The Development of Water Resources Planning in the United States From

Planning Methods in an Era of Challenge and Change

The Development of Water Resources Planning in the United States from the Reclamation Act of 1902 to the Principles and Standards of 1973 An Assessment Based on Original Sources

By David C. Major and Eugene Z. Stakhiv

IWR Maass-White Series

The Institute for Water Resources (IWR) is a U.S. Army Corps of Engineers (USACE) Field Operating Activity located within the Washington DC National Capital Region (NCR), in Alexandria, Virginia and with satellite centers in New Orleans, LA; Davis, CA; Denver, CO; and Pittsburg, PA. IWR was created in 1969 to analyze and anticipate changing water resources management conditions, and to develop planning methods and analytical tools to address economic, social, institutional, and environmental needs in water resources planning and policy. Since its inception, IWR has been a leader in the development of strategies and tools for planning and executing the USACE water resources planning and water management programs.

IWR strives to improve the performance of the USACE water resources program by examining water resources problems and offering practical solutions through a wide variety of technology transfer mechanisms. In addition to hosting and leading USACE participation in national forums, these include the production of white papers, reports, workshops, training courses, guidance and manuals of practice; the development of new planning, socio- economic, and risk-based decision-support methodologies, improved hydrologic engineering methods and software tools; and the management of national waterborne commerce statistics and other Civil Works information systems. IWR serves as the USACE expertise center for integrated water resources planning and management; hydrologic engineering; collaborative planning and environmental conflict resolution; and waterborne commerce data and marine transportation systems.

The Institute’s Hydrologic Engineering Center (HEC), located in Davis, CA specializes in the development, documentation, training, and application of hydrologic engineering and hydrologic models. IWR’s Navigation and Civil Works Decision Support Center (NDC) and its Waterborne Commerce Statistical Center (WCSC) in New Orleans, LA, is the Corps data collection organization for waterborne commerce, vessel characteristics, port facilities, dredging information, and information on navigation locks. IWR’s Risk Management enter is a center of expertise whose mission is to manage and assess risks for dams and levee systems across USACE, to support dam and levee safety activities throughout USACE, and to develop policies, methods, tools, and systems to enhance those activities.

Other enterprise centers at the Institute’s NCR office include the International Center for Integrated Water Resources Management (ICIWaRM), under the auspices of UNESCO, which is a distributed, intergovernmental center established in partnership with various Universities and non-Government organizations; and the Conflict Resolution and Public Participation Center of Expertise, which includes a focus on both the processes associated with conflict resolution and the integration of public participation techniques with decision support and technical modeling. The Institute plays a prominent role within a number of the USACE technical Communities of Practice (CoP), including the Economics CoP. The Corps Chief Economist is resident at the Institute, along with a critical mass of economists, sociologists and geographers specializing in water and natural resources investment decision support analysis and multi-criteria tradeoff techniques.

The Director of IWR is Dr. Joe Manous. Additional information on IWR can be found at: http://www.iwr.usace.army.mil. IWR’s NCR mailing address is:

U.S. Army Engineer Institute for Water Resources 7701 Telegraph Road, 2nd Floor Casey Building Alexandria, VA 22315-3868

The Development of Water Resources Planning in the United States from the Reclamation Act of 1902 to the Principles and Standards of 1973

An Assessment Based on Original Sources

David C. Major and Eugene Z. Stakhiv

No copyrighted part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical or photocopying, recording or otherwise without prior written permission of the publisher.

Library of Congress Control Number: 2019900025

ISBN 978-0-578-44105-4

Published by IWR Press Casey Building 7701 Telegraph Road Alexandria VA 22315

The Maass-White Library Series in Water Planning and Management, Volume 5

Interior design by Patricia Hart

Contents The Maass-White Library Series v Preface ix Acknowledgments xi Chapter 1 Introduction 1 Section I: Three Key Laws Prior to World War II 5 Chapter 2 The Reclamation Act of 1902 7 Planning Methods and Procedures 9 Development and Results of Reclamation Programs 11 Perspectives 16 Chapter 3 The 308 Reports 19 Authorization and Reports 19 Plans and Planning Methods 22 Perspectives 28 Chapter 4 The Flood Control Act of 1936 31 Flood Control as a Federal Responsibility 31 Benefits and Costs 33 Local Contributions 34 Projects 35 Perspectives 35 Section II: The Classic Model of River Basin Planning 39 Chapter 5 Gilbert White, Human Adjustment to Floods 41 The Context 42 Content and Approach 44 Perspectives 47 Chapter 6 Arthur Maass, Muddy Waters 51 The Ickes Foreword 52 Muddy Waters 54 Perspectives 60 Maass and White 61 Chapter 7 The Green Book 63 The Economics of Project Evaluation 65 Objectives 68 Perspectives 69 Chapter 8 John V. Krutilla and Otto Eckstein, Multiple Purpose River Development 71 Economic Analysis and the Economics of River Basin Development 72 The Cost of Capital 72 Case Studies: Economics 73 Case Studies: Distribution of Costs and Gains 77 Perspectives 81

iii

Chapter 9 Delaware River Basin Report 83 Planning Methods 84 Recommendations and Results 92 Perspectives 97 Section III: Development of New Methods 99 Chapter 10 The U.S. President’s Water Resources Council Report 101 Objectives and Project Formulation 104 Perspectives 108 Chapter 11 Arthur Maass et al., Design of Water-Resource Systems 111 Research of the Harvard Water Program 112 New Investment Criteria 113 Conventional Techniques, Mathematical Modeling, and the Design Process 115 Political Decision Making 119 Perspectives 120 Chapter 12 The North Atlantic Regional Water Resources Study 123 Institutional Arrangements 125 Regional Subdivisions for Planning 126 Planning Period and Forecast Years 127 New Methods: Multiple Objectives 128 New Methods: Mathematical Models 130 Needs, Sources, and Devices 133 Forecasts and Recommendations 134 Perspectives 136 Chapter 13 The Principles and Standards, 1973 139 Background 139 Authority 141 Objectives and Criteria 142 Perspectives 148 Section IV: Perspectives 151 Chapter 14 A Summing Up and Conclusions 153

References 157 Index 169

Foreword The Maass-White Library Series in Water Planning and Management

Robert A. Pietrowski, Former Director, Institute for Water Resources

I’m truly delighted to share my thoughts on this volume, the fifth in the series of books published by the U.S. Army Corps of Engineers’ (USACE) Institute for Water Resources (IWR) in its Maass-White Library Series in Water Planning and Management. The MaassWhite Library Series is devoted to the advancement of water resources planning and management practices along the continuum that was broadly influenced by the pioneering research of Drs. Arthur Maass and Gilbert White. USACE, through its Office of History and IWR, is honored to serve as the custodian and repository for the personal libraries of these two great scholars, and the Institute was most fortunate to have both Arthur Maass and Gilbert White actively participate in the dedication ceremony of the Maass-White Library at IWR in 2001.

The true measure of the contributions of Professors Maass and White is that their influence reached far outside their classrooms and academia. In fact, the intellectual foundations provided by Drs. Maass and White and other 20th century scholars have proved essential to the maturation of the practice of planning and managing water resources both in the U.S. and around the world. It speaks to the foresight of their scholarship that the underlying principles that flowed from their collective works remain highly relevant today, and which, to Maass and White’s credit, were gradually adopted and then applied by several generations of water resources practitioners, policymakers and decision-leaders over the last half-century.

One benefit of their interaction with the engineering profession and the public sector was the profound influence that

v Professors Maass and White have had on the way that USACE embraced change during the evolution of Federal water management decision-making in the decades following their seminal works, which culminated in the Federal Government’s adoption of the United States Water Resourcees Council (WRC) Principles and Standards (P&S) [1973], and the concurrent integration of the P&S into the Corps national policy guidance for all planning studies. One clear example was the immediate implementation of the principles in the North Atlantic Regional (NAR) [1972] and Northeastern Water Supply (NEWS) studies [1976], which the Corps conducted, as both Maass and White were on the oversight coordinating committees for both studies.

Not only is this a tribute to the legacy of these two scholars -- and to the past leadership of USACE -- but significantly, it also serves as a reminder of the need for the continued injection of scholarship into water resources planning practice as evaluation tools and our understanding of the hydrologic, geomorphic, biologic and social sciences have continued to advance. This need remains as compelling as ever today, in this new century, as the Nation faces a new set of water resources challenges both on the domestic front and around the world. In fact, as the pace of technological and social change has accelerated, the infusion of contemporary engineering and scientific scholarship into public policy is critically important to the future of U.S. water resources and to the future of the Corps’ Civil Works mission itself. Facilitating a continuing intersection of scholarship and practice within the Civil Works program remains the essence of the Institute's purpose.

The present volume fits squarely within the Maass-White tradition – it provides a review of the development of water resources planning in the United States from the end of the 19th century through the 1970’s, based on original sources. The authors have undertaken this study to provide planning practitioners, especially newer planners, with concise reviews of the original sources, assessments of their meaning and relevance to the present, and how they will help to provide a basis for further development of 21st century water resource planning methods. It is fitting, in including this volume within the Maass-White Library Series, to note that both authors worked with Professors Maass and White over significant portions of their careers and gratefully acknowledge the lasting influence of these distinguished scholars on their work.

vii

viii

Preface

Water resource planning methods have a long history in the United States, much of which remains highly relevant today. Valuable perspectives from the past, however, are often not fully incorporated into current planning; this is in part because of a lack of familiarity with key materials in the development of water planning methods and procedures. We hope that this volume will go some way toward resolving this problem and thereby will provide insights into the further development of planning criteria. This volume continues our effort to preserve for younger planners key elements of water planning methods and their development that have relevance in the present day. In that respect, it is a companion to another collaboration (with DCM as editor and ES as principal advisor), Harry E. Schwarz and the Development of Water Resources and Environmental Planning, volume 3 of the U.S. Army Corps of Engineers Institute for Water Resources Maass-White Library Series in Water Planning and Management (Alexandria, VA: IWR Press, 2010). That volume illuminates how a single innovative planner can have a significant impact on the advancement of methods. In doing so it also makes available a selection of some of the most innovative water resources planning efforts in the second half of the twentieth century. A companion to the present book that focuses on other key documents before and during the period covered here would be welcome, including, for example, the Pick-Sloan plan for the Missouri River Basin (U.S. Congress, Flood Control Act of 1944, P.L. 78-534, sec. 9). Further, a worthwhile third volume could deal with water resources planning in the context of the many new considerations that have entered into public policy, laws, and other documents since the latter part of the twentieth century, such as the National Environmental Policy Act of 1969. We hope that these and perhaps other related volumes will be forthcoming from other students of water resources planning.

ix With this volume, we acknowledge our productive professional friendship of nearly fifty years, beginning with the early days of the North Atlantic Regional Study, described in chapter 12.

New York Alexandria, Virginia 2018

Acknowledgments

We wish to acknowledge the contributions of the distinguished experts who have reviewed the volume: Dr. John S. Major, our former colleagues at the Institute for Water Resources, Dr. Martin Reuss, former Director Kyle Schilling, and the encouragement of Robert A. Pietrowski, former Director of the Institute for Water Resources, as well as other colleagues at the institute, especially Robin L. Baird for her assistance with figures. We also thank Dr. Dana Neascu of the Columbia University Law Library for assistance with locating legal documents. In addition, we are grateful for support from the U.S. Army Corps of Engineers Institute for Water Resources, and Columbia University. We both remember with gratitude our professional interactions with Arthur Maass and Gilbert White, mentors and friends, for whom a library at the Institute for Water Resources has been named. We are grateful for permission to reprint material in the works on which the book is based, as follows: From the Committee on Geographical Sciences, Univer- sity of Chicago, for quotations and an illustration from the work reviewed in chapter 5. From Harvard University Press, for quotations and illustrations from the work reviewed in chapters 6 and 11.

xii Chapter 1 Introduction

This book provides an assessment of the development of water resources planning in the United States from the Reclamation Act of 1902 to the Principles and Standards of 1973 based on original sources. The twelve essays each focus on a significant law, original report, or book. They aim to provide a concise description of the works and their background and explain their importance both at the time they were written and later. Figure 1-1 shows the principal water planning regions of the United States.

Figure 1-1: Water resource regions of the United States Source: U.S. Geological Survey (n.d.), https://water.usgs.gov/wsc/map_index.html

The book is organized into four sections. In the first section, three important laws prior to World War II introduce the subject: the Reclamation Act of 1902, the “308” Act of 1926, and the Flood Control Act of 1936.

1 In the second section, the emphasis is on the beginning of standard multipurpose planning with benefit-cost analysis just before World War II and its development thereafter. This section covers the development of river basin planning from key critiques and analyses by Gilbert White (1945), Arthur Maass (1951), and John Krutilla and Otto Eckstein (1958), through the Green Book (1958) and a typical example of classic river basin planning of the time, the Delaware River Basin Report (1960). The third section covers the development of new methods through 1973. It includes works from 1962 through 1973, beginning with well-known discussions of criteria in Senate Document 87-97 (1962), through Design of Water-Resource Systems (1962), the North Atlantic Regional (NAR) Study (1972), and the U.S. Water Resources Council’s New Standards (1973). This era marks a decisive break with the classic economics- focused river basin planning discussed in the second section. A short final section offers perspectives on the work. Here we provide a summary of the key developments treated in the book, knowledge of which may contribute to dealing effectively with the complex challenges of the present time. The sections focus on documents relating to different eras of water resource planning, with the material overlapping in time where this appears appropriate. The selection criteria for the works discussed include the extent to which each represents an important element of the water resources planning of its time and the breadth of its influence on research, planning, and operations both earlier and today. Some of the documents (e.g., the Flood Control Act of 1936) were or are binding on the agencies that use them; others provide recommendations; and some, although not official documents, such as Design of Water-Resource Systems, have shaped the whole field of water resource planning for decades. Each essay is designed to be succinct but not simply a summary: the document and the ideas in it are placed in context and explained conceptually, and the document’s importance to later developments is delineated. Within a focus on general

2 guidelines and normative and applied frameworks, material especially relevant to the Corps and other U.S agencies is examined. The essays also reference many important works related to those under discussion, so that the book covers the field to some degree, although a historian of public works would argue, with some justification, that each of the essay subjects in the volume deserves one or more monographs to capture the full complexity of its context and meaning. The figures in each essay evoke the contributions of the documents under review. Chapter references are incorporated into a comprehensive list at the back of the book. As the intended audience of the book covers a wide range of professional fields, some of the essays include explanations of concepts with which not all readers may be familiar, such as the economic efficiency and multiobjective approaches. We hope the book will be useful as an adjunct to upper-level undergraduate and graduate courses, as a reference work, and as part of individual professional libraries. The book as a whole, with the references, can serve as a compact, albeit partial, library of key documents of water resources planning in the United States. As explained in the preface, additional volumes on other important documents would be welcome.

Section I Three Key Laws prior to World War II

The three chapters in this section each focus on a key law from the first decades of the twentieth century. The Reclamation Act of 1902, discussed in chapter 2, is a landmark in the history of water planning in the United States. It provided for single- purpose development for an objective that, broadly defined, was the continued development of the American West utilizing smallholder agriculture. The act, which followed decades of less comprehensive activity relating to irrigation at various levels of government, covers all or parts of seventeen western states. From the standpoint of water planning in the United States, the Reclamation Act is notable for the federal assumption of substantial obligations in the developing West, as opposed to the national remit of the Corps of Engineers. The “308” reports referred to in chapter 3 are named for H. Doc. 69-308, in which the Corps of Engineers and the Federal Power Commission provided cost estimates for surveying streams throughout the country for multipurpose development. The 1927 law authorizing the surveys of the streams in the 308 reports (U.S. Congress 1927) provided for the first time clear authority for the Corps of Engineers to develop multipurpose river basin plans nationwide. These studies resulted in large investments in dams and other water infrastructure throughout the country. The Flood Control Act of 1936, the subject of chapter 4, is one of the most important federal acts in the history of water resources planning. It established flood control as a federal responsibility and thereby contributed to the development of multipurpose planning. It presented, for the first time in an easily accessible public document, the requirement that, for flood control projects authorized in the act, “the benefits to whomsoever they may accrue are in excess of the estimated costs,” and it established requirements for local contributions.

Chapter 2 The Reclamation Act

U.S. Congress, An Act Appropriating the Receipts from the Sale and Disposal of Public Lands in Certain States and Territories to the Construction of Irrigation Works for the Reclamation of Arid Lands, P.L. 57-161, 57th Congress, 1st Session, 1902 (“The Reclamation Act”)

The Reclamation Act of 1902 (fig. 2-1) was a landmark in the development of water planning in the United States: it provided for single-purpose development for an objective that, broadly defined, was the continued development of the American West, utilizing smallholder agriculture. The area ultimately covered by the reclamation program (fig. 2-2) includes thirteen states that had obtained statehood by 1902 (nine after the Civil War) and three territories that later became states (Oklahoma in 1907 and New Mexico and Arizona in 1912). Texas, which had no federal lands, was added in 1906 (U.S. Congress 1906). The act, which followed decades of less comprehensive activity relating to irrigation at various levels of government (Brookings Institution 1919, 2–8), provided that the secretary of the interior was authorized to identify and construct irrigation projects, the cost to be covered initially from the sale of public lands. Irrigators were to repay a proportionate share of the costs of the works over a period of no more than ten years, when the works (with the exception of dams) would be turned over to the irrigators and operated by them. The irrigators were to have land not exceeding 160 acres, the amount designated in the homestead laws (that is, one quarter of a “country section,” or square mile; the first Homestead Act is U.S. Congress, May 20, 1862). Private lands could be included, with a proportionate share of costs allocated to them (U.S. Congress 1902a; the provisions are summarized in Brookings 1919, 21–23; a brief summary of the early years of the reclamation program is in Holmes 1972, 8).

Figure 2-1: Reclamation Act of 1902, sections 1 and 2

Source: U.S. Congress (1902a)

Within a few weeks of passage of the act, the secretary of the interior created the Reclamation Service, under the juris-diction of the director of the Geological Survey, but not as a part of the survey (Brookings 1919, 23). Thus an organizational framework was quickly set up for single-purpose irrigation planning with a repayment mechanism and a regional objective.

Figure 2-2: The seventeen western states covered by the Reclamation Act

The service became an independent bureau within Interior in 1907, and its name was changed to the Bureau of Reclamation in 1923 (http://www.usbr.gov/history). (The act is sometimes referred to as the “Newlands Act,” after Representative Francis G. Newlands of Nevada, a principal sponsor; see Brookings 1919, 17–21, for a description of Newlands’s role).

Planning Methods and Procedures As noted, the broad objective of the reclamation program was the continued development of the American West, and this was to be accomplished through small-scale irrigated farming. This should be taken in historical context: the development of the West was seen as a work in progress, with much to be accomplished. To take just three indicators, the Morrill Act (U.S. Congress, July 2, 1862) provided for the land-grant universities of the nation, which were of particular importance in the West; the first transcontinental railroad was completed in 1869; and the Apache

9 wars in the Arizona Territory ended only in 1886. At the time of the act, in other words, the American West was not the West of today’s great cities, towns, highways, and airports. The objectives of the Reclamation Act were well described in President Theodore Roosevelt’s message to Congress in 1901 supporting the passage of the program. He compared the importance of federal efforts in irrigation to those for rivers and harbors; the aim was to reclaim land and provide homes: “Our aim should be not simply to reclaim the largest area of land and provide homes for the largest number of people, but to create for this new industry the best possible social and industrial conditions” (Roosevelt 1901, cited in Brookings 1919, 21). The reclamation program would fit into the national effort to develop the whole territory of the country. The president noted that the program “will enrich every portion of our country, just as the settlement of the Ohio and Mississippi valleys brought prosperity to the Atlantic States” (Roosevelt 1901, cited in Brookings 1919, 19). It was not, as might be thought today, simply a regional program, for the president’s argument had merit within the historical context. Purpose. The Reclamation Act provided for a single- purpose program of investment—irrigation—although over the years, as noted below, the Bureau of Reclamation became a multipurpose agency. It should be observed that in 1902 water in the West was already used for multiple purposes, especially mining, navigation, and water supply. Project selection. Project selection was based on the adequacy of water supplies and the possibility of controlling them; the quality of soils; and market conditions, including distance from markets (Brookings 1919, 42–43). A substantial amount of information on these matters was available in some areas as a result of Geological Survey and other investigations. From this point of view, the reclamation program was a scientific and technical effort. On the institutional side:

In the initial selection of irrigable areas the choice of projects was governed less by the question of ease or cheapness of work than by the expressed desire on the part of public men that the Reclamation Service should confine its efforts as largely as possible to those works which for one cause or another could not readily be built by corporate effort, namely, those involving interstate or international relations or legal complications which put them beyond the reach of either private or state enterprise. (Brookings 1919, 43)

Development and Results of Reclamation Programs Early challenges and changes. There were some changes relatively early in the reclamation program. These are summarized in Brookings (1919, 26–32). Among the most important was the recognition that more private lands would have to be incorporated into projects than had been foreseen. This was because although there were huge amounts of unsettled acreage, in many areas the best acreage had already been taken up privately. The complexity of making financial arrangements with a large number of such private users led to the development of water users associations with which the Reclamation Service could deal. The coverage of the act was extended in 1905 to the part of Texas that would permit a project in New Mexico on the Rio Grande (Brookings 1919, 28), and to all Texas in 1906 (U.S. Congress 1906). In 1908 an act provided the secretary with authorization to negotiate with Native American tribes regarding the reclamation of lands allotted to them (U.S. Congress 1908). Among other changes, provision was made for advances from the Treasury to the Reclamation Fund (U.S. Congress 1910), to be repaid on specified terms. The greatest early change in terms of the bureau’s authority was in the Reclamation Extension Act of 1914 (U.S. Congress 1914), which provided for congressional appropriations for individual projects, the selection of which had earlier been vested first in the secretary and later in the president. Also, the period for repayment of construction charges was

11 extended to twenty years, and authority was granted to impose charges for operation and maintenance (Brookings 1919, 31–32). First projects. In March 1903 the secretary of the interior authorized the first five projects: Truckee-Carson (later renamed Newlands) in Nevada, Milk River in Montana, Sweetwater (North Platte) in Wyoming and Nebraska, Gunnison (Uncompahgre) in Colorado, and the Salt River Project in centralsArizona (Rowley 2006, 119–20). Progress on these and other projects could be rapid because of the information on western conditions already accumulated by the Geological Survey and others. As an example, the Pathfinder Dam on the North Platte was completed in 1909 (fig. 2-3).

Figure 2-3: Pathfinder Dam in Wyoming under construction in June 1908, part of one of the first U.S. Reclamation Service Projects

Source: U.S. National Park Service (n.d.), https://www.nps.gov/nr/testing/ReclamationDamsAndWaterProjects/Pathfinder_Dam .html

The project, 214 feet high, in a narrow canyon, was a combination of an arch and a gravity dam, resulting in both horizontal and vertical load distribution, a method the bureau later used in the construction of much larger dams, such as the Hoover Dam (http://www.nps.gov). The program grew apace: by 1917 the bureau had completed or was constructing facilities for 3,072,795 acres, to supply a total of 62,451 farms (Brookings 1919, app. 7, 129). At the same time, progress in dam construction techniques and materials provided the ability to build larger dams (Billington, Jackson, and Melosi 2005, 62–84). Multiple purposes and large projects. “The basic policy enacted by the Reclamation act of 1902 has persisted without change up to the present time” (Brookings 1919, 26; some modifications are described above and in Brookings 1919, 26–32). It was not long, however, before the basic approach of the bureau changed dramatically. New sources of revenue in addition to public land sales were required, and the utility of the multipurpose approach to water development became more obvious—the original single-purpose approach was no longer adequate. Backed by generous emergency funds from the New Deal, the principle of multiple-use came to dominate the Bureau of Reclamation's large projects in the 1930s. … Multiple-use offered a rationale to build projects without repayment guarantees, while agricultural development was expected to return costs to the Treasury. … [T]he cost of a project could be charged off to flood control and navigation, while revenue to pay for construction could be expected from the sale of hydroelectric power and water to urban users. … Emphasis upon multiple-use marked a change that ranked in importance with earlier decisions by Congress to permit delivery of federal water to private lands. (Rowley 2006, 343–44)

This was the era of the largest bureau projects, such as the most famous, Hoover Dam on the lower Colorado (fig. 2-4). This huge dam was to be multipurpose, as envisaged in the Boulder Canyon Project Act of 1928 (U.S. Congress 1928): flood control, navigation, flow regulation, irrigation, domestic water supply, and power generation (Rowley 2006, 286). It is 726.4 feet from foundation rock to the roadway on the crest of the dam. The towers and ornaments on the parapet rise 40 feet above the crest. Work began in 1931 and was completed on March 1, 1936 (http://www.usbr.gov/lc/hooverdam/faqs/damfaqs.html).

