of $149 billion in services, up from $69 billion in 2003. During that same time frame, U.S. exports of business, professional, and technical services grew by 85 percent. In the variety of forms described above, the U.S. exchanges of value in global chains occur most frequently with Canada, Mexico, the European Union, Japan, and China. Foreign direct investment by U.S. multinationals has also contributed to the generation of other global supply chains in East Asia, China and Mexico, and elsewhere.

U.S.-linked Production Networks. Supply chains in high-end products—electronics, semiconductors, computers, pharmaceuticals, among others—are led by a small number of highly integrated multinationals, typically based in the United States, Japan, or Europe. These lead firms concentrate on R&D, novel intellectual property, design, coordination of complex production processes, and market knowledge. As noted above, the actual production process is often fragmented, with the work carried out in different countries and with components and semifinished products passing back and forth across borders. As an example for the United States, one oft-cited supply chain revolves around the Apple iPod. There are 451 identified parts and components packed into the iPod, with Apple’s large profit stemming from the conception and design of the product, coordination of the production process, and final distribution and marketing. Research by three economists at the University of California, Irvine—Greg Linden, Kenneth L. Kraemer, and Jason Dedrick (2007)—has gone some way toward accounting for who actually makes the most important components and where they are manufactured (table 8-6 lists the top ten inputs). Summarizing findings

HHassett.indbassett.indb 261261 111/8/121/8/12 9:279:27 PMPM 262 RETHINKING COMPETITIVENESS

TABLE 8-6 MOST EXPENSIVE INPUTS IN THE 30GB THIRD-GENERATION IPOD, 2005

Cost as Company Manu- Estimated % of Est’d HQ facturing Factory All iPod Value Component Supplier Location Location Price Parts Capture Hard drive Toshiba Japan China $73.39 51% $19.45 Display module Toshiba- Japan Japan $20.39 14% $5.85 Matsushita Video/multimedia Broadcom US Taiwan or $8.36 6% $4.39 processor Singapore Portal player CPU PortalPlayer US US or $4.94 3% $2.21 Taiwan Insertion, test, and Inventec Taiwan China $3.70 3% $0.11 assembly Battery pack Unknown $2.89 2% $0.00 Display driver Renesas Japan Japan $2.88 2% $0.69 Mobile SDRAM Samsung Korea Korea $2.37 2% $0.67 memory - 32 MB Back enclosure Unknown $2.30 2% Mainboard PCB Unknown $1.90 1% Subtotal for 10 most $123.12 85% $33.37 expensive inputs All other inputs $21.28 15% Total all iPod inputs $144.40 100%

SOURCE: Linden et al. (2007).

from their paper, $163 of the $299 retail value of the 30-gigabyte video iPod was captured by Apple and a few much smaller U.S. companies. For the production costs alone, Apple retained about $80 of the total, with much of the rest captured by Japanese companies (total productions inputs were valued at about $144). Table 8-6 itself represents only an aggregate picture, behind which are many additional networks and supply chains. As we saw in figure 8-5, the production of hard drives is the result of a highly complex value chain that spans much of East Asia and includes inputs from the United States. Furthermore, Portal Player’s controller chip is likely manufactured by a Taiwanese company, as are components of Broadcom’s video processor. Finally, though not shown in the tables, the iPod is assembled by Taiwanese firms in mainland China (U.S. International Trade Commission 2011).

HHassett.indbassett.indb 262262 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 263

FIGURE 8-8 VALUE ADDED PERCENTAGES FOR IPHONE 4

Cost of inputs: China labor Cost of inputs: Non-China labor 1.8% 3.5%

Cost of inputs: materials 21.9%

Apple profits 58.5% Unidentified profits 5.3% S. Korea profits 4.7% Japan profits 0.5% Taiwan profits 0.5% EU profits 1.1% Non-Apple U.S. profits 2.4%

SOURCE: Linden, Kraemer, and Dedrick (2011).

Recently, Linden, Kraemer, and Dedrick (2011) updated their research, using the iPhone 4 as the basis for their calculations. As figure 8-8 illustrates, Apple still captures a very large portion of value added: 58.5 percent, with South Korea (4.7 percent), the European Union (1.1 percent), Taiwan (.5 percent), and Japan (.5 percent) following in order. Once again, although the iPhone is assembled in China, in factories owned by Taiwanese firms, Chinese value added is negligible. China is only represented in labor costs of about 1.8 percent.

Buyer-Driven Chains. Apparel represents a labor-intensive sector where a national supply chain moved internally from New England to the South and recently became global in its span and reach. In general, three types of U.S. lead firms are represented in buyer-driven supply chains: retailers and mass merchants, such as Walmart, Target, Sears, Gap, and American Eagle; brand marketers, such as Levi’s and Polo; and brand manufacturers, such as Hanesbrands and Fruit of the Loom (Gereffi and Frederick 2010).

HHassett.indbassett.indb 263263 111/8/121/8/12 9:279:27 PMPM 264 RETHINKING COMPETITIVENESS

FIGURE 8-9 APPAREL GLOBAL SUPPLY-CHAIN PROCESS

Research & Design Production Logistics Marketing Services Value-Adding Development Activities

Inputs Textiles Final Products Distribution & Sales

SOURCE: Gereffi and Frederick (2010).

Figure 8-9 lays out the steps in the apparel supply-chain process. From the outset, U.S. lead firms specialized in the front- and back-end activities: R&D, design, marketing, and sales. (R&D is often outsourced to other U.S. manufacturing firms for the development of new fibers, coatings, and fabric finishes, though the apparel firm remains heavily involved in the process.) Actual production is entirely outsourced, and logistics is split between U.S. and foreign firms. As the apparel supply chains have evolved in recent years, various permutations of tasks have proliferated. Originally, apparel firms attempted to control most supply-chain activities, from design, to selection of raw materials and fabrics, to ownership of foreign production or directly contracted production. Today, there are gradations of outsourc- ing, but at the farthest end, the lead firm (Walmart or Liz Claiborne, for example) outsources most of the steps in the supply chain, including design, production, and logistics, to a “full-package” coordinator. For instance, Hong Kong–based Li & Fung provides virtually all services to apparel firms, including product design and development, raw and fabric material sourcing, production, and logistics. The firm, and others like it, have developed a network of fabric suppliers and manufacturers around the world and thus can provide the resources for just-in-time orders in wide-ranging quantities and differing national consumer tastes (Gereffi and Frederick 2010).

Comparing U.S. and East Asian Supply Chains. On the policy front, there has been a hot debate over the economic and social consequences of the new trade paradigm. Critics have argued that U.S. multinationals

HHassett.indbassett.indb 264264 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 265

FIGURE 8-10 SHARE OF U.S. MULTINATIONAL PARENTS IN EMPLOYMENT, OUTPUT, CAPITAL INVESTMENT, AND R&D

90 80 70 60 50 40

Parent Shares 30 20 10 0 Employment Output Capital Research and Investment Development

SOURCE: Slaughter (2010).

are “abandoning” the home country and moving technology and jobs offshore through foreign operations and participation in global supply chains. Actually, the reverse is true: through increased firm-level competiveness enhanced by the key U.S. role in global supply chains, U.S productivity and living standards are higher today than they would have been otherwise. And of equal importance, though the new trade paradigm is still evolving, to this point U.S. multinationals are overwhelmingly “American.” As figure 8-10 demonstrates, U.S. parent corporations’ domestic operations account for about two-thirds of all worldwide U.S. multinational employment, about 70 percent of worldwide output, just over 70 percent of capital investment, and 85 percent of research and development (Slaughter 2010). With regard to the trade consequences, it is important to note that most sales of foreign affiliates do not come back to the United States. In 2007, U.S. foreign affiliates had total sales of $4.7 billion, of which only $500 million came back to the U.S. as imports (Slaughter 2010).

HHassett.indbassett.indb 265265 111/8/121/8/12 9:279:27 PMPM 266 RETHINKING COMPETITIVENESS , 2004 (%) ECTORS S Latin America Europe Other Total ELECTED , S MPORTS U.S. I IN East Asia Canada Mexico 8-7

ABLE DDED T A ALUE V OF

OURCES U.S. S Returned China Japan EGIONAL R OR

OUNTRY C : USITC (2011); commission estimates. OURCES Sector S TotalSelected Sectors ApparelChemicals, rubber, and plasticsMotor vehicles and parts 6.3Electronic equipmentMachinery and equipment 5.0Business services 19.1 9.7 11.3 8.3 2.5 8.6 8.7 10.1 11.0 23.0 7.7 14.4 12.0 17.2 11.2 10.4 7.2 19.0 1.5 2.5 9.7 2.4 12.0 16.0 29.6 1.3 27.8 11.0 3.6 6.9 3.8 2.4 42.8 6.2 2.4 4.9 4.7 1.9 9.3 12.7 9.4 2.0 6.3 23.1 2.9 1.3 100.0 8.8 10.4 26.1 32.1 3.4 11.4 11.4 13.2 0.2 100.0 5.1 3.9 100.0 21.4 100.0 2.7 100.0 100.0 55.5 11.3 100.0

HHassett.indbassett.indb 266266 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 267 , 2004 (%) ECTORS S Latin America Europe Others Total ELECTED , S XPORTS U.S. E IN East 8-8 Asia Canada Mexico

ABLE DDED T A ALUE V OF

OURCES S EGIONAL R OR

OUNTRY C : USITC (2011); commission estimates. OURCES SectorTotalSelected sectors ApparelChemicals, rubber, and plasticsMotor vehicles and parts 85.5Electronic equipmentMachinery and equipment 0.5Business services U.S. 81.5 1.0 ChinaS 87.1 89.4 Japan 1.3 76.9 1.1 88.5 0.8 1.0 3.0 2.7 2.2 0.7 1.3 1.2 95.6 1.9 3.7 0.7 0.8 1.5 1.3 0.2 3.0 5.1 1.5 1.5 1.7 1.3 0.6 1.3 1.6 4.4 1.3 0.9 0.8 0.7 1.3 2.2 3.0 0.6 1.1 0.8 100.0 0.5 4.7 1.0 0.7 3.3 2.8 0.2 2.0 4.7 100.0 3.3 2.1 1.4 0.2 2.0 100.0 100.0 100.0 2.5 1.4 100.0 0.6 100.0

HHassett.indbassett.indb 267267 111/8/121/8/12 9:279:27 PMPM 268 RETHINKING COMPETITIVENESS

Observations and Conclusions for Policy

One strong theme developed throughout this chapter was that the increasing complexity of global supply chains has wreaked havoc with traditional trade accounting by national governments. In general, national statistics treat flows of trade and investment as if they were arms-length transactions of end products across national borders. The reality, however, is far different as intermediate parts and components cross and recross borders, and U.S. and other multinationals outsource both to affiliates and through independent contractors. National trade balances, particularly in the United States, which has run substantial trade deficits for many years, have become highly charged political issues. But as we have noted, many nations add value to intermediate products as they pass through national borders, and attributing the entire value to the last exporting country provides an incorrect picture of the value of U.S. (and other nations’) trade. This is particularly true with regard to China, which has specialized as the final assembler of a large number of global value chains and utilizes components from East Asia and around the world in the production of its exports. Economists at the U.S International Trade Commission (2011) have estimated that the U.S.-China trade deficit calculated on a value-added basis is 40 percent smaller (2004 numbers) than the common official basis for gross trade. Moreover, the concern in the United States over whether another country’s firms might reside at the top of global value chains is misplaced, for it implies that the United States is involved in a competition for valuable industries. Another theme this chapter discusses is that it is nearly impossible for countries to gain from attempting to compete for industries, and most often they, and the rest of the world, will lose if they try to do so. The developments in trade theory from the 1980s and 1990s stating that a nation might gain did not account for a world with global value chains. It is ironic, then, that the concerns over China’s alleged rapid climb up the technology ladder may not even reflect reality. And there are dangers of a different kind that could lead to greater trade friction in the future. If indeed Chinese officials conclude that the three-decade-long policies of open investment and lowering trade barriers have failed to advance Chinese

HHassett.indbassett.indb 268268 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 269

“competitiveness,” then a shift to so-called indigenous innovation, implemented through subsidy, protection, and a fruitless search for spillover technologies, will certainly bring China into conflict with the United States and other members of the World Trade Organization. Beyond concerns over China, there are additional lessons in trade policy for the United States and other major trading nations. Important among these is the fact that some traditional trade policy instruments, such as antidumping and countervailing duties, can have wide and quite negative economic consequences. Potential disruption can occur both upstream and downstream in global supply chains—and thus are not confined to the target nation or industry. As an example, posit an antidumping duty applied to steel imported to produce parts for engines manufactured in a second country and installed in a third country. The duty would protect the steel producers of the first country but raise the costs of both the engine manufacturers in the second country and the assemblers in the third country, ultimately making the finished auto less price competitive. This is a simple example, but the effect would be amplified if the supply chain included other steel-related parts. Finally, the action might come back to haunt the first nation where the duty was collected, in that over time there would be a strong inducement to eliminate participation by firms in countries where trade-remedy actions are frequent. On a more positive note, it was undoubtedly reverse action—that is, East Asian nations unilaterally lowering their duties and other barriers— that was one important driver behind the explosion of supply chains in the region over the past two decades. Local firms were aware of the negative effects of keeping or raising barriers and lobbied hard against such action. There is one final lesson that represents policy continuity with more traditional trade of the nineteenth and twentieth centuries. That is, trade policy per se is not the source of a nation’s competitiveness. As this chapter has shown, the United States’ comparative advantage places it at the head of the global supply chains in which it participates. To retain that place and to achieve future prosperity, the nation must look to its internal policies. This means —among other things—continued high investments in basic research, both public and private; it means shoring up a weak education system in order to produce an ever-higher-skilled workforce; it means adopting and enforcing regulations that bind but do not undercut

HHassett.indbassett.indb 269269 111/8/121/8/12 9:279:27 PMPM 270 RETHINKING COMPETITIVENESS

enterprise; and similarly it means enforcing antitrust laws and regulations that will foster competition among firms. Most of all, in the near term, it means getting the U.S. fiscal house in order and restoring a sensible balance among savings, investment, and consumption.

Notes 1. Grossman and Rossi-Hansberg (2006) coined the term “trade in tasks” in their famous Jackson Hole paper, “The Rise of Offshoring: It’s Not Wine for Cloth Anymore.” 2. In the same footnote, Krugman recites this story: “Richard Feynman once told a reporter who wanted a 5-minute summary of his work that if it could have been summarized in 5 minutes, it wouldn’t have warranted a Nobel.” Perhaps it is fitting that Alfred Nobel never established what is commonly referred to as the Nobel Prize in Economics. We interpret Krugman generously to allow us five minutes per Big Idea, rather than five minutes for all of trade theory. 3. Deflating Douglass North’s (1958) seminal freight rate index to the U.S. general price index, O’Rourke and Williamson (1994) find that Atlantic shipping costs declined 41 percent between 1870 and 1910. An older index in Isserlis (1938) that includes non-Atlantic routes found a smaller but still drastic decline of 25 percent. The price of shipment from the coast to the interior was also declining during this period due to the expansion of the railroad. 4. U.S. import tariffs dropped from around 6 percent in the 1950s to around 1.5 percent in the 2000s. Worldwide average import tariffs dropped from 8.6 percent to 3.2 percent between 1960 and 1995. 5. In actuality, the exact price would be determined by demand. 6. Typically, perusal of any U.S. president’s jobs plan will provide abundant fodder for this critique. See, for example, White House (2011). 7. This is not to say that gains in productivity around the world are always positive sum. It is common to hear new initiates to comparative advantage claim that productivity gains in foreign countries are always good for their own country as well, even if the other country’s productivity growth is faster than their own country’s. This is not true, however, when the foreign country’s productivity gain is in an industry that the home country exports. 8. Bela Balassa’s study follows a 1951 study by G. D. A MacDougall. 9. This model was originally expounded in Ohlin’s 1933 book, Interregional and International Trade. The ideas, however, were developed while Ohlin was a graduate student supervised by Heckscher at the Stockholm School of Economics. Several important extensions have been conducted by Paul Samuelson, Ronald Jones, and Jaroslav Vanek. 10. A thorough review of the empirical evidence for the H-O model can be found in Leamer and Levinsohn (1995).

HHassett.indbassett.indb 270270 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 271

11. That is certainly not enough to settle a debate among economists, though. Paul Krugman and Maurice Obstfeld (1999) respond that “these findings do not, however, contradict the observation that overall the Heckscher-Ohlin model does not seem to work very well, because North-South trade in manufactures accounts for only about 10 percent of total world trade.” 12. The H-O model also provides a strong prediction for the distributional effects of trade liberalization. Simply put, it says that gains will accrue to owners of a country’s most abundant resources. For the United States, this has often meant that high-skilled workers will gain from globalization and that low-skilled workers will lose. In particular, high-skilled workers in industries that employ many high-skilled workers will gain the most. Policy response, therefore, has often centered on training low-skilled workers in new skills to attempt to shift them into higher-skilled jobs. This discussion, however, rests outside of the scope of this chapter. 13. The seminal papers in this field were Dixit and Norman (1980), Lancaster (1980), Krugman (1979, 1980, 1981), Helpman (1981), and Ethier (1982). These economists relied heavily on the work of Dixit and Stiglitz (1977), which provided an easy way to model imperfect competition. 14. Even Ohlin’s 1933 book that is the basis for the Heckscher-Ohlin model men- tions increasing returns to scale as a motivation of trade. 15. See Krugman and Obstfeld (1999) for an excellent and approachable description of imperfect competition and the modeling of increasing returns. 16. The ability of firms to grow large enough to exploit increasing returns to scale is incompatible with the perfect competition market structure, where firms are price- takers, which earlier trade models rely on. 17. The answer to this question is surely knowable, but it remains unknown to us. Our guess is that it’s because the 1ZZ-FE, the engine used in the Corolla, was made in the same Cambridge, Ontario, plant before it was discontinued in 2007. 18. Krugman (1991) argues for a definition of competitiveness under these terms in his article “Myths and Realities of U.S. Competitiveness.” In a more pop- literature setting, Michael Porter (1990) relies partly on external economies of scale to define competitiveness in his book, The Competitive Advantage of Nations. Although the idea that external economies might contribute to trade and be an area over which nations compete goes all the way back to Alfred Marshall’s 1920 book, Principles of Economics. 19. A movie company founded in Hollywood, for example, will have a much easier time finding talented actors than if it had decided to be on the forefront of the movie industry in South Africa and had begun there. Likewise, if a young woman wants to become a movie star, she’ll move to Los Angeles. This is a particularly salient example because of the turnover in the project-oriented movie industry; actors and cinematographers and set designers are always completing one job and trying to find the next.

HHassett.indbassett.indb 271271 111/8/121/8/12 9:279:27 PMPM 272 RETHINKING COMPETITIVENESS

20. One reason a boutique investment firm might locate in New York City, for example, is that there is already access to securities lawyers, representatives from specialized software firms, and plenty of steakhouses for celebrating deals. 21. Rosenthal and Strange’s (2004) chapter in the Handbook of Urban and Regional Economics provides an excellent review. 22. Relevant inquiries along these lines would include questioning whether knowledge spillovers extend from software firms to hardware firms; whether the U.S. car industry being in Detroit, relatively close to Canada, contributed to the rise of automobile manufacturing in Canada; and whether a learning curve exists in watchmaking: would the six hundred watchmaking firms in Switzerland make such excellent watches if the Watchmakers’ Guild of Geneva hadn’t been established in 1601 and if the Swiss watchmaking industry hadn’t been refining its body of knowledge ever since? 23. Other examples abound: New York is now a global financial center, but the city was founded because of a canal that is not used anymore. There are no more silicon mines in Silicon Valley than there are in London, but Stanford University’s Dean of Engineering Frederick Terman nurtured Hewlett-Packard in the 1940s and 1950s and from there the industry snowballed. 24. Baldwin (1969) writes, “If after the learning period, unit costs in an industry are sufficiently lower than those during its early production stages to yield a discounted surplus of revenues over costs (and therefore indicate a comparative advantage for the country in the particular line), it would be possible for firms in the industry to raise sufficient funds in the capital market to cover their initial excess of outlays over receipts” (quoted in Pack and Saggi 2006, p. 297). 25. This is certainly not impossible, though, and can be seen happening on a smaller level with consortiums of real estate developers being financed to build new neighborhoods. 26. Several papers have completed the exercise of showing that Ricardian models of trade have multiple equilibria when external IRS are present, with some equilibria exhibiting losses from trade. See, for example, Ethier and Ruffin (2009). 27. Arguments that do not rely on these assumptions have been made to support industrial policy in developing countries. Many of these rely on the fact that developing countries do not have well-developed financial markets, so even firms wishing to enter internal IRS industries will not be able to gain enough financing to compete with already established firms. Although this debate is outside of the scope of this chapter, we offer a brief rejoinder: if access to financing is the weak link, subsidize the development of private-sector financial markets, not the internal IRS industry. 28. This is loosely borrowed from Krugman (1991): “This is the real competitiveness issue: The possibility that international competition will exclude the United States from some industries in which it could or should have had a comparative advantage” (p. 254). 29. For example, it is probably most efficient globally for the eastern part of North America to have only one financial center because external IRS drive down costs, but

HHassett.indbassett.indb 272272 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 273

Canada still might gain individually if it supported the development of a financial center in Toronto. 30. In the 1984 U.S. presidential election, for example, Democratic contenders Gary Hart, Ernest Hollings, and Walter Mondale were all strong proponents of enacting industrial policy on a large scale in the United States. 31. Measuring success is difficult, given that there is never a good counterfactual comparison, and an industry’s success does not necessarily mean that subsidies or protection was worth the cost. Also, few countries that have conducted industrial policy have stuck to the recommendation to invest only in industries where external IRS might exist. Last, the tools that many countries use to protect industries can be hard to quantify and are wide ranging. Largely due to these problems, there is little consensus in the economics literature about the role of industrial policy in any recent economic successes. 32. Global value chains and global supply chains are largely indistinguishable in the popular and academic literature; we choose “global value chain” here because it emphasizes the fact that each country is adding value in the production process. Many products that both begin with research and end with consumption in the United States, for example, have much of their value added in other countries. These cannot be thought of as solely American goods. 33. This term used to apply only to the distribution of tasks to affiliates, but we will use it to apply to both affiliates and other firms. 34. Air freight prices fell at an annualized rate of 12.8 to 16.6 percent (depending on methodology) from 1957 to 1972, according to Gordon (1990). Hummels (2007) finds an increase in price from 2.87 percent annually from 1973 to 1980, but then prices continued to fall at 2.52 percent annually from 1980 to 1993 and 2.1 percent annually from 1991 to 2004, although prices began to rise after 2001. 35. Much more complicated models of trade in tasks exist, some of which show losses from trade in special circumstances, but, overall, trading tasks are considered to fit well within this framework. 36. Baldwin (2011) provides an excellent explanation of this process and makes much of this argument. Fifarek and Veloso (2010) note how this hollowing out will occur less in industries that exhibit large “vertical spillovers” either between R&D and production or between production and retail. 37. Grossman and Rossi-Hansberg make this point in their 2010 article, “External Economies and International Trade Redux.”

References Amiti, Mary, and Caroline Freund. 2008. The Anatomy of China’s Export Growth. Policy Research Working Paper Series 4628. World Bank, Washington, D.C. Athukorala, Prema-chandra, and Nobuaki Yamashita. 2006. “Production Fragmenta- tion and Trade Integration: East Asia in a Global Context.” North American Journal of Economics and Finance 17(3):233–56.

HHassett.indbassett.indb 273273 111/8/121/8/12 9:279:27 PMPM 274 RETHINKING COMPETITIVENESS

Balassa, Bela. 1963. “An Empirical Demonstration of Classical Comparative Cost The- ory.” Review of Economics and Statistics 4:231–38. Baldwin, Richard. 2006. “Globalisation: The Great Unbundling(s).” In Globalisation Challenges for Europe, ed. Secretariat of the Economic Council. Helsinki: Prime Minister’s Office Publications. ———. 2011. “Integration of the North American Economy and New-Paradigm Glo- balization.” In Global Value Chains: Impacts and Implications, ed. Aaron Sydor. Min- istry of Public Works and Government Services, Canada. Baldwin, Richard, and Philippe Martin. 1999. Two Waves of Globalisation: Superficial Similarities, Fundamental Differences. NBER Working Paper 6904. National Bureau of Economic Research, Cambridge, MA. Baldwin, Robert E. 1969. “The Case against Infant-Industry Tariff Protection.” Journal of Political Economy 77(3):295–305. Blonigen, Bruce, and Alyson Ma. 2007. Please Pass the Catch-up: The Relative Perfor- mance of Chinese and Foreign Firms in Chinese Exports. NBER Working Paper, vol. 13376. National Bureau of Economic Research, Cambridge, MA. Branstetter, Lee, and Nicholas Lardy. 2006. China’s Embrace of Globalization. NBER Working Paper, vol. w12373. National Bureau of Economic Research, Cambridge, MA. Dean, Judith M., and Mary E. Lovely. 2008. Trade Growth, Production Fragmentation, and China’s Environment. NBER Working Paper 13860. National Bureau of Eco- nomic Research, Cambridge, MA. Dixit, Avinash, and Victor Norman. 1980. Theory of International Trade. Cambridge: Cambridge University Press. Dixit, Avinash, and Joseph Stiglitz. 1977. “Monopolistic Competition and Optimum Product Diversity.” American Economic Review 61:297–308. Ethier, Wilfred. 1982. “National and International Returns to Scale in the Modern Theory of International Trade.” American Economic Review 72:389–405. Ethier, Wilfred J., and Roy J. Ruffin. 2009. External Economies of Scale and Com- parative Advantage. PIER Working Paper 08-008. Penn Institute for Economic Research, Philadelphia. Ferrantino, Michael J., Robert Koopman, Zhi Wang, and Falan Yinug. 2010. “The Nature of US–China Trade in Advanced Technology Products.” Comparative Eco- nomic Studies 52(2):207–24. Fifarek, Brian J., and Francisco M. Veloso. 2010. “Offshoring and the Global Geogra- phy of Innovation.” Journal of Economic Geography 10:559–78. Fontagné, Lionel, Michael Freudenberg, and Guillaume Gaulier. 2006. “Systematic Decomposition of World Trade into Horizontal and Vertical IIT.” Review of World Economics 142(3):459–75. Gaulier, Guillaume, Francoise Lemoine, and Deniz Unal-Kesenci. 2011. China’s Inte- gration in East Asia: Production Sharing, FDI and High-Tech Trade. CEPII Working Paper 2005–09. CEPII, Paris.

HHassett.indbassett.indb 274274 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 275

Gereffi, Gary, and Stacy Frederick. 2010. The Global Apparel Value Chain, Trade and the Crisis: Challenges and Opportunities for Developing Countries. Policy Research Working Paper 5281. World Bank, Washington, D.C. Gordon, Robert. 1990. The Measurement of Durable Good Prices. Chicago: University of Chicago Press. Grossman, Gene M., and Esteban Rossi-Hansberg. 2006. The Rise of Offshoring: It’s Not Wine for Cloth Anymore. Paper presented at Kansas Fed’s Jackson Hole conference for Central Bankers, Jackson Hole, Wyoming, August 25. ———. 2008. “Trading Tasks: A Simple Theory of Offshoring.” American Economic Review 98(5):1978–97. ———. 2010. “External Economies and International Trade Redux.” Quarterly Jour- nal of Economics 125(2):829–58. Haddad, Mona. 2007. Trade Integration in East Asia: The Role of China and Production Networks. Policy Research Working Paper 4160. World Bank, Washington, D.C. Helpman, E. 1981. “International Trade in the Presence of Product Differentiation, Economies of Scale, and Monopolistic Competition: A Chamberlinian-Heckscher- Ohlin Approach.” Journal of International Economics 2:305–40. Hummels, David. 2007. “Transportation Costs and International Trade in the Second Era of Globalization.” Journal of Economic Perspectives 21:131–54. Hummels, David, Jun Ishii, and Kei-Mu Yi. 2001. “The Nature and Growth of Vertical Specialization in World Trade.” Journal of International Economics 54:75–96. Isserlis, Leon. 1938. “Tramp Shipping Cargoes and Freights.” Part 1. Journal of the Royal Statistical Society 101:53–14. Koopman, Robert, William Powers, Zhi Wang, and Shang-Jin Wei. 2010. Give Credit Where Credit Is Due: Tracing Value Added in Global Production Chains. NBER Work- ing Paper 16436. National Bureau of Economic Research, Cambridge, MA. Koopman, Robert, Zhi Wang, and Shang-Jin Wei. 2008. How Much of Chinese Exports Is Really Made In China? Assessing Domestic Value-Added When Processing Trade Is Pervasive. NBER Working Paper 14109. National Bureau of Economic Research, Cambridge, MA. Krugman, Paul. 1979. “Increasing Returns, Monopolistic Competition, and Interna- tional Trade.” Journal of International Economics 9:469–79. ———. 1980. “Scale Economies, Product Differentiation, and the Pattern of Trade.” American Economic Review 70:950–59. ———. 1981. “Intra-Industry Specialization and the Gains from Trade.” Journal of Political Economics 89:959–73. ———. 1991. “Myths and Realities of U.S. Competitiveness.” Science, November 8. ———. 1993. “The Current Case for Industrial Policy.” In Protectionism and World Welfare, ed. D. Salvatore. Cambridge: Cambridge University Press, pp. 160–79. ———. 1999. Was It All in Ohlin? Paper presented for the Centennial Celebration of Bertil Ohlin. Stockholm, October.

HHassett.indbassett.indb 275275 111/8/121/8/12 9:279:27 PMPM 276 RETHINKING COMPETITIVENESS

Krugman, Paul, and Maurice Obstfeld. 1999. International Economics Theory and Pol- icy. 5th ed. Boston: Addison-Wesley. Lancaster, K. 1980. “Intra-Industry Trade under Perfect Monopolistic Competition.” Journal of International Economics 10:151–75. Leamer, Edward E., and James Levinsohn. 1995. “International Trade Theory: The Evidence.” In Handbook of International Economics, vol. 3, ed. G. M. Grossman and K. Rogoff. Amsterdam: Elsevier, pp. 1339–94. Leontief, W. 1953. “Domestic Production and Foreign Trade: The American Capital Position Re-Examined.” Proceedings of the American Philosophical Society 97:331–49. Linden, Greg, Kenneth L. Kraemer, and Jason Dedrick. 2007. Who Captures Value in a Global Innovation System: The Case of Apple’s iPod. Personal Computing Industry Center, University of California, Irvine. ———. 2011. “Innovation and Job Creation in a Global Economy: The Case of Apple’s iPod.” Journal of International Commerce and Economics 3(1):223–39. Ma, Alyson, Ari Van Assche, and Chang Hong. 2009. “Global Production Networks and China’s Processing Trade.” Journal of Asian Economics 20:640–54. MacDougall, G. D. A. 1951–52. “British and American Exports: A Study Suggested by the Theory of Comparative Costs.” Parts 1 and 2. Economic Journal 61(244):697– 724; 62(247): 487–521. Marshall, Alfred. 1920. Principles of Economics. 8th ed. London: Macmillan. Marshall Alfred, and Mary Paley Marshall. 1879. The Economics of Industry. London: Macmillan. Montout, S., J. L. Mucchielli, and S. Zignago. 2002. “Regionalization and Intra-Indus- try Trade: An Analysis of Automobile Industry Trade in NAFTA” Revue Région et Développement 16:137–60. Ng, Francis, and Alexander Yeats. 1999. Production Sharing in East Asia: Who Does What for Whom, and Why? Policy Research Working Paper Series 2197. World Bank, Washington, D.C. ———. 2003. Export Profiles of Small Landlocked Countries: A Case Study Focusing on Their Implications for Lesotho. Policy Research Working Paper Series 3085. World Bank, Washington, D.C. North, Douglass C. 1958. “Ocean Freight Rates and Economic Development, 1750– 1913.” Journal of Economic History 18:537–55. O’Rourke, Kevin, and Jeffrey G. Williamson. 1994. “Late Nineteenth-Century Anglo- American Factor-Price Convergence: Were Heckscher and Ohlin Right?” Journal of Economic History 54(4):892–916. Pack, Howard, and Kamal Saggi. 2006. “Is There a Case for Industrial Policy? A Criti- cal Survey.” World Bank Research Observer 21:267–97. Porter, Michael. 1990. The Competitive Advantage of Nations. New York: Free Press. Preeg, E. H. 2004. The Threatened U.S. Competitive Lead in Advanced Technology Prod- ucts (ATP). Manufacturers Alliance for Productivity and Innovation, Arlington, VA.

HHassett.indbassett.indb 276276 111/8/121/8/12 9:279:27 PMPM GLOBAL VALUE CHAINS AND THE CASE FOR FREE TRADE 277

Ricardo, David. 1817. The Principles of Political Economy and Taxation. New York: Dover. Rodrik, D. 2006. “What’s So Special about China’s Exports?” China and World Econ- omy 14(5):1–19. Rosenthal, Stuart S., and William C. Strange. 2004. “Evidence on the Nature and Sources of Agglomeration Economies.” In Handbook of Urban and Regional Econom- ics, ed. J. V. Henderson and J.-F. Thisse. Amsterdam: Elsevier, 2119–71. Schott, P. K. 2008. “The Relative Sophistication of Chinese Exports.” Economic Policy 23(53):5–49. Slaughter, Matthew J. 2010. How U.S. Multinational Companies Strengthen the U.S. Economy: Data Update. Business Roundtable and U.S. Council for International Business, New York. U.S. International Trade Commission. 2011. “Global Supply Chains.” In The Economic Effects of Significant U.S. Import Restraints. Investigation No. 332–325. Washington, D.C.: U.S. International Trade Commission. White House. Office of the Press Secretary. 2011. Facts Sheet and Overview. http:// www.whitehouse.gov/the-press-office/2011/09/08/fact-sheet-and-overview. Wood, Adrian. 1994. “Give Heckscher and Ohlin a Chance!” Review of World Econom- ics [Weltwirtschaftliches Archiv] 130(1):20–49. Yi, Kei-Mu. 2003. “Can Vertical Specialization Explain the Growth of World Trade?” Journal of Political Economy 111:52–102.