Figure 2-4: Hoover Dam, the largest reclamation project

Source: U.S. Bureau of Reclamation (n.d.), https://www.usbr.gov/lc/hooverdam/

There is one dam in the United States that is taller than Hoover Dam, and that is the Oroville Dam on the Feather River in California, which stands 770 feet tall. Constructed by the state of California, it is an earthfill dam, not a concrete structure like Hoover (http://www.usbr.gov/lc/hooverdam/History/essays/biggest.html). (The Oroville Dam experienced a serious spillway event in February 2017, leading to an emergency evacuation. An independent team of consultants attributed the incident to “a long-term systemic failure of the California Department of Water Resources, regulatory, and general industry practices” [Independent Forensic Team Report 2018, S-1]). By the time the large projects were under way, the original idea of distributing public lands to smallholders had been largely ignored. William D. Rowley notes that “Bureau of Reclamation leaders indicated [to Secretary Ickes in 1934] that it [the 160 acre rule] had been ‘a dead letter for years’ and it was best ‘to let sleeping dogs lie.’ By the 1930s, various methods had been applied to circumvent the law and allow water to flow to larger holdings, especially private lands” (2006, 345). The era of environment and management. The era of large, multipurpose bureau projects came to an end with the rise of the environmental movement, the passage of the National Environmental Policy Act (1969) and other laws and regulations. At that time and since, the reclamation program was subject to heavy criticism on economic and environmental grounds. To cite just a few sources, Mark Kanazawa (1993) and Richard W. Wahl (1989) view the issue from an economic standpoint; among more general and popular critiques, the best known is Marc Reisner (1993), which deals also with Corps of Engineers programs. Along with environmental considerations, there was also pressure on funding, given the Vietnam War and domestic priorities. According to Rowley, The combination of reduced funding and a growing environmental opposition to blocking free flowing rivers turned the Bureau of Reclamation on to a new course that demanded new outlooks, goals, and even a cultural

15 change within the organization. … It came as little surprise when the New York Times, on October 2, 1987, ran a modest announcement that the Bureau of Reclamation’s dam building days were over. “Instead of constructing big water and power projects, the Federal agency will concentrate on managing the existing projects, conserving water and assuring good water quality and environmental protection,” said the article. (2006, 40–41) With this change, the reclamation program of 1902 completed two transformations: first to a large, multipurpose dam program, and second to being a manager and conservator of western water resources.

Perspectives From the standpoint of the development of water planning in the United States, the Reclamation Act is notable for the federal assumption of substantial obligations in the developing West (as opposed to the national remit of the Corps); the initial use of a defined payment mechanism in conjunction with smallholding; and the development of (initially) single-purpose projects to meet the program’s objective. The analysis of projects was on an engineering, scientific, and financial basis, with project selection, soon after the program’s initiation, left with Congress. Purposes. The original program was single-purpose, as it would now be described; that it developed into a multipurpose program was perhaps inevitable, and indeed the multipurpose approach may be regarded as inherent in water resources development. In the case of reclamation, it is an illustration of changes in conditions (large cities) and the consequent need for reallocation to new purposes (e.g., urban water supply). Using modern terms, in general, single-purpose planning results in “misallocation” of water. Again from the modern standpoint, two purposes that are entirely absent in the early discussions are recreation and flow for the maintenance of ecological conditions. One interesting point about implementation of the single purpose of the original program is that for the first years of the

16 program the selection of sites was left to the agency, with approval by the secretary of the interior. This is a partial version of the approach recommended in Design of Water-Resource Systems (see chapter 11), that is, to set standards and then to select the best projects to meet them, rather than to make choices on a more ad hoc, project-by-project basis. Objectives. The objective of the reclamation program was, broadly defined, the continued development of the American West, and this was to be accomplished through small-scale irrigated farming. This can be thought of, in the context of the time, as really two objectives—the development of a huge, undeveloped part of the country and a social objective relating to the Homestead Act (U.S. Congress 1862a). From the present perspective, the regional objective of the act might be thought of in terms of regional income. However, this both would neglect the historical setting of the act and would be conceptually anachronistic. The welfare model underlying the theoretical basis of economic allocation, and later the regional income idea, was first developed in 1938 (Bergson 1938; see also Boulding 1952; Marglin 1967), as were the national income accounts (Kuznets 1937). Moreover, if the regional objective is thought of as regional income, this would be different from President Roosevelt’s view that the development of the arid West would help the entire nation: the regional income objective, in modern parlance, is redistributive in nature, without any necessary claim for overall benefit. If this objective is phrased as regional development (see chapter 12), this would be closer to President Roosevelt’s concept. In terms of later developments in planning (see chapter 13), the emphasis on small-scale agriculture could be thought of at least partially as a social well- being objective. From the modern standpoint, perhaps the biggest difference in objectives is the complete absence of any attention to ecological or environmental issues in the Reclamation Service’s early programs (the Brookings [1919] report has no mention of these at all). This suggests the importance of

17 flexibility in objectives (as well as purposes of programs) that are intended to last for many decades.

Chapter 3 The 308 Reports

U. S. Congress. An Act Authorizing the Construction, Repair, and Preservation of Certain Public Works on Rivers and Harbors, and for Other Purposes. March 3, 1925 (“The 1925 Rivers and Harbors Act”).

Section 3 of the Rivers and Harbors Act of 1925 (U.S. Congress 1925), authorized and directed the Corps of Engineers and the Federal Power Commission (created by Congress, 1920b; now the Federal Energy Regulatory Commission) to provide cost estimates for surveying streams throughout the country for multipurpose development. The Corps and the Commission provided these estimates in H. Doc. 69-308 (fig. 3-1; U.S. Congress 1926). Surveys of the streams in the 308 reports were authorized by Congress and the president in 1927 (U.S. Congress 1927), providing for the first time clear authority for the Corps to develop multipurpose river basin plans nationwide. (An earlier act, the River and Harbor Act of 1920 [U.S. Congress 1920a], authorized a study of the Tennessee River and its tributaries for navigation and power.) The Corps completed some two hundred 308 river basin planning reports studying navigation, power, flood control, and irrigation (U.S. Army Corps of Engineers 2010, 6) over the decades beginning in the late 1920s, and these studies resulted in large investments in dams and other water infrastructure throughout the country. It is of interest that the results of the 1925 act are better known than the act itself, in something of a contrast to the Reclamation Act and the Flood Control Act of 1936.

Authorization and Reports The immediate origin of the 308 reports lies in the 1925 act: The Secretary of War, through the Corps of Engineers of the United States Army and the Federal Power Commission are jointly hereby authorized and directed to prepare and submit to Congress an estimate of the cost of making such examinations, surveys, or other investigations as, in their opinion, may be required of those navigable streams of the United States, and their tributaries, whereon power

19 development appears feasible and practicable, with a view to the formulation of general plans for the most effective improvement of such streams for the purposes of navigation and the prosecution of such improvement in combination with the most efficient development

Figure 3-1: H. Doc. 69-308

Source: U.S. Congress (1926)

of the potential water power, the control of floods, and the needs of irrigation: Provided, That no consideration of the Colorado River and it problems shall be included in the consideration or estimate provided herein. (U.S. Congress 1925, section 3) This authorization provided the basis for truly national, multipurpose basin planning by the Corps. Congress (1927) authorized the Corps to do the planning, although the estimates were jointly developed with the Federal Power Commission. This authorization is a landmark in the development of water planning methods and procedures in the United States. The official history of the Corps situates it within a wider move toward multiple purposes: “The 1925 Rivers and Harbors Act accelerated the movement toward multipurpose water management” (Baldwin 1998, 130), a movement also followed by the Bureau of Reclamation (chapter 2). The Colorado River was not included in the 308 program; it was the focus of the seven-state Colorado River Compact of 1922 (U.S. Bureau of Reclamation 1922). The Corps and the commission responded quickly to the 1925 authorization, providing the information included in H. Doc. 69-308. The river basins for which estimates are provided are shown in figure 3-2; they range around the country from “streams draining to the Atlantic Ocean north of Cape Cod,” to “streams draining into Pacific Ocean north of Columbia River,” followed by “rivers in Hawaiian Islands.” The total estimated cost of the surveys is given as $7,322,400, with the largest amount for “streams except the Mississippi draining to Gulf of Mexico,” at $909,000, and the smallest for the Raritan River (New Jersey), at $19,400. The approximately two hundred 308 reports that followed in the next decades provided the basis for a significant range of investments in dams and other water infrastructure on rivers throughout the nation, including, for example, the Tennessee (U.S. Congress, H. Doc. 71-328, 1930), the Missouri (U.S. Congress, H. Doc. 73-238, 1934), and the Columbia (U.S. Congress, H. Doc. 73-103, 1933).

Figure 3-2: Some studies and their estimated costs, H. Doc. 69-308. The total cost of all studies from Maine to Hawaii was estimated as $7,322,400

Source: U.S. Congress (1926, section 5)

The 308 reports “became basic planning documents for many of the multipurpose projects undertaken by the federal government just before and after World War II” (U.S. Army Corps of Engineers 1998, 130.) With respect to the Tennessee, it is worth noting that the Tennessee Valley Authority Act of 1933 (U.S. Congress 1933a) provided not only for multipurpose development but for what might later be regarded as multiple objectives: “The establishment of the TVA marked the first time that an agency was directed to address the total resource development needs of a major region” (https://www.ourdocuments.gov).

Plans and Planning Methods The Corps carefully described what needed to be done in the 308 reports, including both the recognition of the substantial time and effort required and the importance of multiple purposes:

It is not believed necessary or desirable to provide at once for a comprehensive program of surveys or investigations covering all or even a large part of the streams included in this estimate, but the importance of this work is so great and the length of time required to complete it is such that it should be inaugurated as soon as practicable. There are evidently two principal purposes for which investigations of this nature would be useful; either for the preparation of plans for improvements to be undertaken by the Federal Government alone or in connection with private enterprise, or to secure adequate data to insure that waterway developments by private enterprise would fit into a general plan for the full utilization of the water resources of a stream. This department is now charged with examinations and surveys for navigation and flood-control improvements and with the construction of such projects as are authorized by Congress. In both classes of investigations the department must, by law, give consideration to the development of potential water power. In investigations for flood control it must also consider other possible utilization of water resources. Irrigation projects constructed by the Federal Government have not heretofore affected to any considerable extent the navigable capacities of any waterways which the Government has undertaken to improve for navigation, nor does it appear probable that they will do so to any considerable extent in the future. Inasmuch as consideration must be given to irrigation requirements in flood control investigations it would seem that Congress has made adequate provision for the proper coordination of all interests in those projects for the utilization of waterways with which this department is interested. … [T]he laws for the protection and preservation of navigable waterways are sufficiently broad to permit an equitable adjustment of conflicting interests and the investigations necessary to arrive at a

decision can now be made by the department without congressional authorization. (U.S. Congress, H. Doc. 69- 308, 1926, section 6) It is noteworthy from the present perspective that recreation as a purpose is not included, nor is environmental protection, including maintenance of ecologic systems; these purposes came later in the development of water resources planning methods and procedures. The planning processes for rivers extended over time, and the 308 reports, while providing a basis, were often just one in a series of reports that led to river development. An example of the 308 planning process and the work that followed is that of the Columbia River (dated 1932 and printed as U.S. Congress, H. Doc. 73-103, 1933), fully described in Willingham (2006, chap. 1). The report states that “The Columbia River and its tributaries are susceptible of being developed into the greatest system for water power to be found anywhere in the United States” (section 7). It further states that “The cost of this development will exceed that of any other single development of any kind for power that has ever been made” (section 11). One of several dams recommended was the Grand Coulee (fig. 3-3), which was estimated at $204,500,000; construction (by the Bureau of Reclamation) started in 1933 and was completed in 1942 (U.S. Bureau of Reclamation n.d.). Willingham describes the complex relationships of the Corps and the Bureau of Reclamation with respect to irrigation on the Columbia. The 308 reports laid the basis for planning for many other rivers, for example, the lengthy study for the Tennessee (U.S. Congress, H. Doc. 71-328, 1930): “The 308 plan for the Tennessee River Basin provided the basis for the Tennessee Valley Authority (TVA), a hallmark of New Deal resource planning” (National Research Council 1999, 12). The 308 planning efforts, like most planning, were not free of contention. One example is the Connecticut River project (U.S. Army Corps of Engineers, New England Division, 1934), described in

Figure 3-3: Grand Coulee Dam proposed in the Corps’ 308 report on the Columbia River Source: U.S. Bureau of Reclamation (n.d.), https://en.wikipedia.org/wiki/Grand_Coulee_Dam detail by Eve Vogel and Alexandra Lacy (2012). Local interests fought long and successfully to preserve hydropower development for state and private interests, as opposed to the comprehensive plans of the federal government (Vogel and Lacy 2012, 66). It should also be noted that many of the 308 reports did not recommend any federal development. One example is the Androscoggin River in Maine (fig. 3-4; U.S. Congress, H. Doc. 71-646, 1930). An extensive study was done of the river for flood control, power, navigation, and irrigation. No development was recommended, with costs expected to exceed any probable benefits. The river has a significant drop in elevation, and at the time of the survey it had been in large part developed by private interests for power. Costs and benefits for multiple purposes were not, as would later be the case, compared in detail; rather, available information was reviewed and general conclusions were reached as to the desirability of improvements. For example,

25 flood damages for certain years are given (4), as well as prevailing market rates for power and some costs of installation (22). It was only with the Flood Control Act of 1936 (see chapter 4) and its injunction that the federal government should participate in flood control projects on navigable waters “if the benefits to whomsoever they may accrue are in excess of the estimated costs” (U.S. Congress 1936, sec. 1) that the basis was provided for benefit-cost analysis to become a regular part of water planning. Two other acts especially affected 308 planning by the Corps during the 1930s and beyond. In 1935 Congress authorized the Corps to supplement the 308 reports with studies “to take into account important changes in economic factors as they occur, and additional streamflow records, or other factual data” (U.S. Congress 1935, sec. 6). “This authority charged the Corps with a broad responsibility to undertake continuing river basin planning, with the emphasis on navigation and flood control” (Baldwin 1998, 130). In 1936 the Flood Control Act stated that “it is the sense of Congress that flood control on navigable waters and their tributaries is a proper activity of the Federal Government in cooperation with States, their political subdivisions, and localities thereof” (U.S. Congress 1936, sec. 1). “Responsibility for federal flood control projects was given to the Corps of Engineers, while projects dealing with watershed run-off and soil erosion were assigned to the Department of Agriculture. This law made the Corps responsible for flood control throughout the nation, working in cooperation with the Bureau of Reclamation” (Baldwin 1998, 51). These two acts resulted in changes in Corps 308 procedures and recommendations: District engineers concluded that construction of reservoirs along such streams as the Yazoo and St. Francis [along the

Figure 3-4: The Lower Androscoggin Watershed, Maine. The Corps’ 308 report on this river did not recommend federal development. Source: U.S. Environmental Protection Agency (n.d.), https://cfpub.epa.gov/surf/huc.cfm?huc_code=01040002

27 lower Mississippi], while contributing to local flood control, would not be cost effective. This position proved increasingly politically unpopular in the midst of growing unemployment resulting from the Great Depression. Public works projects, once considered uneconomical, began looking very attractive as a means of employment. Moreover, many politicians felt that flood control was essential to protect human life no matter what the economists said. Mainly reacting to this political interest, the Corps reversed its position on a number of flood control projects. Revised reports concluded that the necessity for “public-work relief” and the suffering caused by recurring floods provided grounds for construction. (Baldwin 1998, 51) Such changes were not undertaken in the orderly manner that would have been required under later regulations; from a modern point of view, they referred to two elements of planning: the addition of the objective of saving lives and the use of underemployed or otherwise unemployed labor. The latter can be seen either as an objective in itself (increased employment in the midst of the Great Depression) or, from a purely economic point of view, as the utilization of “secondary benefits,” or, more theoretically, applying a zero or low shadow price to certain labor inputs.

Perspectives The 308 plans both resulted in a significant change in water planning methods, for the first time providing clear authority for the Corps to develop multipurpose river basin plans nationwide, and also had practical results, the development of a large number of dams and other water resources infrastructure. From a modern standpoint, the original objective of the 308 reports was, formally at least, economic, although in fact many if not most projects were selected at least in part based on some regional or other criteria. With the 1935 and 1936 acts, consideration of employment and lives saved were also brought into play; the idea of an environmental

28 objective was in the future. In terms of purposes, the general idea of multipurpose planning was firmly implanted, although it would now be said that additional purposes such as recreation, water quality, and maintenance of flows for ecological purposes would be added. The nationwide remit of the 308 program required continued interaction between the Corps and the Reclamation Bureau in the latter’s territory; this interaction was sometimes friendly and sometimes not (see chapter 6). In sum, the 308 planning effort resulted in a long-term change in water planning: the continued use of basin plans for multiple purposes throughout the United States.

Chapter 4 The Flood Control Act of 1936

U.S. Congress, Flood Control Act of 1936, P.L. 74-738, 1936

The Flood Control Act of 1936 was a landmark in the development of water planning methods, for three reasons: (1) it definitely established flood control as a federal responsibility (section 1); (2) it presented, for the first time in an easily accessible public document, the famous requirement that, for flood control projects authorized in the act, “the benefits to whomsoever they may accrue are in excess of the estimated costs” (section 1); and (3) it established the “a-b-c” requirements for local contributions (section 3). The act further provided for interstate compacts to share local contributions and operating costs (section 4); authorized more than 250 specific projects (section 5); and authorized preliminary examinations of other projects and problems (sections 6, 7). In addition, the Department of Agriculture’s watershed program was included in the act. A significant impetus for the act was the devastating flooding in the Northeast in 1936 (fig. 4-1). The act as passed was developed under the guidance of Dr. Royal S. Copeland (fig. 4-2), a Democratic senator from New York who chaired the Senate Commerce Committee, with the technical advice of Major General Edward M. Markham (fig. 4-3), the chief of engineers, and a young officer, Captain (later General) Lucius D. Clay, who went on to a brilliant military career in World War II and after. The Flood Control Act has been widely written about (Arnold l988a, 1988b; Schad 1988; Shabman 1988), and its influence has been clearly noted.

Flood Control as a Federal Responsibility The act states in section 1 that “it is the sense of Congress that flood control on navigable waters and their tributaries is a proper

Figure 4-1: Flooding in Pittsburgh, 1936

Source: Photo by Frank E. Bingaman. From the Collections of the Pennsylvania Department, Carnegie Library of Pittsburgh, Lorant Collection, http://www.info- ren.org/projects/btul/exhibit/neighborhoods/downtown/down_n254.html. activity of the Federal Government in cooperation with States, their political subdivisions, and localities thereof.” This was an important change, because for a long period in the nineteenth century (Arnold 1988a, 28), in accord with a strict interpretation of the Constitution, flood control was considered a local responsibility except when required to promote navigation and commerce. An important shift came with the Flood Control Act of 1917 (U.S. Congress, 1917; see Wright, 2000, p. 7); it provided for some federal responsibility for flood control in the lower Mississippi and the Sacramento basin in California. Prior to the Flood Control Act of 1928 (U.S. Congress 1928), the Corps had followed a “levees only” policy for the Mississippi (Wright 2000, 8). The 1928 act, passed following devastating floods in 1927, recognized the limits of relying on levees alone and authorized “floodways, spillways, and channel improvement operations, as well as continued work on levees” (Reuss 1988, xi). The 1936 act further expanded the role of the Corps, bringing it into the business of dam building in a big way (Arnold 1988a, 29–30).

Figure 4-2: Senator Royal S. Copeland. Copeland (1868–1938) was an American academic, physician, and politician. He represented New York as a Democrat in the United States Senate from 1923 until 1938 and was chairman of the Senate Commerce Committee, which developed the Flood Control Act of 1936 with advice from the chief of the U.S. Army Corps of Engineers, Major General Edward Markham. Source: U.S. Senate Historical Office

Benefits and Costs The statement in section 1 that the federal government should participate in flood control projects on navigable waters “if the benefits to whomsoever they may accrue are in excess of the estimated costs” is the first easily accessible reference to benefit- cost analysis in the water planning documentary history. It has been taken as the origin of formal economic benefit-cost analysis (Marglin 1967, 16; Major 1977, 2). The statement led directly to agency efforts to better define the meaning of benefits and costs and to establish the criteria that should govern project evaluation (see chapter 7 on the Green Book). It is one of the cases in the water resources planning record that presents a clear origin for an approach to evaluation. It is often forgotten, however, that this sentence is followed by the clause “and if the lives and social security of people are otherwise adversely affected,” that is, the objectives of such projects are not simply economic. The precise

Figure 4-3: Major General Edward Markham (1877–1950), chief of engineers from 1933 to 1937. Markham advised Senator Copeland on the Flood Control Act of 1936.

Source: U.S. Army Corps of Engineers (n.d.), https://web.archive.org/web/20050306124500/http://www.hq.usace.army.mil/history/ coe3.htm#36 provenance of the statement regarding the comparison of benefits and costs is unknown, although it may simply have been derived from developments in the 308 reports (chapter 3) and earlier efforts (see Shabman 1988, 116–19). The general idea of considering both the beneficial effects and the costs of proposed projects dates at least to the law authorizing the Corps’ Board of Engineers for Rivers and Harbors (U.S. Congress 1902b, section 3). Nonetheless, it is the clear and precise statement in the Flood Control Act of 1936 to which formal benefit-cost analysis can properly be traced.

Local Contributions Section 3 of the act sets out the “a-b-c” requirements that local interests must meet before the construction of an authorized

34 project can begin. Local interests are required to “(a) provide without cost to the United States all lands, assessments, and rights-of-way necessary for the construction of the project; (b) hold and save the United Sates free from damages due to the construction works; and (c) maintain and operate all the works after completion in accordance with regulations prescribed by the Secretary of War.” In the event that local contributions exceed the construction costs of the project, the federal government will reimburse half of this excess. These provisions were changed in several ways in the Flood Control Acts of 1938 and 1941, with the result that, while contributions were still required for local flood protection and channel improvement, the costs of dams became the sole responsibility of the federal government (Schad 1988, 33).

Projects Many of the projects authorized in the Flood Control Act of 1936 (and in later acts) were ultimately constructed; one example is the controversial Kinzua Dam in Pennsylvania (fig. 4-4), one of the largest dams east of the Mississippi. The chief of engineers in 1988 referred to the “hundreds of reservoir, levee, and channelization projects that resulted from the 1936 act and subsequent amendments” (Heiberg 1988). According to Theodore Schad, “Disruptions resulting from World War II essentially froze the nation’s flood control program into the form it had taken in 1944. The Flood Control Act of 1944 authorized over a billion dollars’ worth of projects, including the first eleven watershed programs recommended by the Secretary of Agriculture under the provisions of the 1936 Act” (1988, 34).

Perspectives The Flood Control Act of 1936 had great influence on the water resources programs of the Corps in the decades that followed.

Figure 4-4: The Kinzua Dam. This dam in the Allegheny National Forest in Warren County, Pennsylvania, is one of the largest in the United States east of the Mississippi River. Authorized by the Flood Control Acts of 1936 and 1938, actual construction was begun by the Corps of Engineers in 1960 and completed in 1965. The dam and the associated reservoir submerged over 9,000 acres of the Seneca Nation’s habitable land, their ancestral homes, farms, community centers, and burial plots (Rosier 1995, 345). The main purpose of the dam is flood control on the Allegheny River. Kinzua controls drainage on a watershed of 2,180 square miles, an area twice the size of Rhode Island. Side benefits include drought control, hydroelectric power production, and recreation.

Source: U.S. Army Corps of Engineers (n.d.), http://www.lrp.usace.army.mil/Missions/Recreation/Lakes/Kinzua-Dam-Allegheny- Reservoir/

Perhaps the case is best summed up by a 1988 quotation from the then assistant secretary of the army for civil works: Few laws have had such an impact on the United States Army Corps of Engineers as the 1936 Flood Control Act. For over 50 years before passage of this act, Congress had been periodically charging the Corps with flood control responsibilities. However, lawmakers generally justified the work on the basis of aiding navigation. It was only in the 1936 act that Congress stipulated that flood control

was an appropriate federal activity. The act authorized hundreds of flood control projects and established policies that endure to this day. (Page 1988) This influence has been felt in the establishment of flood control as a federal function, in the relationship of benefits to costs, and in the local contribution requirements. Indeed, according to Joseph Arnold, the Flood Control Act of 1936 “was unquestionably a major step forward in the conservation and development of the nation’s water resources” (1988a, 24). Specifically relating to benefits and costs, the act is referred to in appendix 1 to the Green Book of 1950 (not included in the 1958 version): “This latter act also establishes the principle of comparing benefits to whomsoever they may accrue with the estimated costs.” It is also stated there that the Department of Agriculture has watershed treatment programs under various laws for soil conservation and the Flood Control Act of 1936, but only the latter requires a definite analysis of benefits and costs (61). (See chapter 7 for more on the Green Book.) It is difficult to differentiate between the direct influence of the act, which was considerable, and the influence of the (social) welfare economics model of optimal resource allocation in this regard. The welfare model is the explicit basis of standard treatments of benefit-cost analysis, such as the classic work of Otto Eckstein (1961, vii). Undoubtedly the techniques would have developed without the act, but also without question it was a spur to action. It is an interesting question why the benefit and cost procedures were not more clearly spelled out in the act. According to Arnold (1988a, 20), this may have been because the extensive battles (including ideological wrangles about the role of the federal government) over the local interest provisions of the act took away the appetite of the Senate Commerce Committee for another fight over the nature of benefits and costs. It simply expected the Corps of Engineers to lead the way. It is also of interest to consider the fact that the benefit-cost statement in the act led primarily to assessments of economic benefits and costs. The context of the times made it clear that more was at

37 stake (Maass 1988), but the logic of planning for water resources management and development with multiple objectives was made clear only many years later (Maass et al. 1962; see chapter 11 below). Many observers, however, did not view the act itself favorably. Arthur Maass referred to it as this “confused and confusing piece of legislation” (de Roos and Maass 1949, 23, cited in Arnold 1988a, 24). One problem was that it moved planning away from the expanding idea of multipurpose development to single-purpose, primarily structural solutions. Arnold (1988a, 24) says that it had the negative impact of postponing for years the implementation of a more comprehensive river basin development program favored by the Natural Resources Committee and President Franklin D. Roosevelt. It is intriguing that Gilbert White worked in Washington for, among other entities, the Water Resources Committee during the time of the Flood Control Act of 1936 and while he was developing his Ph.D. thesis, submitted in 1942 (White 1945; see chapter 5 below). He helped draft a veto message for the act, part of a routine in which President Roosevelt was presented with all options (Moore and Moore 1989, 35).

Section II The Classic Model of River Basin Planning

The essays in this section deal with both government and academic work that provided an overall framework for the development of federal multipurpose water resource planning beginning in the 1930s. Gilbert White’s 1945 analysis of human adjustment to floods, the subject of chapter 5, was developed in a period when water resources were one of the most significant federal domestic responsibilities. It is a wide-ranging assessment of the full array of methods of adjustment and includes the well-known observation that, unless planning was comprehensive, some flood control measures could potentially increase, rather than decrease, damages. Arthur Maass’s work Muddy Waters: The Army Engineers and the Nation’s Rivers (1951) is the focus of chapter 6. The book assessed the performance of the Corps of Engineers against performance criteria regarding interest groups, the executive and legislative branches, and the profession. Maass found the Corps wanting in most respects, especially in its overly close relations with Congress. This and White’s study have formed part of the framework for thinking about federal water resource planning to the present day. Chapter 7 deals with the “Green Book,” the U.S. Inter- Agency River Basin Committee’s report formally entitled Proposed Practices for Economic Analysis of River Basin Projects (revised 1958; originally published 1950). This is a landmark in the development of water resources planning methods in the United States. It is generally considered to be the foundation of applied economic benefit-cost analysis in federal water planning. In addition, it contains important references to objectives other than the economic objective.

The work of John Krutilla and Otto Eckstein, Multiple Purpose River Development (1958), discussed in chapter 8, is an effort to analyze and clarify important issues of multipurpose water project planning, with emphasis on the economic benefits and costs of projects and their impacts on income distribution. While the authors focus on economics, they are clear in their discussion that other objectives are important in water planning. Chapter 9 concludes this section with a discussion of the Delaware River Basin Report, prepared by the U.S. Army Corps of Engineers, Philadelphia District (December 1960; rev. May 1961). This massive eleven-volume report represents the full effort of the Corps on multipurpose river basin planning as it was understood at the time. It can be seen as a culmination of the processes that included, among other key legislation, the 308 reports (chapter 3) and the Flood Control Act of 1936 (chapter 4).

Chapter 5 Gilbert White, Human Adjustment to Floods

Gilbert Fowler White, Human Adjustment to Floods, Department of Geography Research Paper no. 29, University of Chicago, 1945

Gilbert White’s groundbreaking doctoral thesis in the Department of Geography at the University of Chicago was completed in 1942 and published as a research paper in 1945 (fig. 5.1). Like Design of Water-Resource Systems (chapter 11), it is an academic research enterprise that has had a significant impact on water planning over a long period of time. The overall structure of the work is a thorough look at the various ways, and their pros and cons, in which humans can and do adjust to floods. It may be said to be based on White’s insights, as a geographer, about human interaction with the land, or, as he phrased it, “Floods are ‘acts of God,’ but flood losses are largely acts of man” (2; page references in this chapter are to White, 1945 unless otherwise specified). Its particular contributions include: the breadth of view; what would now be called structural and nonstructural measures; its concern with proper benefit-cost accounting (then just in development); and finally and perhaps most well-known, White’s observation that some flood protection measures, unaccompanied by an assessment of all potential measures (such as land use controls), could actually increase rather than decrease flood damages.

White (figs. 5-2, 5-3) completed the formal requirements for the Ph.D. degree by 1934 and then spent eight years in Washington, D.C., with the Mississippi Valley Committee, the National Resources Committee, the National Resource Planning Board, and the Bureau of the Budget. His interest in the flood management problem developed from this (Reuss 1993, ix, 5–6, 24).

Figure 5-1: University of Chicago, Geography Research Paper no. 29, l942

Thus, at the time when he was working on and completing his thesis, he was already an experienced researcher and analyst with a deep knowledge of the flood management problem. He wrote in a remarkably clear and effective style that marked his work throughout his career.

The Context White wrote after the Flood Control Act of 1936 (chapter 4) had been enacted, when questions of benefit-cost measurement and the federal role in flood management had come to the fore. White cited not only the act but also the TVA program:

The Flood Control Act of 1936 marks the initiation of flood protection on a national scale, and of surveys of landmanagement problems in connection with engineering surveys. In these respects it undoubtedly owes much to the

operations of the Tennessee Valley Authority. Under the terms of the Tennessee Valley Act of May 18, 1933, the Authority had launched the construction

Figure 5-2: Gilbert F. White as President of Haverford College, 1946-56

Figure 5-3: Gilbert F. White receiving the National Medal of Science, 2002

Source: National Science and Technology Medals Foundation

of a system of reservoirs for the joint purposes of flood control, navigation, and hydroelectric power development. Those reservoirs were visible demonstrations of the practicability of flood protection by such means, and they were undertaken solely at Federal expense. The Authority had been directed to combine studies of forest and soil conservation with its engineering work, and for the first time in legislative history the desirability of integrated development of resources on a regional basis had been recognized. These accomplishments and other conditions combined to promote public support for Federal participation in somewhat similar types of work in other sections of the country. (11–12) It is of interest that in his thesis White cites works in engineering journals but not the classic study of welfare economics, the optimal allocation of economic resources (Bergson 1938). Bergson’s work would have been relatively obscure at the time, although the welfare model appears likely to have affected later work, such as the Green Book (chapter 7).

Content and Approach White later expressed this considered view of his thesis: My judgment, which may not be that of others, is that the dissertation suggested a simple but fundamental concept that, for any mode of resources management, finding the optimal use of a resource theoretically involves canvassing the whole range of alternatives that are open to society, and then trying to estimate what the consequences would be, both favorable and unfavorable, of undertaking any one of those alternatives or a combination of them. This was applied to floods. The same notion would apply to any other kind of resource management. (Reuss 1993, 27)

Within this context, White said in his thesis that adjustments to floods may be divided usefully into eight major classes, as follows: Land elevation. The elevation (up-building) of the surface of a flood plain so as to render it less susceptible to flooding. Flood abatement. The application of land-management measures upstream from a flood plain to prevent the accumulation of all or part of a flood flow in a stream channel. Flood protection. The use of levees, channel improvements, cut-offs, floodways, reservoirs, and other engineering devices to reduce flood crests in a stream channel or to prevent floods from overflowing flood plains. Emergency measures. The temporary removal or protection of property and persons, and the temporary rescheduling of human activities. Structural. The arrangement of physical structures, such as buildings, roads, and communication systems, and of institutional structures, such as production schedules and marketing procedures. Land use. The arrangement of the pattern of land use of a flood plain. Relief. The granting or loaning of private or public assistance to flood sufferers. Insurance. The accumulation of premium payments from property owners in order to compensate them for losses resulting from floods. (47) There follows a careful analysis of the types of data then available on flood losses: Because the systems for collecting data with respect to the amount and character of flood losses in the United States are incomplete, inaccurate and inconsistent, it is impossible to estimate with accuracy the full extent of such losses. While precise measurements of the flow and peak of floods have been instituted with meticulous care under the guidance of the U. S. Geological Survey, the

social impacts of those events have been canvassed lamely and inadequately. (53) Under these circumstances there is no assurance that all flood losses will be canvassed in detail each year, inasmuch as field surveys are made in only a few areas and questionnaires do not yield returns from the remainder of the country in every instance. Moreover, even if all areas experiencing flood loss were covered by surveys of some type, the questionnaire method of appraisal used by the Weather Bureau is dependent upon the judgment of so many individuals working under such vague instructions and with criteria so lacking in precision and uniformity that the accuracy of the results would be open to grave doubt, particularly because the Weather Bureau makes no systematic provision for verifying the returns. These weaknesses relating to coverage and to accuracy can be corrected in large measure by administrative changes aided by additional funds. However, new personnel, better organization, and larger funds might not result in an adequate data- collection system. That can come only as sound criteria for evaluating losses are generally adopted. (55–56) Given the problems with flood loss data, White’s famous observation that some flood management efforts could increase rather than decrease losses is carefully phrased: “In the light of meager evidence it seems possible that reservoirs and channel improvements unless supplemented by landuse measures may induce or promote further encroachment upon a flood plain, and so may increase rather than decrease mean annual losses” (206). As part of a comprehensive review of adjustments in the conclusion to the thesis, there are several that relate to the issue of losses increasing rather than decreasing: (a) All possible adjustments except those in land use and insurance tend to favor the preservation of existing land occupance.

(b) Public relief favors further encroachment upon a flood plain by bearing part of the costs of such encroachment. (c) Effective emergency measures also may favor encroachment in less degree by reducing the hazard of flood loss. (d) Encroachment upon flood plains is likely to continue so long as the riparian doctrine is not modified by public regulation. (209) It should be noted that these observations illustrate the point at hand but do not dominate the wide range of analyses in the concluding section.

Perspectives White’s thesis is an example of the work of an individual mattering a great deal in water resources planning (see Major 2010, with respect to Harry Schwarz); it also illustrates the potential impact of academic work on water resource planning methods. Another example is the multiple-participant Harvard Water Program (see chapter 11). While, as Macdonald et al. (2012) point out, it is difficult to trace exact influences from White’s work throughout the gray literature and forgotten agency discussions about flood management, it seems clear that White’s study stood for a long time as the most comprehensive review of flood management in the United States and that it was often and widely consulted: “The University of Chicago at that time had a policy of simply publishing a few copies of a doctor’s dissertation, putting it in a few libraries around. And that was what was done with mine. Then people began hearing about it and asking for copies, and so they published it in what came to be a monograph series. So you’ll find different dates ranging from ’42 to ’45 depending on when people got access to it” (Reuss 1993, 27). As one example of the impact of White’s work, figure 5-3 shows a page annotated by Arthur Maass on his own copy of the thesis. Overall, White’s landmark work has been influential for many years in shaping requirements that planners look at nonstructural as well as structural measures, and especially

47 that they should consider land-use regulation in conjunction with structural measures to prevent increases in damages.

Figure 5-4: An example of Arthur Maass’s comments on Gilbert White’s 1945 thesis. Maass was an exceptionally careful reader; on this page of his personal copy of the thesis he noted his approval (one vertical line next to a sentence or paragraph) and strong approval (two lines) of statements in White’s “Outline of a Geographical Approach” (White 1945, 34). Maass’s copy of White’s thesis is in the U.S. Army Corps of Engineers Institute of Water Resources at Fort Belvoir, VA.

Source: IWR Maass-White Library

The outbreak of World War II of course shifted public policy away from some domestic issues for a time. (White himself was in Europe for the Friends Service Committee and then was interned in Germany for thirteen months [Reuss 1993, 22–23].) As a result, much consideration of flood management policy and water resource policy more generally awaited the end of the war.

Chapter 6 Arthur Maass, Muddy Waters

Arthur Maass, Muddy Waters: The Army Engineers and the Nation’s Rivers (Cambridge, MA: Harvard University Press, 1951) Muddy Waters (fig. 6-1) is based on the Ph.D. thesis at Harvard by Arthur Maass (fig. 6-2), “a scholar who wanted not only to find out how the institutions of government actually work, but also what could be done to make them better serve the common good” (Mansfield et al. 2007). This assessment fits well with the intent of Muddy Waters. The book sets out criteria for gauging administrative responsibility and evaluates the Corps’ work against these criteria in terms of the adjustment of group interests, responsibility to the executive and legislative branches, and responsibility to the profession. The book concludes with a lengthy case study of the Kings River Project in California. Maass’s main fields of research and teaching over a long career, public administration and the legislative process, are both

Figure 6-1: Muddy Waters Source: Harvard University Press

Fig. 6-2: Arthur Maass (1917–1984), as Frank G. Thomson Professor of Government, Emeritus, Harvard University

Source: Harvard Gazette reflected in this work. It will be no surprise to many readers that Maass finds the Corps wanting in most respects, especially in its overly close relations with Congress. In particular, Maass identified and analyzed the complex networks of influence among interest groups, congressional committees, and bureaucratic agencies, a situation that, although he did not use the term, became known as an “iron triangle.” The book is notable not only for its careful scholarship but also for an “explosive” (Mansfield et al. 2007) foreword by Harold L. Ickes, the former secretary of the interior. Accordingly, this chapter includes a review of Ickes’s piece, in addition to an assessment of Maass’s work itself. The chapter ends with a brief summary of some parallels between the careers of Maass and Gilbert White.

The Ickes Foreword Harold Ickes served as United States secretary of the interior for thirteen years, from 1933 to 1946, and in that capacity was well

52 stocked with interagency grievances against the Corps. In his own preface to the book, Maass is clear about his intentions in asking Ickes to contribute the foreword. After expressing his admi- ration for Ickes and his career, Maass says: “I am intensely interested in doing what little I can to bring about a change in the Army's programs for resources development. A foreword to the book by Harold Ickes increases its effectiveness manifold” (vi; page references in this chapter are to Maass, 1951, unless otherwise specified). When the book was published, Maass was already well versed in the ways of Washington and no doubt felt that Ickes’s contentious foreword would help to publicize what might otherwise be taken as just another academic study. Ickes, in writing of the Corps, pulls no punches: “No more lawless or irresponsible Federal group than the Corps of Army Engineers has ever attempted to operate in the United States, either outside of or within the law” (xiv). He adds: “It is to be doubted whether any Federal agency in the history of this country has so wantonly wasted money on worthless projects as has the Corps of Army Engineers. (xii). Ickes then cites chapter and verse. For example: The power-hungry Army Engineers will commit Federal funds without interest and without thought of repayment for projects that it has been settled National policy to build only upon an obligation to repay. For instance, they fought brazenly for the right to build Pine Flat Dam on the Kings River in California and won. Originally, this was to be a reclamation project. Under the Reclamation Law, the water users benefited by Pine Flat Dam would repay in installments the estimated value of the water used by them for irrigation purposes. This would go into a revolving fund, out of which additional reclamation projects in future might be financed without undue drains upon the Treasury. Under the pretext of “flood control” and despite both oral and written orders to the contrary from President

53 Roosevelt, the Army Engineers defiantly lobbied through the Congress a bill giving it the authority to build this dam, although its real purpose was to provide irrigation water and generate power. (xi–xii) The intemperance (and perhaps partial truth) of Ickes’s writing contrasts starkly with the carefully modulated scholarship of Maass both in this book and in his later scholarly publications. In a later interview (Reuss 1989, 11–12), Maass discussed his dealings with the strong-minded Ickes in producing the foreword.

Muddy Waters Maass’s work is structured in a powerful manner. He sets out in chapter 1, “Gauging Administrative Responsibility” (written with Laurence Radway, then of Dartmouth College), criteria by which the adequacy of administrative performance may be measured. These relate to adjustment of group interests, responsibility to the executive and legislative branches, and responsibility to the profession, which are the topics of the chapters that follow. Throughout the work there is a prodigious amount of assessment of documentary material relating to Corps methods, projects, and recommendations, of which Maass would have been aware and to which he would have had access, based on both his academic training and his service in Washington. The proposed guidelines are worth setting out: Responsibility for What? Working Bias: An administrative agency should be responsible for formulating as well as executing public policy. (3) Maass and Radway point out here that, while “the distinction between policy making and policy execution may have a great deal of practical value … [p]olitics and administration are not two mutually exclusive processes; they are, rather, two closely linked aspects of the same process” (3).

Responsibility to the People at Large: Working Bias: An administrative agency cannot and should not normally be held directly responsible to the people at large. (4) This is explained in this manner: “In general it is becoming clear that direct control by the public at large cannot insure administrative responsibility and that the influence of John Doe can be exerted effectively only through the legislature, the executive, and special interest associations.” (5) Responsibility to the People—Pressure Groups: Working Bias: An administrative agency should be responsible to pressure groups so far as necessary to equalize opportunities for safeguarding interests, to acquire specialized knowledge, and to secure consent for its own program. (5) Here Maaas and Radway opt for a limited, preferably formalized, advisory relationship of interest groups and stress the importance of an agency neither suppressing group demands nor being unduly responsive to these demands. (8) Responsibility to the Legislature: Working Bias: An administrative agency should be responsible to the legislature, but only through the chief executive, and primarily for broad issues of public policy and general administrative performance. (8) This is at the heart of Maass’s critique of the Corps and was a continuing element of his work throughout his career. “System Design and the Political Process: A General Statement,” chapter 15 of Design of Water-Resource Systems (see chapter 11 below), has a model that embodies this principle; see also Maass (1983). As Maass and Radway expand on this, the responsibility of administrative agencies to the legislature through the chief executive should be “primarily for the broader questions which arise in formulating and executing policy” (11).

55 The final three guidelines are as follows: Responsibility to the Chief Executive: Working Bias: An administrative agency should be directly responsible for conforming to the general program of the chief executive and for coordinating its activities with other agencies of the executive branch. (13) Responsibility to Political Parties: Working Bias: An administrative agency cannot be held independently responsible to the organization or policies of political parties. (16) Responsibility to Profession: Working Bias: An administrative agency should be responsible for maintaining, developing, and applying such professional standards as may be relevant to its activities. (16) One further possible administrative responsibility, to the courts, is not discussed in detail. Maass and Radway note that they could add little to the many specialized studies of administrative law then available (18). The body of the work evaluates the Corps against these criteria (each of which is developed in more detail) in chapters on the adjustment of group interests, responsibility to the executive and legislative branches, and responsibility to the profession. The book concludes with a lengthy case study of the Kings River Project (which so exercised Ickes) that takes into account the analysis of the earlier chapters. As to adjustment of group interests, Maass notes at the end of chapter 2 that conclusions are tentative until the remainder of the analysis in the book, but he nonetheless states: “If, as seems to be the case, the Engineer procedure requires or allows a continued emphasis on the localized aspects of individual water projects, an avoidance of responsibility on controversial projects, and a shifting of responsibility to others where its assumption might mean loss of some support, then it would appear that the public interest is not attained by the Corps” (60).

In the Kings River Project case study, Maass notes an inappropriate relation to interest groups: The procedure whereby the Corps of Engineers and the Bureau of Reclamation have announced to local interests the tentative recommendations contained in field survey reports, before these recommendations are fully reconciled with those of other Federal agencies, has made effective reconciliation more difficult. The recommendations acquire partisan support and opposition among local interests who attach themselves to, and provide support for, one Federal agency or the other. (257) With respect to responsibility to the executive and legislative branches, Maass’s criticism is unstinting: A realistic appraisal of the situation in 1949, however, must lead to the conclusions that (1) the USED [United States Engineer Department] is still responsible directly to committees and to Members of Congress to a significant degree; (2) the USED is responsible to the chief executive to a much greater extent than in 1934, but not to a point which approaches the set standard; (3) the desired procedure for planning water developments has not been realized; (4) the USED has not assisted in developing legislative standards by recommending national water policies. The Corps still regards the development of policy as the sole duty of Congress and regards itself as no more than the executing agent of specific Congressional directives. Finally (5), this very pattern of legislative responsibility has continued to guarantee for the Army Corps of Engineers an unusual degree of Congressional support. (132–33) The “desired procedure for planning water developments” to which Maass refers here is strikingly similar to the Corps’ work in the Delaware River Basin Report (see chapter 9 below). It is described in detail on pages 69–71 of Muddy Waters. As with

57 his career-long concern with appropriate agency relations to the executive and legislative branches, Maass continued to ponder the issue of effective planning methods throughout his work, with the high point perhaps reached in Design of Water-Resource Systems. Throughout Muddy Waters, the level of detailed analysis is impressive, especially given that Maass worked exclusively on hard copy in traditional library and archival settings. And, amid a welter of congressional, executive, and other documents, Maass somehow managed to unearth the memorable and unexpected, for example, a short note from President Franklin Roosevelt dated August 12, 1937, to “Dear Uncle Fred,” the president’s uncle, the businessman and civic leader Frederic Adrian Delano. Delano was chairman of the National Planning Board and later of the National Resources Planning Board. This note concerns the Corps’ relation to the Congress, the subject of chapter 3 of Muddy Waters. A joint resolution had been passed that, unbelievably, would have had the secretary of war report directly to Congress, rather than to the executive, with regard to water planning. The National Resource Council was strongly opposed to this and other aspects of the joint resolution and recommended that President Roosevelt veto it. The president wrote to his uncle to assure him that he both had constitutional authority to veto a joint resolution and would in fact veto this one (93). With respect to responsibility to profession (chapter 4 of the book), Maass’s conclusions are also hard-hitting: Using as a guide a composite of professional views, such, for example, as that found in the reports of the Water Resources Committee of the National Resources Planning Board, we have attempted to hold the Corps of Engineers to a detailed accountability for Federal executive policy for flood control and navigation. With less precision, we have held the agency accountable for the manner in which it has conformed to approved power and irrigation policies in planning its multi-purpose developments. That the Engineer Department fails to live up to a great many of the

accepted standards of professional responsibility in all of these respects has been pointed out. (207) The Kings River Project case study, chapter 5 of Muddy Waters (see fig. 6-3), reports on a highly contentious set of agency and user battles. Maass frames this with the following observation: The basic cause of failure [to resolve the Central Valley conflict] lies in the fact that two Federal water resource agencies are operating in the same river basin, each planning multiple-purpose water control projects. It may well be that in situations such as that in the Central Valley, where two agencies are doing the same job in the same area, coordination cannot succeed. Coordination implies the relating of similar, but not wholly duplicative, tasks. There can be coordination when two agencies are doing the same or similar jobs in different places. (254) This is the conflict described by Ickes in his foreword (xi–xii).