HHassett.indbassett.indb 277277 111/8/121/8/12 9:279:27 PMPM 9

International Competitiveness Phillip Swagel

It is common for policymakers to link proposals to notions of international competitiveness by asserting that a particular proposal will boost a nation’s ability to compete in the global economy.1 The straightforward meaning of international competitiveness means that policymakers see a competition in which a nation wins and others lose. As noted by Krugman (1994), however, “It is simply not the case that the world’s leading nations are to any important degree in economic competition with each other, or that any of their major economic problems can be attributed to failures to compete on world markets” (p. 30). Krugman famously terms the obsession (as he puts it) with competitiveness as “misguided and damaging,” notably because a focus on competitiveness can reduce the quality of the policy formation process and lead to improper policy choices (p. 44). Many others have echoed this assessment, including most recently Toder (2012). Indeed, economic considerations often imply that nations would be better off cooperating on economic policies rather than competing—though there are exceptions. International trade, for example, is usually seen as providing benefits for both sides rather than being characterized by welfare gains resulting from winning a competition against another nation. In a macroeconomic context, strong economic growth in a trading partner generally has a positive rather than negative impact on a nation’s own economic well-being as measured by GDP or employment growth. For the United States, it is tenuous to associate such contemporary economic problems as high unemployment to an inability to compete against other nations—just as it was for Europe when Krugman complained about the earlier European policy focus on a lack of competitiveness. Appropriate policy responses to

278

HHassett.indbassett.indb 278278 111/8/121/8/12 9:279:27 PMPM INTERNATIONAL COMPETITIVENESS 279

high U.S. unemployment might encompass measures to boost supply, such as improving incentives and productivity, or to increase demand, such as by expansionary fiscal action (the appropriate choice of such measures is beyond the scope of this chapter). But these responses would be appropriate in their own right and not with reference to international competition. Competitiveness does have a precise meaning when talking about the competitiveness of firms in particular industries or of countries on international markets, and in these contexts there are policies that could affect competitiveness and thus welfare. Similarly, competitiveness has a precise meaning in other arenas, such as financial markets, where policies can make a country more or less attractive as a destination for activities that are relatively flexible in terms of their location. For example, tax or regulatory policy can affect the locational decisions of activities such as firm startups and initial public offerings (IPOs). Changes in measures of competitiveness, such as relative costs or prices, would be reflected in outcomes such as export growth, market share, or in some cases the trade balance, or employment in tradable industries. Mea- sures related to financial sector competitiveness might translate into outcomes such as the number of public offerings. The catch, however, is that such outcomes, though connected to well-defined measures of competitiveness, are not always straightforwardly associated with underlying indicators of well-being. For example, a move toward a trade surplus or trade deficit is not necessarily connected to an improvement or decline in well-being; the U.S. trade deficit narrowed sharply during the recent recession, but this likely reflects changes in demand rather than a decline in U.S. competitiveness. And it would be hard to see the narrower trade deficit as a good thing. By the same token, policies that boost the international competitiveness of a country’s manufacturing sector or that lead to a more positive trade balance could actually be welfare decreasing (a recession would likely lower costs and weaken a nation’s currency, improving competitiveness as measured). And similarly, an increased trade deficit or slower export growth is not necessarily a problem—a country with relatively strong economic growth would naturally be expected to move toward a trade deficit. Policy proposals aimed at industry- or firm-specific competitiveness can be evaluated using cost-benefit tests or other forms of welfare calculations. Such calculations would balance any costs of the policies against

HHassett.indbassett.indb 279279 111/8/121/8/12 9:279:27 PMPM 280 RETHINKING COMPETITIVENESS

benefits for firms that become more competitive and gain market share at home or abroad at the expense of rivals from another country. Interna- tional competitiveness of this sort can be clearly understood in the context of particular industries and activist policies formulated in accordance with standard economic analysis—this would be along the lines, for example, of the economic literature on strategic trade policy. At the same time, policies based on a partial equilibrium approach might be good for an industry but not necessarily welfare improving for a nation as a whole. This means that a focus on well-defined measures of competitiveness can lead to unsatisfac- tory policy outcomes. This chapter reviews different concepts for international competitiveness and discusses their connection to measures of economic growth and national prosperity and to economic policies. A first variety of competitiveness indicators relates to performance in trade, including measures such as real exchange rates based on consumer prices, export prices, or relative labor costs; market shares or export growth; and other measures. These can be related to changes in trade flows, but the relationship with growth and well-being is not always straightforward. A second class of indicator relates more broadly to aspects of national economic well-being, such as growth of productivity, output, or employment, or to factors that are thought to support growth. Such inputs to growth could include a broad variety of concepts, such as the quality of a nation’s educational system, infrastructure, the number of engineers (or even just of female engineers, as argued recently by the U.S. Department of Commerce [2012a]), and so on. These can then be aggregated into synthetic measures, such as the Global Competitiveness Report of the World Economic Forum (WEF) (2011), that seek to assess the overall status of an economy or to gauge a nation’s economic prospects. In contrast to the first class of trade-related measures, these indicators might be seen as better connected to overall economic growth but often barely related to a competition between nations. This makes the label of competitiveness somewhat awkward, as noted by Krugman and others. It might be better instead to think of this second class of indicators as useful for diagnos- ing economic shortcomings—areas that merit policymaking attention— rather than as saying something about competition. The international comparison could still be useful as a form of benchmarking, in that the

HHassett.indbassett.indb 280280 111/8/121/8/12 9:279:27 PMPM INTERNATIONAL COMPETITIVENESS 281

experience of other countries suggests what might be possible, while keeping in mind that some of the areas under examination are not arenas for actual competition. In a sense, the real competition, if it can be called a competition, is for each nation to strive to be the best it can be—to compete with itself to boost productivity or increase the number of engineers or strengthen innovation or so on. But this can all be motivated internally rather than as a competition against other nations. Such actions might be connected to underlying concepts of well-being or growth, even if loosely or inaccurately (such as with an ineffective policy that is advocated under the rubric of this second form of competitiveness). This contrasts with the first type of competitiveness indicators, which are connected to competition (or at least interactions) with other countries but do not always have a causal connection to well-being (such as a wider trade deficit reflecting strong economic growth rather than lost competitiveness). This circles back to the point made by Krugman, that with nations not in competition with each other on most economic dimensions, the connection between measures of international competitiveness and underlying indicators of welfare or other economic fundamentals is often unclear, making it difficult to assess proposals and giving rise to the concern that a focus on competitiveness can lead to muddled thinking or unconstructive policies. There is a sense in which the present-day usage of competitiveness to motivate policy has become more of an exhortation to “do good things” (in the eye of the proposer) for one’s own economy and not necessarily in competition with another nation. In this parlance, competitiveness is often taken as a synonym for productivity (implicitly or explicitly). This is despite the fact that proposals to boost productivity are often not linked to competition with another country. Consider, for example, the idea of increasing government expenditures on infrastructure (which is among the proposals discussed below). This might boost productivity and increase the level of economic activity (subject to the usual cost-benefit test for government policies), but such a proposal can be motivated just as well by purely domestic considerations, such as a desire to raise standards of living—there is no need to invoke international competition. The downside of this usage is that an appeal to competitiveness might obscure the policy debate on the fundamentals—competitiveness becomes

HHassett.indbassett.indb 281281 111/8/121/8/12 9:279:27 PMPM 282 RETHINKING COMPETITIVENESS

a slogan rather than an objective. To be sure, sometimes the use of competitiveness can be a verbal tic or empty political rhetoric—and this is especially worrisome when the focus on competitiveness is used to bolster the case for policies that are likely misguided according to fundamental measures such as notions of well-being or better-defined economic concepts such as income growth. But as shown below, the modern use of competitiveness as a word that relates to a need to “do good things” has become pervasive and well understood. This use of competitiveness could be seen as a new vari- ant of the English language rather than as simply a meaningless word or perversion of the language. Just as we do not speak the same English as Chaucer or even Keynes, economic phrasing evolves, and this use of competitiveness might be seen in that light. This is certainly grating for people who grew up admiring the clarity of Krugman’s admonishment of competitiveness. But use of the word has become commonplace in the nearly two decades since that article was published. And as I discuss below, the meaning of other words has similarly evolved. I thus conclude less critically in some respects than Krugman. The key is for policymakers and the broad public to recognize when discussions over competitiveness refer to well-grounded measures relating to international trade or actual competition over limited resources and dif- ferentiate these instances from the latter-day usage of competitiveness as a synonym for “relating to good things.” In all circumstances it is essential to evaluate any policy proposals based on the eventual impact on measures of well-being rather than on the intermediate steps of exports or costs or productivity or so on. This chapter next discusses the uses and meanings of competiveness more specifically, including making connections with economic outcomes and measures of well-being. The chapter then examines policy uses of competitiveness. Many of the recent uses of this term could be seen as misuses—unless the language has been understood to have evolved. The chapter then considers this possibility of a new usage and discusses the implications for policymaking of competitiveness. Throughout, a focus is to provide policymakers with a skeptical guide by which to evaluate exhortations to undertake policy actions based on notions of international competitiveness.

HHassett.indbassett.indb 282282 111/8/121/8/12 9:279:27 PMPM INTERNATIONAL COMPETITIVENESS 283

Uses and Meanings of “Competitiveness”

Measures of international competitiveness can be broadly split into indicators that are reasonably well connected to a competition between nations in trade or perhaps investment and indicators that are associated more with broad notions of growth and well-being but not with international competition per se. Confusingly, the first type of measures, while related to international competition, is often not monotonically connected to underlying concepts of welfare or well-being, while the second type of indicator does relate to welfare but is often equally applicable to a closed economy (one without international trade or investment). In this case, competitiveness is naturally focused on metrics tied to growth or welfare and not related to international competition. The challenge for policymakers is to discern the appropriate meaning.

Trade- and Investment-Related Measures of International Competi- tiveness. Trade-related measures of international competitiveness connect prices or costs at a national or industry-specific level to economic outcomes such as imports and exports or the balance of trade, investment across borders, or investment by firms facing international competition. Trade and investment outcomes in turn affect measures such as profitability, employment, and so on. These indicators are related to international competitiveness in that the competitive interaction with other countries or with foreign firms impacts the domestic economy. Changes in these measures of competitiveness, however, are not uniformly good or bad in terms of welfare—a nation can appear to lose competitiveness in a well-defined sense as an artifact of developments that are beneficial. Productivity growth in a nation’s manufacturing sector, for example, can lead to a real exchange rate appreciation and thus a loss of competitiveness as typically measured, but without any actual impairment of the ability of firms to compete in global markets (this is the case with the Balassa-Samuelson effect). Moreover, changes in the outcomes, such as the trade balance, that are affected by these measures of competitiveness might result more from factors having nothing to do with competitiveness per se. For example, the U.S. trade deficit might narrow considerably even when U.S. competitiveness is not improving, but it could be because U.S. import

HHassett.indbassett.indb 283283 111/8/121/8/12 9:279:27 PMPM 284 RETHINKING COMPETITIVENESS

demand has fallen with a slowing economy (as was the case in late 2008). The use of even well-defined competitiveness indicators thus requires caution in interpretation on behalf of the policymaker. Durand and Giorno (1990), Turner and Van ’t dack (1993), and Marsh and Tokarick (1994) review measures of trade competitiveness based on relative prices and costs. Among the chief measures under assessment are real exchange rates based on consumer prices, unit labor costs in tradable goods such as manufacturing, and export unit values (export prices). These three measures relate to trade performance in that they gauge costs or prices in sectors of the economy exposed to international competition, using exchange rates to convert prices or costs into a common currency. The real exchange rate based on consumer prices is typically calculated as the nominal exchange rate (XR) multiplied by the ratio of consumer price indices (CPI); for example, the real exchange rate between the dollar and the euro would be calculated as follows: Real exchange rate = Nominal XR in dollars per euro × CPI in Europe / CPI in United States Analogous measures would be constructed by replacing the two consumer price indices with either unit labor costs in each country or export values. Each of these three measures of real exchange rates has a somewhat different coverage basis and thus implication and interpretation, but all are well connected to notions of competitiveness. The real exchange rate based on the relative CPI reflects overall prices for the goods and services purchased by households, which include both items that are traded and subject to international competition and products and (especially) services that are not traded. A larger value for the real exchange rate as defined above would correspond to a loss in U.S. competitiveness, in that the prices of the overall basket of U.S. items is higher relative to the price of foreign items when measured in a common currency. As discussed below, however, this apparent loss of competitiveness in some cases can come about for reasons that signal economic improvements rather than the opposite. An analogous finding would hold for relative unit labor costs—a larger value for U.S. relative labor costs would indicate that higher costs for U.S. firms would generally be expected to translate into higher prices and

HHassett.indbassett.indb 284284 111/8/121/8/12 9:279:27 PMPM INTERNATIONAL COMPETITIVENESS 285

thus a decline in U.S. firms’ competitive positions as reflected in output, sales, revenue, employment, and so on. Such a loss in competitiveness could come about through a nominal exchange rate appreciation or by an increase in wages relative to productivity. Indeed, higher labor compensation alone would not correspond to a loss of international competitiveness if these wages reflected increased productivity—unit labor costs would be unchanged. The coverage of unit labor costs in the measure of competitiveness can be narrower than with consumer prices, since the labor costs considered can include only industries that are exposed to international competition. As noted by Neary (2006), Cerra, Soikkeli, and Saxena (2003) assert that measures of competitiveness based on relative unit labor costs are preferred to measures based on consumer prices or export prices because they cover a greater share of trade than consumer prices and avoid some of the endogeneity problems inherent in measures of export prices based on unit values. This still leaves the question of how to aggregate wages in different industries within a country in order to calculate a single measure of unit labor costs. The perhaps natural approach of simply adding together all wages and all output (for the respective numerator and denominator of unit labor costs) would implicitly give equal weights to wages and the number of workers within a sector (since the wage bill is the product of the two) and more weight to sectors within a country with the highest value of production and the largest wage bill. One could imagine a situation, however, in which the policy concern is over the impact of changes in external competitiveness on employment rather than output. In this case, it would be appropriate to use employment shares as the weights rather than output shares—indeed, Soikkeli and Cerra (2002) show that this makes a considerable difference in assessing competitiveness in Ireland, because a few sectors have extremely high output per worker and thus get a large weight with the output-weighted measure of unit labor costs but not with employment weights. If the concern is the impact of exchange rate changes on employment, then the usual weighting scheme based on output might be misleading. Measuring competitiveness with relative export prices converted to a common currency would have a narrower basis than the CPI and a different one than relative unit labor costs. Export prices would be focused more

HHassett.indbassett.indb 285285 111/8/121/8/12 9:279:27 PMPM 286 RETHINKING COMPETITIVENESS

squarely on tradable items (even within manufacturing, items face differing degrees of international competition). As noted by Bennett et al. (2008), export prices in practice are often measured using export unit values—the value of exports divided by the quantity or volume. Changes in export value as measured could reflect the quality of exports, with rising export prices then reflecting quality upgrading rather than a loss of competitiveness. Neary (2006) devises a version of a real exchange rate index that measures the exchange rate needed to restore the initial level of GDP (or employment) after a set of changes in prices for tradable goods and changes in domestic wages and other factor costs. This gives a precisely defined measure of competitiveness in terms of the exchange rate. The downside is that the measure requires an extensive apparatus by which to relate changes in output or employment to changes in prices, wages, and the nominal exchange rate (the components of the real exchange rate). While difficult to construct and sometimes unclear in their implication, the measures of competitiveness discussed above are readily used in economic analysis. Carlin, Glyn, and van Reenen (2001), for example, examine the impact of changes in relative costs on export market shares in twelve manufacturing industries across fourteen countries in the Organisation for Economic Co-operation and Development. They find that export market shares are more sensitive to relative costs in industries that are characterized by increased product market competition over time. An implication of this result is that in industries marked by relatively fierce competition so that price more nearly equals cost, changes in costs matter more than in industries with greater price-cost markups that can adjust to absorb changes in relative costs. These indicators of international competitiveness can change both with prices and costs and through changes in the nominal exchange rate. Indeed, real exchange rates are often constructed to gauge the appropriateness or the impact of a particular value for a country’s nominal exchange rate. This is especially apt in the case of a country with a fixed exchange rate regime. A typical motivation for such a regime is for the exchange rate to provide a nominal anchor by which to stabilize inflation and inflationary expecta- tions. A country with a fixed nominal exchange rate can be said to suffer a loss of competitiveness if it experiences higher inflation (measured with the relative CPI measure) or higher unit labor cost growth than its trading

HHassett.indbassett.indb 286286 111/8/121/8/12 9:279:27 PMPM INTERNATIONAL COMPETITIVENESS 287

partners. Over time, this loss of competitiveness would likely result in lower net exports and reduced output and employment at exporters and import-competing firms. Ultimately, a nation might not be able to sustain the exchange rate peg. This describes the situation of Argentina in the mid- 1990s leading up to that nation’s crisis in 2001, as inappropriately loose fiscal policy was associated with chronic inflation above the rate in the United States, the currency of which the peso was pegged. The real appreciation trend of the peso (despite the fixed nominal exchange rate) was matched by rising unemployment rates and eventually contributed to a recession that began in 1998 and culminated in a debt default and end of the peg. The experience of Argentina is one in which a measure of competitiveness provides a useful framework with which to evaluate economic policies. Mussa (2002) discusses the adverse economic impacts of the fixed exchange rate regime in affecting external competitiveness and real economic indicators, such as growth and unemployment, that are clearly related to well-being. This example of Argentina illustrates that measures of international competitiveness with precise meanings such as the real exchange rate or relative unit labor costs in certain circumstances can be connected to welfare outcomes. Indicators of international competitiveness then translate into policy considerations—the impact of Argentina’s real exchange rate appreciation implying the need to consider changes in fiscal and monetary policies. Unfortunately, measures of international trade competitiveness do not always give a clear signal. A real exchange rate appreciation, for example, might suggest that firms in tradable goods industries, such as manufacturers, are losing competitiveness against foreign firms, as was the case with Argentina. But it can also be the case that elevated inflation relative to a nation’s trading partners reflects the impact of uneven productivity growth through the so-called Balassa-Samuelson effect. This is the phenomenon in which productivity gains in tradable industries lead to higher wage for workers in nontradable sectors—this comes about because of pressures within a nation’s internal labor market. The resulting inflation translates into a real exchange rate appreciation in terms of a CPI-based real exchange rate—but this is an artifact of productivity growth rather than a loss of competitiveness. Indeed, this sort of real exchange rate appreciation is more accurately seen as a symptom of success—of higher productivity in sectors exposed to international competition. Indeed, an irony is that other

HHassett.indbassett.indb 287287 111/8/121/8/12 9:279:27 PMPM 288 RETHINKING COMPETITIVENESS

policy-oriented measures that purport to measure competitiveness can be seen as boiling down to productivity so that stronger productivity growth can lead to both improvements in and deteriorations of competitiveness across different measures. The general point, as stressed by Durand and Giorno (1990), is that a nation could seem to lose competitiveness in trade-related measures of international competitiveness but not have a similar deterioration in trade flows themselves or in terms of measures of welfare. Improved productivity in a country as discussed above could lead to a stronger exchange value of a country’s currency and thus to improved terms of trade (a move toward lower prices for imports relative to exports) that would lead to greater purchasing power and a better standard of living but not an improved trade balance. Changes in global demand patterns also confound the relationship between measures of competitiveness and welfare. An increased preference by foreign residents for U.S.-produced goods and services, for example, could lead to an appreciation of the dollar and higher prices for U.S. exports, but the appearance of a loss of competitiveness would not constitute a cause for concern—the loss of competitiveness does not translate into reduced well-being, meaning that it is vital to understand the causal factors behind the change in competitiveness. Quantity indicators for imports and exports or the trade balance are sometimes used as measures of competitiveness in addition to relative prices and costs. The idea here is that relatively strong export growth compared to other countries and thus gains in international market shares reflect an improved competitiveness position, whether driven by changes in relative prices and costs or by other factors, such as changes in preferences. These measures of competitiveness suffer problems along the lines of price- and cost-based measures that make it difficult to directly connect changes in competitiveness based on trade flows to welfare. As noted by Bennett et al. (2008), trade shares are affected by price movements (including those reflecting exchange rate shifts) and by income growth. Strong U.S. income growth, for example, might translate into rising net imports even if U.S. firms become more productive. Rising imports or a wider trade deficit by themselves are thus not necessarily good indicators of the competitive status of the U.S. economy. In the broader macroeconomic context, a

HHassett.indbassett.indb 288288 111/8/121/8/12 9:279:27 PMPM INTERNATIONAL COMPETITIVENESS 289

nation’s overall trade balance reflects many factors, including both the price and demand effects discussed here and the interrelationships between trade and capital flows, which in turn affect exchange rates and growth. Increased saving by U.S. residents relative to investment, for example, would be associated with changes that lead to a narrower current account deficit and likely a narrower external deficit. A weaker dollar would normally be expected as part of these changes, and this might translate into improvements in measures of competitiveness based on prices or quantities. But the driving force in this scenario would be the change in saving and investment rather than in competitiveness per se. Baily and Lawrence (2006) assess the impact of changes in trade flows on U.S. living standards, with a particular focus on the terms of trade—the quantity of imports that can be purchased from exports. Stronger productivity growth might tend to increase the U.S. supply of goods and services and thus lower their price. This would lead to lower terms of trade and make it relatively more expensive for Americans to obtain imports—or, equivalently, require a greater quantity of exports for Americans to obtain a given amount of imports. Conversely, higher terms of trade would be expected to be associated with a larger trade deficit, since U.S. export prices would be relatively high. Baily and Lawrence disentangle changes in the relationship between the terms of trade and the trade balance from changes in the terms of trade itself that then translate into changes in the trade balance but without a shift in the underlying relationship between the two. They estimate an econometric relationship between exchange rates and the trade balance in which a 10 percent weaker dollar translates into a 12.5 percent narrowing of the trade deficit after three years (reflecting the usual adjustment process by which prices affect trade flows) and then calculate that a 22 percent depreciation of the dollar would have been needed to close the U.S. trade deficit in 2005. The estimated relationship between the exchange value of the dollar and the trade balance includes a shift in 1994, with any level of the trade balance being associated with an 8 percent weaker value for the real exchange rate from 1994 to 2005 compared to previous years. As they note, “With imports equal to about 15 percent of U.S. incomes, 0.15 × 0.8 equals 1.2 percent of GDP” (p. 231)—that is, the deterioration in the relationship between the dollar and trade corresponds to a loss of purchasing power equal to 1.2 percent of U.S. GDP, reflecting

HHassett.indbassett.indb 289289 111/8/121/8/12 9:279:27 PMPM 290 RETHINKING COMPETITIVENESS

the fact that the dollar must be weaker from 1994 on to maintain a given trade balance (or equivalently, the trade deficit would be wider after 1994 if there is no change in the value of the dollar). At the same time, however, Baily and Lawrence note that the stronger value of the dollar in this latter period itself accounts for three times as much of the wider U.S. trade deficit after 1994 than the change in the relationship between the U.S. terms of trade and trade flows. That is, U.S. trade performance deteriorated in the sense that a weaker dollar was needed to achieve a given trade balance, but this still explains only about one-quarter of the actual widening of the deficit since the early 1990s. The stronger dollar and the host of changes that are associated with it instead account for wider U.S. deficits. Whether this reflects a loss of competitiveness depends on what has brought about the stronger dollar. To the extent that the dollar is stronger because foreign investors seek to hold U.S. assets, this more likely reflects a welcome development for the United States. Again, the conclusion is that it is necessary to assess the causal factors behind a change in a measure of competitiveness to assess the relationship between competitiveness and well-being. Feenstra and Rose (1997) put forward a novel measure of competitiveness that relates to the development of a country’s export sector, in which countries that export fined-grained categories of goods to the United States relatively early are said to be more competitive than countries that export the same type of goods later—the idea being that early export is a sign of progress up a technological ladder. Early technological development seems like a good thing, but it is unclear how to relate this measure directly to well-being. This means that the policy implications are indistinct, since there is no way to evaluate the benefits of actions that lead to improved competitiveness measured in this way. One could also imagine nations competing over locational decisions for certain types of investment, such as the construction of factories or foreign direct investment more broadly. Indeed, inflows of foreign direct investment could improve productivity through technology transfer, the spread of managerial best practices, or other channels. The location of foreign direct investment could be affected by a range of indicators and policy levers such as trade and tax policies, transportation costs, or other factors such as importance of on-the-ground knowledge of local preferences and institutional factors.

HHassett.indbassett.indb 290290 111/8/121/8/12 9:279:27 PMPM INTERNATIONAL COMPETITIVENESS 291

Toder (2012) looks at competitiveness in a particular way by assessing tax policies in terms of the effect on the location of factors of production such as capital and labor.2 The competition between nations is then over these factors—for example, he compares the impact of several tax policy options on the location of firm investment. This gives a precise meaning to competitiveness, but as Toder notes, these locational impacts do not match up neatly with welfare measures. For example, a particular tax policy choice might lead to greater domestic investment but raise tax-inclusive costs for U.S. multinational firms and thus put them at a disadvantage against overseas competitors. It is not clear which of these outcomes is best for the United States—competitiveness is well defined if the desired outcome is the location of investment, but this does not translate straightforwardly into well-being. Even looking more narrowly than overall welfare, we might find that domestic employment (which is likely closer to the desired outcome of the policymaker than investment per se) would be higher with tax policies that boost U.S. firms’ global competitive positions rather than their domestic investment—after all, U.S. firms that invest in other countries tend to create jobs in the United States as well. The overall point is that there are well-defined measures of international competitiveness that have measurable links to outcomes such as trade flows. But the policy implications of measures of international competitiveness depend crucially on an understanding of the causal influences at play and the relationship between welfare and measures of international competitiveness.

Competitiveness Measures Based on Overall National Living Stan- dards. A different approach to competitiveness looks more broadly at indicators that appear to relate to living standards or various aspects of economic growth and development, including the sustainability of growth. These indicators often have meaningful connections to well-being, typically through measures of productivity. They thus provide some guidance for policymakers looking to boost growth, job creation, and so on: policies should be chosen to increase productivity. The irony, though, is that this second class of competitiveness measures is only modestly connected to international competitiveness according to the usual definition relating to competition with other countries. Higher productivity growth is good but generally for its own sake rather than in relation to a competition with other countries.

HHassett.indbassett.indb 291291 111/8/121/8/12 9:279:27 PMPM 292 RETHINKING COMPETITIVENESS

A hint of the lack of connection to international competitiveness is the difficulty in setting down a meaningful definition of competitiveness. A recent competitiveness report from the U.S. Department of Commerce (2012a) acknowledges this: “Similar to innovation, ‘competitiveness’ has also proved difficult to define and measure” (p. 2-2). The report goes on to reference a definition of competitiveness for countries from the WEF Global Competitiveness Report 2011–2012 (World Economic Forum 2011) as “the set of institutions, policies, and factors that determine the level of productivity of a country” (p. 4). As noted above, however, productivity is not necessarily determined through competition with other countries. The WEF’s Global Competitiveness Report 2011–2012 uses a vast array of indicators to measure competitiveness as both an aggregate ranking and through detailed subrankings. The report states that it “aims to capture the complexity of the phenomenon of national competitiveness, which can be improved only through an array of reforms in different areas that affect the longer-term productivity of a country” (p. 44). The report organizes its measures of competitiveness into twelve areas (pillars) seen as broad determinants of competitiveness and productivity. These areas include institutions, infrastructure, macroeconomic environment, health and primary education, higher education and training, goods market efficiency, labor market efficiency, financial market development, technological readiness (the “agility with which an economy adopts existing technologies to enhance the productivity of its industries”), market size, business sophistication, and technological innovation. There are numerous measures within each of these areas, and the overall global competitiveness index includes interactions between indicators within the twelve pillars. The report further notes that different factors matter for countries at different stages of development. The mass of information is boiled down to a rank comparison. The United States, for example, is ranked as the fifth most competitive nation, after Switzerland, Singapore, Sweden, and Finland. This is even though U.S. per capita income was higher than all of these countries except Switzerland. Even with the numeric competition, Krugman’s original point remains: countries do not compete with one another on most aspects of the WEF indicators. More advanced financial market development or improved business sophistication in China, for example, is, if anything, more likely to be

HHassett.indbassett.indb 292292 111/8/121/8/12 9:279:27 PMPM INTERNATIONAL COMPETITIVENESS 293

connected to stronger U.S. growth because these factors would strengthen China’s growth and thereby increase demand for U.S. goods and services. To be sure, it is possible to relate some of the WEF pillars to competition between nations in the sense that the choice of investment by a multinational firm might depend on the quality of one nation’s infrastructure or trade or tax policy or growth prospects relative to these indicators in another country. Countries might then win a competition based on these dimensions to attract global investment. As in Toder (2012), this is not necessarily connected to well-being in a straightforward way. One could imagine a country driving away investment through onerous regulation or chronic corruption, or attracting global-investment-based pecuniary incentives for multinational firms. But a welfare calculation (that is, a cost-benefit analysis) might still indicate a net negative if there is a cost to the policies needed to win the competition. A competition-based view might be more sensible in the context of industries with limited numbers of firms or resources. For example, one could view the training of engineers in a country as relevant to attracting investment. To the extent that global investment is a fixed amount, attracting investment through a policy of improved technical education could be seen as part of a competition between nations for limited investment and resources. A useful way to consider macro-oriented competitiveness measures is for countries to identify strengths and weaknesses and to pinpoint factors constraining their own growth and development. Cross-country comparisons are then useful for benchmarking rather than a competition. For the United States, for example, the global competitiveness index identifies the most problematic factors in doing business as involving tax rates, inefficient government bureaucracy, access to financing, and tax regulations. It is hard to conclude that these are the main policy challenges facing the United States, but this information might still be useful to highlight these areas for policymakers, and of course such a benchmark and spotlight could be more useful for countries other than the United States.

Capital Markets Competitiveness Assessments of capital markets competitiveness have focused on the desirability of countries for attracting initial public offerings. Zingales (2007) and

HHassett.indbassett.indb 293293 111/8/121/8/12 9:279:27 PMPM 294 RETHINKING COMPETITIVENESS

a number of reports from the Committee on Capital Markets Regulation (CCMR) have quantified that the United States attracts a smaller share of IPOs than was the case in 2000 or before.3 A January 25, 2012, report from the CCMR notes that “the proportion of the global IPO market captured by U.S. markets shrank to 9.8% during the third quarter of 2011”—a decline from a share of 14 to 17 percent in 2009 and 2010 and “significantly lower than the 1996–2006 average of 28.7%.” Zingales links this trend both to the improved attractiveness of equity markets in other countries (especially Europe) and to the impact of increased regulatory and compliance costs in the United States—notably from the 2002 Sarbanes-Oxley Act. Zingales (2007) notes that the U.S. share of IPOs can be seen as a “canary in the mine shaft,” with a smaller share of IPOs a potential warn- ing sign of reduced economic dynamism. As discussed by Daniels (2011), the 2010 Dodd-Frank Wall Street Reform and Consumer Protection Act (Dodd-Frank) sought to mitigate some of the impacts of Sarbanes-Oxley in raising compliance costs but added other requirements that might go in the other direction. More recently, the Jumpstart Our Business Starts Act (JOBS Act) enacted in April 2012 aims to make it easier for small business to raise capital and reduce the regulatory burden on IPOs for companies with up to $1 billion in revenues. New firms are especially mobile in terms of their locational decision, and even existing firms can change their domicile. These locational outcomes thus correspond to a type of international competition over financial resources, meaning that such indicators are useful for gauging capital markets competitiveness. As with other indicators of competitiveness, however, caution must be used in translating changes in capital markets competitiveness into policies—an improvement in European capital market performance does not necessarily signal a problem in the United States, though it could be cause for reassessment. According to Zingales (2007), changes in U.S. regulation since 2002 have made the United States relatively more attractive to firms “coming from countries with poor corporate governance standards and less favorable to countries with good corporate governance standards” (p. 3). He finds that relatively fewer dollars of global IPOs are in the United States since 2002 and that a markedly smaller share of firms are choosing to cross-list their shares on U.S. markets. In terms of the international competitiveness of U.S. capital markets, Zingales

HHassett.indbassett.indb 294294 111/8/121/8/12 9:279:27 PMPM INTERNATIONAL COMPETITIVENESS 295

concludes that increased regulation does not match up well with a cost- benefit standard. The direct impact of these changes can be measured in revenues and employment among market participants and analysts, but the more salient reason for considering such measures of financial market competitiveness is as an indicator of the health of capital markets and the underlying economy.

Policy Appeals to Competitiveness Appeals to competitiveness are commonplace in the discussion of policy proposals. This section discusses several such usages and analyzes them in the context of the discussion of measures of competitiveness above. Krug- man (1994) and Toder (2012) similarly provide examples of policymakers’ uses of competitiveness. Perhaps most famous among recent uses was the focus by President Obama in his January 25, 2011, State of the Union address on the concept of “winning the future” by competing in a global economy. The president’s budget focused on “Competing and Winning in the World Economy,” with an array of proposals, such as increased infrastructure investment, teacher funding, and many more (a huge list). A memo to cabinet members highlighted the need for government reforms to boost competitiveness and innovation: “By out educating, out innovating, and out building our competitors, we will enable our Nation to grow, create jobs, and thrive in the years ahead” (White House 2011a). The proposals arising from the State of the Union address purport to boost U.S. productivity, but are not well connected to international competitiveness. From high-speed rail to teachers and wasteful duplication, these proposals could be evaluated for the sake of the U.S. economy and not because of global competition. Other recent government efforts focus as well on competitiveness. The President’s Council on Jobs and Competitiveness was established as part of an effort to “help ensure that the United States has the most competitive and innovative economy in the world” (Immelt 2011). Among the proposals of the council were for U.S. policies to focus on manufacturing exports, to expand free trade, and to incentivize investment in innovation. The council organized a pledge by private companies to take on 6,000 new engineering interns and has suggested that the United

HHassett.indbassett.indb 295295 111/8/121/8/12 9:279:27 PMPM 296 RETHINKING COMPETITIVENESS

States streamline regulatory reviews, focus on improving education and infrastructure, take steps to boost international investment in the United States, and help startup firms (White House 2011b, c). These ideas are all motivated by an appeal to international competitiveness, but absent specifics on costs and projected outcomes, it is difficult to assess their impact. What is clear, however, is that these measures have little to do with competition with other countries. These ideas might be good or bad, but the focus is more accurately characterized as on domestic productivity than on international competition. A similar approach can be seen in a June 24, 2011, White House report titled “Materials Genome Initiative for Global Competitiveness,” which asserts that accelerating the process by which newly discovered advanced materials move from the laboratory to the commercial market place could significantly improve U.S. global competitiveness and ensure that the nation remains at the forefront of the advanced materials marketplace (White House 2011d). Despite the gaudy connection to global competition, the actual policy initiative involves better sharing of information among researchers. This seemingly modest initiative might be good for knowledge generation, but it is not clear how this activity relates to an international competition. Imagine if, say, British researchers advance the state of the art in materials science. Presumably this would be beneficial to the United States, since there is open commerce between the two nations. To be sure, U.S. incomes would be yet higher if the United States invented new materials and developed other innovations, but this is only modestly related to an international competition per se. A January 6, 2012, a Commerce Department report titled “A Roadmap for Strengthening U.S. Competitiveness” was meant to serve “as a call to arms, highlighting bipartisan priorities to sustain and promote American innovation and economic competitiveness” (U.S. Department of Commerce 2012b). Along the same lines as the president’s earlier proposals, the report focuses on the role of the government in improving innovation, investment, and ultimately growth, with calls for increased government spending on research, education, and infrastructure and further support for the manufacturing sector. These proposals are motivated by the assertion that other countries have become better educated that the United States and that this loss of the U.S. competitive edge means that future generations will not

HHassett.indbassett.indb 296296 111/8/121/8/12 9:279:27 PMPM INTERNATIONAL COMPETITIVENESS 297

enjoy a higher living standard than today. Again, however, it is unclear how any of this connects to notions of international competitiveness. If the American education system improves, that would undoubtedly be beneficial for the U.S. economy and for the individuals receiving a better education. But it is not clear how this matters in relation to the education levels of other countries in a way in which global commerce leads to the spread of innovation. To be sure, increased innovation would be beneficial for the United States. It is not obvious, however, that less innovation in other countries—as one might imagine if this is truly a competition—would be helpful as well. It is worth noting that the focus on competitiveness is long-standing. President George W. Bush proposed the so-called American Competitive- ness Initiative in February 2006 as part of his fiscal year 2007 budget proposal, with a call for increased government spending on research and development, education, and innovation. President Bush advanced the proposals with a nod to international competition, stating that “with the right policies, we will maintain America’s competitive edge, we will create more jobs, and we will improve the quality of life and standard of living for generations to come” (White House 2006). What is striking about these examples is that the use of competitiveness has evolved to encompass proposals that are not really based on international competition but instead focused on the goal of improving domestic performance such as by strengthening productivity or growth. As noted by Krugman (1994), a danger of a policy focus on global economic competitiveness when there is none is the possibility that policy discussions will become muddled and lead to support for inappropriate policies. On the other hand, it is possible that some of the proposals discussed above could be helpful, even if they are motivated by an inappropriate focus on international competitiveness. Many of the proposals amount to recommendations to boost spending on infrastructure or other areas of government investment. This might be reasonable, even if it is motivated inappropriately. The question then is whether it is acceptable to carry out the right policy for the wrong reason. The problem with the competitiveness approach might be stronger on an industry or product level, where a comparison with other countries could lead to actions that fail cost-benefit tests. Such an idea would be the notion of spending billions of dollars on high-speed rail systems in the

HHassett.indbassett.indb 297297 111/8/121/8/12 9:279:27 PMPM 298 RETHINKING COMPETITIVENESS

United States because China has done so and threatens U.S. leadership in transportation. It is correct, of course, that the infrastructure investment in China will affect global trade flows, but the dominant impact is likely to be a lowering of Chinese costs (assuming that exporters are not required to involuntarily pay for the rail line construction). This would then tend to lower the price of Chinese exports—a terms of trade increase and welfare improvement for the United States. Consideration of construction of high- speed rail systems in the United States should then be based on appropriate cost-benefit tests without reference to developments in China. The danger of a focus on international competitiveness is one of confusion—that the policy development process will be guided by inappropriate considerations. This is the case especially for competitiveness indicators that are well connected to international trade but that do not straightforwardly relate to well-being. The problem in this case is that policies will be favored because they improve measures of competitiveness, such as by promoting exports or narrowing the trade deficit, even if the costs of the policies outweigh the benefits. It remains useful to consider so-called international competitiveness indicators that amount to diagnostics of the domestic economy, even when presented as benchmarks against other countries. For example, policymakers might use indicators from the WEF to consider the potential outcomes if the United States were to adopt economic and regulatory policies like those in Europe. Similarly, a comparison with measures of competitiveness in China could indicate what might be possible in terms of growth, even if the comparison does not have immediate policy implication for the United States. China’s remarkable growth performance has reflected in various parts the catch-up of that country to the market frontier, including in terms of technology, capital, and labor. These changes include the massive movement of workers from low-productivity activities in the interior to the market-oriented economy of the coast; the equally massive increase in capital stock driven by investment; and the imitation (or theft) and deployment of more advanced technologies and production techniques. A welfare evaluation is more complicated. Output and income have risen in China, but growth has been driven by investment rather than consumption, implying that individual households have not received the full benefit of rising incomes as consumption. Indeed, Kahneman and Krueger

HHassett.indbassett.indb 298298 111/8/121/8/12 9:279:27 PMPM INTERNATIONAL COMPETITIVENESS 299

(2006) summarize the results of polling in China and Japan that show that indicators of life satisfaction or happiness actually decreased during those countries’ respective periods of strong growth. This could be related to the “adaptation” effect discussed by Kahneman and Krueger that rising expec- tations affect self-reported measures of happiness. The Chinese experience does not necessarily provide straightforward policy lessons for the United States, and a policy formation process based on competitiveness with China might well be inappropriate. This does not mean that there should be no focus on economic growth—not by any stretch. But the quality and composition of growth matter. In sum, notions of international competitiveness can be useful in some ways, but the limitations should be kept in mind for U.S. policymaking.