Figure 6-3: Existing and proposed major water developments in the state of California, used to illustrate the analysis of the Kings River Project, Muddy Waters, chapter 5

Source: Harvard University Press

Maass’s overall conclusions with respect to the Corps, found on the final page of the book, are as follows: There is no question that significant improvements have been made; and, it is hoped, these improvements have been pointed up. But the progress has in no way kept pace with the increasing complexities of multiple-purpose river-basin development. If the major water projects under development today were similar to the singlepurpose projects of fifteen years ago, the present procedures of the Corps of Engineers would approach more nearly the requirements of administrative responsibility. Since 1936, however, the Engineer Department has become increasingly involved in very complex operations the manner of whose execution has the most profound effects upon the social structure and economic welfare of large regions and of the nation as a whole—“a seamless web: the unity of land and water and men.” The Corps has failed to grow to the task. (259)

Perspectives The quote by Mansfield et al. (2007) with which this chapter began is an apt summary of Maass’s efforts in Muddy Waters: “a scholar who wanted not only to find out how the institutions of government actually work, but also what could be done to make them better serve the common good.” In this work, Maass succeeded to a significant degree. There are two main themes in the work, dealing with planning procedures and with the Corps’ relationship to the executive and the legislative branches. With respect to the first, Maass’s influence was felt earlier, and in fact what the Corps has now done for many years fits well with the “Procedure for Coordinated Planning,” described in detail on pp. 69–71 of the book. The agency wars described in the Kings River Project case study are also largely over in that form, although of course bureaucratic jockeying remains an integral part of Washington today. It is perhaps with respect to the Corps’

60 relationship to the executive and the legislative branches that Maass’s influence has been most pervasive: it is the mark of a classic study that no one now can imagine thinking about the Corps without thinking about this issue. In this regard, Maass’s influence is like that of White (see chapter 5), in that it would be difficult to think about adjustments to floods without the pervasive effect of White’s work.

Maass and White Maass and White had interesting parallels in their careers. White was born a few years before World War I (1911), and Maass during that conflict (1917). Both worked extensively in Washington agencies, White during the 1930s and early 1940s, and Maass during World War II. These were times when Washington was truly the center of intellectual excitement and decision-making power. Maass spent the war years and immediately thereafter serving with the Office of the Secretary of the Navy, the Natural Resources Planning Board, the Bureau of the Budget, and the water resources staff of the Natural Resources Task Force of the first Hoover Commission. White was a conscientious objector who served in France and was later interned in Germany; Maass was a naval occupation officer in Japan after the war. Both went on to tenured positions at great research universities: Maass at Harvard and White at Chicago. They were, respectively, the preeminent political scientist and geographer of water resources (and much else) in the latter half of the twentieth century. In both cases their doctoral theses— White’s published as a thesis, Maass’s rewritten as Muddy Waters—had a remarkable effect on water resources planning that continues into the twenty-first century. Maass, in fact, used White’s work in his own thesis (vi). Maass and White had cordial professional relations over the course of their careers, and, among other things, they served together on the North Atlantic Regional Study Board of Consultants (see chapter 12). They are jointly honored by, and many of their books, reports and studies are kept

61 at, the Maass-White Library at the Corps’ Institute for Water Resources at Fort Belvoir, Virginia.

Chapter 7 The Green Book

United States Inter-Agency River Basin Committee, Subcommittee on Benefits and Costs, Proposed Practices for Economic Analysis of River Basin Projects, Report to the Federal Inter-Agency River Basin Committee, rev. ed. (Washington, DC, 1958, originally published 1950)

The report discussed in this chapter is one of the most important landmarks in the development of water resources planning methods in the United States. It is generally considered to be the foundation of applied economic benefit-cost analysis in federal water planning; in addition, it contains important references to objectives other than the economic objective. It has been universally known as the Green Book, after the color of its cover (fig. 7.1). The report is in fact a short text on applied benefit-cost analysis, covering most of the issues that relate to that analysis as it was understood at the time. The report is divided into six sections: (1) an introduction; (2) basic principles and concepts; (3) project and program formulation; (4) measurement of benefits and costs; (5) application of principles to various project purposes; and (6) cost allocation for multipurpose projects. The first version of the Green Book appeared just after World War II (1950); a second edition, with agency comments, was issued in 1958. The relatively minor revisions in the 1958 version are not marked in the text of that edition but are described in agency comments on the revision that are included in an appendix. The Green Book was never adopted as legally binding, but its influence has been significant. It followed in time two notable events: the Flood Control Act of 1936 (see chapter 4) and the full mathematical development of the “welfare model” of

Figure 7-1: The Green Book

Source: U.S. Inter-Agency River Basin Committee, Subcommittee on Benefits and Costs (1958) economics (Bergson 1938). In that model, resources are mobile and prices reflect input and output values, resulting in an optimal allocation of economic resources. (The reference to “welfare” is to social—that is, economic—welfare.)

There is no clear indication in the Green Book of whether it was considered a direct response to the Flood Control Act of 1936, or of how or whether the welfare model of economics influenced the work. However, it is clear that the authors of the Green Book made themselves thoroughly familiar with agency practices, which would include the benefit-cost injunction of the Flood Control Act of 1936; in fact a study of agency practices was the first task the authors set themselves (1958, 1), and the act was referred to in appendix 1 to the 1950 edition (not included in the 1958 version). The welfare model was well known at the time of the Green Book’s development. Kenneth Boulding’s (1952) clear geometrical explication was available, as was the still helpful summary of welfare economics by J. de V. Graaff (1963 [1957]).

The Economics of Project Evaluation The authors of the Green Book framed the needs of economic analysis well (fig. 7-2). According to the authors: To be most effective, the economic analysis must be oriented to and be consistent with the following principles: (1) The goods or services to be produced by a project have value only to the extent that there will be need and demand for the product. (2) The most effective use of economic resources required for a project is made if they are utilized in such a way that the amount by which benefits exceed costs is at a maximum rather than in such a way as to produce a maximum benefit-cost ratio or on some other basis. Maximization of net benefits is a fundamental requirement for the formulation and economic justify- cation of projects and programs. (3) The project as well as any separable segment or increment thereof selected to accomplish a given purpose should be more economical than any other actual or potential available means, public or private, of accomplishing that

specific purpose. The cost of making the product or service available by alternative means establishes a limit to the justified project investment for accomplishing a specific purpose. (4) From an economic standpoint the order in which a number of projects should be undertaken should be based on their relative efficiency in use of economic resources. (5; all page references below are to the 1958 edition unless otherwise specified)

Figure 7-2: Optimization of benefit-cost analysis

Source: Green Book, 1958, p. 13, upper half of figure 1.

The measure of benefits that the authors rely on is generally market price (17), although alternative cost is also recommended where appropriate. While emphasizing the importance of market price as the best framework for consistent evaluation, the authors provide an insightful review of perspectives on the adequacy of market price:

[T]he use of market price as a criterion for determining the worth of public projects may involve certain deficiencies as well as present difficulties of measurement. Specifically, the price of a product or service in the private market may inadequately reflect its value from a public viewpoint. Rigidities in the economic system, the lack of complete consumer knowledge, patterns of income distribution, and the absence of an entirely satisfactory means of expressing the public viewpoint through purchase decisions, limit the area in which the private market can provide completely satisfactory standards for the evaluation of project effects. (6–7) However, the authors go on to say that “despite the limitation of the market price system in reflecting values from a public viewpoint, there is no other suitable framework for evaluating the effects of public works projects in common terms. Accordingly, the market price system is the starting point for formulation of principles for benefit-cost evaluation” (7). In this respect, the Green Book does not fully develop the willingness to pay criterion that, in principle, allows for the application of the welfare model to nonmarket goods such as recreation. (This approach is explained in detail in Design of Water-Resource Systems [see chapter 11]; it was developed in applied form for water resource planning quite soon after the Green Book. See, for example, Merewitz 1966.) The authors follow this up with a correct explanation that the quantity to be maximized in establishing the scale of a project is net benefits rather than the benefit-cost ratio. This is shown in figure 1 of the book (see figure 7.2, above), which also indicates that the optimal point is where marginal benefits equal marginal costs. While this point seems obvious today and is a routine part of textbooks, it was not universally understood at the time. Much of the report deals with the details of prices, periods of analysis, and methods for estimating costs and benefits for various purposes. In section 5, for example, the many purposes

67 examined reflect the range of agencies and programs with which the authors were dealing. The purposes include irrigation and related drainage, flood and erosion control for agricultural land, flood control, navigation, electric power, municipal and industrial water supply, recreation, fish and wildlife, and water pollution control. The last, as noted in comments by the Department of Health, Education, and Welfare (55), was an add- on to the revised report. There is extensive discussion of secondary benefits, a continuing issue for the agencies (9–10). In this context, these are benefits that represent income to a region but are not additions to the national product; they are therefore not countable in the national economic development account, but they may be considered where there is a regional objective (10). There is also a discussion of what is now called “shadow pricing”: where there is heavy unemployment labor may be valued at less than market rates for economic project evaluation purposes (9). An extensive and thoughtful discussion of interest rates, a still unresolved issue in water resources planning, is provided (22– 24), and the authors conclude that, pending further investigation, an average rate of return on federal bonds is appropriate, providing that adequate allowance for risk and uncertainty is made in project formulation (24).

Objectives Although the Green Book is now generally remembered for its detailed development of economic methods, there is perhaps surprising attention paid to what would now be called multiple objectives. This is both because of an awareness of the importance of such objectives to society and because of the extent to which they are embodied in some of the laws and regulations governing the agencies, such as the Reclamation Act of 1902 (see chapter 2). The importance of multiple objectives is made clear at the beginning of the work: The public policies governing the development of the Nation's water and related land resources are not necessarily determined solely on the basis of economic

considerations. Thus, for example, regional development and national defense have been objectives of various resource development programs. Even in such cases, economic analyses may serve a valuable purpose by showing the extent to which costs must be incurred to accomplish expected tangible or intangible results. (3) This is further emphasized with the following statement: [F]rom the standpoint of society as a whole there may be beneficial or adverse effects that would escape consideration in a summation of individual evaluations, as for example, effects on health and welfare, improvement of underdeveloped areas, value of resources conservation to future generations, and effects on national security. In applying the public viewpoint to economic analysis of projects it is essential that consideration be given to all effects of a project and that such effects be evaluated as completely as possible. (6) The report concludes that, where project sizing and formulation are based at least in part on other factors than economic benefits and costs, this should be made clear and the implications for economic benefits and costs spelled out. This is where the consideration of multiple objectives was left in the report; it would await the more formal analysis of Design of Water-Resource Systems (chapter 11, below) to provide a suitable logical framework for such considerations. This relationship of economic and other objectives is one that has continued as a central challenge in water planning to the present day.

Perspectives The Green Book was a very substantial effort in the context of the time and has had a lasting influence as a resource, although it was not adopted by the Federal Interagency Committee because of agency differences; Caulfield (2000, 14) refers in particular to disagreements over the inclusion of secondary benefits in Bureau

69 of Reclamation projects. Its relationships to one contemporary document, U.S. Bureau of the Budget Circular A-47 (1952), and one document that soon followed it, the U.S. President’s Water Resources Council Report (1962), are of interest. A memoir by Henry Caulfield (2000), who chaired the committee that prepared the latter report, describes these two documents in the context of the Green Book. Circular A-47, the publication of which was noted in the Green Book (2), was an attempt by the Bureau of the Budget to impose strict economic criteria on water projects, in particular requiring, aside from exceptional circumstances, that economic benefits exceed economic costs. This was “in contrast to the Green Book, which called for the application of its criteria and principles within the framework of an agency’s particular program and responsibilities” (Caulfield 2000, 15). Con- gressional dissatisfaction with Circular A-47 led ultimately to the U.S. President’s Water Resources Council Report, described in chapter 10 below. This document, which replaced Circular A-47, was a multiobjective document that appeared in the same year as Design of Water-Resource Systems. The Green Book, which did not ultimately resolve some of the continuing challenges in water planning, including objectives and the interest rate, nonetheless remains a landmark in the development of applied benefit-cost analysis within the framework of agency obligations.

Chapter 8 John V. Krutilla and Otto Eckstein, Multiple Purpose River Development

John V. Krutilla and Otto Eckstein, Multiple Purpose River Development (Baltimore: John Hopkins University Press, 1958)

Multiple Purpose River Development is a largely forgotten gem that illustrates the high level of sophistication in analysis in the period following the Green Book. John V. Krutilla (1922–2003) and Otto Eckstein (1927–1984) present a clear discussion of the economics of river basin development; departures from competitive assumptions; and assessments of costs, benefits, and income distribution in case studies. Perhaps most remarkably, the authors make it clear that their economic analysis is only a part of a fully developed and carefully explained range of multiple objectives in public expenditure. It is of interest that this presentation preceded Design of Water-Resource Systems (see chapter 11) by four years. Krutilla, principal author of the volume, was for most of his career (1955–1988) a staff member at Resources for the Future in Washington, DC, where he helped to pioneer the field of natural resource economics. Eckstein was a professor of economics at Harvard and also served as a member of the President’s Council of Economic Advisors (1964–68). The author of a standard study of water resource development and benefit- cost analysis (1961), he was the founder of Data Resources, Inc., one of the first firms to make digital government data widely available. Martin Reuss (1992) refers to both Krutilla and Eckstein in his insightful analysis of the roles of social scientists, engineers, and other experts in the development of water resources planning since the beginning of the twentieth century.

71 Economic Analysis and the Economics of River Basin Development The book begins with a well-written minicourse in economic analysis, including, for example, a graph of consumer optimization with two goods, their prices, and a fixed budget that will be familiar to all who have taken an elementary economics course (23; page references are to Krutilla and Eckstein 1958 except where otherwise specified). The development of the market model is followed by a critical evaluation of some assumptions, in particular the existence of public goods and externalities, natural monopolies, departures from competitive assumptions in capital markets, and issues of income distribution and full employment. In chapter 3 of the book, “Market Mechanics, River Basin Development, and Efficiency,” issues with the competitive model particular to the water resources sector are described and analyzed. This is an elegant way of transitioning from pure theory to the authors’ case studies. The typical example of a public good in water resources is flood control (54–56); it is difficult to protect a single consumer, as per the economic efficiency model, without protecting others as well. With respect to externalities, Krutilla and Eckstein offer the case of gravity flow irrigation (56), which recharges aquifers, thereby reducing pumping costs for those who do not pay for the irrigation system as such. They present detailed analyses of other interactions, such as those between storage dams that can contribute to hydrogeneration downstream at locations operated by different entities; in such cases, special institutional arrangements may be required to optimize system outputs. These and other concepts are used to evaluate the case studies.

The Cost of Capital A key element of the case studies is hydropower, which is a marketable output and is therefore well suited to economic analysis. In chapter 4, “The Social Cost of Federal Financing,” Eckstein’s principal contribution to the study (vii), the authors

72 first explain why available market rates do not represent the true cost of capital. In an integration of the theoretical and practical typical of the volume, they explain that the American economy does not closely fit the competitive model. For example, there are risk premiums in the terms on which businesses can borrow, and credit rationing to some businesses and to most consumers. These prevent consideration of any single existing interest rate as a measure of the social cost of capital. But in evaluating the economic value of projects, the authors correctly state that reasonable estimates of the social cost of federal funds are needed (79). (Note that other approaches, generally developed shortly after Krutilla and Eckstein’s book was written, take a broader approach to discounting; see, for example, Marglin 1967; Dasgupta et al. 1972.) The authors then describe their detailed empirical methods. They consider which taxes are imposed, on which businesses and consumers, and the interest rates at which these actors operate. (For small changes, either a tax cut or a tax increase will yield the same rates.) They use two models, A and B, the first examining the effects of a tax cut stimulating consumption and the second stimulating investments. The results of this empirical effort are, for Model A, an interest rate of 5.79 percent (93), and for Model B, a rate of 5.44 percent (102). The authors also discuss some alternatives to their approach, including specifying the risk premium (122–24), and social rates of discount (125–27). The latter anticipates more precise treatment of that possibility in Design of Water-Resource Systems, Marglin (1967), and Dasgupta et al. (1972), though these issues are still unresolved in practice. Armed with this information and the applied analytics of the previous chapters, the authors embark on their case studies.

Case Studies: Economics The first case study is chapter 5, “The Hells Canyon Case: Comparative Efficiency of Alternative Approaches to Development.” As the authors note, the biggest difficulty in

73 achieving efficient development is with multipurpose systems where problems of direct interdependence, indivisibility, and nonmarketable project outputs are most significant. It is in such systems that the federal government has found the need to participate actively: “This participation has taken two forms: The federal government, through one of its several water resource agencies, has directly developed streams for multiple purpose objectives. It has licensed others, consistent with provisions of the Federal Water Power Act of 1920, to undertake such development” (134). The authors take advantage of the flexibility of the 1920 act to analyze cases in which different power configurations, as well as other multipurpose elements, have been proposed. In Hells Canyon on the Snake River, a tributary of the Columbia, the Federal Power Commission (FPC) licensed the Idaho Power Company (IPC) to build three dams: Brownlee, Oxbow, and Hells Canyon, all of which were constructed (fig. 8- 1). One alternative to this plan was the part of the Columbia River comprehensive study by the Corps of Engineers that provided for the Hells Canyon High Dam (136, 139). A third alternative was proposed by an expert witness at FPC hearings (149) and involved a two-dam solution. The authors find that the IPC’s three-dam proposal and the Corps’ High Dam plan are roughly comparable at an interest rate of 5.5 percent (within the range of the rates they develop), and that there are no strong grounds for favoring one plan over another. At the more conventional interest rate of 2.5 percent, the High Dam “appears more efficient” (147). On the other hand, comparing the IPC plan and the two-dam solution, and taking into account both costs and benefits for the Idaho Power Company and the larger range of economic benefits to society from joint operation of the entire river system, the authors find that the IPC development is too small for maximum net benefits (156), and that the two-dam solution is superior. Finally, comparing the two-dam plan with integrated operation and the three-dam plan of the IPC operated in isolation, the authors find that the two-dam proposal is superior, generating a

74 net annual gain of almost $2.7 million (158, table 26). The authors note that this suggests that an alternative approach to development could have allowed those who gained to compensate those who lost and still have a net gain remaining (159). This is what economists refer to as a Pareto improvement, where winners gain enough to compensate losers (i.e., there is a net improvement in the system). The reason for the superiority of the two-dam plan is that the IPC plan calls for less than the optimal amount of development because it takes into account institutional limitations (156). The authors say that the smaller total output that the IPC felt was appropriate and that was licensed by the FPC (162) is explained by: the more favorable terms under which publicly developed sources can be obtained; the limitation on the marketing territory that Idaho Power Company is franchised to serve exclusively; and the large amount of hydroelectric potential in the Hells Canyon Reach (156).

Figure 8-1: Hydroelectric projects in the Columbia Basin

Source: Columbia Basin Research, http://www.cbr.washington.edu/hydro, based on U.S. Army Corps of Engineers and other sources

This is thus a case of market failure, which would have to be compensated for by institutional arrangements, which were not put forward. Such possibilities would include: complete federal development; development by a local body that would improve prospects for marketing a larger bloc of power than the IPC would develop; and licensing a private developer with the proviso that the development fit within comprehensive development of the Columbia (160–61). The FPC did not follow any of these approaches, for a variety of institutional reasons, including that, in this instance, “the FPC valued private development quite highly” (162). The authors stress the value of analyzing economically more favorable alternatives but recognize also that there are other elements of decisions such as the licensing of the IPC’s three dams. Chapter 6, “The Alabama-Coosa River System: Inte- grated System Development by a Single Licensee,” focuses on the Coosa River, which drains an area of 10,250 square miles (map on p. 179) and is part of the larger Alabama-Coosa system. In this case, some of the conditions for economic efficiency are met for unified development by a single operator: the operator can run the dams in series, and the amounts of power generated are small relative to the entire market served (171). The Alabama Power Company (APC) applied for permission to develop the river after Congress passed and the president signed legislation taking away permission for federal power development on the river (175). The APC’s proposal was for four dams, roughly corresponding to previous plans developed by the Corps of Engineers (176). If these were only for power, an efficient allocation of resources could be achieved through market forces. However, the legislation that provided for private development also required the provision of amounts of a public good—primarily flood control—and it is this issue that makes the case study of particular interest (184–85). The provision of a public good by the APC would require certain analyses to achieve efficiency: that the right amount was provided, given willingness of beneficiaries to pay, and that no price distortions

(for example, by incorporating the extra costs in APC regulated rates) occurred. The APC itself assured the Congress that it would not ask for an appropriation to cover these costs, and it is unclear whether the APC, by way of compensation, wanted to reduce payments to federal headwater releases that would increase power production (192–96). At the time of analysis, no satisfactory solution was available; the authors thus discuss an instance of potentially incorrect allocation of resources in a situation where an optimal allocation would in principle be possible. While this study is a matter of substantial theoretical interest, the authors note that the problems of planning for the Coosa are simpler than those of Hells Canyon, as they deal with a relatively small system from which the predominant share of project services are revenue producing and investment reimbursing (199).

Case Studies: Distribution of Costs and Gains The final two case studies (“The Willamette River Case: Analysis of the Distribution of Costs,” and “The Willamette River Case: Analysis of the Distribution of Gains”) deal with the distributional aspects of water resource development. The Willamette is a relatively small river in Oregon, which flows north to join the Columbia River at Portland (203, map); it drains an area of 11,200 square miles (200). For the authors, the significance of this work is that, although the main part of the book is on economic efficiency, the redistributive (and equity) consequences of projects cannot be ignored. In making these comments, they once more anticipate works a few years in the future: Design of Water Resource Systems (chapter 11, below), Marglin (1967), Dasgupta et al. (1972), and Major (1977). It should be noted also that the authors identify the difference between “accounting” costs and the true efficiency (or economic) costs that are the object of benefit-cost analysis (200n.1). In examining gains and losses, Krutilla and Eckstein, while of course dealing with laws, regulations, and data different from those that present investigators would consider, provide a

77 model of how to make a detailed examination of distributive effects, both by income class and by region. They do not do a complete assessment of gains and losses in the Willamette but rather provide selective analyses that show how such overall approaches can be undertaken. The authors study both the Corps’ original plan for integrated operation and alternatives that were floated legislatively for essentially the same development but with the feature that marketable outputs would be undertaken by nonfederal entities that could recoup costs in the market (206). To enable the analysis of the distribution of costs (chapter 7), the benefits considered that are “susceptible of appropriation by pricing mechanics” are of three types: power, irrigation and domestic water supply; the majority of benefits, principally flood control, are not (204). To begin their applied analysis, the authors construct a hypothetical but typical small Willamette hydropower installation. They then calculate the total annual charges that would be made to customers (that is, the accounting costs of the alternative developments), taking into account interest rates, taxes, and undistributed profits for federal, nonfederal public, and private development. These are estimated as $1,554,402, $1,871,101, and $3,043,789 respectively (208). “The problem may be stated thus: With public development taxes would have to be increased (or could not be reduced), and with private development taxes would not need to be increased (or could be reduced). Under these circumstances, who bears the burdens not borne by power customers when public development is undertaken?” (215–16). The results of their calculations, using Tax Model A of chapter 4, are shown in their table 29 (216). For federal development, the results of Model A primarily affecting consumption indicate that the costs would be borne mainly by low-income persons; the results of a tax following Model B (primarily affecting investment) would be felt mainly by higher- income persons (218–19).

To examine regional impacts (as opposed to income classes), the authors use a similar approach. They calculate the amounts that would be raised from taxes using Model A by region for the cases of federal development, nonfederal public development, and private development; in the public cases, the amounts raised by taxes are those not covered by fees for energy (in the case of private development there is tax shifting via rapid amortization). They find that for federal development, and taking into account revenues from energy sales, approximately 69 percent of costs are borne by the Pacific region and 31 percent by the rest of the country (226, table 42). In the case of nonfederal public development, approximately 76 percent of costs would be borne by the Pacific region and 24 percent by the rest of the country (226, table 43). For private development, the Pacific region would bear 12.49 percent of the costs of the shifted tax burden and the rest of the country 87.51 percent. For Model B, the authors show that for federal development, the Pacific region bears the same amount of costs as it does for Model A (69%) (232, table 46), but the distribution of costs by other regions differs—the Middle Atlantic states, for example, contribute about 10 percent more than for Model A because of the concentration of stockholders there at the time of analysis (233). This interesting result indicates the power of the authors’ analytic techniques to contribute to public policy. Chapter 8 contains the analysis of the distribution of gains. Here the data are not so plentiful as in the case of costs, and the authors note that this, as well as the complexity of the problem, “restricts the effort to rough approximations only, in contrast with our quantitative analysis in connection with the distribution of costs” (234). They approach the problem using the same hypothetical Willamette hydropower installation as in chapter 7. They consider power development from the same three sources—federal, nonfederal public, and private—and then trace out insofar as possible the distribution of gains from the project. The most detailed assessment is of the federal case. Here the gains are taken as power sold to customers “at less cost than

79 would be possible from the next most attractive alternative” (235). Federal power was governed by the preference clause, which in the case of the hypothetical project requires that public bodies and cooperatives be given preference in the distribution of electrical energy from federally developed projects (235). Taking this into account, the authors use distributive shares from the Bonneville Power Authority to structure the analysis, with shares going to municipalities, public utility districts, and private utilities. After detailed assessment of the gains to each of these, as well as federal agencies and electroprocess industries, the authors find annual gains of $653,900 to the Pacific Coast and $539,500 to other regions (254). Gains from nonfederal public development are estimated at $844,400 for the Pacific Coast and $256,800 for other regions; the local gains are larger than in the federal case because more of the power goes to local households (259). Finally, the gains from private development are linked to the tax shift from accelerated amortization, with a small gain on a net basis for the Pacific Coast and much larger gains for New England and the Middle Atlantic areas, where stock ownership was more common, and losses everywhere else. Summarizing the results of chapters 7 and 8 and the many assumptions involved in them, the authors say: “On balance, after all shifts in federal and local tax burdens are accounted for, the residents of the region in which federal hydroelectric development takes place would enjoy a net income gain; residents in other regions of the nation would suffer some net income loss. This suggests a net income transfer from other regions to the one in which the project is undertaken” (263). In addition, to the degree that federal power provides a greater supply at lower rates than would otherwise be available, resources would be attracted to the region, adding to a more rapid economic growth rate. With nonfederal (local) public development, the authors’ work indicates that greater net gains remain in the region; the impacts of private development are smaller and relate primarily to tax issues. With private

80 development, “all of the costs of the hydroelectric development would be borne by the power consumers in the region” (263–64).

Perspectives Multiple Purpose River Development repays reading by present- day planners. It moves smoothly from basic economic theory to the conditions tending toward economic optimization in water resource planning and the issues that might prevent these. These concepts are applied in case studies of both economic return and income distribution. In the course of their work, the authors anticipate much of what followed in water resource planning, in particular multiple objectives. As they say, Values in addition to economic efficiency are at stake in water resources development; we have given them only passing consideration. Since water development has been an instrument for attaining certain social goals, water programs include numerous intangibles. Among these are the protection of human life in the flood plains, the preservation of scenic areas because of their aesthetic appeal, the improvement of public health and welfare through provision of recreation facilities, the assurance of security from vagaries of weather through irrigation agriculture—these, and many others. In addition our society has used water resources projects as a means of providing employment and settlement opportunities, of fostering the growth of underdeveloped or depressed regions, and of promoting the widespread use of electric power. These goals are commonly accepted—although individuals attach varying weights to each. (265) The use of the word “weights” is striking. It became a common term just a few years after the publication of this work. The authors emphasize the importance of both economic analysis and distributive analysis in the context of other possible objectives. In a way, this volume is a descriptive version of what

81 was later put in mathematical terms (e.g., Marglin 1967; Dasgupta et al. 1972) with respect to objectives. One area that remains unsettled in water planning is that of the interest rate; the authors, although using opportunity cost rates in their own work, refer in this regard to social rates of interest (125–27). Given data limitations and the predigital era of the study, the detail provided is a model for other assessments.

Chapter 9 Delaware River Basin Report

U.S. Army Corps of Engineers, Philadelphia District, Delaware River Basin Report, December 1960; rev. May 1961, vol. 11, printed as H. Doc. 522, 87th Congress, 2nd Session, August 16, 1962

The Delaware River Basin Report (DRBR; fig. 9-1) can be seen as a culmination of the planning methodology of the U.S. Army Corps of Engineers following the developments described in the earlier chapters of this section, as well as the Corps’ extensive earlier basin planning work (see, for example, chapter 3 on the 308 reports). It was a substantial, multiagency professional effort at a time of changing methods and challenges, within a complex multistate institutional setting. The DRBR, largely because of its timing, did not include the developments in method that were in progress at the time of planning (e.g., Maass et al. 1962 [see chapter 11 of this volume]; see Major 1965 for a review of the DRBR and a comparison of DRBR planning and new methods). Moreover, the DRBR also was issued before the enactment of the principal environmental legislation of the 1960s and early 1970s—for example, the National Environmental Policy Act (U.S. Congress 1969) and the Federal Water Pollution Control Act Amendments of 1972 (U.S. Congress 1972)—and the concomitant rise in environmental concerns. The DRBR is thus a marker in the history of planning change between two significantly different eras. The volumes of the DRBR are available online at websites such as ProQuest (http://www.proquest.com). References to the various volumes of the report are made here as follows: material in volume 1 that precedes the main report is cited as “1:xx,” references

Figure 9-1: Delaware River Basin Report Source: U.S. Army Corps of Engineers, Philadelphia District (1960) to the main report are cited as “Report, xx,” and references to material in the appendices are cited, for example, as “Q-x.”

Planning Methods The planning methods used by the Corps in the DRBR included the full range of methods and criteria that were appropriate at the time of planning. Planning area and time horizon. The DRBR is a river basin plan (fig. 9-2). As was appropriate, however, the planners took into account that for some water uses (domestic, industrial,

84 hydroelectric, recreation), demand is in part from outside the basin (Q-7), and of course diversions to the New York City system are large (aspects of the system are described at www.nyc.gov/dep). An interbasin physical linkage was also

Figure 9-2: THE DELAWARE RIVER BASIN Source: Delaware River Basin Source Water Collaborative (n.d.), http://www.delawarebasindrinkingwater.org/about-the-drbswc/ studied, a diversion from the Susquehanna (S-19, S-24), but this was very expensive and was rejected (Major 1965, 87). The planning horizon, which was to be fifty years from 1960, was instead from 1955 to 2010 because of issues of data availability (Q-6).

85 Institutions. The DRBR, consisting of the main report and twenty-four appendices (fig. 9-3), is a “review of the Delaware River and Tributaries, requested by a resolution of the Committee on Public Works, United States Senate, adopted 13 April 1950, and other resolutions of that Committee and of the Committee on Public Works, House of Representatives, listed in the report” (1:v). The report was undertaken by the Philadelphia District of the U.S. Army Corps of Engineers with significant planning input from federal agencies, which produced many of the appendices, and additional input from other governmental and private agencies and companies (Report, 5–8). The DRBR did not have the highly developed public participation that came later in the evolution of water planning methods (for public participation approaches, see Langton 1996, appendices B, C). The principal institutional framing elements for the study were the Supreme Court decision of 1954 allocating substantial diversions to the New York City system, with smaller diversions to New Jersey (details are at U.S. Geological Survey n.d., Office of the Delaware River Master), and the development of the Delaware River Basin Compact in 1961 (U.S. Congress 1961) and its operating agency, the Delaware River Basin Commission. The latter replaced the Interstate Commission on the Delaware River Basin (INCODEL), founded in 1936 (INCODEL 1958, 13). In 1963 the functions and remaining funds of this agency were transferred to the Delaware River Basin Commission. Objectives. The stated goal of Corps investment in water resource development at the time of planning was the maximization of “net economic gains and human satisfactions from the use of economic resources in a project.” Project design was to be modified to take into account the effects of benefits and costs that cannot be evaluated in monetary terms. These intangible benefits and costs include elements such as human life and health and national defense (U.S. Army Corps of Engineers 1959, 11–12). The DRBR planners followed this objective (Report, 141) and these procedures. .And, of course, there were required regional objectives given the Supreme Court decision.

Figure 9-3: List of appendices of the Delaware River Basin Report Source: U.S. Army Corps of Engineers, Philadelphia District (1960, 1:iii)

87 Multiple purposes. The DRBR was not a multiobjective planning effort, although non-economic objectives had some relevance to the work. On the other hand, it was a full multipurpose study for the purposes that the authorizations and the planners regarded as most important: flood control, water supply, recreation, and hydroelectric power generation. With respect to water quality needs, it was concluded that the water of the basin is of high quality and that no needs for water quality control would arise that would impose requirements for development within the scope of the plan (Report, 71–74, 76). Interest rate. The planners used a “long-range, risk-free” interest rate, which was a weighted average of historical federal borrowing rates on bonds of relatively long maturities. The interest rate used for initial benefit computations was 2.5 percent (Report, 143; Major 1965, 67); for consistency with the federal rate at the time, future benefit estimates for federal components were recomputed at 2 5/8 percent; other, higher interest rates were used for the nonfederal parts of the proposed plan (Report, 152). Demand projections. Demand (or “need”) projections for the outputs of water control projects and other measures were based on the Economic Base Survey prepared by the Office of Business Economics of the U.S. Department of Commerce (app. B). This study projected basic parameters of the economy for the nation as a whole and derived from these national parameters projections of personal income, population, number of households, and employment according to broad industry subgroups for the Delaware River Service Area, shown in figure 9-4. Projections for the nation and for the Delaware River Service Area are for 1965, 1980, and 2010. The projections for the given years are point rather than range projections. Gross national product was projected to grow at 3 percent, a rate approximating the historical trend (B-9, B-11). The projected percentage rate of population growth was, taking the Base Survey figures for 1958 and 2010 (B-13), implicitly approximately 1.1 percent on a straight-line basis (Major 1965, 62n.8). This is a “median expectation in line with the historical trend” (B-11).

Figure 9-4: The Delaware River Basin service area

As summarized in Major (1965, 63), The projections of needs for the outputs of water control projects rest for the most part on the projections for the Delaware Service Area given in the Economic Base Survey. One exception is flood control, the requirements for which were formulated by analyzing aerial photographs of sample flood reaches (D-20 to D-21). Water supply needs were derived from the basic projections by applying to

89 them use factors for industrial and domestic-municipal water supply (p. P-8). Recreation needs were projected by applying an empirically derived visitor-day per 1000 population factor to the Economic Base Survey population projections (pp. W-5, W-11, W-12). Power needs were also based on the projections of the Economic Base Survey (p. F-8). Plan formulation methods. The planning objective was to be achieved by accepting projects with benefit-cost ratios greater than one, and adjusting project size by increments so as to maximize net project benefits. (U.S. Army Corps of Engineers 1959, 11). The planners reviewed the idea of a mathematical model for DRBR planning but recognized that “it was not possible to develop a mathematical model for complete solution of all or any substantial part of the planning job for the Delaware River” (Q-14). This reflected the situation at the time; instead, the planners opted for a simplified sequential appraisal method (Q-15). To estimate benefits from fulfilling these needs, alternative cost measures were generally used; for flood benefits, established methods of damages prevented were used (Q-10). Benefits and costs for the water control elements of the plan of development are given in table VIII-9 (144). Prices at the early 1959 level were attached to project costs (Q-9). The initial screening of fifty sites with the potential for major development (Q-60) used generalized relationships of inactive storage, long-term storage for water supply, low flow control, and flood control (Q-62–63). Then, cost indices were developed in some detail (Q-63–64), and benefit estimates were based on alternative costs for average cost of surface water in the basin (Q-64–65). These were combined into “Indices of Merit” as shown in table Q-13 (Q-66a). The planners then developed nine alternative plans for the basin. Some of these were dropped, and others were added based on the proposals of an informal working group for the project (Q-68–69). Analysis of these in some detail (Q-69–77) produced the basic plan for initial and later development.

The planners then proceeded to the next step, the discussion of points of maximum net benefit. This is worth quoting at length because it illustrates the significant level of effort that the planners made. The next phase of analysis in forming the proposed plan of development was the determination of that scale of development for each impoundment site in a system which would produce the greatest net benefits in realizing the balanced program objectives of river basin development. This was achieved in a two phase analysis. First, each of the selected sites was individually analyzed to identify the area of its optimum scale of development in terms of net benefit maximization with consideration given to all water resources products that could be produced by a project at that site. In the second phase, modifications were made at each site’s optimum level of development so that the scale of development of each site as an integral part of the balanced program for water resource development would produce only the resource products for which foreseeable markets existed and where the system net benefits would be at a maximum. It is in this latter analysis that the fundamental goal for comprehensive river basin planning is achieved in that the production of required water resource goods and services is accomplished in a manner that cannot be equaled, with regard to balanced production and investments, by any other allocation of productive resources. (Q-78; plate Q-8) The eleven multipurpose projects recommended for early implementation are listed by their estimated required date of implementation (Q-94). It is also recommended that thirty-nine upstream projects move ahead “at earliest date all requirements are met to permit construction under existing programs” (Q-94).

91 Recommendations and Results The recommended Comprehensive Plan (fig. 9-5) included eleven major multipurpose projects, of which eight were to be authorized by the federal government and the other three by nonfederal interests. Eight additional projects were to be developed initially for recreation and later for other purposes. Of the thirty-nine additional small upstream projects, only three required federal authorization (Major 1965, 60). The presentation of the Board of Engineers for Rivers and Harbors (BERH) constitutes the official Corps recommendation: The District Engineer recommends: That the 58-reservoir plan be adopted as the Comprehensive Plan for development of the water resources of the Delaware River Basin; at a total cost presently estimated at $591 million, of which $232.1 million would be Federal, with $89.1 million reimbursable by non-Federal interests; continuous and vigorous action by Federal and non- Federal agencies, separately and cooperatively under authorities that may exist or be provided, in prosecution of programs for land management, controlling and regulating the use and development of flood plains, preservation and development of recreation and fish and wildlife resources, conservation of ground and surface waters, and preservation of sites for the projects that comprise the Comprehensive Plan; that the units of the Comprehensive Plan designated as Beltzville, Blue Marsh, Trexler, Tocks Island (without the pumped- storage power feature), Aquashicola, Maiden Creek, and modification of the existing Prompton and Bear Creek projects, be authorized for construction at a presently estimated total Federal cost of $224 million, of which $89.1 million would be reimbursable. (1:15)

Figure 9-5: The Recommended Plan

Source: Plate Q-7

93 The Comprehensive Plan takes into account the construction of the last of the New York City Delaware System reservoirs, Cannonsville, which was placed in service in 1964. The earlier reservoirs are Rondout in 1950, Neversink in 1954, and Pepacton in 1955. (Rondout is actually located within the Hudson River Basin but is operationally part of New York City’s Delaware system; http://www.nyc.gov/html/dep/html/drinking_water/wsmaps_wide.shtml.) In his letter to Speaker of the House John W. McCormack, Secretary of the Army Cyrus Vance noted: The Director of the Bureau of the Budget in his letter to me dated 17 July 1962 notes that in view of recent legislation approving the Delaware River Basin Compact and creating the Delaware River Basin Commission, the formal adoption by Congress of the recommended comprehensive plan for the Delaware River Basin is unnecessary. Also, in connection with the complete development of the recreation potentials of the Tocks Island project, wholly at Federal expense, the Director finds that such development would be appropriate and in accord with the program of the President, provided suitable entrance, admission, and other user fees are established. (1:v) Seven of the eight projects for federal financing are shown in table 9-1 with their storage for individual site maximization; the eighth, Prompton, then under construction, had a proposed storage capacity of 51,000 AF, of which 28,000 AF were to be for the water supply modification (Q-96). Blue Marsh and Maiden Creek are in the Schuylkill subbasin; Tocks Island is on the main stem; Beltsville, Bear Creek, Aquashicola, and Trexler are in the Lehigh subbasin; and Prompton is on the Lackawaxen River. All except Tocks Island are west of the Delaware.

Table 9-1: Scales of Development for Seven Proposed New and Modified Federal Projects

In their comments, New York and New Jersey approved the eight main reservoirs for federal funding and recommended that other actions wait upon a full review by the new Delaware River Basin Commission (1:x). Pennsylvania also approved the eight reservoirs (1:xiii) and devoted substantial commentary to water quality issues, noting that the transmittal letter of the BERH included a reference to the new 1961 law relating to these (1:xii). Delaware provided detailed reviews of the report and appendices (1:xviiff.); in particular, there was a concern that the DRBR focuses on surface waters, whereas most of Delaware was then dependent on groundwater (1:xxviii–xxx). Of the eight projects, two, Bear Creek (renamed Francis E. Walter in 1963) and Prompton, were underway at the time of planning. These projects were completed in 1960 and 1961, respectively (U.S. Army Corps of Engineers, Philadelphia District, n.d., 84), without, however, the modifications proposed; these were not included largely for financial reasons, including changes in cost-sharing procedures under the Water Resource Development Act of 1986 (P.L. 99-662) (U.S. Army Corps of

95 Engineers, Philadelphia District, n.d., 86–89). The Beltzville Lake Project was completed in 1971, and Blue Marsh Lake was dedicated in January 1979 (U.S. Army Corps of Engineers, Philadelphia District, n.d., 65). When Congress passed the Water Resources Development Act of 1986, it officially deauthorized construction of the Trexler Lake Dam, and Aquashicola and Maiden Creek were removed from the Delaware River Basin Commission plan in 1989 (“Aquashicola Dam Removed From Regional Planning List” 1989). The biggest controversy with the recommended plan was with the proposed Tocks Island impoundment (fig. 9-6). Although this was just one of the eight federal projects proposed for modification or new construction, its storage capacity dominated the plan. Tocks Island would have had 62 percent of the total storage foreseen for the first eight reservoirs requiring federal authorization, including the storage for Prompton. (The construction of Tocks Island was assumed in the NAR study [see chapter 12]). The rejection of Tocks Island is summarized in DelawareWaterGap.org (n.d.): Opposition to the project began almost immediately among landowners on both sides of the Delaware whose properties were targeted for condemnation. The Delaware Valley Conservation Association was established in 1965 with more than a [sic] 1,000 members to fight the project. On July 31, 1975 the Delaware River Commission voted 3–1 against the immediate construction of the Tocks Island Dam. Pennsylvania was the only state to approve the dam, while the federal government abstained. The Commission stopped the dam, but the true end of the project came in 1978 when Congress designated the section of the river that is within the recreation area as a “Wild and Scenic River,” in effect barring the construction of any dams at the Tocks Island site or anywhere along this section of the river. In 1992, the Tocks Island Dam Project was officially de-authorized by Congress.

Figure 9-6: Tocks Island on the Delaware River, site of the proposed dam Source: New Jersey Department of State

Perspectives The Delaware River Basin Report represents the state of Corps basin planning just before the transition period to new mathematical methods and multiple objectives described in the next section of this volume. The DRBR was a complex professional effort using the best methods at the time. It used an economic base study on which to base needs, applied benefits (using primarily the alternative cost measure) and costs, related the needs to basin hydrology, utilized a sequential preliminary screening process, and then maximized net benefits by project and by recommended plan. Institutionally, the plan was undertaken in cooperation with other federal agencies; comments were primarily after the completion of the plan from federal, state, and other agencies, without public involvement. The plan resulted in two

97 new federal multipurpose projects, with others (in particular, Tocks Island) stopped by environmental and financial considerations. At the same time, other measures were implemented to some degree, and a substantial body of information and analysis was made available to the newly created Delaware River Basin Commission. The Delaware is one of the most closely managed basins in the nation because of its importance to the basin states and the New York City water supply system. A few years after the DRBR, the Potomac River Basin plan (U.S. Army Corps of Engineers, Baltimore District, 1963) “was one of the first to move beyond the standard approach. The innovations of the plan included: the explicit consideration of the distribution of the economic and social impacts of projects; the use of alternative plans to deal with the fundamental uncertainty of planning; the exploration of the links between water quality and water quantity; and the use of early electronic computers for hydrologic calculations (a Burroughs E101 was used for the work in the Potomac planning)” (Major 2010, 44). This plan also foundered on environmental opposition, and the experiences of the Corps with these two basin plans and other planning efforts led to substantial efforts beginning in the late 1960s to increase public involvement; these were led by the Corps’ new Institute for Water Resources, founded in 1969 and located at Fort Belvoir, Virginia. The new developments to come, described in the next section, include multiple objectives and the use of mathematical models in planning. For a chronology and bibliography of the Corps’ work in public involvement from the late 1960s, see Langton (1996, app. B, C).

Acknowledgement: Much of the detailed analysis of the DRBR in this chapter is based on Major, 1965, chapter III.

Section III Development of New Methods

The essays in this section deal with the development of new methods, including multiple objectives and other evaluation criteria, the integration of mathematical modeling into water resource planning, and framework planning. Chapter 10 focuses on the U.S. President’s Water Resources Council’s Policies, Standards, and Procedures in the Formulation, Evaluation, and Review of Plans for Use and Development of Water and Related Land Resources (1962). Senate Document 87-97, as the 1962 standards are generally called, marks a genuine change in evaluation standards for water resources planning. It is the first official set of federal criteria that moved in a comprehensive way to adopt multiobjective planning. Design of Water-Resource Systems: New Techniques for Relating Economic Objectives, Engineering Analysis, and Governmental Planning, also produced in 1962, is the subject of chapter 11. This volume by Arthur Maass and others is the most influential book on water resources planning of the second half of the twentieth century. It explores the interactions of economics, engineering (including then newly developed mathematical modeling techniques), and political decision making in the context of multiple objectives. The impact of this work has been pervasive and long-lasting. Chapter 12 turns to the North Atlantic Regional Water Resources Study by the U.S. Army Corps of Engineers, North Atlantic Division, New York (1972). This was one of some twenty “framework” studies designed to provide a regional planning context within which detailed, consistent plans for water resources management and development could be implemented. (Eleven of the studies were underway when the program was later superseded by the National Assessments, the first of which was U.S. Water Resources Council 1968). It was highly innovative in the use of multiple objectives and new

99 institutional arrangements as well as the implementation of large- scale mathematical modeling. The Water Resources Planning Act of 1965 established the U.S. Water Resources Council, which had among its other duties the development of standards and criteria for planning and evaluating water resource projects. The extensive reports of its Special Task Force (1970) ultimately resulted in the Principles and Standards of 1973, the subject of chapter 13. This document broke new ground in investment criteria for federal water resource planning, committing federal agencies to a multiobjective approach for federal water project evaluation and review.