International Competitiveness in the Modern Vernacular Reference to international competitiveness remains a feature of policymaking discussions. The continued use of international competitiveness in non- competitive situations is so common that it is difficult to believe that it is meant to be intentionally misleading. President Obama might well have set forth his State of the Union address and February 2011 (fiscal year 2012) budgetary proposals with a reference to international competitiveness, but he surely made these proposals because he thought they would improve U.S. well-being (whether they actually would do so is another matter). Indeed, contemporary usage of the word “competitiveness” has evolved to refer to well-being rather than solely to competition with another nation or another firm. Policies to boost competitiveness can then be understood as policies that are meant to improve productivity or growth, even though this usage of the language does not accord with the traditional meaning of competition. The meanings of other words have similarly changed in policy discussions and in common parlance. The word “literally,” for example, has come to denote a point of emphasis rather than indicate that something should actually be taken literally.4 Similarly, the word “frankly” has taken on almost a reverse meaning: a policymaker saying that he or she is speaking frankly more usually provides a signal that a contrived or calculated message is about to be delivered. Other recent changes in language include use of the

HHassett.indbassett.indb 299299 111/8/121/8/12 9:279:27 PMPM 300 RETHINKING COMPETITIVENESS

phrase “in person,” which has come to include interactions that are merely synchronous rather than actually in person—a conversation in this usage would switch from e-mail or text to an “in person” phone call. Similarly, the phrase “off line” has come to indicate communications that occur in an alternate setting in addition to the standard description of the status of a machine—a conversation taken off line is one carried out separately from the current context (such as a group discussion). The new usage of competitiveness as referring to well-being rather than to competition per se should be seen as one of many language changes. The key for policymaking is to keep this new meaning in mind and to focus on the ultimate impacts of policies, and not on their relation to other countries in the many instances when such a comparison is of limited importance. The focus of economic policymaking would then remain on the ultimate objectives, such as growth of income, consumption, and employment; distribution; and stability.

Notes 1. For readability, I do not enclose each use of the word competitiveness in quo- tation marks. 2. Toder looks at competition for what he terms “five possible areas in which zero-sum competition may exist: competition for (1) labor supply, (2) financial and physical capital, (3) corporate residence and intangible capital, (4) tax revenues from multinational corporations, and (5) natural resources” (21). 3. Reports from the CCMR can be found at http://capmktsreg.org. 4. An example of this can be seen in the statement by Autor and Katz (1999) that research on wages and inequality had “literally exploded” in the previous ten years (1465). The authors obviously do not mean to describe a situation of considerable danger to economics graduate students but instead to emphasize the enormity of the increase in research.

References Autor, David, and Lawrence Katz. 1999. “Changes in the Wage Structure and Earn- ings Inequality.” In Handbook of Labor Economics, vol. 3, ed. O. Ashenfelter and D. Card. Amsterdam: Elsevier. Baily, Martin N., and Robert Z. Lawrence. 2006. “Competitiveness and the Assessment of Trade Performance.” In C. Fred Bergsten and the World Economy, ed. Michael Mussa. Washington, D.C.: Peterson Institute for International Economics. http:// www.piie.com/publications/chapters_preview/3979/10iie3979.pdf.

HHassett.indbassett.indb 300300 111/8/121/8/12 9:279:27 PMPM INTERNATIONAL COMPETITIVENESS 301

Bennett, Herman, Julio Escolano, Stefania Fabrizio, Eva Gutiérrez, Iryna Ivaschenko, Bogdan Lissovolik, Marialuz Moreno-Badia, Werner Schule, Stephen Tokarick, Yuan Xiao, and Ziga Zarnic. 2008. Competitiveness in the Southern Euro Area: France, Greece, Italy, Portugal, and Spain. IMF Working Paper 08/112, International Mon- etary Fund. Washington, D.C., April. Carlin, Wendy, Andrew Glyn, and John van Reenen. 2001. “Export Market Perfor- mance of OECD Countries: An Empirical Examination of the Role of Cost Com- petitiveness.” Economic Journal 111:128–62. Cerra, Valerie, Jarkko Soikkeli, and Sweta C. Saxena. 2003. “How Competitive Is Irish Manufacturing?” Economic and Social Review 34(2):173–93. Daniels, David. 2011. “In Dodd-Frank’s Shadow: The Declining Competitiveness of U.S. Public Equity Markets.” Harvard Business Law Review 1:56–59. Durand, Mattine, and Claude Giorno. 1990. Indicators of International Competitiveness: Conceptual Aspects and Evaluation. Organisation for Economic Co-operation and Development, Paris. http://www.oecd.org/dataoecd/40/47/33841783.pdf. Feenstra, Robert C., and Andrew K. Rose. 1997. Putting Things in Order: Patterns of Trade Dynamics and Growth. NBER Working Paper 5975. National Bureau of Eco- nomic Research, Cambridge, MA. Immelt, Jeffrey R. 2011. “A Blueprint for Keeping America Competitive.” Wash- ington Post, January 21. http://www.washingtonpost.com/wp-dyn/content/article/2011/01/20/AR2011012007089.html. Kahneman, Daniel, and Alan B. Krueger. 2006. “Developments in the Measurement of Subjective Well-Being.” Journal of Economic Perspectives 20(1):3–24. Krugman, Paul. 1994. “Competitiveness: A Dangerous Obsession.” Foreign Affairs 73:28–44. Marsh, Ian W., and Stephen P. Tokarick. 1994. Competitiveness Indicators: A Theoreti- cal and Empirical Assessment. IMF Working Paper 94/29. International Monetary Fund, Washington, D.C., March. Mussa, Michael. 2002. Argentina and the Fund: From Triumph to Tragedy. Policy Analy- ses in International Economics No. 67. Washington, D.C.: Institute for Interna- tional Economics. Neary, J. Peter. 2006. Measuring Competitiveness. IMF Working Paper 06/2009. Inter- national Monetary Fund, Washington, D.C., September. Soikkeli, Jarkko, and Valerie Cerra. 2002. How Competitive Is Irish Manufacturing? IMF Working Paper 02/160. International Monetary Fund, Washington, D.C., September. Toder, Eric, 2012. International Competitiveness: Who Competes against Whom and for What? Urban-Brookings Tax Policy Center, Washington, D.C., January 10. http:// taxpolicycenter.org/UploadedPDF/412477-international-competitiveness.pdf. Turner, Philip, and Jozef Van ’t dack. 1993. Measuring International Price and Cost Competitiveness. BIS Economic Paper No. 39. Bank for International Settlements, Basel, Switzerland, November.

HHassett.indbassett.indb 301301 111/8/121/8/12 9:279:27 PMPM 302 RETHINKING COMPETITIVENESS

U.S. Department of Commerce. 2012a. The Competitiveness and Innovative Capacity of the United States. http://www.commerce.gov/sites/default/files/documents/2012/ january/competes_010511_0.pdf . ———. 2012b. A Roadmap for Strengthening U.S. Competitiveness. http://www.c ommerce.gov/Competes. White House. 2006. American Competitiveness Initiative. http://georgewbush-white house.archives.gov/stateoftheunion/2006/aci/aci06-booklet.pdf. ———. 2011a. Presidential Memorandum: Government Reform for Competitive- ness and Innovation. http://www.whitehouse.gov/the-press-office/2011/03/11/ presidential-memorandum-government-reform-competitiveness-and-innovation. ———. 2011b. President’s Council on Jobs and Competitiveness Announces Indus- try Leaders’ Commitment to Double Engineering Internships in 2012. http://www. whitehouse.gov/the-press-office/2011/08/31/president-s-council-jobs-and- competitiveness-announces-industry-leaders. ———. 2011c. President Obama Presented Ideas to Accelerate Job Growth and America’s Competitiveness at Jobs Council Meeting. http://www.whitehouse.gov/ blog/2011/06/13/president-obama-presented-ideas-accelerate-job-growth-and- america-s-competitiveness. ———. 2011d. Materials Genome Initiative for Global Competitiveness. http://www. whitehouse.gov/sites/default/files/microsites/ostp/materials_genome_initiative- final.pdf. World Economic Forum. 2011. Global Competitiveness Report 2011–2012. Zingales, Luigi. 2007. Is the U.S. Capital Market Losing Its Competitive Edge? ECGI Work- ing Paper 192. European Corporate Governance Institute, Brussels. November.

HHassett.indbassett.indb 302302 111/8/121/8/12 9:279:27 PMPM Index

A Nation at Risk (report), 69 American Competitiveness Initiative, “A Pure Theory of Local Expenditures” 297 (Tiebout), 6–8 American Recovery and Reinvestment “A Roadmap for Strengthening U.S. Act, 69 Competitiveness” (report), 296 Anderson, Patricia M., 177 Absolute advantage, 101, 227, 229, 230 Anthony, Denise L., 164 See also Comparative advantage Antidumping duties, 227, 269 theory Apparel industry supply chains, 263 Academic Ranking of World Universities Apple, 141, 225–26, 261–63 (ARWU), 81 Approaches to Improve the Competitive- Accelerated depreciation, 48, 49, 61 ness of the U.S. Business Tax System Accidents and life expectancy, 210–11 for the 21st Century (Treasury Achievement gaps, educational, 71, report), 34 72–78, 83–86 Argentina, exchange rate example, 287 See also Education and global Arrow, Kenneth, 144 competitiveness Atkinson, Anthony B., 15 “Adaptation” effect, 299 Automation machinery and investment Administrative costs in health care, in labor, 112 168–69 Aydemir, Abdurrahman, 111 Advanced-technology products (ATP), 257–60 Baicker, Katherine, 156, 166, 174 Affordable Care Act, 177, 180 Baily, Martin N., 289–90 AFTA (ASEAN Free Trade Agreement), Balassa, Bela, 230 245, 250 Balassa-Samuelson effect, 283, 287 Agglomeration forces v. dispersion Baldwin, Richard, 227, 241 forces, 236–38, 243–44 Banking systems and tax havens, 21, 22 Allowance for Corporate Equity system, Barﬁ eld, Claude, 225, xvi 47 Base broadening (tax), 38–39, 43–44

303

HHassett.indbassett.indb 303303 111/8/121/8/12 9:279:27 PMPM 304 RETHINKING COMPETITIVENESS

Bayer, Patrick, 11 Business taxation (continued) Behavioral effect of modern health care, sector-specifi c benefi t arguments, 162–63, 176, 220n34 43–44 Bennett, Herman, 286, 288 U.S. corporation income tax, 44–45 Berger, Thomas, 3 See also Corporate tax competition, Bertrand competition, 102 global; Tax policy perspectives “Bid function” and household sorting, Buyer-driven supply chains, 260, 13–15 263–64 Biggs, Stephen, 9 Bird, Richard, 42 California and immigrant labor, 95, Birth, life expectancy at, 210–13 105–6 Birth rates and immigration policies, Campaign fi nance restrictions, 201 126–27 Canada Blanco, Luisa, 22 cross-national group ties and trade, Bleakley, Hoyt, 176 104–5 Blonigen, Bruce, 274–75 health care system, 165, 168, 182–83 Bogart, William, 15 immigrant skill composition, 125, Bonus depreciation, 48, 55 126 Borjas, George J., 111 postsecondary degree completion Bracey, Gerald, 82 rates, 80 Bradbury, Katharine L., 14 U.S. trade with, 232, 233 Bribery in health care systems, 208, wage effects of immigration, 111 209 Canada Health Act, 183 Brill, Alex, 60 Cancer survival rates, 163, 208–9, Brinkley, Garland L., 176 214–15, 216t Bristow, Gillian, 3 Capital, human, and economic growth, Buchmueller, Thomas, 177 79, 82–83, 85–86 Burden, tax Capital as factor endowment, 232 defi nitions and perspective, 42–43, Capital assets, tax treatments of, 49 47 Capital markets, competition for, and global location choice, 150 293–95 and immigration, 98–99, 113, 114, Capital tax competition, 21–23 122 Capitalization of tax and service differ- Bush, George W., 297 entials in Tiebout model, 8–11, Business service tasks, 260–61 12–13 Business taxation Card, David, 111 alternatives to U.S. system, 46–47 Carlin, Wendy, 286 defi nitions and measurement of, Cascade dynamic of innovations, 42–43 139–40 income v. consumption basis argu- Casella, Alessandra, 104 ments, 43 Cebula, Richard J., 10 and investment disincentives, Cell phone technology, 143, 241, 263 47–49 Cellini, Roberto, 2

HHassett.indbassett.indb 304304 111/8/121/8/12 9:279:27 PMPM INDEX 305

Cerra, Valerie, 285 Communication technology, advances Chandra, Amitabh, 156, 166, 174 in, 5, 225, 227, 240, 241, 252 Chen, Duanjie, 55–56 Comparative advantage theory Chernew, Michael E., 156, xvi health care perspective, 158–59, China 173, 217n5 global trade, role in Ricardian model, 1, 101, 229–30, cross-national group ties and 231f, 242 trade, 101, 105 trade perspective, 229–30, 230–32, high tech arena, 257–60 233, 234–38 policy approaches, 245, 252–53, v. absolute advantage, 101, 227, 257–58, 268–69 229, 230 processing trade regime, Comparative effectiveness review 253–57 (CER), 218n14 immigrants to U.S., 120 Compensation effects (health insur- innovation adoption in, 151–52 ance), 170–74 and intellectual property rights, 136, Competition 149 among health care providers, postsecondary enrollment rates, 81 167–68 Shanghai province, student achieve- among localities (See Tiebout comment in, 68, 70, 77 petition model) Christensen, Kevin, 42 Bertrand competition, 102 Chronic disease management, 176 and innovation adoption, 141, Clausing, Kim, 60, 61 143–48 Cline, Robert, 42 See also Firms, competition among Clustering, industry, 235, 236–37, 238, Competitive advantage, concept of, 2 243 The Competitive Advantage of Nations COBRA, 177 (Porter), 238 Coinsurance, See Cost sharing in public Competitiveness, global health systems capital markets, competition for, College enrollment rates, 79 293–95 See also Postsecondary education, deﬁ nitions global comparisons education perspective, 70 Commerce, U.S. Department of, 292, health care perspective, 157–58, 296 159–61, 196–97 Commission on National Aid to Voca- immigration perspective, 95–96 tional Education, 69 overview, 2–4 Committee on Capital Markets Regula- and policy choices, 278–82 tion (CCMR), 294 tax perspective, 32–36 Common Travel Area (UK), 5 trade paradigm perspective, Commonwealth Fund analysis of 227–28, 234–38 health care systems, 198–99 interjurisdictional, 18–23 Communication services business, internal motivations for, 281–82 260–61 for key industries, 227–28

HHassett.indbassett.indb 305305 111/8/121/8/12 9:279:27 PMPM 306 RETHINKING COMPETITIVENESS

Competitiveness, global (continued) Corporate tax competition, global measuring corporate profi t and policy deci- indicators for, overview, 280–82, sions, 41–42 xvii GDP, effect on, 56–58 investment-related indicators, marginal effective tax rate on invest- 290–91 ment (METR), 54–56 living standard indicators, revenues, effect of taxation on, 58, 291–93 60–62 measuring, 2, 35–37, 279–80 statutory rates, effects and compari- trade-related indicators, sons, 54, 59t 283–90 Tiebout model, 18–23 national level v. fi rm level, 1–3 U.S. corporation income tax, 44–45 and productivity, 1–3, 281–82 (See also Business taxation) theory, historical overview, 1–5 See also Business taxation; Tax policy See also Education and global perspectives competitiveness; Health care Corrado, Carol, 134–35 and global competitiveness; Cortes, Patricia, 112 Immigration and global Cost of capital and tax systems, 45, competitiveness; Intellec- 47–48, 53, 55 tual property and economic See also Investment incentive and competitiveness; Tax policy tax systems perspectives and competitive- Cost sharing in public health systems ness; Tiebout competition France, 185–86 model (foot voting); Trade Japan, 184 theory and competitiveness; Medicare (U.S.), 179 Zero-sum v. mutual benefi t United Kingdom, 181 concept Counterfeiting and piracy, 138 “Competitiveness: A Dangerous Countervailing duties, 227, 269 Obsession” (Krugman), 1–2 Country–country comparisons and Components, trade in, See East policy decisions, 41, 55, 72, 293, Asia and parts/components 297–98 trade Couverture Maladie Universelle (CMU), Comprehensive business income tax 185–86 (CBIT), 46–47 Cross-national group ties and trade, 104–5 Consumer price indices (CPI), 284 “The Current Case for Industrial Policy” Coordination services business, 260, (Krugman), 238 261, 264 Coordination technology, advances in, Daniels, David, 294 227, 240, 241, 244, 264 Danzon, Patricia M., 165 See also Communication technology, Darby, Michael R., 118 advances in Data-management software, 241 Copayments, See Cost sharing in public De Mooij, Ruud A., 19 health systems Death rates, 210–11, 212–13