100

Chapter 10 The U.S. President’s Water Resources Council Report

U.S. President’s Water Resources Council, Policies, Standards, and Procedures in the Formulation, Evaluation, and Review of Plans for Use and Development of Water and Related Land Resources, Senate Doc. 87-97, 87th Congress, 2nd Session, May 19, 1962

Senate Document 87-97 (fig. 10-1), as the 1962 standards are generally called, marks a genuine change in evaluation standards for water resources planning. The document was explicitly a statement of executive requirements for water resource planning: The purpose of this statement is to establish Executive policies, standards, and procedures for uniform application in the formulation, evaluation, and review of comprehensive river basin plans and individual project plans for use and development of water and related land resources. … These provisions shall govern, insofar as they are consistent with law and other applicable regulations, all formulation, evaluation, and review of water and related land resources plans. Any proposed variation from these policies and standards shall be specified in planning reports and the reasons therefor [sic] indicated. (1; page references in this chapter are to S-Doc. 87-97 unless otherwise specified.) A short document of just thirteen pages, it includes many of the themes that have continued to animate discussion of water resources planning since its publication: multiple objectives; the use or nonuse of an efficiency requirement for evaluation; the interest rate; and Congress versus the executive branch.

101

Figure 10-1: S. Doc 87-97

S. Doc. 87-97 developed in opposition to a strict efficiency emphasis promulgated by the U.S. Bureau of the Budget in its Budget Circular A-47 (1952). The Bureau of the Budget attempted rigorously to apply “A -47” to all projects presented to it for review in the 1950s. This effort led to great dissatisfaction with “A-47” within the Congress beginning about 1956. Few, if any, in the Congress called for abandonment of benefitcost

102

analysis per se, but there was a widespread call for its “liberalization.” Democrats in the Congress declared that the Eisenhower administration had a “no new starts policy.” (Caulfield 2000, 15) This background is made clear in the statement of Senator Clinton P. Anderson in the preface to the new standards: The recommendations of the four Department heads, and the President’s approval of them, are consistent with the established policies of the Senate. They are, in fact in compliance with requests of the Senate. The new policies and standards, established in an agreement of the four Department heads, replace Budget Bureau Circular A-47 which caused considerable contention, both as to content and as to the propriety of its source. (S. Doc. 87-97, iii) As a result, Policies, Standards, and Procedures in the Formulation, Evaluation and Review of Plans for Use and Development of Water and Related Land Resources, an interdepartmental agreement approved by President Kennedy for application by the federal departments concerned and the Bureau of the Budget, replaced Budget Bureau Criteria A-47 on May 15, 1962. (Caulfield 2000, 15)

It should be noted that the transition from the Green Book and the single-objective orientation of Bureau of the Budget Circular A-47 to S. Doc. 87-97 was preceded by a report (Hufschmidt et al. 1961), published just before the latter, of a panel of experts to the Bureau of the Budget. This document provided a comprehensive rationale for multiobjective planning for water resources (Major 1977, 4–5) but was never officially released by the bureau.

103 Objectives and Project Formulation Senate Document 87-97 was the joint effort of four departments: Army; Agriculture; Health, Education, and Welfare; and Interior. It was prepared by the Interdepartmental Staff Committee of the President’s Water Resources Council (Caulfield 2000, 15), of which Henry P. Caulfield, Jr., a distinguished political scientist and public servant (fig. 10-2), was the chair. This is the first official set of federal criteria that moved in a comprehensive way to adopt multiobjective standards. The document begins with a statement of the overall objective of water resource development: “The basic objective in the formulation of plans is to provide the best use, or combination of uses, of water and related land resources to meet all foreseeable short- and long-term needs. In pursuit of this basic conservation objective, full consideration shall be given to each of the following objectives and reasoned choices made between them when they conflict” (1).

Figure 10-2: Henry P. Caulfield, Jr. (1915–2002), the first director of the U.S. Water Resources Council and chair of the Interdepartmental Staff Committee that produced S. Doc. 87-97.

Source: Creative Commons License,

104

The objectives are then stated (1-2): A. Development National economic development, and development of each region within the country, is essential to the maintenance of national strength and the achievement of satisfactory levels of living. Water and related land resources development and management are essential to economic development and growth, through concurrent provision for— Adequate supplies of surface and ground waters of suitable quality for domestic, municipal, agricultural, and industrial uses—including grazing, forestry, and mineral development uses. Water quality facilities and controls to assure water of suitable quality for all purposes. Water navigation facilities which provide a needed transportation service with advantage to the Nation's transportation system. Hydroelectric power where its provision can contribute advantageously to a needed increase in power supply. Flood control or prevention measures to protect people, property, and productive lands from flood losses where such measures are justified and are the best means of avoiding flood damage. Land stabilization measures where feasible to protect land and beaches for beneficial purposes. Drainage measures, including salinity control where best use of land would be justifiably obtained. Watershed protection and management measures where they will conserve and enhance resource use opportunities. Outdoor recreational and fish and wildlife opportunities where these can be provided or enhanced by development works.

105 Any other means by which development of water and related land resources can contribute to economic growth and development. B. Preservation Proper stewardship in the long-term interest of the Nation's natural bounty requires in particular instances that— There be protection and rehabilitation of resources to insure availability for their best use when needed. Open space, green space, and wild areas of rivers, lakes, beaches, mountains, and related land areas be maintained and used for recreational purposes; and Areas of unique natural beauty, historical and scientific interest be preserved and managed primarily for the inspiration, enjoyment and education of the people. C. Well-being of people Well-being of all of the people shall be the overriding determinant in considering the best use of water and related land resources. Hardship and basic needs of particular groups within the general public shall be of concern, but care shall be taken to avoid resource use and development for the benefit of a few or the disadvantage of many. In particular, policy requirements and guides established by the Congress and aimed at assuring that the use of natural resources, including water resources, safeguard the interests of all of our people shall be observed. With respect to objectives, S. Doc. 87-97 can be taken as a fully multiobjective document. In academic writing and standards to follow, the “development” objective has generally been divided into national and regional development as separate objectives. (The four objectives in U.S. Water Resources Council [1973] are national economic development, environmental quality, regional development, and social well-being, with the

106 last two being designated display rather than design objectives; see chapter 13 below). It is clear throughout S. Doc. 87-97, however, that regional objectives can be different from, and may conflict with, the national economic objective. As summarized by Caulfield, “Regional, state, and local points of view or objectives were to be considered as well as national points of view in terms of criteria of national economic efficiency or other national policy” (2000, 16). The statement of objectives is followed by sections on national, regional, state, and local interests; multipurpose planning; river basin planning; individual project planning; coordination within the federal government and with nonfederal agencies; and relation to existing law and executive orders (2–4). Standards for the formulation of plans include, first, an economic plan with net benefits maximized, and then the development of other plans for other objectives as may be suitable. Comprehensive plans shall be formulated initially to include all units and purposes which satisfy these criteria in quantitative economic terms: (a) Tangible benefits exceed project economic costs. (b) Each separable unit or purpose provides benefits at least equal to its costs. (c) The scope of development is such as to provide the maximum net benefits. (7) However, other pertinent objectives are to be considered, and appropriate plans drawn up with respect to them: When there are major differences among technically possible plans conceived as desirable on the basis of consideration of intangible benefits and costs, in comparison with optimum plans based on tangible benefits and costs, alternative combinations of projects within a river basin or alternative projects, giving expression to these major differences, shall be planned. (6)

107 Further, Reports and plans shall also indicate the extent to which departures from that scale of development are proposed in order to take into account intangibles or other considerations warranting a modification in scale not reflected in the tangible benefits and project economic costs. For example, a higher degree of flood protection, particularly in urban areas, than is feasible on the basis of tangible benefits alone may be justified in consideration of the threat to lives, health, and general security posed by larger floods. (8) These standards represent a comprehensive commitment to a multiobjective approach to water resources planning. And remarkably, no criterion is set forth that requires that approved projects have a benefit-cost ratio great than 1: “In contrast to ‘A- 47’ no requirement was set forth that projects, generally, must have a benefit-cost ratio greater than 1.0 to 1 as a basis for recommendation to the Congress” (Caulfield 2000, 16). In addition to objectives and project formulation standards, S. Doc. 87-97 details definitions of benefits, types of primary benefits and costs, the treatment of secondary benefits, the interest rate, which is taken as an average rate of interest on Treasury securities with maturities of fifteen years or more, and the relation of the standards to cost allocation, reimbursement, and cost-sharing (8–13).

Perspectives S. Doc. 87-97 is the first multiobjective set of standards adopted by the federal government to govern water planning. It gives a broad statement of objectives, includes the development of alternative plans, and explicitly provides for comparison and choice among them. The objectives were not as fully detailed as in later academic and government documents, such as Design of Water-Resource Systems (chapter 11) and the 1973 Principles and Standards (chapter 13). The former provided the full logic of

108 multiobjective planning, and the latter had the advantage of further developments in benefit estimation and the consideration of the environment, especially in the National Environmental Policy Act (U.S. Congress 1969). In addition, the NAR study (chapter 12) required the development of plans emphasizing alternative objectives from the beginning, while S. Doc. 87-97 asked first for the maximum net economic benefits plan and then for alternatives to that, although this may not be considered a major difference. More important, in contrast to Budget Circular A-47, no requirement was set forth that projects generally must have a benefit-cost ratio greater than 1.0 to 1 as a basis for recommendation to Congress. At the same time, the ongoing debates about standards for water resources planning were not stilled by S. Doc. 87-97: the 1973 Principles and Standards (U.S. Water Resources Council 1973) followed in a direct line, whereas the 1983 Principles and Guidelines (U.S. Water Resources Council 1983, iv–v) marked an effort to return to planning based on the economic objective. The interest rate also remained a continuing matter of contention. Moreover, there can be a difference between formal and informal standards. Caulfield notes that, while the standards of S. Doc. 89- 97 superseded Bureau of the Budget Circular A-47 in not requiring that economic benefits exceed economic costs, “[such a] requirement was the unwritten rule [in the Bureau of the Budget] since the promulgation of Senate Document 97 on May 15, 1962” (2000, 16). Nonetheless, S. Doc. 87-97 may be taken as a distinct change in the federal approach to water resources planning as compared to the Green Book (chapter 7) and especially Budget Circular A-47.

109

110

Chapter 11 Arthur Maass et al., Design of Water-Resource Systems

Arthur Maass, Maynard M. Hufschmidt, Robert Dorfman, Harold A. Thomas, Jr., Stephen A. Marglin, and Gordon Maskew Fair, Design of Water-Resource Systems: New Techniques for Relating Economic Objectives, Engineering Analysis, and Governmental Planning (Cambridge, MA: Harvard University Press, 1962)

Design of Water-Resource Systems is the most influential book on water resources planning of the second half of the twentieth century. It embodies the results of a carefully structured research enterprise, the Harvard Water Program (HWP), which proceeded by stages to explore the interactions of economics, engineering (including then newly developed mathematical modeling techniques), and political decision making. The impact of this work has been pervasive and long-lasting. The Harvard Water Program was supported by grants from the Rockefeller Foundation and benefited from agency support, in particular from the Bureau of Reclamation and the U.S. Army Corps of Engineers. The book describes how the program developed: There have been three stages in the history of the Harvard Water Program: one year (1955-56) of exploration to determine whether or not a study should be undertaken in this field, and, if so, what type of study it should be; three years (1956–57 to 1958–59 of combined training and research, during which senior employees of federal and state water-resource agencies came to Harvard to assist in the research and at the same time to prepare themselves for positions of greater responsibility in the public service; and one final year (1959–60) of research and writing. (10–11; page references in this chapter are to Maass et al. 1962 unless otherwise specificed.)

111 The presence at Harvard for extended periods of highly qualified agency personnel who participated fully in the research and writing of reports was a notable element. Other personnel came for shorter periods for review meetings. These contacts provided a rich network for future water planning efforts (Schwarz and Major 1993). This supported research reflected a different era at research universities, when professors were fully paid by the university and research grants were not an essential part of academic life. This circumstance allowed the HWP to proceed at what, by present standards, would be a leisurely pace. Arthur Maass and his colleagues saw a moment when techniques and knowledge in economics, engineering, and politics could be brought together to transform the possibilities of effective water planning. The success of the program was in providing a complete approach to water planning, with significant advances in each area.

Research of the Harvard Water Program The framework of the book begins with the four steps of design: “The methodology of system design, as defined in this book involves four related steps: identifying the objectives of design; translating these objectives into design criteria; using the criteria to devise plans for the development of specific water-resource systems that fulfill the criteria in the highest degree; and evaluating the consequences of the plans that have been developed” (2). The structure of the book follows these steps. It should be noted that the approach used throughout is an optimizing approach— searching for the best available system designs. The research results of the program are for the most part presented in Design of Water-Resource Systems, but there were other related publications as well (e.g., Hufschmidt and Fiering 1966). The book as a whole is a detailed, integrated study, but it is possible to review briefly several main parts: new investment criteria, engineering and mathematical modeling, and political decision making.

112

New Investment Criteria One of the main contributions of the book is the presentation of a set of economic investment criteria that goes substantially beyond the welfare economics formulation (the microeconomic model of resource allocation described in the essay on the Green Book; see chapter 7 above). The fundamental change is one in which decisions are looked at as social decisions, at least in part, rather than as the purely decentralized decisions of the microeconomic model. This is the change that brought about the development of what has come to be known as the multiobjective approach to planning (see chapter 12 below on the NAR study and chapter 13 on the 1973 Principles and Standards). The Green Book essentially adopts the goal of maximizing economic output, the standard efficiency objective of economic benefit-cost analysis (although other objectives are discussed), while in Design of Water-Resource Systems a second objective is introduced: the income distribution objective (62–87). In this multiobjective approach, improved distributional results, for example, are traded off against higher national economic benefits; in other words, a social decision is made that is not solely the result of decentralized individual decisions, as in a market. Peter O. Steiner (1969) provides a good overview of social decision making; the fundamental nature of such decisions is a reflection of such important works of the second half of the twentieth century as Arrow (1963) and Rawls (1999). Describing the application of redistribution goals, the authors state that “the goal … is to translate a general governmental policy of using water-resource developments to redistribute income into a criterion for the design of specific systems” (62). “Income distribution is always circumscribed by the community’s notion of how much redistribution it wants, how much efficiency it is willing to sacrifice in order to obtain redistribution, or both” (70). The possible redistributions discussed are to a group of individuals, to a region, or within a group or region (84). Criteria for accounting for redistribution of benefits and costs are

113 described; these accounting rules differ from the rules for the efficiency objective (22–62; see also Marglin 1967). Distributional effects have always been important in water planning, but the discussion in Design of Water-Resource Systems was the first to provide a formal framework for taking them (and other objectives) into account. Presentations of the multiobjective approach in an accessible manner, including economic, distributional and other objectives, can be found in Marglin (1967) and Major (1977), and in extensive detail in Dasgupta et al. (1972). The second area of fundamental change is the choice of the interest rate to be used in water resource planning, still a vexed and unsettled area of decision making (compared to multiple objectives, which in one way or another pervade system design now). The standard economic (efficiency) model uses an interest rate (an interest rate incorporates weights on the present as compared to the future), derived like market prices from decentralized individual decisions. There are many approximations and substitutes for this rate (see Chapter 8, above). In U.S. federal water resources planning it has long been common to use a rate based on returns on U.S. bonds (see Chapter 9, above). However, in Design of Water-Resource Systems, the concept of time weights that reflect social decisions (analogous to the social decisions of multiple objectives), called the social rate of discount, is analyzed in detail and recommended for government investments (194ff.). Such a rate, if it could be decided on, might be lower than the market rate and so would have to be used with factors called shadow prices to ensure that government investments would not displace private investments that would have a higher social return. A good overview is in Dasgupta et al. (1972); shorter treatments are in Marglin (1967) and Major (1977). While with the development of concerns with global warming and sustainability such approaches seem more relevant than ever, there has been little agreement on how to proceed since Design of Water-Resource Systems. Nor is the theoretical perfectly competitive interest rate actually known;

114 what are known are various rates available from markets that do not fully reflect the economic assumptions of optimal resource allocation (see chapter 8 above). Other elements not customarily taken into account in standard benefit-cost analysis that are discussed in Design of Water-Resource Systems include the integration of budget constraints into system design, so that the same gains at the margin are achieved in all projects (159–77; see also Eckstein 1961), and the dynamics of design, whereby the conditions under which benefits are maximized by choosing the optimal year of implementation are examined with and without budget constraints (177–92). It is notable that neither of these procedures has been implemented on a large scale in U.S. water planning, although both are technically possible.

Conventional Techniques, Mathematical Modeling, and the Design Process The research on conventional techniques, mathematical modeling, and the design process forms a central part of Design of Water-Resource Systems (chaps. 5–14). The presentation begins with a demonstration that a minimally complex river system can be planned with graphical techniques (226; see fig. 11-1). At the same time, this discussion makes it clear that more complex systems require more powerful techniques. To study the application of these, a simplified but still substantial river basin planning problem was constructed, with four reservoirs and three purposes (irrigation, electric power, and flood protection). The hydrologic inputs for this system were taken from U.S. Geological Service records for the Clearwater River and its tributaries in Idaho, and costs and benefits are used that are hypothetical but based on real experience (265; see also fig. 11- 2). This system was analyzed both with the application of conventional techniques and with the use of mathematical simulation and sampling of the net-benefit response surface. A “best” solution to the design problem was found using the two approaches. Analysis of the simplified system with conventional

115 techniques yielded “best” net benefits of $724 million versus net benefits of $811 million for simulation (254–55). (The simulation results were also studied using synthetic streamflow to test the stability of net benefits; 256–57.)

Figure 11-1: Simplified river system used for analysis

Source: Maass et al. (1962, 226)

The same problem was then studied with newly developed mathematical programming models: Broadly speaking, we have employed two types of techniques. The first one simulates the behavior of a water-resource system on a high-speed digital computer, observes its response over extended periods of time to alternative combinations of structures and levels and purposes of development, and selects the best combination. The second technique adapts the design problem so that it can be solved by certain mathematical methods which proceed automatically to the optimal

116

Figure 11-2: Second simplified river system for analysis Source: Maass et al. (1962, 265)

solution for relatively simple problems. Both techniques are designed to allow simultaneous consideration of a large number of alternatives, conjoining of engineering and economics, and adaptation to any reasonable objective and institutional constraint. (247) Mathematical programming (257–61) is the second technique referred to above (Dantzig 1963) to study the simplified river basin. Several versions of mathematical programming models were developed, using different approaches (such as critical-year values to deal with the problem of overyear storage) to try to overcome the simplifications required for optimization.

117 Mathematical programming models differ from simulation models in that they can in principle generate optimal solutions to problems such as those posed by river basin planning. However, this optimizing capacity is gained at the cost of considerable simplification in the mathematical depiction of the system being studied. This in turn means that a formally optimal solution may not in fact be optimal or even hydrologically realistic. Thus the authors recommend combining the optimization search from mathematical programming with the detailed testing of configurations that is possible with simulation models. In summary: We have concluded that when mathematical models of this kind [i.e. mathematical programming models] are perfected, we shall be in possession of an adequate strategy for the design of simple river-basin systems. For more complex developments, however, perhaps even for the simplified river-basin system, the mathematical models will probably continue to give only an approximate answer or a good first fit. With this we can turn to simulation. The operating procedures that we program will then be optimal or near optimal, because of knowledge obtained from the mathematical model; similarly, the combinations of variables that we select for computer studies will be in the region of the optimum, because that region will have been identified by the model. (261–62) In other words, the two types of models will be used together, taking advantage of the strengths of each. In retrospect, given the computational power and user- friendly interfaces of today, the talent and care of the researchers in developing their results and conclusions is remarkable, and this part of the volume repays reading by those who would like to have a sense of the dawn of the computer era in water resource planning.

118

Political Decision Making The political decision framework proposed for the design process is in chapter 15 of the volume. This was designed in part to remedy the inadequacies of traditional Corps procedures that Arthur Maass analyzed in Muddy Waters (1951; see chapter 6 above): that the Corps saw itself as responsible to Congress rather than to the executive, and that as one result there were not consistent national policies and criteria for water resources development (Maass 1951, 132–33). This model starts with a discussion of the role of community and state and then passes more specifically to the roles of the executive and the legislature in decision making. The authors argue that in both the legislative and administration processes, the executive should take the initiative and Congress should perform an oversight role (581; see fig. 11-3 below). The objectives of water resource development should be developed in the legislative discussion process, and the design criteria should reflect these. The idea is that this approach will provide a consistent water resources development program without the ad hoc and inconsistent quality of the past, which prevented nationally agreed objectives from being achieved efficiently. Obviously the approach being advocated is the opposite of a pure “pork barrel” one; in actual practice there tends to be a blend of national policy and pork barrel.

Figure 11-3: The relationship of institutions, processes, and roles in government decision making

Source: Maass et al. (1962, 581)

119

Perspectives The influence of Design of Water-Resource Systems has been wide and continuing, and indeed it is so pervasive that it would be difficult to trace in its entirety. Following the publication of the book, Maass (1966, 1970) made further assessments. Journal articles and reports worldwide reflect the work, doubtless often without the authors realizing where their ideas and techniques originated. Two landmark studies using the multiobjective and multiple-model techniques of Design of Water-Resource Systems followed directly from the study. One was the North Atlantic Regional (NAR) Study (U.S. North Atlantic Regional Water Resources Study Coordinating Committee 1972 and Major and Schwarz 1990; see chapter 12 below). This was one of some twenty framework studies that were scheduled to be completed for the U.S. Water Resources Council, of which 11 were underway before the system was superseded by national assessments (U.S. Water Resources Council 1968 xii). The NAR study used three objectives—the efficiency (national income) objective, regional development, and environmental quality— and the models used included mathematical programming, simulation, and an input-output system for forecasting water demands. A second study was the MIT-Argentina study (Major and Lenton 1979). This was a study of the Rio Colorado in Argentina, which was apparently the first in which a river basin was planned with a linked set of mathematical models, the approach recommended by the authors of Design of Water-Resource Systems. The techniques used included mathematical programming and simulation models, and the objectives included national income and several objectives relating to the distribution of water and income to five provinces with interests in the basin. In both of these studies, personnel associated with the Harvard Water Program participated, and in the NAR study Maass, the faculty research leader of the HWP, served on the Board of Consultants. Many other models and studies followed that used one or another of the techniques pioneered in Design of Water-Resource

120

Systems. The book was also a key inspiration for the 1973 Principles and Standards (chapter 13, below), the multiobjective federal water planning standards based on an expanded view of multiple objectives (not simply the efficiency vs. income distribution of Design of Water-Resource Systems). As would be expected, however, the entire program of the book has not been reflected in water planning—institutional entanglements, new planning and modeling developments, and regulation have all had effects. Perhaps the most interesting is the role of regulation. In mathematical terms, regulation and optimization can in principle attain the same results, but in practice regulation tends to heavily constrain attempts at optimization, the underlying principle of Design of Water- Resource Systems. This situation may, however, change as techniques improve and computers grow faster, making optimization a more appealing alternative to simple regulation. As for objectives, most water planning problems are essentially multiobjective, but they are not necessarily or even mostly analyzed in the orderly way laid out in Design of Water- Resource Systems; various approximations to the ideal are often used. This situation is explored in Stakhiv (1986) as it relates to federal criteria. With respect to the models, simulation models have now become very widely used, beginning with Hufschmidt and Fiering (1966), among others. Mathematical programming is somewhat less used as compared to what was foreseen in Design of Water-Resource Systems; more powerful computers have permitted simulation models to review many more elements of the response surface, somewhat blurring the distinction between optimization and simulation models in terms of their results. Political decision making on water resources development has moved more toward national criteria since Design of Water- Resource Systems, but it remains a complex mix of standards and specific project decisions.

121

122

Chapter 12 The North Atlantic Regional Study

U.S. North Atlantic Regional Water Resources Study Coordinating Committee, North Atlantic Regional Water Resources Study, U.S. Army Corps of Engineers, North Atlantic Division, 1972

The North Atlantic Regional Water Resources (NAR) Study was a large-scale study for the region stretching from Maine to Virginia and from the Atlantic coast to the Appalachians. Some twenty “framework” studies were originally scheduled for completion; the NAR study was one of them. The idea of the framework studies was to provide a regional planning context within which detailed, consistent plans for water resources management and development could be implemented. (Eleven of the studies were underway when the program was later superseded by the National Assessments, the first of which was U.S. Water Resources Council (1968). The NAR study is, from the standpoint of planning methods, the most significant of these studies. It embodies many of the most important characteristics of the era of water planning that commenced in the early 1960s, the paradigm and techniques of which remain central to the field today. The NAR study was the first, and one of the most successful, attempts to put the theory, philosophy, and principles of the Harvard Water Program (chapter 11) into practice, and the study was widely followed in the water resources profession. It was undertaken at the time when large-scale mathematical modeling became practical for water planning, and it also broke new ground in its assessment of visual and cultural factors and the potential impact of water resources development on them. The most detailed assessment of the NAR study is that of Major and Schwarz (1990), which is quoted in this essay; some personal reflections on the work are in Schwarz and Major (1993) and in an oral history interview with the chief planner, Harry E. Schwarz, reprinted in Major (2010). The breadth and comprehensiveness of

123 the study are illustrated by the list of study reports in table 12-1. The study volumes are on deposit at the library of the U.S. Army Corps of Engineers Institute for Water Resources in Fort Belvoir, Virginia. In this essay, references to the NAR study are in two forms: (25) means the main report of the study, p. 25; and (T-30) refers to appendix T, p. 30. Other appendices and annexes are referenced in the same way.

Figure 12-1: REGIONS FOR PLANNING Source: Major and Schwarz (1990, 27)

Table 12-1: Principal Reports of the NAR Study Main Report Annex 1: Area Summaries Annex 2: State Summaries Appendices

A. History of Study B. Economic Base C. Climate, Meteorology, Hydrology D. Geology and Groundwater E. Flood Damage Reduction F. Upstream Flood Prevention G. Land Use and Management H. Minerals I. Irrigation J. Land Drainage K. Navigation L. Water Quality and Pollution M. Outdoor Recreation N. Visual and Cultural O. Fish and Wildlife P. Power Q. Erosion and Sedimentation R. Water Supply S. Legal and Institutional Environment T. Plan Formulation U. Coastal and Estuarine Areas V. Health Aspects

Selected Other Reports Plan of Study, Vols. I, II Special Publication: Manuals for Computer Programs of NAR Study: Storage-Yield, Demand and Supply Models Source: Major and Schwarz, 1990, p. 3 In addition to marking a break with more traditional methods of basin planning, the NAR study looked forward to,

124 and influenced, the multiobjective Principles and Standards (P&S) of 1973 (see chapter 13 below), representing as it did a leading example of the feasibility of the approach embodied in the standards. The study was largely completed prior to the enactment and implementation of the National Environmental Policy Act (U.S. Congress 1969) and the Federal Water Pollution Control Act Amendments of 1972 (U.S. Congress 1972), which changed many of the key elements in planning and brought about an additional set of priorities, some of which were foreshadowed in the NAR methods. Illustrative of the influence of the methods of the study are the many scientific papers based on them, including multiple objectives, the use of mathematical models, and the institutional methods used (e.g., Schwarz 1972a, 1972b; Schaake and Major 1972; DeLucia and Rogers 1972; Major 1972; Schwarz, Major, and Frost 1975). This essay provides an overview of and perspectives on the planning methods used in the NAR study and its place in the development of these methods.

Institutional Arrangements The main institutional measures developed and applied in the NAR study were shaped by the planning context: a framework plan; a commitment to multiobjective methods; significant quantities of data from different agencies; and the use of many standard methods of economics, engineering, and water planning (Major and Schwarz 1990, 23–24). These institutional methods encompassed first the development of a planning structure specific to the NAR. This comprised not only the prescribed Coordinating Committee but also an outside Board of Consultants (of which both Gilbert White and Arthur Maass were members) and a Plan Formulation Work Group. Second, there was provision for continued interaction and iteration throughout the planning process among the Coordinating Committee, the planning staff, and the staffs of the cooperating agencies (Major 1977, 56–57). Third, new reporting forms were developed for use by all participants throughout the planning process. These were designed to ensure that, as far as possible, information would be

125 focused on the needs of the study rather than presented in terms of standard (and possibly unrelated) agency procedures. Fourth, the use of the models themselves was an institutional method that contributed to the coherence and cohesiveness of the study (Major 1972). All these methods were utilized within a multistage planning process (Major and Schwarz 1990, chap. 8).

Regional Subdivisions for Planning For the plan the North Atlantic region was divided into geographical units at three different scales. The basic unit of planning for the NAR was the area, of which there were twenty- one (fig. 12-1). Areas were defined primarily on the basis of hydrologic criteria. They vary in size from the largest, the Susquehanna, Area 17, with 25,510 square miles, to the smallest, New York City, Long Island, and Coastal Westchester, Area 13, with 1,900 square miles. Some of the areas have more than one drainage basin; others only one, such as the Susquehanna; and still others have part of a drainage system (NAR 1972, 18; descriptions of the areas on 20–21). For some purposes the areas were aggregated into larger units, the six subregions (A–F in fig. 17-1). For other planning purposes the areas were subdivided into subareas (also called subbasins); these are shown as 21a, 21b, etc. in figure 12-1. For example, these smaller units were used in the supply model, described below. Economic and demographic data were generally available by county; for the purposes of using these data, area and subarea hydrologic boundaries were approximated by relevant county boundaries. Plan results for the NAR were presented for the whole region (NAR 1972, chap. 8) and for the areas (chap. 8 and annex 1); as well as for the states (annex 2).

126

Figure 12-1: Regions for planning Source: Major and Schwarz (1990, 27)

Planning Period and Forecast Years The NAR planning period extended to 2020, with intermediate benchmark years of 1980 and 2000. The planning activities in the study and the final forecasts of needs, sources, and devices for different objectives took place in terms of these three benchmark years. The study was intended to cover a period of fifty years from the initially scheduled date of completion; it was delivered to the Water Resources Council in 1972, so the actual forecast time from date of delivery was forty-eight years. The first benchmark year, 1980, was chosen to provide a relatively near-term target that was nonetheless sufficiently far

127 off that some implementation activities could take place by that date. The second benchmark year, 2000, was intended to provide for guidance in planning for, among other measures, large structural elements if these were recommended; the planning and implementation of these is a process that covers many years. The final benchmark year, 2020, provided a distant marker for generalized long-range planning: “This set of three benchmark years was judged sufficient to provide suitable planning guidelines at the framework level” (Major and Schwarz 1990, 26).

New Methods: Multiple Objectives Multiobjective analysis is a generalization of traditional cost- benefit analysis (see chapters 10 and 11 above). Traditional analysis focuses on the national income (or “efficiency”) objective of water planning. Multiobjective analysis, by contrast, emphasizes the design of water projects and programs in terms of all relevant objectives: environmental, regional, social, and others, including the national income objective. In practice, as in the NAR study, multiobjective analysis is concerned with the choice of objectives, the development of alternative feasible plans responsive to objectives, and the final choice of a plan. Works that describe the theoretical structure of multiobjective planning include Maass et al. (1962; see chapter 11 of this volume), Marglin (1967), Dasgupta, Sen, and Marglin (1972), Major (1977), and Major and Lenton (1979). NAR planning took place in terms of three objectives: national income (the traditional objective of benefit-cost analysis), regional development, and environmental quality (A- 10). The approach to plan formulation for multiple objectives that was employed in the study included the development of three alternative plans, each with a somewhat exaggerated emphasis on a single objective. This approach is shown in figure 12-2, which illustrates the geometry of planning for the NAR. These plans were examined by the governing federal-state Coordinating Committee, which then used them to decide on a final recommended mixed objective plan for the NAR. Multiobjective

128 plan formulation in turn shaped both the models and the institutional methods used in the study.

Figure 12-2: The geometry of planning for the NAR Source: Major and Schwarz (1990, 36)

The level of detail at which each multiobjective effect was examined varied, depending on its importance for decision making; available staff, time, and other resources; and the methods available. Relatively detailed work was done for economic cost accounting and for visual and cultural effects. With respect to the latter, a landscape quality survey of the entire North Atlantic region was undertaken as part of the environmental assessments in the study. Figure 12-3 shows the results of this innovative study. By contrast, economic benefits were discussed in terms of general magnitude, and regional effects were discussed broadly. The multiobjective approach was

129 thus used as a conceptual framework to guide the organization and evaluation of information rather than as a consistently detailed guide to the evaluation of individual projects. In retrospect, the three objectives used in the study, national income, regional development, and environmental quality, appear to represent well the objectives relevant to water and related land planning in the NAR. The extent of iteration about preferences toward objectives among staff and Coordinating Committee representatives seems also to have been generally adequate. The level of detail at which the multiobjective approach was used in the study could have been usefully increased. Even with the constraints on the planners' abilities to handle large amounts of information, given the limited computational capabilities available at the time of planning, it would have been appropriate to attempt further detailed assessment of multiobjective impacts, particularly in the areas of national income benefit assessment and the measurement of regional impacts. (Major and Schwarz 1990, 168)

New Methods: Mathematical Models In the NAR planning process three models were essential components: a demand (or requirements) model based on input-output methods; a mathematical programming supply model; and a storage- yield model used to generate hydrologic inputs to the supply model. These models were used together to forecast requirements and to estimate supply costs. The three models are treated in detail in Major and Schwarz (1990) and the papers cited below.

130

Figure 12-3: Landscape quality ranking in the NAR Source: Major and Schwarz (1990, 59)

The demand model (Schaake and Major 1972) is a group of subroutines based on forecasting relationships designed to yield estimates of demands for water based on specific assumptions. The mathematical components of the model are an input-output table for the NAR region, a regression estimator for publicly supplied municipal and industrial water, and a group of arithmetic operations associated with these components. The principal inputs to the model are projections of regional economic product, population, personal income, and water withdrawal coefficients, together with projections of the geographic distribution within the region of these variables. The

131 model acts on the inputs to produce estimates by benchmark years of water flow demands by economic sector, by type of water quality, and by subbasin, basin, state, or area. These flow demands are inputs to the supply model. Many different runs, responsive to alternative assumptions about forecasts and objectives, were made with this model; these are summarized and evaluated in NAR appendix T-271–92. The supply model is a mathematical programming (optimizing) model (DeLucia and Rogers 1972). It is designed to permit evaluation of the sources and costs of supplies of water required to meet various subarea water requirements as specified by the demand model. The basic inputs to the model are withdrawal, instream, and consumption requirements for the fifty NAR subareas and combinations thereof; data on existing and prospective intrabasin and interbasin transfers; and estimates of the costs of development. The formal objective of the model (the objective function) is to minimize the costs of supplying specified requirements. Provision is made, however, for inserting and varying parametrically constraints on sources, degrees, and types of development in the model, and in this way the influence of planning objectives other than cost minimization can be brought to bear. Supply model runs are described and analyzed in NAR appendix T-331–35. The storage-yield relationships used in the supply model are based on analyses conducted with the aid of the third model. This hydrologic model was designed to determine the amount of storage required and the risk of failure of that storage at various demand rates at locations where the streamflow is known, and to express these failures by means of a shortage index. In the model, monthly historic streamflows for groups of stations are extended to equal length, and then streamflow traces of 100 year length per series are generated. These streamflows are routed past a given station at predetermined storage increments to estimate the shortages obtained given preset yield requirements. Deficiencies are expressed in terms of an index computed by the model. The storage-yield model is described in NAR (Special Publication, 1–28).

132

The three principal computer models used in the study— the demand, supply, and storage-yield models—thus deal, respectively, with demand (or needs) forecasting, the analysis of supply alternatives, and the assessment of the stochastic physical resource. The models were all developed at a relatively early stage in the application of computer methods to water resources. With the availability of vastly increased computational power and modeling experience since the NAR, more detailed models can be chosen and developed. The models did not incorporate all of the multipurposes of planning, for example navigation and flood control and there were institutional and conceptual problems in linking these to the purposes such as industrial water supply that were included in the models. The planning process as a whole would have been improved had more been done to incorporate one purpose, water quality, into the models. The difficulties encountered in doing this in the NAR study were in part due to the institutional stress at the time of what is now the Environmental Protection Agency; during the planning period the relevant office was part of, successively, the Public Health Service, the Federal Water Pollution Control Administration, and the Environmental Protection Agency. However, another reason was simply that water quality was a relatively new consideration in framework planning at the time, and the range of problems of integrating any new factor into a complex process were encountered with respect to it. (Major and Schwarz 1990, 169–70)

Needs, Sources, and Devices The examination of the interrelationships among alternative needs (outputs), sources, and devices occupied an important role in NAR planning from the beginning of the planning process. The analysis of these elements was undertaken in an explicit and systematic way in the NAR study and with an awareness of the

133 possible relationships of each of the elements to the planning objectives (T-26, 27, 31–33). Fifteen water and water-related needs were examined in the NAR plan (123). The needs are outputs of water and related land systems and are thus part of the physical relationship known as the production function in economics, as are also the sources and devices. The sources that were examined in the NAR planning process include surface waters (rivers, lakes, fresh and saline estuaries, oceans); groundwater (fresh, mineralized, and saline); and atmospheric water (chap. 4; T-36). Land resources (chap. 4), discussed particularly in connection with visual and cultural needs, are described in Major and Schwarz (1990). Twenty-three major categories of devices were considered in the NAR (134–35). To examine the possible interrelationships among needs, sources, and devices (including nonstructural devices) systema- tically, a three-dimensional matrix was developed to relate the three elements (T-35, 36). In the matrix each cell represented one possible interrelationship among the needs, sources, and devices. The matrix has 4,050 cells representing fifteen needs, nine sources, and thirty devices (as compared to the list of twenty-three devices used later in planning). An early analysis concluded that there were 655 relationships of interest, or about 16 percent of the possible relationships. During the planning process, these were examined to focus on the most important relationships and to eliminate the least important from further systematic consideration (T-27).

Forecasts and Recommendations The outputs of the NAR study are described and summarized by Major and Schwarz (1990, chap. 9); an appendix to that chapter provides rankings of needs, devices, benefits and costs for the twenty-one NAR areas, and also, as an example, the Area Program for Area 9, Southeastern New England. Overall, the NAR plan includes a wide range of forecasts and recommendations. As with essentially all plans, some of these were accurate and timely in retrospect, and some were not.

134

Among the many forecasts and recommendations of the NAR that were accurate and timely were the forecast of increasing suburbanization of office activities in the region, with its implications for water supply and sewage systems (47) and the plan’s call for a substantial program to maintain high-quality landscape areas in the densely populated NAR region (195). Many other forecasts and recommendations appear in retrospect to be inaccurate. One such is the forecast of substantially increased use of power cooling water (most of it saline or brackish) in the region, to a level of 84,000 cfs in 1980 from 45,000 cfs at the time of planning. This forecast was too high; the planners did not foresee the substantial impact on the U.S. energy sector of the real increases in petroleum prices that occurred following the completion of the plan. … Any recommended plan, which is essentially a point forecast, is nearly certain to be both “right” and “wrong” in different respects. These convergences and divergences of forecasts and recommendations from actuality do not in themselves indicate that a plan is adequate or inadequate. The question is whether the planning process as a whole was a useful mechanism for aiding decisions. Here, two things seem clear. First, the idea of alternative plans is a good one, and in fact is central to such legislation as NEPA [National Environ- mental Policy Act of 1969]. These permit the illustration both of alternative possibilities, and of alternative preferences among objectives. In this regard, perhaps the alternative plans in the NAR study could have been more fully elaborated, given additional resources; and perhaps one or two more alternative plans could have been prepared. (Major and Schwarz 1990, 166, 167) In addition to alternatives in the planning process itself, there is a need for a continuous planning process (Schwarz and Major 1971, 6; Kamarck 1983, 129). The report recommended

135 updating the NAR plan every ten years in the year immediately following each census year (223).

Perspectives The NAR study has seen extensive use as a guide for some regions and purposes and as a source of methods. The use most in line with the framework idea was the employment of the demand model to produce forecasts in the U.S. Water Resources Council’s Second National Assessment (1979; the National Assessment program superseded the framework planning program). For this assessment, the demand model was rerun to provide forecasts of water needs for the Mid-Atlantic region (Schilling n.d.). The methods of the NAR also influenced its regional companion plan, the Northeastern United States Water Supply (NEWS) Study. This influence included the use of the multiobjective approach to characterize alternative plans (for example, U.S. Army Corps of Engineers, 1975, ch. 4), and the application of other methods such as the demand model regression equation (Anderson-Nichols 1971, 33). As another example, the NAR study’s region 9 (eastern Massachusetts, most of Rhode Island, and a small portion of Connecticut) formed the basis of the New England River Basins Commission’s Southeastern New England Study. The conceptual basis of the visual and cultural ranking system was used in somewhat different form in the Souris-Red-Rainy basins study (Souris-Red- Rainy River Basins Commission 1972, I-164ff.). (Major and Schwarz 1990, 165) Although widely used as a guide for regions and purposes and as a source of methods, the NAR study was not consistently used as a framework plan as originally intended by the U.S. Water Resources Council. The idea of the framework studies was to provide a regional planning context within which detailed, consistent plans for water resources management and development could be

136 implemented. This use did not develop in a consistent way for funding and institutional reasons. Funding for federal water development was much reduced in the 1970s and 1980s because of controversies over local funding for projects and environmental considerations. In addition, funding for the Water Resources Council, under whose aegis the framework plans were undertaken, ceased in the early 1980s. Finally, Institutional mechanisms to implement the framework plans after their completion were not fully provided at the time the framework planning idea was developed, and thus there was no automatic or simple way available to insure their use. (One method of doing this would be to create and staff, during the study, a small continuing work group to serve as a source of post-study guidance to planning agencies on the methods and conclusions of the study; see the suggestion in Major and Lenton, 1979, p. 227, regarding the importance of continued post-study liaison between planners and the sponsoring agency.) (Major and Schwarz 1990, 168) In overview, the NAR study can be seen as a landmark effort. It embodied, more than any other study of the time, the new methods of water resources planning developed in the 1960s, including especially multiple objectives and the applied use of mathematical models run on digital computers. The NAR study influenced much of what followed and looked forward to a wider emphasis on the environment. It illustrates what can be accomplished in water planning given adequate resources and freedom to innovate.

137

138

Chapter 13 The Principles and Standards, 1973

U.S. Water Resources Council, “Water and Related Land Resources: Establishment of Principles and Standards for Planning,” Federal Register 38 (174) (1973): 24,778–869

The Water Resources Planning Act of 1965 (U.S. Congress 1965) established the U.S. Water Resources Council (WRC), which had among its other duties the development of standards and criteria for planning and evaluating water resource projects. Beginning with a Special Task Force in 1969, the WRC followed an extensive process of development that resulted in the Principles and Standards (P&S) of 1973. This document broke new ground in investment criteria for federal water resource planning, committing federal agencies to a detailed multiobjective approach. In this respect it is as notable as the Green Book, an earlier federal effort that codified planning primarily for the economic objective.

Background The P&S was based on a theoretical development in planning: the generalization of benefit-cost analysis from a strictly economic method to include what were seen as the multiple objectives of public expenditure, such as increasing regional income or protecting the environment (see chapters 11 and 12). In some earlier documents, including the Flood Control Act of 1936 (chapter 4) and the Green Book (chapter 7), noneconomic objectives were mentioned, but not in the explicit way in which they are treated in multiobjective analysis, and no attempt was made to develop detailed methods of assessing positive and negative effects on these other objectives. The change in method is most associated with Design of Water-Resource Systems (see chapter 11) and other publications related to that book (Marglin 1967 and Major 1977, in both of which the theory is presented graphically). Multiobjective analysis considers all relevant social

139 objectives (one among them being the national income [economic] objective) in a way that permits the analysis and design of projects and programs to reflect an appropriate balance among objectives. The theoretical underpinning of the Green Book approach is the welfare model of economics (Bergson 1938), a model of optimal allocation based on individual economic preferences. While not neglecting the economic approach, multiobjective analysis reflects the need for a political process for making some decisions; it relates to the fundamental work of Kenneth J. Arrow (1963), dealing with the impossibility of always aggregating individual preferences successfully, and John Rawls (1999), concerned with the welfare of the least advantaged. The Green Book, Bureau of the Budget Circular A-47, and other related directives made federal water resource planning in the period from 1950 through the early 1960s a system of planning in which projects and programs were undertaken for multiple objectives according to the most basic standards in the legislative histories of the programs but in which they were evaluated in terms of only one objective. This made the already complicated decision process used to approve federal water resource projects even more complex. “The inadequacy of single- objective analysis for multiobjective problems is illustrated by the fate of the U.S. Army Corps of Engineers (1970) report for the Potomac River basin, which was not approved in part based on environmental considerations which the Corps had no mandate, under the then-prevailing standards, to analyze in a systematic way” (Major 1977, 4). An important element of the transition from the single- objective orientation of the Green Book and Circular A-47 to multiple objectives is marked by a prescient report from a panel of experts to the Bureau of the Budget (Hufschmidt et al. 1961). This document provided the rationale for multiobjective planning for water resources. It was not officially released by the bureau, although it was disseminated informally. An official set of criteria that moved toward multiobjective standards was a report

140 by the U.S. President’s Water Resources Council (S. Doc. 87-97, 1962; see chapter 10). While this document was sufficiently multiobjective in its approach to provide some of the impetus for multiobjective planning (e.g., the NAR study; see chapter 12), a full statement of multiobjective planning guidelines for federal agencies awaited the work of the Special Task Force of the Water Resources Council.

Authority The secretary of the interior had earlier drafted plans for a Department of Natural Resources, but this idea was dropped and responsibility for water planning was left with the several agencies involved. However, a Federal Water Resources Council was created (Reuss 1991, 27). The new council’s authority was carefully limited in section 3 of the Water Resources Planning Act of 1965 (U.S. Congress 1965): Nothing in this Act shall be construed— (a) to expand or diminish either Federal or State jurisdiction, responsibility, or rights in the field of water resources planning, development, or control; nor to displace, supersede, limit or modify any interstate compact or the jurisdiction or responsibility of any legally established Joint or common agency of two or more States, or of two or more States and the Federal Government; nor to limit the authority of Congress to authorize and fund projects; (b) to change or otherwise affect the authority or responsibility of any Federal official in the discharge of the duties of his office except as required to carry out the provisions of this Act with respect to the preparation and review of comprehensive regional or river basin plans and the formulation and evaluation of Federal water and related land resources projects;

141 On the other hand, the council was given the significant duty of establishing investment criteria for federal projects; this is the origin of the P&S. Sec. 103. The Council shall establish, after such consulta- tion with other interested entities, both Federal and non- Federal, as the Council may find appropriate, and with the approval of the President, principles, standards, and procedures for Federal participants in the preparation of comprehensive regional or river basin plans and for the formulation and evaluation of Federal water and related land resources projects. Such procedures may include provision for Council revision of plans for Federal projects intended to be proposed in any plan or revision thereof being prepared by a river basin planning commission.