HHassett.indbassett.indb 306306 111/8/121/8/12 9:279:27 PMPM INDEX 307

Dedrick, Jason, 261–63 Durand, Mattine, 284, 288 Deductions, tax, 39, 44–45, 46, 47, 55, 61 Duties, 227, 253, 269 Deferral of residual tax, 52 East Asia and parts/components trade “Defi cit commission,” 32 China Degree completion rates, 79–80 cross-national group ties and Demand patterns and competitiveness, trade, 101, 105 288–89 high tech arena, 257–60 Denison, Edward, 134 policy approaches, 245, 252–53, Depreciation and tax policies, 47, 257–58, 268–69 48–49, 55, 61 processing trade regime, 253–57 Desai, Mihir, 22, 53 overview of production networking, Devereux, Michael P., 20, 61 245–52 “Diaspora externality,” 104 and U.S., comparisons, 257–60, Differentials, capitalization of in 264–67 Tiebout model, 8–11, 12–13 East Asian economies and innovation, Disaggregation, 228 136 See also Unbundling and evolution Eaton, Jonathan, 97, 100, 103 of trade Economic growth, as gauge for com- Discrimination (racial) and migration, 10 petitiveness, 280–82 Disease management, chronic, 176 Economic health indicators, 33–35, Disincentives, tax 288–89 balancing, 40–41 See also GDP (gross domestic product) to investment, 47–49, 54–56, 58–62 The Economist, 34 Dispersed sourcing, defi nition, 250 Edel, Matthew, 9 Dispersion forces v. agglomeration Ederveen, Sjef, 19 forces, 236–38, 243–44 Edmonston, Barry, 114 Distributional effects of trade, 231 Education and global competitiveness Diversity and educational performance, attainment 77–78 global comparisons, 78–82 Dividends, taxation of, 46–47 (quantity) v. quality, 83–84, 87 Dixit, Avinash, 233 U.S. employment share, 106–8, “Do good things” principle, 282 109f, 110f Dodd-Frank Wall Street Reform and fi nancial expenditures, 75–77, 87 Consumer Protection Act (Dodd- high school graduation rates, 79, 111 Frank), 294 math and science achievement gaps, Dollar, value of, 289–90 71, 72–78, 83–86 Domestic Production Activities Deduc- overview, 68–72, xiv–xv tion, 39 policy perspective Dortum, Samuel, 97 lessons learned, 86–88 Double taxation, 45 recommendations, 88–90 Dowding, Keith, 9 postsecondary education Duncan, Arne, 68, 69 degree completion rates, 79–80 Duncombe, William D., 14 enrollment rates, 81–82

HHassett.indbassett.indb 307307 111/8/121/8/12 9:279:27 PMPM 308 RETHINKING COMPETITIVENESS

Education and global competitiveness Europe/EU (continued) postsecondary education (continued) immigrants to, skill distribution, foreign-born students and U.S. 124–26 productivity, 99, 115–22 innovation and productivity v. U.S., and immigrant impact, 106–8, 146–47, 149 109f, 110f passport-free zones, 5 and public services, effi ciency of, 14 policy focus and competitiveness, quality of education and economic 126, 278–79, 284 growth, 82–86 tax and welfare policies, 126 socioeconomic factors, 77–78 See also individual countries Emanuel, Rahm, 156 Exchange rates and competitiveness, Employer-provided health care 284, 286–90 and competitiveness, 157 Exchange value of dollar, 289–90 in Japan, 184 Exit exams in education, 88 labor market, impact on, 170–74 Expense write-offs in tax systems, 47, macroeconomic effects, 176–78 48, 55 management costs of, 168 Export market and competitiveness, overview (U.S.), 180 285–86 See also Health care and global Export supply capacity, 102–3, 104, 122 competitiveness; Insurance, External IRS (Increasing Returns health care to Scale), 234, 235, 236–37, Employer-sponsored permanent visas 242–44 (green cards), 115, 120–22, Externalities, See Spillover effects 128 Employment share of immigrants, Face the Nation, 156 106–8, 109f, 110f Factor content/endowments and eco- Engineering fi elds, and foreign-born nomic growth, 85, 231–32 students, 117–18 Falck, Oliver, 14 Enrollment rates, postsecondary, 79 Family sponsored visas, 121–22, 123, See also Postsecondary education, 127–28 global comparisons Feenstra, Robert C., 290 Enrollment rates, presecondary, 113, Ferrantino, Michael J., 259 176 Ferreira, Fernando, 11 Entrepreneurship, 14, 44, 149 Fertility rates and immigration policies, Equity, taxation of, 45, 47 126–27, 213 Equity in health care quality, 198–99 Firms, competition among Europe/EU innovation-based v. price-based, components trade in, 250, 251 143–48, 151 exit exams, 88 measuring, 279–80 health care in, 169, 213, 214t national level v. fi rm level, 1–3 high school graduation rates, 79 size of fi rm high-tech market in, 144 and innovation propensity, 144–45 immigrants from, 105 and per-unit cost of goods, 233

HHassett.indbassett.indb 308308 111/8/121/8/12 9:279:27 PMPM INDEX 309

Firms, competition among (continued) Formulations, prescription drug, 165, and tax systems, 52–53 207 See also Intellectual property and Foshay, Arthur W., 71 economic competitiveness; Fox, William, 16 Locational choices France, health care system in, 185–86, First unbundling, See Unbundling and 206–7 evolution of trade Free trade areas, 5, 245, 250 Fiscal policy discipline, 35, 40 Freeman, Richard, 81, 85, 117 Fischel, William A., 9 “Full package” coordination, 264 Fisher, Elliott S., 164 Furukawa, Michael F., 165 Flat tax systems, 46, 47 “Flattening” of world Gaulier, Guillaume, 256–57 communication technology, Gauthier-Loiselle, Marjolaine, 118 advances in, 5, 225, 227, 240, GDP (gross domestic product) 241, 252 and education achievement, 77, coordination technology, advances 83, 85 in, 227, 240, 241, 244, 264 growth comparison, 212, 213t and locational choices, 4–5, 16–17, and health care spending, 156–57, xiii, xiv 169, 198–99, 209, 212 transportation, advances in, 5, 225, and innovation, investment in, 227, 228, 240–41 147, 151 See also Unbundling and evolution as prosperity indicator, 37, 278 of trade and tax systems, 56–58, 59t, Foley, Fritz, 22, 53 60–62 “Foot voting,” 6–8, xiv General Educational Development See also Tiebout competition model (GED) certiﬁ cates, 79 Foreign Affairs, 1 General equilibrium effects, 97, 100, Foreign direct investment (FDI) 158, 173 in China hi-tech trade regime, General Motors, 175 257–60 Giannetti, Mariassunta, 14 and competitiveness measures, Gilmour, Alan, 170 290–91 Giorno, Claude, 284, 288 and innovation, 134, 136, 143–44 Gittlesohn, Alan, 164 and intellectual property protection, “Give Heckscher and Ohlin a Chance!” 138 (Wood), 232 and tax systems, 19, 56 Global Competitiveness Index (GCI), Foreign market pricing, 102 35–37, 70 Foreign-born students and U.S. pro- Global Competitiveness Report, 2–3, 280, ductivity, 99, 115–22 292–93 See also Postsecondary education, Global general equilibrium framework, global comparisons 97 Foreign-invested enterprises (FIEs), Global integration, See “Flattening” of 255, 258, 260 world

HHassett.indbassett.indb 309309 111/8/121/8/12 9:279:27 PMPM 310 RETHINKING COMPETITIVENESS

Global supply chains, See Global value Health care and global competitiveness chains (GVCs) economic impacts of U.S. policy Global value chains (GVCs) (continued) evolution of, 225–27, 226–27, 239– overview, 169–70 41, 260–61, xvi–xvii retiree costs, 175 and high tech innovation, 143–44, worker health and productivity, 149 176 policy implications and New Trade effi ciency/performance issues Theory, 232–38 administrative spending, 168–69 U.S production networks, 260–67 contemporary analyses, 197–202 See also Trade theory and “medical arms race,” 168 competitiveness outcomes comparisons, 162–64, Glyn, Andrew, 286 209–15 Goldhaber, Dan, 89 overviews, 157–58, 161–62, 196 Goldin, Claudia, 79 price disparities, analysis of, Gompers, Paul, 14 164–68 Gordon, Roger H., 57 productivity v. competitiveness Gould, David M., 104 concepts, 159–61 Gravelle, Jane G., 61 spending/service comparisons, 199– Greece, health care system in, 208 200, 202–9 Green cards, 115, 120–22, 128 structure/services, system profi les Grogger, Jeffrey, 125–26 Canada, 182–83 Gross domestic product (GDP), See France, 185–86, 206–7 GDP (gross domestic product) Greece, 208 Gross receipts tax, 38–39 Italy, 207 Gruber, Jonathan, 172, 177 Japan, 183–85 quality, 163 Haddad, Mona, 250–51 Spain, 207–8 Hall-Rabushka fl at tax, 46, 47 spending, 156–57, 162–63 Hamilton, Bruce W., 12–13 United Kingdom, 181–82, 208–9 Hanson, Gordon H., 95, 114, 125–26, U.S., 178–80 xv U.S. system and competitiveness, Hanushek, Eric, 83–85 overview, 156–59, 187–88, xvi Hassett, Kevin, 1, 60, xiv utilization, 164, 165–66 Havens, tax, 21–22, 50–52 Health effect of modern medicine on Hawkins, Brett W., 10 behavior, 162–63, 176, 220n34 Head, Keith, 104–5 Heblich, Stephen, 14 Health care, public, See Public health Heckscher-Ohlin (H-O) model of trade care theory, 230–32 Health care and global competitiveness High school graduation rates, 79, 111 economic impacts of U.S. policy Higher education, global expansion of, labor costs, 170–74 81–82 macroeconomic effects, 176–78 See also Postsecondary education

HHassett.indbassett.indb 310310 111/8/121/8/12 9:279:27 PMPM INDEX 311

High-tech industry Idea/information exchange China’s trade regime, 257–60 idea-based economies, 146–48, 151 employer-sponsored visas, 119–20 and immigration, 99–100 global trade in, 257–60 and innovation, 133, 136 and globalized production, 143–44 knowledge spillover, 235 and immigration, dependence on, Illegal immigrants, impact on U.S. 95, 122–23 economy, 98–99, 123, xv and innovation, 101, 149 Immigration Act of 1990, 121 innovation rates in, 136 Immigration and global and U.S. education system, 86 competitiveness See also Technology fi scal impacts Hines, James R., Jr., 19, 22, 53 illegal immigrants, impact on Homeland Security, U.S. Department U.S. economy, 98–99 of, 119–20 low-skilled labor and market Hong Kong and global trade, 245, share, 100–103 255–56 overview/implications, 113–14, Hookworms, 176 122–23, 127–28 Horizontal intra-industry trade, 232– prices, 112–13 33, 234f, 236 skill distribution and employ- See also Vertical intra-industry trade ment share, 106–8 Horizontal v. vertical integration and taxes and social services, 113 tax incentives, 38 trade costs, 104–6 Hospitals foreign-born distributions, 123, 124 beds, number of, 204, 205t illegal immigrants, 123, xv Canada system profi le, 183 impact and implications for U.S. competition among, 168 policy, 97–100 fee comparisons, services, 203–4, 205t innovation and economic growth, France system profi le, 186 103–4 Japan system profi le, 185 overview and theory, 95–97, 100, xv UK system profi le, 182 policy perspectives, 103–4, 114, U.S. system .profi le, 180 124–28 Housing prices visas and immigration, 98, 112–14, 122 employer-sponsored permanent labor supply effect, 172 (green card), 115, 120–22, and migration choices, 8–11, 128 12–13, 18t F-1 (student), 115–16 Hoxby, Caroline M., 11, 85 H-1B (work), 19, 95, 115, Hubbard, Glenn, 1, xiv 118–20 Hulten, Charles, 134–35 number issued in U.S., 115–17, Human capital and economic growth, 118–20, 121–22 79, 82–83, 85–86, 114 policy recommendations, 127–28 Hungerford, Thomas L., 61 wages, effects on, 107f, 108, 111–12 Hunt, Jennifer, 118 Immigration shocks, 96–97

HHassett.indbassett.indb 311311 111/8/121/8/12 9:279:27 PMPM 312 RETHINKING COMPETITIVENESS

Income and desired public service Innovation and economic growth level, 9, 13 (continued) Income shifting, 50–52, 53, 58, 60, 61, overview and conclusions, 148–52, 150 xv–xvi Incorporated businesses and U.S. tax and patenting, 99 system, 44–45 R&D approach and immigration, India, 81, 120 81, 103–4, 117–18, 122, 149 Industrial policy, strategic See also Intellectual property and of China, 257 economic competitiveness and competitiveness theory, 226, Innovation-based v. price-based compe- 228, 234–38, 239, 243–44 tition, 143–48, 151 and tax systems, 39–40 Input pricing and immigration, 101–2 Industry clustering, 235, 236–37, 238, Input sharing, 235, 238, 243 243 Insurance, health care Industry-speciﬁ c competitiveness, administrative costs, 169 39–40, 97, 279–80 Medicaid (U.S.), 179 Infant mortality comparisons, 212–13 Medicare (U.S.), 164, 167, 178–79 Inﬂ ation and competitiveness indica- price offset effects, 165, 167–68 tors, 286–88 See also Employer-provided health care Information exchange, See Idea/infor- Intangible assets and economic growth, mation exchange 134–35, 151 Information technology investment, See also Intellectual property and 99–100, 133, 134–35, 136, economic competitiveness 147–48 Integration, global, See “Flattening” of Initial public offerings (IPOs) and world competitiveness indicators, Intellectual property and economic 293–95 competitiveness Innovation and economic growth innovation, value of, 133–35 development and adoption knowledge economy and productiv- and competitive advantage, ity growth, 146–48 143–48 overview, 132, xv–xvi dynamics of, 139–42 patenting economic value of innovation, cross-border, 150 133–35 and economic growth, 136 economic value of intellectual immigrant impact on, 99, 118, 120 property rights, 135–38 and IP-dependent industries, and education level, 83 145–46 immigration, impacts of rights enforcement foreign-born students and U.S. U.S., 101 productivity, 99, 115–22 value of to innovation, 135–38, overview, 103–4, 114 144–45, 149–50 “lotteries” concept and productivity, See also Innovation and economic 101–2 growth

HHassett.indbassett.indb 312312 111/8/121/8/12 9:279:27 PMPM INDEX 313

Interest, taxation of, 46–47 Kafoglis, Milton Z., 10 Interest groups and health care policy, Kahn, Matthew, E., 12 200–201 Kahneman, Daniel, 289–99 Internal IRS (Increasing Returns to Katz, Lawrence, 79 Scale), 232–33, 234f Kerr, William, 119–20 International Institute for Management Knight, Brian G., 11–12 Development (IMD), 2 Knowledge capital, investment in, International rivalry v. competitiveness, 99–100, 133, 134–35, 136 161 See also Innovation and economic Internet, 139–40, 144, 241 growth Invention, See Innovation and eco- Knowledge transfer, See Idea/ nomic growth information exchange Investment incentive and tax systems, Koopman, Robert, 250, 253, 255 47–49, 53, 54–56, 58–62 Kortum, Samuel, 97, 100, 103 Investment-related competitiveness Kraemer, Kenneth L., 261–63 indicators, 283–91 Kreuger, Alan B., 289–99 IPad/iPod/iPhone (Apple), 225–26, Krugman, Paul, 1–2, 70, 158, 160–61, 261–63 228, 238, 278, 281, 292, 297 Ireland, tax rates in, 51 IRS (Increasing Returns to Scale) Labor external, 234, 235, 236–37, cost of, 48, 101, 166 242–44 costs as measure of competitiveness, internal, 232–33, 234f 284–85 and policy implications of trade demographics of labor force, theory, 237–38 126–27 Italy, health care system in, 207 as factor endowment, 232 health care impacts on, 170–74, Jacobellis v. Ohio, 34 176–77 Japan high-skilled v. low-skilled immi- birth rates and working-age popula- grants, effects of, 96–99 tion, 126 market experience and employment and global trade, 245, 250–51, 262 share, 106–8, 109f, 110f health care in, 183–85 regulation and innovation, 142, and innovation-based competition, 146, 150 147–48, 149 See also Wages market shares, 144 Labor, U.S. Department of, 120 regulatory environment, 146 Labor market pooling, 235 Jensen, Matthew H., 1, 225, xiv, xvi Ladd, Helen F., 14 Job lock, 177 Laffer, Arthur, 58, 60 John, Peter, 9, 10 Lall, Sanjaya, 3 Jumpstart Our Business Starts Act Land prices and immigration, (JOBS Act), 294 112–13 “Just in time” manufacturing, 241, 264 Language, global integration of, 5

HHassett.indbassett.indb 313313 111/8/121/8/12 9:279:27 PMPM 314 RETHINKING COMPETITIVENESS

Language, modern vernacular in policy Ma, Alyson, 274–75 discussion, 299–300 Madrian, Brigitte, 177 Latin American economies, lack of Manton, Kenneth G., 211 growth of, 136 Marginal effective tax rate on invest- Lawrence, Robert Z., 289–90 ment (METR), 54–56, 59t Lee, Young, 57 Market bidding and household sorting, Lemoine, Francoise, 256–57 13–15 Leontief, Wassily, 231–32 Market power in U.S. health care, 167 Leontief Paradox, 231–32 Marsh, Ian W., 284 Lerner, Josh, 14 Marshall, Alfred, 235 Levy, Philip I., 156 Marshall, Mary Paley, 235 Lewis, Ethan, 112 Maskus, Keith, 117 Li & Fung, 264 “Materials Genome Initiative for Life expectancy comparisons, 163, Global Competitiveness” 210–12 (report), 296 Lincoln, William, 119–20 Maternity benefi ts and wages, 172 Linden, Greg, 261–63 Mathematics, achievement gaps in, Living standard indicators of competi- 72–78, 83–86 tiveness, 291–93 McCain, John, 156 See also GDP (gross domestic product) McGlynn, Elizabeth A., 176 Locational choices McKinsey Global Institute, 162–63, agglomeration forces v. dispersion 165 forces, 236–38, 243–44 McMillan, Robert, 11 and capital market competition, Medicaid (U.S.), 179 293–95 “Medical arms race” phenomenon, and comparative advantage, 237 168 “fl attening” of world, 4–5, 16–17, Medical “tourists,” 216 xiii, xiv Medicare (Canada), 182–83 and intellectual property concerns, Medicare (U.S.), 164, 167, 178–79 132, 150 Medicare Advantage program, 179 investment incentive and tax sys- Medications, adherence to, 176 tems, 47–49, 53, 54–56, 58, Medicine, See Health care and global 61, 62 competitiveness; Prescription and tax environment, 279 drugs/pharmaceuticals See also Clustering, industry; Mercantilism, 1, 253, xiii Tiebout competition model Migration choices, global (foot voting) patterns of, modern, 5 “Location-specifi c rents,” 64n15 and Tiebout model, implications of, Lockwood, Ben, 20 18–23 Lotteries, innovation, 100–101 trade, impact on, 104–6 Lotteries, state, and foot voting, 11–12 See also Immigration and global Lotteries, visa, 119–20 competitiveness; Locational Luedemann, Elke, 14 choices

HHassett.indbassett.indb 314314 111/8/121/8/12 9:279:27 PMPM INDEX 315

Migration choices, local No Child Left behind Act, 90 and externalities, effect of, 13–14 Nominal exchange rates, 284 and housing market, 8–11 Nonprice rationing in health care, housing prices, 12–13, 18t 202–3, 206, 208, 209, 216 spillover effects, 14–15, 17 “Nonrival goods,” 137–38 Military superiority and global compe- North American Free Trade Agreement tition, 33 (NAFTA), 5 Mintz, Jack, 55–56 Mobarak, Ahmed Musfi q, 117 Oates, Wallace, 8–9, 14 The Moment of Truth (report), 32 Obama administration, 68–69, 295– Monopolies and innovation, 142, 144 97, 299 Montout, S., 232 Obesity, 162 Motivations for trade, See Comparative OECD analysis of health care systems, advantage theory; IRS (Increasing 199–200 Returns to Scale) Offshoring of production, 239 Mucchielli, J. L., 232 See also Global value chains (GVCs) Musgrave, Richard, 6 Ohsfeldt, Robert L., 210 Mussa, Michael, 287 Oligopolies and innovation, 142, 144 Mutual benefi t v. zero-sum concept Omitted variable bias in tax compari- competitiveness theory, 1–2, 3, 5, sons, 60–61 16–17, 33, 159–60 Organization for Economic Co-opera- education perspective, 69–70, 86 tion and Development (OCED), health care perspective, 187–88 19 immigration perspective, 96–97 Out-of-school factors, 77–78 tax competition perspective, 22–23 trade paradigm perspective, 227–28, Pareto-optimality/effi ciency, 4, 15 278 See also Zero-sum v. mutual benefi t concept National Commission on Fiscal Parts subindustry in global value Responsibility and Reform, 44 chains, 243 National Defense Education Act, 69 Parts/components trade and global National Health Service (NHS) (UK), value chains, See East Asia 181–82, 208–9 Patenting Neary, J. Peter, 285, 286 cross-border, 150 Neoclassic economic theory, 4 and economic growth, 136 Network effects and technology inno- immigrant impact on, 99, 118, 120 vation, 139 and IP-dependent industries, Networks, production, See Global value 145–46 chains (GVCs) See also Intellectual property and Neubig, Thomas S., 42 economic competitiveness Neutral tax systems, 39–40 Pauly, Mark V., 166, 168 New Trade Theory, 232–38 Pecuniary external IRS, 235 9/11 and U.S. visa issuance, 116–17 Peer effects and entrepreneurship, 14

HHassett.indbassett.indb 315315 111/8/121/8/12 9:279:27 PMPM 316 RETHINKING COMPETITIVENESS

Per capita income as measure of pros- Policy perspectives (continued) perity, 37 industrial policy, strategic, 39–40, See also GDP (gross domestic 226, 228, 234–38, 239, 243– product) 44, 257 Percy, Stephen L., 10 IRS (Increasing Returns to Scale), Personal tax and corporate-source 237–38, 242–44 income, 46 language, modern vernacular in Pharmaceuticals, See Prescription policy discussion, 299–300 drugs/pharmaceuticals and New Trade Theory, 232–38 Physicians recent approaches, overview and Canada system profi le, 182–83 limitations, 295–99 fee comparisons, 203–4 See also Tax policy perspectives and France system profi le, 186, 207 competitiveness Italy system profi le, 207 Politicization of policy decisions, 201– numbers of, 204, 205t, 207 2, 238, 282 Spain system profi le, 207–8 Population growth rates and immigra- specialty v. primary care, distribution policies, 126–27 tions, 166–67, 180, 183 Porter, Michael, 2, 145, 238 UK system profi le, 182 Postsecondary education, global U.S. delivery system, 157, 178–80 comparisons “Pillars” of competitiveness, 3, 36, degree completion rates, 79–80 292–93 enrollment rates, 81–82 Piracy and counterfeiting, 138 foreign-born students and U.S. pro- PISA (Program for International Stu- ductivity, 99, 115–22 dent Assessment), 68, 72, 73 and immigrant impact, 106–8, 109f, Policy perspectives 110f China’s trade policy regime, 245, See also Visas 252–53, 257–58, 268–69 Potential productivity concept, 101 competition-based view, dangers of, Prescription drugs/pharmaceuticals 293, 297–99 formulary restrictions, 207 and competitiveness concepts/defi - plan comparisons nitions, 278–82 Canada, 183 education perspective, 86–88, France, 185 88–90 Japan, 184 Europe/EU policy focus and com- Medicare system, 178–79 petitiveness, 126, 278–79, United Kingdom, 181 284 price comparisons, 165 fi scal policy discipline, 35, 40 time to market, 163, 215–16 health care, impacts of U.S. policy President’s Council on Jobs and Com- on, 169–70, 170–74, 175, petitiveness, 295–96 176–78 Price regulation and innovation, 172 immigration, 97–100, 103–4, 114, Price suppression in health care systems, 124–28, 127–28 199–200, 206, 208, 209, 216

HHassett.indbassett.indb 316316 111/8/121/8/12 9:279:27 PMPM INDEX 317

Price-based v. innovation-based compe- Public health care (continued) tition, 143–45, 143–48, 151 France, 185–86 Pritchard, Robert S., 164 Greece, 208 Private health insurance immigrant impact on, 113 Canada, 183 Japan, 183–85 France, 186 Medicare/Medicaid (U.S.), 164, 167, Spain, 207–8 178–79 UK, 182 public hospitals (U.S.), 180 U.S., 179, 180 spending comparisons, 199–200 See also Employer-provided health United Kingdom, 181–82 care v. employee-provided, 177–78 Private v. public school competition, See also Health care and global 88–89 competitiveness “Prize” concept in competition, 160, Public services and migration/immigra- 173–74 tion, 6–8, 13, 98, 113 Procedure-oriented health care, 167 Processing trade regime of China, “Quality ladder,” 103 253–57 Production networks, See Global value Race to the Top, 69 chains (GVCs) Racial considerations, 10, 71, 73, 78 “Productivity ladder,” 103 Rationing (nonprice) in health care, Productivity v. competitiveness, deﬁ ni- 202–3, 206, 208, 209, 216 tions, 1–3, 35–36, 158, 159–61, Rauch, James E., 104, 105 281–82, 292 R&D (research & development), 81, Program for International Student 103–4, 117–18, 122, 149 Assessment (PISA), 68, 72, 73 See also Intellectual property and Property taxes/values and migration, economic competitiveness See Housing prices Real exchange rates, 284 Prosperity and competitive tax policy, Redoano, Michela, 20 35–37 Regulatory environment Protectionist trade barriers, 237–38 and innovation, 132, 133, 142, 144, Public corporations and tax systems, 45 146 Public education and locational choices, 173, 279 competition and productivity, 11 and migration choices, 12 competition with private education, policy implications for, 296–98 88–89 and productivity, 2–3 historical perspective, 78–79 and trade activity, 227, 294 immigration impact on, 113 Reinert, Erik S., 3–4 and innovation, 149 Remittance (tax) responsibility, deﬁ ni- strategies, overview, 71 tions and perspective, 42–43 Public health care Rents and intellectual property, 138 Canada, 182–83 Repatriation and residual tax payment, fee comparisons, 203 52

HHassett.indbassett.indb 317317 111/8/121/8/12 9:279:27 PMPM 318 RETHINKING COMPETITIVENESS

Research productivity, 81, 103–4, 117– Schumpeter, Joseph, 144 18, 122, 149 Schwager, Robert, 11 See also Intellectual property and Science, 117 economic competitiveness Science, profi ciency in, 72–78, 83–86, Residual tax deferral, 52 117–18 Resource utilization and factor endow- Sclar, Elliott, 9 ment, 231 Scott, Bruce, 4 Retail sales tax, 38 Second unbundling, See Unbundling Retiree health benefi ts, 175 and evolution of trade Revenues, effect of taxation on, 58, Secondary effects, See Spillover effects 60–62 Secondary schooling, evolution of Ricardian comparative advantage (U.S.), 78–79 model, 1, 101, 229–30, 231f, “Secret” inventions, 137 242 Sector-specifi c competitiveness, 39–40, Ricardo, David, 1, 228 97, 279–80 Rice, Thomas, 167 Self-insured health plans, 180 Ries, John, 104–5 Service tasks and global value chains, Rising Above the Gathering Storm 260–61 (report), 81–82 Shanghai province, student achieve- “Rival goods,” 137–38 ment in, 68, 70, 77 Rockefeller Sanitary Commission, Shapiro, Robert J., 132, xv 176 Sheiner, Louise, 172 Rogers, Cynthia, 22 Shifting of income, 50–52, 53, 58, 60, Romer, Paul, 133 61, 150 Rose, Andrew K., 22, 290 Shipping costs, 228 Ross, Stephen L., 13–14 See also Transportation, advances in and global integration Safety net providers, health care, 180, Sichel, Daniel, 134–35 181, 184, 186 Silicon Valley and immigration, 105–6 Saiz, Alberto, 113 Simonov, Andrei, 14 Sala-i-Martin, Xavier, 70 Single Market (free trade area, EU), 5 Sales tax, 38 Single-payer health care systems, 208 Samuelson, Paul, 6 Skill distribution of immigrants, 96–99, Sarbanes-Oxley Act, 294 100–103, 106–8, 124–26 Satterthwaite, Mark A., 168 Slaughter, Matthew J., 114 Saxena, Sweta C., 285 Slemrod, Joel, 32, 56–57, xiv Scharfstein, David, 14 “Slicing” of production process, 226 Schengen Agreement (EU), 5 Small open-economy taxation, 50 Scheve, Kenneth, 114 Smith, Adam, 1, 13 Schiff, Nathan, 11–12 Smith, James P., 114 Schneider, John E., 210 Smith-Hughes Act, 69 Schneider, Mark, 16 Smoking, 162 Schott, P. K., 257 Soci, Anna, 2

HHassett.indbassett.indb 318318 111/8/121/8/12 9:279:27 PMPM INDEX 319

Social services and migration/immigra- Supply chains, globalization of, 225 tion, 6–8, 13, 98, 113 See also Global value chains (GVCs) Socioeconomic status and educational Swagel, Phillip, 278, xvii achievement, 77–78, 86 Swenson, Deborah, 104–5 Soikkeli, Jarkko, 285 Solow, Robert, 134, 135, 140 Tanner, Michael, 207 Somin, Ilya, 5, 18–19 Tariffs, 102, 104, 228 Sood, Neeraj, 156, 174 Tax base Sørensen, Peter Birch, 60 broad-base systems, 38–39, 43–44 Sorting (group) in housing choice, business taxation, 47, 62 13–15 erosion of by tax havens, 21–22, 51 South Asian economies, 136 investment disincentives, 58–61, 60 South Korea, innovation adoption, 134 territorial v. worldwide taxation, 52 South Korea-United States Free Trade Tax burden Agreement, 5 defi nitions and perspective, 42–43, Spain, health care system in, 207–8 47 Special interest groups, 238 and immigration, 98–99, 113, 114, Specialization rationales in trade per- 122 spective, 229–30, 233 and intangible asset location, 150 Spiegel, Mark, 22 in small open economy, 50 Spillover effects Tax havens, 21–22, 50–52 and health care costs, 174 Tax policy perspectives and and innovation, 40, 118, 147–48 competitiveness knowledge, 235 arguments, unpersuasive, 40–42 and migration among localities, business taxation 14–15, 17 alternatives to U.S. system, State lotteries and foot voting, 11–12 46–47 Statutory corporate tax rates, 54, 59t defi nitions and measurement of, See also Corporate tax competition, 42–43 global income v. consumption basis Stewart, Potter, 31 arguments, 43 Stiglitz, Joseph E., 15, 233 and investment disincentives, Student achievement, measuring, 68, 47–49 72 sector-specifi c benefi t arguments, Stuen, Eric T., 117 43–44 Subindustries in global value chains, U.S. corporation income tax, 243–44 44–45 Subsidies, government competitiveness implications over- health insurance, 180 view, 37–40 industry protection policies, 237–38 corporate tax competition, global private education, 88–89 corporate profi t and policy deci- Supplemental health coverage, 179, sions, 41–42 182, 183, 185–86, 207, 208 GDP, effect on, 56–58

HHassett.indbassett.indb 319319 111/8/121/8/12 9:279:27 PMPM 320 RETHINKING COMPETITIVENESS

Tax policy perspectives and Technology (continued) competitiveness high-tech industry corporate tax competition, global China’s trade regime, 257–60 (continued) employer-sponsored visas, marginal effective tax rate on 119–20 investment (METR), global trade in, 257–60 54–56 and globalized production, revenues, effect of taxation on, 143–44 58, 60–62 and immigration, dependence statutory rates, effects and com- on, 95, 122–23 parisons, 54, 59t and innovation, 101, 149 Tiebout model, 18–23 innovation rates in, 136 U.S. corporation income tax, and U.S. education system, 86 44–45 information technology investment, defi nitions, 32–35 99–100, 133, 134–35, 136, and “foot voting,” 8–11, 16, 19–23 147–48 and GCI, 36 and investment in labor, 112 global perspective, 49–53 medical resources, 204, 205t and health insurance, 171, 177–78 transfer, 138, 145 and immigration, 98–99 See also Innovation and economic objectives, 37–38 growth overview/conclusions, 32, 62–63, Terms of trade and competitiveness xiv indicators, 288–90 qualifi cations, 35–37 Territorial v. worldwide taxation sys- Tax Relief, Unemployment Insurance tems, 52–53 Reauthorization and Job Creation Tessada, Jose, 112 Act of 2010, 55 Testing programs, student, 68, 72 Teacher quality improvement, 89 Tiebout, Charles, 5 Technological external IRS, 235 Tiebout competition model (foot voting) Technology capitalization of differentials, 8–11, advances in and globalization, 5, 12–13 225, 227, 228, 240–41, 244, and competition among localities, 252, 264 5–8, xiv cell phone, 143, 241, 263 and effi ciency of services, 11–12 China’s trade regime, 257–60 external effects of housing choice, coordination services business, 260, 13–15 261, 264 housing market and migration data-management software, 241 choices, 8–11, 12–13, 18t development of and competitive- implications for global arena, 17–23 ness, 290 locational decisions by fi rms, 15–16 and education development, 78, and zero-sum concept, 16–17 85 Time horizons and health care policy, and health care effi ciency, 163 200–202

HHassett.indbassett.indb 320320 111/8/121/8/12 9:279:27 PMPM INDEX 321

Toder, Eric, 291 Transportation, advances in and global Tokarick, Stephen P., 284 integration, 5, 225, 227, 228, Total compensation theory, 171 240–41 Total factor productivity (TFP), 96–97, Trauma death rates, 210–11 99, 100–101, 102, 103 Travel, global integration of, 5, 139 Trade Travel zones, passport-free, 5 costs and immigration, 104–6 Trends in Mathematics and Science export supply capacity, 102–3, 104, Study (TIMSS), 72 122 Trindade, Vitor, 105 free, multi- and bi-lateral agree- Turner, Philip, 284 ments for, 5, 245, 250 Twelve pillars of competitiveness, 3, 36, growth of, global, 5 292–93 immigrant knowledge transfer and Two-country, two-intermediate good, U.S. investment, 99–100 one-fi nal good model, 242 measures of competitiveness, 283–91 Two-good, two-country trade model, Trade balance 229–30, 242 as economic health indicator, 33, Two-way trade, 232–33, 234f 288–89 and immigration impact, 128n2, Unal-Kesenci, Deniz, 256–57 128n4 Unbundling and evolution of trade U.S.-Asia, 257–60, 264–68 fi rst (mid 1800s - mid/late 1900s) Trade barriers, protectionist, 237–38 disaggregation overview, 228 Trade in tasks, 242 Heckscher-Ohlin model, Trade theory and competitiveness 230–32 global value chains New Trade Theory and internal evolution of, 225–27, 226–27, IRS, 232–33, 234f 239–41, 260–61, xvi–xvii and policy implications of trade and high tech innovation, 143– theory, 234–38 44, 149 Ricardian model, 229–30 policy implications and New overview and policy implications, Trade Theory, 232–38 227–28 U.S production networks, 260–67 second (mid/late 1900s – present) IRS (Increasing Returns to Scale) and global value chains, 239–41 external, 234, 235, 236–37, trade in tasks and external IRS, 242–44 242–44 internal, 232–33, 234f trade in tasks and motivations of and policy implications of trade trade, 242 theory, 237–38 See also Trade theory and policy outlook, 268–70 competitiveness U.S. participation profi le, 260–67 Unemployment See also East Asia and parts/compo- and health care spending, 156, 174 nents trade; Unbundling and policy responses to, 278–79, 287 evolution of trade Unintentional injury deaths, 210

HHassett.indbassett.indb 321321 111/8/121/8/12 9:279:27 PMPM 322 RETHINKING COMPETITIVENESS

Unit labor costs and competitiveness, Wages 254–58 and competitiveness measurement, United Kingdom, health care system in, 285–86 181–82 and education development, 78 Universities, competition among, 11, health care beneﬁ t offsets, 171–72 81, 85, 101 and immigration, effects of, 97–98, See also Postsecondary education, 106, 107f, 108–9, 111–12, global comparisons 114, 122 U.S.-South Korea Free Trade Agree- and productivity potential, ment, 5 101–2 Waiting lists, medical, 202, 207, 208 Valletta, Robert, 177 Wang, Zhi, 253, 255 “Value gap” in health insurance, 172 Washington Post, 170, 173 Value of dollar, 289–90 Wayslenko, Michael, 16 Value-added tax, 38 Wealth segregation, 13–15 Van Reenen, John, 286 Wei, Shang-Jin, 253, 255 Van’t dack, Jozef, 284 Welfare programs, 9–10, 113, 126 Vaupel, James W., 211 Well-being as competitiveness indica- Venture capital markets, 101, 293–95 tor, 2, 159–61, 291, 298–99, Verdecchia, Arduino, 214 xvii Vernacular in policy discussion, Wennberg, John E., 164 299–300 West, Martin, 68, xiv Vertical intra-industry trade, 234f, 236, “Winning the future” concept, 295 239, 241 Woessman, Ludger, 83–85 See also Horizontal intra-industry Women, 70, 112 trade Wong, Herbert S., 168 Vertical specialization, 250, 251f, Wood, Adrian, 232 253 Woolhandler, Stefﬁ e, 169 Vertical v. horizontal integration Work group management, global, and tax incentives, 38 241 Visas Work visas (H-1B), 19, 95, 115, employer-sponsored permanent 118–20 (green card), 115, 120–22, World Competitiveness Yearbook, 2, 3 128 World Economic Forum (WEF), 2, F-1 (student), 115–16 35–37 H-1B (work), 19, 95, 115, 118–20 World Health Organization (WHO) number issued in U.S., 115–17, analysis of health care systems, 118–20, 121–22 197–98 policy recommendations, 127–28 World Intellectual Property Organiza- Vladeck, Bruce C., 167 tion (WIPO), 137 Voting, “foot,” 6–8, xiv Worldwide v. territorial taxation sys- See also Tiebout competition model tems, 52–53 (foot voting) Write-offs, tax, 47, 48, 55

HHassett.indbassett.indb 322322 111/8/121/8/12 9:279:27 PMPM INDEX 323

X-tax system, 47 Zero-sum v. mutual beneﬁ t concept (continued) Yi, Kei-Mu, 241 tax competition perspective, Yinger, John, 13–14 22–23 trade paradigm perspective, Zero-sum v. mutual beneﬁ t concept 227–28, 278 competitiveness theory, 1–2, 3, 5, Zignago, S., 232 16–17, 33, 159–60 Zingales, Luigi, 293–95 education perspective, 69–70, 86 Zoning, neighborhood, 12–13, 14–15 health care perspective, 187–88 Zucker, Lynne G., 118 immigration perspective, 96–97 Zycher, Benjamin, 196, xvi

HHassett.indbassett.indb 323323 111/8/121/8/12 9:279:27 PMPM HHassett.indbassett.indb 324324 111/8/121/8/12 9:279:27 PMPM About the Authors

Claude Barfield, a former consultant to the office of the U.S. Trade Rep- resentative, is a resident scholar at the American Enterprise Institute, where he researches international trade policy (including trade policy in China and East Asia), the World Trade Organization (WTO), intellectual property, and science and technology policy. His many books include SWAP: How Trade Works (AEI Press, 2011) and Free Trade, Sovereignty, Democracy: The Future of the World Trade Organization (AEI Press, 2001), in which he identifies challenges to the WTO and to the future of trade liberalization.

Michael Chernew is a professor in the department of health care policy at Harvard Medical School. His research activities focus on several areas, most notably the causes and consequences of growth in health care expenditures, geographic variation in medical spending, and use and value-based insurance design. Mr. Chernew is a member of the Medicare Payment Advisory Commission, an independent agency established to advise Congress on issues affecting the Medicare program. He is also a member of the Con- gressional Budget Office’s Panel of Health Advisors and Commonwealth Foundation’s Commission on a High Performance Health System. In 2000, 2004, and 2011, he served on technical advisory panels for the Centers for Medicare and Medicaid Services that reviewed the assumptions used by the Medicare actuaries to assess the financial status of the Medicare trust funds. He is a faculty research fellow of the National Bureau of Economic Research, coedits the American Journal of Managed Care, and is a senior associate editor of Health Services Research. In 2010, Mr. Chernew was elected to the

325

HHassett.indbassett.indb 325325 111/8/121/8/12 9:279:27 PMPM 326 RETHINKING COMPETITIVENESS

Institute of Medicine of the National Academies and served on the Commit- tee on the Determination of Essential Health Benefits.

Gordon Hanson is director of the Center on Emerging and Pacific Econo- mies and professor of economics at the University of California, San Diego (UCSD), where he holds faculty positions in the School of International Relations and Pacific Studies and the department of economics. He is a research associate at the National Bureau of Economic Research, a member of the Council on Foreign Relations, and a coeditor of the Review of Econom- ics and Statistics. Before joining UCSD in 2001, he was on the economics faculty at the University of Michigan (1998–2001) and the University of Texas (1992–1998). In 2011, Mr. Hanson received the Chancellor’s Associates Award for Excellence in Research in Social Science and the Humanities from UCSD. He specializes in the economics of international trade, international migration, and foreign direct investment. He has published extensively in the top academic economics journals, is widely cited for his research by scholars across the social sciences, and is frequently quoted in major media outlets. His current research examines the international migration of skilled labor, border enforcement and illegal immigration, the impact of imports from China on the U.S. labor market, and the determinants of comparative advantage. His most recent book is Regulating Low-Skilled Immigration in the United States (AEI Press, 2010).

Kevin A. Hassett is the director of economic policy studies and a senior fellow at the American Enterprise Institute (AEI). Before joining AEI, he was a senior economist at the Board of Governors of the Federal Reserve System and an associate professor of economics and finance at the Graduate School of Business of Columbia University, as well as a policy consultant to the Trea- sury Department during the George H. W. Bush and Clinton administrations. He served as an economic adviser to the George W. Bush 2004 presidential campaign, chief economic adviser to Senator John McCain during the 2000 presidential primaries, and senior economic adviser to the McCain 2008 presidential campaign. Mr. Hassett also writes a column for National Review.

R. Glenn Hubbard, a former chairman of the President’s Council of Economic Advisers, is currently the dean of Columbia Business School.

HHassett.indbassett.indb 326326 111/8/121/8/12 9:279:27 PMPM ABOUT THE AUTHORS 327

He specializes in public and corporate finance and financial markets and institutions. He has written more than ninety articles and books, including two textbooks on corporate finance, investment decisions, banking, energy economics, and public policy, including Seeds of Destruction: Why the Path to Economic Ruin Runs Through Washington, and How to Reclaim American Prosperity (FT Press, 2010, with Peter Navarro). He has served as a deputy assistant secretary at the Treasury Department and consultant to the Federal Reserve Board and the Federal Reserve Bank of New York, among others.

Matthew H. Jensen is a research assistant at the American Enterprise Insti- tute (AEI) with an active research agenda focused on public finance and trade. In addition to AEI publications, his work has appeared in Tax Notes, the Wall Street Journal, Real Clear Markets, and others.

Philip I. Levy researches international trade policy, political economy, economic development, and international finance. Levy has taught most recently at Columbia University’s School of International Public Affairs and served as a resident scholar at the American Enterprise Institute. He previously handled international economic issues for the U.S. Secretary of State’s policy planning staff and served as senior economist for trade for the President’s Council of Economic Advisers. Prior to that, Levy served on the economics faculty at Yale University, where he was a member of the Economic Growth Center. He also served one year as academic director of the Yale Center for the Study of Globalization. Levy’s scholarly work has appeared in journals such as the American Economic Review, the Journal of International Economics, and Economic Journal. He is a regular contributor to Foreign Policy and has published in the Wall Street Journal, Investor’s Busi- ness Daily, and American.com, among other outlets.

Robert J. Shapiro is the chairman of Sonecon LLC, a private firm that provides advice and analysis to senior executives and officials in U.S. and foreign businesses, governments, and nonprofit organizations. He is also adviser to the International Monetary Fund, a senior policy fellow at the Georgetown University McDonough School of Business, chairman of the U.S. Climate Task Force, and director of the Globalization Initiative at the New Democrat Network. Mr. Shapiro was undersecretary of commerce for

HHassett.indbassett.indb 327327 111/8/121/8/12 9:279:27 PMPM 328 RETHINKING COMPETITIVENESS

economic affairs from 1997 to 2001. Before that, he was cofounder and vice president of the Progressive Policy Institute and the Progressive Foun- dation. He was the principal economic adviser to Bill Clinton in his 1991– 1992 campaign and a economic adviser to the Gore, Kerry, and Obama presidential campaigns. Mr. Shapiro also served as legislative director and economic counsel for Senator Daniel Patrick Moynihan (D-NY) and as a fellow of Harvard University, the Brookings Institution, and the National Bureau of Economic Research.

Joel Slemrod is the Paul W. McCracken Collegiate Professor of Business Economics and Public Policy at the Stephen M. Ross School of Business at the University of Michigan, and professor and chair in the department of economics. He also serves as director of the Office of Tax Policy Research, an interdisciplinary research center housed at the Ross School of Busi- ness. He has been a consultant to the U.S. Department of the Treasury, the Canadian Department of Finance, the New Zealand Department of Treasury, the South African Ministry of Finance, the World Bank, and the Organisation for Economic Co-operation and Development. From 1992 to 1998, Mr. Slemrod was editor of the National Tax Journal and coeditor of the Journal of Public Economics from 2006 to 2010. He is the coauthor with Jon Bakija of Taxing Ourselves: A Citizen’s Guide to the Debate over Taxes, the fifth edition of which will be published by MIT Press in 2013.

Phillip L. Swagel is a professor at the University of Maryland School of Public Policy and a visiting scholar at the American Enterprise Institute. Mr. Swagel was Assistant Secretary for Economic Policy at the Treasury Depart- ment from December 2006 to January 2009. Mr. Swagel previously worked at the White House Council of Economic Advisers, the International Mon- etary Fund, and the Federal Reserve, and taught economics at Northwestern University, the University of Chicago Booth School of Business, and the McDonough School of Business at Georgetown University.

Martin R. West is an assistant professor of education at the Harvard Graduate School of Education; deputy director of Harvard’s Program on Education Policy and Governance; and an executive editor of Education Next, a journal of opinion and research on education policy. He is also a

HHassett.indbassett.indb 328328 111/8/121/8/12 9:279:27 PMPM ABOUT THE AUTHORS 329

research affiliate of the Taubman Center for State and Local Government at Harvard Kennedy School and of the CESifo Research Network. Mr. West’s research focuses on the politics of K–12 education policy in the United States and institutional and policy effects on student achievement and noncognitive skills. His most recent book (coedited with Joshua Dunn), From Schoolhouse to Courthouse: The Judiciary’s Role in American Education (Brookings Institution Press), examined the increase in court involvement in education policymaking over the past half-century. He has also published widely in academic journals and outlets including Education Next, Education Week, Vox, and the Wall Street Journal. Before joining the Harvard faculty, Mr. West taught at Brown University and was a research fellow in governance studies at the Brookings Institution.

Benjamin Zycher is the president of Benjamin Zycher Economics Associ- ates Inc., a visiting scholar at the American Enterprise Institute, and a senior fellow at the Pacific Research Institute. He is also an associate in the Intel- ligence Community Associates Program of the Office of Economic Analysis, Bureau of Intelligence and Research, U.S. Department of State. He served as a senior staff economist for the President’s Council of Economic Advisers from July 1981 to July 1983. He is the author of the recent AEI Press book Renewable Electricity Generation: Economic Analysis and Outlook.

HHassett.indbassett.indb 329329 111/8/121/8/12 9:279:27 PMPM Board of Trustees The American Enterprise Institute Kevin B. Rollins, Chairman for Public Policy Research Senior Adviser TPG Capital Founded in 1943, AEI is a nonpartisan, nonprofit research and educational organization based in Washington, DC. Tully M. Friedman, The Institute sponsors research, conducts seminars and Vice Chairman Chairman and CEO conferences, and publishes books and periodicals. Friedman Fleischer & Lowe, LLC AEI’s research is carried out under three major programs: Economic Policy Studies, Foreign Policy and Defense Studies, Clifford S. Asness Managing and Founding Principal and Social and Political Studies. The resident scholars and AQR Capital Management fellows listed in these pages are part of a network that also Gordon M. Binder includes adjunct scholars at leading universities throughout Managing Director the United States and in several foreign countries. Coastview Capital, LLC The views expressed in AEI publications are those of the Arthur C. Brooks authors and do not necessarily reflect the views of the staff, President advisory panels, officers, or trustees. American Enterprise Institute The Honorable Matthew K. Rose Council of Academic Richard B. Cheney Chairman and CEO Advisers BNSF Railway Company Peter H. Coors George L. Priest, Chairman Vice Chairman of the Board Edward B. Rust Jr. Edward J. Phelps Professor of Law Molson Coors Brewing Company Chairman and CEO and Economics Harlan Crow State Farm Insurance Companies Yale Law School Chairman and CEO D. Gideon Searle Alan J. Auerbach Crow Holdings Managing Partner Robert D. Burch Professor of Ravenel B. Curry III The Serafin Group, LLC Economics and Law Chief Investment Officer University of California, Berkeley Eagle Capital Management, LLC Mel Sembler Founder and Chairman Eliot A. Cohen Paul H. Nitze School of Advanced Daniel A. D’Aniello The Sembler Company Cofounder and Managing Director International Studies The Carlyle Group Wilson H. Taylor Johns Hopkins University John V. Faraci Chairman Emeritus Eugene F. Fama Cigna Corporation Chairman and CEO Robert R. McCormick Distinguished International Paper Company William H. Walton Service Professor of Finance Booth School of Business Managing Member Christopher B. Galvin University of Chicago Chairman Rockpoint Group, LLC Harrison Street Capital, LLC Martin Feldstein William L. Walton George F. Baker Professor Raymond V. Gilmartin Rappahannock Ventures, LLC of Economics Harvard Business School Marilyn Ware Harvard University Harvey Golub Ware Family Office Aaron L. Friedberg Chairman and CEO, Retired Professor of Politics and American Express Company International Affairs Chairman, Miller Buckfire Emeritus Trustees Princeton University Robert F. Greenhill Richard B. Madden Robert P. George Founder and Chairman McCormick Professor of Jurisprudence Greenhill & Co., Inc. Robert H. Malott Director, James Madison Program Frank J. Hanna Paul F. Oreffice in American Ideals and Institutions Princeton University Hanna Capital, LLC Henry Wendt Bruce Kovner Gertrude Himmelfarb Chairman Distinguished Professor of History Caxton Alternative Management, LP Emeritus Officers City University of New York Marc S. Lipschultz Arthur C. Brooks R. Glenn Hubbard Partner President Kohlberg Kravis Roberts Dean and Russell L. Carson Professor David Gerson of Finance and Economics John A. Luke Jr. Columbia Business School Executive Vice President Chairman and CEO MeadWestvaco Corporation Walter Russell Mead Jason Bertsch James Clarke Chace Professor of George L. Priest Vice President, Development Foreign Affairs and the Humanities Edward J. Phelps Professor of Law Henry Olsen Bard College and Economics Yale Law School Vice President; Director, John L. Palmer National Research Initiative University Professor and J. Peter Ricketts Dean Emeritus President and Director Danielle Pletka Maxwell School of Citizenship Platte Institute for Economic Vice President, Foreign and and Public Affairs Research, Inc. Defense Policy Studies Syracuse University

HHassett.indbassett.indb 330330 111/8/121/8/12 9:279:27 PMPM Sam Peltzman Nicholas Eberstadt Norman J. Ornstein Ralph and Dorothy Keller Henry Wendt Scholar in Resident Scholar Distinguished Service Professor Political Economy of Economics Pia Orrenius Jeffrey A. Eisenach Visiting Scholar Booth School of Business Visiting Scholar University of Chicago Richard Perle Jon Entine Fellow Jeremy A. Rabkin Visiting Fellow Professor of Law Mark J. Perry George Mason University Rick Geddes Scholar School of Law Visiting Scholar Tomas J. Philipson Harvey S. Rosen Jonah Goldberg Visiting Scholar John L. Weinberg Professor of Fellow Economics and Business Policy Edward Pinto Aspen Gorry Resident Fellow Princeton University Research Fellow Richard J. Zeckhauser Alex J. Pollock Scott Gottlieb, M.D. Resident Fellow Frank Plumpton Ramsey Professor Resident Fellow of Political Economy Vincent R. Reinhart Kennedy School of Government Kenneth P. Green Visiting Scholar Harvard University Resident Scholar Richard Rogerson Michael S. Greve Visiting Scholar Visiting Scholar Research Staff Michael Rubin Kevin A. Hassett Resident Scholar Ali Alfoneh Senior Fellow; Director, Resident Fellow Economic Policy Studies Sally Satel, M.D. Joseph Antos Robert B. Helms Resident Scholar Wilson H. Taylor Scholar in Health Resident Scholar Gary J. Schmitt Care and Retirement Policy Arthur Herman Resident Scholar; Director, Leon Aron NRI Visiting Scholar Program on American Citizenship; Resident Scholar; Director, Co-Director, Marilyn Ware Center Russian Studies Frederick M. Hess for Security Studies Resident Scholar; Director, Michael Auslin Education Policy Studies Mark Schneider Visiting Scholar Resident Scholar Ayaan Hirsi Ali Claude Barfield Visiting Fellow David Schoenbrod Visiting Scholar Resident Scholar R. Glenn Hubbard Michael Barone Visiting Scholar Nick Schulz DeWitt Wallace Fellow; Resident Fellow Frederick W. Kagan Editor-in-Chief, American.com Roger Bate Resident Scholar; Director, AEI Resident Scholar Critical Threats Project; and Roger Scruton Christopher DeMuth Chair Visiting Scholar Andrew G. Biggs Resident Scholar Leon R. Kass, M.D. Sita Nataraj Slavov Madden-Jewett Chair Resident Scholar Edward Blum Andrew P. Kelly Vincent Smith Visiting Fellow Research Fellow, Jacobs Associate Visiting Scholar Dan Blumenthal J.D. Kleinke Christina Hoff Sommers Resident Fellow Resident Fellow Resident Scholar; Director, John R. Bolton Desmond Lachman W. H. Brady Program Senior Fellow Resident Fellow Michael R. Strain Karlyn Bowman Adam Lerrick Research Fellow Senior Fellow Visiting Scholar Phillip Swagel Alex Brill John H. Makin Visiting Scholar Research Fellow Resident Scholar Marc A. Thiessen James Capretta Aparna Mathur Fellow Visiting Scholar Resident Scholar, Jacobs Associate Stan A. Veuger Lynne V. Cheney Michael Mazza Research Fellow Senior Fellow Research Fellow Alan D. Viard Steven J. Davis Michael Q. McShane Resident Scholar Visiting Scholar Research Fellow Peter J. Wallison Sadanand Dhume Thomas P. Miller Arthur F. Burns Fellow in Financial Policy Studies Resident Fellow Resident Fellow Thomas Donnelly Charles Murray Paul Wolfowitz Scholar Resident Fellow; Co-Director, W. H. Brady Scholar Marilyn Ware Center for Roger F. Noriega John Yoo Security Studies Fellow Visiting Scholar Mackenzie Eaglen Stephen D. Oliner Benjamin Zycher Resident Fellow Resident Scholar NRI Visiting Fellow

HHassett.indbassett.indb 331331 111/8/121/8/12 9:279:27 PMPM HHassett.indbassett.indb 332332 111/8/121/8/12 9:279:27 PMPM HHassett.indbassett.indb 333333 111/8/121/8/12 9:279:27 PMPM HHassett.indbassett.indb 334334 111/8/121/8/12 9:279:27 PMPM