Objectives and Criteria Process. The WRC began its process for formulating principles, standards, and procedures in response to the 1965 act by convening a Special Task Force. The members of the task force were from federal agencies (WRC 1970b, frontispiece); the task force was assisted by many more agency personnel and independent outside reviewers (WRC 1970a, acknowledgments, 1–8). Its first, preliminary report was the “Blue Book” of June 1969 (WRC 1969), and its final proposals were published in 1970 (WRC 1970a, b, c, d). Three of the reports (1970a, b, and c) were published in a Senate committee print, “Procedures for Evaluation of Water and Related Land Resources” (U.S. Senate 1971; fig. 13-1), indicative of Congress’s interest in water resources investment criteria.

142

Figure 13-1: U. S. Senate, Committee on Public Works: 1971, Procedures for Evaluation of Water and Related Land Resources: Findings and Recommendations of the Special Task Force of the United States Water Resources Council.

Following a review period, proposed Principles and Standards were published in the Federal Register of December 21, 1971 (WRC 1971); the final Principles and Standards are in the Federal Register of September 10, 1973 (WRC 1973). The task force report is an impressive effort, running to several hundred pages and covering a full range of project evaluation issues, including the implementation of multiple objectives; it is a handbook of project evaluation and still deserves attention. It is the

143 proper analog of the Green Book, a product of a federal interagency group with the lengthy comment period (and bargaining) that resulted in the P&S. In addition to bringing together agency representatives and outside commentators, the WRC’s process was unusual in providing funding for nineteen tests of the applicability of the proposed new standards (1970a, 3). One published example of a test, comparing national income benefits to acres of an environmentally significant area that would be flooded in part by a proposed dam, is in Major (1977, 69–74). In a summary statement the WRC said of the tests that they “have indicated that the multiobjective approach to planning is practical. Meaningful results can be accomplished and reasonably uniform comparability in application can be achieved by establishing carefully structured principles, standards, and procedures” (1970a, 3–4). Objectives. In considering the P&S, it is useful to focus first on the work of the task force, which is the fundamental set of documents on which the P&S is based. The task force proposed four principal objectives or categories of objectives for water resource planning: (1) national economic development; (2) quality of the environment; (3) social well-being; and (4) regional development (WRC 1970a, 5). The 1970 report represents a full commitment to the use of multiple objectives in the process of project and program formulation: “No one objective has any inherently greater claim on water and land use than any other. These principles do not imply the relative priorities to be assigned among the multiobjectives in plan formulation and evaluation” (1970a, 3). The Standards (WRC 1970b) provide extensive detail on ways of estimating positive and negative effects of projects on objectives. The following principles are proposed for plan formulation (WRC 1970a, 16): l. Specify components of the multiobjectives relevant to the planning setting.

144

2. Evaluate resource capabilities and expected conditions without any plan. 3. Formulate alternative plans to achieve varying levels of contributions to the specified components of the multiobjectives. 4. Analyze the differences among alternative plans which reflect different emphases among the specified components of the multiobjectives. 5. Review and reconsider, if necessary, the specified components for the planning setting and formulate additional alternative plans as appropriate. 6. Select a recommended plan based upon an evaluation of the tradeoffs among the various objectives. This series of steps is essentially a complete presentation of applied multiobjective planning (Major 1977, 6). Objectives in the 1971 proposed P&S and the final 1973 P&S. The choice of objectives in water planning, as well as the interest rate, have been the subject of bargaining and conflict for many decades, both within federal agencies and with other participants in water resource development. A change in the presentation of objectives between the task force report and the 1971 proposed P&S reflects this. In the latter the number of objectives for which projects could be designed was reduced from four to three: national economic development, quality of the environment, and regional development (WRC 1971, 24,145). The proposed P&S then introduced a distinction (not made in the task force report) between design and display objectives. In addition to the three design objectives, another objective was presented as a display objective: “social factors” (24,145–46). The meaning of this distinction is that the effects of alternative projects designed for the three principal objectives could then be displayed for the fourth objective. This distinction marks a retreat from the broader approach of the task force; on the other hand, by allowing the display of effects on the display objective, the proposed P&S ensured that these effects would be part of

145 congressional decision making. The concepts of design and display objectives are explained in figure 13-2.

B 0

Figure 13-2: Design and Display Objectives. If there are two design objectives (national and regional income), the maximum of either on the net benefit transf- ormation curve (the boundary of feasible choices) might be chosen (B or C), or, more likely, some intermediate point (A) would be selected, with net benefits to the two objectives, for example, of N1 and R1. If national income is the only design objective, B would be chosen, and the effects of this choice on regional income (R2) would be displayed. If there is only one design objective (national income) and no display objective, only the vertical axis would exist, and the optimum point would be B0.

Source: Adapted from David C. Major, Multiobjective Water Resource Planning (Washington, D.C.: American Geophysical Union, Water Resources Monograph 4, 1977), p. 11. See the discussion on pp. 17-8 of Major, 1977.

In the final P&S, the number of design objectives is reduced to two: national economic development and environmental quality; the two display objectives are regional development and social well-being (WRC 1973, 24,781–83). It is of interest that the two design objectives bring together classical economic benefit-cost analysis and the environmental objectives of NEPA, although in a more explicit design framework than the latter legislation. The limitation of design objectives in the proposed and final P&S did not sit well with Congress, reflecting a continuing debate about objectives. In the Water Resources Development

146 Act of 1974, section 80c, it is stated that the president should undertake a “full and complete” investigation of principles and standards that “shall include, but not be limited to, consideration of enhancing regional economic development, the quality of the total environment including its protection and improvement, the well-being of the people of the United States, and the national economic development, as objectives to be included in federally- financed water and related resources projects and in the evaluation of costs and benefits attributable to such projects” (U.S. Congress 1974). Other investment criteria. The task force report and the two versions in the Federal Register describe a wide range of investment criteria, including significant amounts of attention to benefit and cost accounting for objectives. An additional criterion of particular importance is the recommended interest rate; this changed from one document to the next. WRC (1970b, 12) states: “A discount rate will be used in evaluation of plans to reflect the relative values placed by society on benefits and costs toward the multi-objectives occurring in the future as compared with benefits and costs occurring in the present. The rate used for this purpose shall reflect the public aspects of the discounting process.” The rate chosen in the P&S is 5.5 percent (WRC 1970b, IV-7, IV-8). The approach in the task force report reflects both the importance of the interest rate in project evaluation and the complex issues surrounding the choice of such a rate. In the development of multiobjective theory, an integral part was treating the interest rate as a series of social weights over time on benefits and costs, expressed as a percentage (Marglin 1967, 47ff.). In this view, the interest rate is a social rate of discount. If this approach is taken, in order to avoid displacing better private investment, a shadow price based on the value of displaced private investment is used to prevent misallocation of resources. An additional consideration is that weights can differ among objectives over time. Implementing this approach would have been beyond the abilities of the agencies at the time and has still

147 not been fully confronted (it has emerged again in discussions of global warming: Nordhaus 2007; Stern 2007, 2009). However, the task force’s acknowledgment of the complex significance of the interest rate was a step forward. In the 1971 proposed P&S, on the other hand, the interest rate is discussed at great length from one perspective, the use of weighted private rates of return, and the resulting value is given as 10 percent (WRC 1971, 24,166–67) Leaving aside the many assumptions as to private rates of return involved, this value would be part of the shadow price in multiobjective terms rather than being the rate of discount. This high rate would not have met with favor in Congress, and in fact the proposed P&S retreated to a rate of 7 percent for the following five years, “recognizing the objective of subsidizing water resource projects” (24,167). After recommendations in the task force report and the proposed P&S for two quite different approaches to the interest rate, the final P&S retreated to the more comfortable position of relating the interest rate to the cost of government borrowing (WRC 1973, 24,784), a view that has a long history in water resources planning.

Perspectives The Water Resource Council’s planning efforts in response to the Water Resources Planning Act of 1965 constitute a significant contribution to the development of applied water resources planning in acknowledging multiple objectives, testing their applicability, and developing methods for counting positive and negative effects toward each objective. In this respect, the task force report is analogous to the Green Book, in that it presented ways of implementing a new theoretical approach. At the same time, it should be noted that despite its valuable efforts on principles and standards, the Water Resources Council lost its backing within the government and from 1982 was no longer funded. While debate on objectives has continued, the proliferation of laws on the environment, and the quality of life

148 tend to make all government actions multiobjective; a main issue is the extent to which planning takes place explicitly in terms of the objectives. From this standpoint, the work of the council represents an important step forward.

Authors’ Note One of the authors (DCM) served as a special consultant to the Water Resources Council and in that capacity was an alternate member of the Special Task Force. Bibliographical Note The official records of the U.S. Water Resources Council, 1965– 1982, are maintained in Record Group 315.3, United States National Archives, College Park, Maryland.

149

150 Section IV Perspectives

This final section provides a summary of the key developments treated in the book. Thesse developments include, among others, those relating to objectives and criteria, methods, organizational arrangements, and funding controversies, decisions on which are all relevant to the present day. A knowledge of how these developments occurred in the past and their impacts can help planners to deal with the complex challenges now facing the Nation.

151 152 Chapter 14 A Summing Up and Conclusions

In the first section of this volume, three key federal laws and programs were considered, each of which focused federal efforts on significant elements of water resources planning. The Reclamation Act of 1902 (chapter 2) was a landmark in the development of water planning in the United States: it provided for single-purpose development for an objective that, broadly defined, was the continued development of the American West utilizing smallholder agriculture. Within a few weeks of the passage of the act, an organizational framework was set up for single-purpose irrigation planning, with a repayment mechanism and a regional objective. The irrigation act illustrates the long time line of policy and infrastructure and how these may no longer serve the objectives of a later day, including environmental objectives and the pressures of drought. The 308 plans (chapter 3) resulted in a significant change in water planning methods, for the first time providing clear authority for the Corps of Engineers to develop multipurpose river basin plans nationwide; it also had practical results. The plans resulted in the development of a large number of dams and other water resources infrastructure. From a modern standpoint, the original objective of the 308 reports was, formally at least, economic, although many if not most projects were selected at least in part based on regional or other criteria. Considerations of employment and lives saved were also brought into play; the idea of an environmental objective was in the future. In terms of purposes, the general idea of multipurpose planning was firmly implanted, although at present additional purposes such as recreation, water quality, and maintenance of flows for ecological purposes would be added. The nationwide remit of the 308 program required continued interaction between the Corps and the Bureau of Reclamation in the latter’s territory; this interaction was sometimes friendly and sometimes not (see

153 chapter 6). The 308 planning effort resulted in a long-term change in water planning: the continued use of basin plans for multiple purposes throughout the United States. The Flood Control Act of 1936 (chapter 4) was a landmark in the development of water planning methods for three reasons: 1. It definitely established flood control as a federal responsibility (section 1). 2. It presented, for the first time in an easily accessible public document, the famous requirement that, for flood control projects authorized in the act, “the benefits to whomsoever they may accrue are in excess of the estimated costs” (section 1). 3. It established the “a-b-c” requirements for local contributions (section 3). The act further provided for interstate compacts to share local contributions and operating costs. This act, along with the 308 reports, paved the way for the classic era of river basin planning after World War II. Two academic critiques affected water planning shortly after the war: Gilbert White’s stress on the importance of integrating all measures, structural and nonstructural, in flood control planning (chapter 5), and Arthur Maass’s analysis of the institutional relationships of the Corps and Congress (chapter 6). These both show the lasting power of outside assessments. It is hard to conceive of water planning in the United States now without the background of these seminal studies. Also at this time, there developed what we call the classic model of river basin planning, with forecasts of needs, least-cost solutions, and the application of benefit-cost analysis (the Delaware River Basin Plan can be seen as a culmination of this; see chapter 9). The developments of benefit-cost analysis, central to this approach, were often quite elaborate, such as the work of an interagency committee in the Green Book (chapter 7) and the work of John Krutilla and Otto Eckstein (chapter 8). (Much of this work is associated with the early incarnation of Resources

154 for the Future in Washington, DC.) At the same time, the economic model of benefit-cost analysis was not in general regarded as the sole criterion for development; other objectives are considered in the documents assessed, including notably the Green Book and the work of Krutilla and Eckstein. The best academic and agency work acknowledged these multiple objectives, even while focusing on the economic objective, but some agency efforts and some narrowly focused academic work did not. In the third section, other substantial changes are recorded: explicit adoption of multiple objectives and the use of mathematical models in planning. The report of the U.S. President’s Water Resources Council (chapter 10) was the first official adoption of multiple objectives, and it reflected Congress’s concern that projects not be evaluated solely in terms of benefit-cost analysis. Design of Water-Resource Systems (chapter 11) was the product of an exceptionally well-organized academic research program that fully incorporated agency personnel in its work. This book developed the conceptual model of multiple objectives and showed how mathematical models could be utilized in applied water resources planning. It also further developed the political decision model of Maass’s earlier work (chapter 6). The volume was perhaps the most distinguished contribution to water planning in the second half of the twentieth century, and its influence is pervasive today. The widely reported NAR study (chapter 12) is the best- integrated effort to implement on a large scale the ideas of Design of Water-Resource Systems and related planning methods in the era of creativity in the 1960s and 1970s. (For a more elaborate scheme of planning development, see Delli Priscoli and Stakhiv 2015.) This era culminated in the planning work of the Water Resources Council, created in the Water Resources Planning Act of 1965 (chapter 13). The Water Resources Council efforts constitute a significant contribution to the development of applied water resources planning: in acknowledging multiple objectives, testing their applicability, and developing methods for counting positive

155 and negative effects toward each objective. In this respect, the council’s task force report is analogous to the Green Book in that it presented ways of implementing a new theoretical approach. At the same time, it should be noted that despite its valuable efforts on principles and standards, the Water Resources Council lost its backing within the government and from 1982 was no longer funded. Looking to the future, the proliferation of laws on the environment and the quality of life tends to make all government actions multiobjective and “increasingly complex” (Schilling, 2002, 1); a main issue is the extent to which planning takes place explicitly in terms of the objectives. From a conceptual point of view, it is clear that most problems are multiobjective, as that term is generally understood. However, it should be said also that, while some of the disagreement on allowed objectives (and the interest rate) is theoretical, much also is instrumental. The number of allowed objectives and the value of the interest rate both affect the number of projects that an observer might regard as appropriate. This is a complex issue on which disagreements will continue. On balance it seems that a recognition of the multiobjective nature of many government investments has become more widely accepted. The issue of the interest rate, on the other hand, remains unresolved. For the future, what these assessments show is both the remarkable development and change of water resources planning methods and the difficult-to- resolve issues of values, which ultimately have to do with objectives and the interest rate (the view of the future). The reports, laws, and books discussed in this volume are focused directly on water and related land planning during the period from 1902 to 1973. As noted in the preface, two worthwhile companion volumes to the present book would be one that focuses on other key documents before and during the period covered, and a second that addresses new environmental and other considerations that significantly expand the context and scope of water resources planning.

156

References

Anderson-Nichols Company, Inc. 1971. “Water Deficits for Urban Metropolitan Areas.” Contract No. DACW 52-71- C-0001, Hartford, CT, May. “Aquashicola Dam Removed from Regional Planning List.” 1989. Morning Call. December 7. Arnold, Joseph L. 1988a. The Evolution of the 1936 Flood Control Act. Fort Belvoir, VA: U.S. Army Corps of Engineers, Office of History. ———. 1988b. “The Flood Control Act of 1936: A Study in Politics, Planning, and Ideology.” In The Flood Control Challenge: Past, Present, and Future, Proceedings of a National Symposium, New Orleans, Louisiana, September 26, 1986, by Howard Rosen and Martin Reuss, 13–27. Chicago: Public Works Historical Society. Arrow, Kenneth J. 1963. Social Choice and Individual Values. 2d ed. New York: Wiley. Baldwin, William. 1998. The History of the U.S. Army Corps of Engineers. 2d ed. U.S. Army Corps of Engineers, Office of History. Bergson, A. 1938. “A Reformulation of Certain Aspects of Welfare Economics.” Quarterly Journal of Economics 52 (2): 310–34. Billington, David P., Donald C. Jackson, and Martin V. Melosi. 2005. The History of Large Federal Dams: Planning, Design, and Construction. Denver: Bureau of Reclamation, October 25. (Project funded by U.S. Bureau of Reclamation and the U.S. Army Corps of Engineers and managed by U.S. National Park Service.) Boulding, K. E. 1952. “Welfare Economics.” In A Survey of Contemporary Economics, edited by B. F. Haley, 2:1–34. Homewood, IL: Richard D. Irwin. Brookings Institution, Institute for Government Research. 1919. U.S. Reclamation Service: Its History, Activities and Organization. Washington, DC. Caulfield, Henry P., Jr. 2000. “Early Federal Guidelines for Water Resource Evaluation.” Water Resources Update 116 (March): 14–17.

157 Dantzig, George. 1963, Linear Programming and Extensions. Princeton, NJ: Princeton University Press. Dasgupta, Partha, Amartya Sen, and Stephen Marglin. 1972. Guidelines for Project Evaluation. United Nations Industrial Development Organization Project Formulation and Evaluation Series, No. 2. New York: United Nations. de Roos, Robert, and Arthur Maass. 1949. “The Lobby That Can’t Be Licked.” Harpers 199 (August). DelawareWaterGap.org. n.d. “Tocks Island Dam Project.” http://delawarewatergap.org/TOCKS_ISLAND_DAM_ PROJECT.html. Delli Priscoli, Jerome, and Eugene Stakhiv. 2015. “Water- Related Disaster Risk Reduction (DRR) Management in the United States: Floods and Storm Surges.” Water Policy 17:58–88. DeLucia, Russell J., and Peter Rogers. 1972. “North Atlantic Regional Supply Model.” Water Resources Research 8 (June): 760–65. Eckstein, Otto. 1961. Water-Resource Development: The Economics of Project Evaluation. Cambridge, MA: Harvard University Press. Graaff, J. de V. 1963 [1957]. Theoretical Welfare Economics. New York: Cambridge University Press. Green Book. See U.S. Inter-Agency River Basin Committee, Subcommittee on Benefits and Costs. Heiberg, E. R., III. 1988. “Preface.” In The Evolution of the 1936 Flood Control Act, by Joseph L. Arnold. Fort Belvoir, VA: U.S. Army Corps of Engineers, Office of History. Holmes, Beatrice Hort. 1972. A History of Federal Water Resources Programs, 1800–1960. U.S. Department of Agriculture Economic Research Service Miscellaneous Publication No. 1233, Washington, DC, June. Hufschmidt, Maynard M., and Myron B Fiering. 1966. Simulation Techniques for Design of Water-Resource Systems. Cambridge, MA: Harvard University Press. Hufschmidt, M. M., J. Krutilla, and J. Margolis, with the assistance of S. A. Marglin. 1961. Report of Panel of Consultants to the Bureau of the Budget on Standards and Criteria for Formulating and Evaluating Federal Water Resources Developments. Washington, DC. June 30.

158

INCODEL. See Interstate Commission on the Delaware River Basin. Independent Forensic Team Report. 2018. Oroville Dam Spillway Incident. January. Interstate Commission on the Delaware River Basin (INCODEL). n.d. http://archives.delaware.gov/collections/aghist/0903.sht ml. ———. 1958. The Delaware River Basin: Control and Utilization of Water Resources. Kamarck, Andrew M. 1983. Economics and the Real World. Oxford: Blackwell. Kanazawa, Mark. 1993. “Pricing Subsidies and Economic Efficiency: The U.S. Bureau of Reclamation.” Journal of Law and Economics 36 (1) (April): 205–34. Krutilla, John V., and Otto Eckstein. 1958. Multiple Purpose River Development. Baltimore: Johns Hopkins University Press. Kuznets, Simon. 1937. National Income and Capital Formation, 1919–1935. Cambridge, MA: National Bureau of Economic Research. Langton, Stuart. 1996. “An Organizational Assessment of the U.S. Army Corps of Engineers in Regard to Public Involvement Practices and Challenges.” Alternative Dispute Resolution Series, IWR Working Paper No. 9. September. Maass, Arthur. 1951. Muddy Waters: The Army Engineers and the Nation’s Rivers. Cambridge, MA: Harvard University Press. ———. 1966. “Benefit-Cost Analysis: Its Relevance to Public Expenditure Decisions.” Quarterly Journal of Economics 80:208–26. ———. 1970. “Public Investment Planning in the United States: Analysis and Critique.” Public Policy 18 (2): 211–43. ———. 1983. Congress and the Common Good. New York: Basic Books. ———. 1988. “Comment.” In The Flood Control Challenge: Past, Present, and Future, Proceedings of a National Symposium, New Orleans, Louisiana, September 26, 1986, by Howard Rosen and Martin Reuss, 144–48. Chicago: Public Works Historical Society.

159

Maass, Arthur, Maynard M. Hufschmidt, Robert Dorfman, Harold A. Thomas, Jr., Stephen A. Marglin, and Gordon Maskew Fair. 1962. Design of Water-Resource Systems: New Techniques for Relating Economic Objectives, Engineering Analysis, and Governmental Planning. Cambridge, MA: Harvard University Press. Macdonald, N., D. Chester, H. Sangster, B. Todd, and J. Hooke. 2012. “The Significance of Gilbert F. White’s 1945 Paper ‘Human Adjustment to Floods’ in the Development of Risk and Hazard Management.” Progress in Physical Geography 36 (1) (February): 125–33. DOI: 10.1177/0309133311414607. Major, David C. 1965. Decision-Making for Public Investment in Water Resources Development in the United States. Cambridge, MA: Harvard University, Graduate School for Public Administration, Harvard Water Program, Contract DA-19-016-XICWNF-62-154. ———. 1972. “Impact of Systems Techniques on the Planning Process.” Water Resources Research 8 (3) (June): 766– 68. ———. 1977. Multiobjective Water Resource Planning. Water Resources Monograph 4. Washington, DC: American Geophysical Union. ———. 2010. Harry E. Schwarz and the Development of Water Resources and Environmental Planning, U.S. Army Corps of Engineers, Institute for Water Resources. Maass-White Library Series in Water Planning and Management, vol. 3. Alexandria, VA: IWR Press. Major, David C., and Roberto L. Lenton. 1979. Applied Water Resource Systems Planning. Englewood Cliffs, NJ: Prentice-Hall. Major, David C., and Harry E. Schwarz. 1990. Large-Scale Regional Water Resources Planning: The North Atlantic Regional Study. Water Science and Technology Library, vol. 7. Dordrecht, The Netherlands: Kluwer Academic Publishers. Mansfield, Harvey C., Jr., Joseph Cooper, Susanne Hoeber Rudolph, and Samuel H. Beer. 2007. Arthur Maass— Memorial Minute. Cambridge, MA: Harvard University, Faculty of Arts and Sciences. Minute placed upon the records, May 15.

160

Marglin, Stephen A. 1967. Public Investment Criteria. Cambridge, MA: MIT Press. Merewitz, L. 1966. “Recreational Benefits of Water Resource Development.” Water Resources Research 2 (4): 625–39. Moore, Jamie W., and Dorothy P. Moore. 1989. The Army Corps of Engineers and the Evolution of Federal Flood Plain Management Policy. Special Publication No. 20. Boulder: University of Colorado, Institute of Behavioral Science, Program on Environment and Behavior. NAR. See U.S. North Atlantic Regional Water Resources Study Coordinating Committee, 1972. National Research Council, Committee to Assess the U.S. Army Corps of Engineers Water Resources Project Planning Procedures, Water Science and Technology Board, Commission on Geosciences, Environment, and Resources. 1999. New Directions in Water Resources Planning for the U.S. Army Corps of Engineers. Washington, DC: National Academy Press. Nordhaus, William. 2007. “Critical Assumptions in the Stern Review on Climate Change.” Science 317 (July 13): 201– 2. Page, Robert W. 1988. “Foreword.” In The Evolution of the 1936 Flood Control Act, by Joseph L. Arnold. Fort Belvoir, VA: U.S. Army Corps of Engineers, Office of History. Rawls, John. 1999. A Theory of Justice. Rev. ed. Cambridge, MA: Harvard University Press. Reisner, Marc. 1993. Cadillac Desert: The American West and Its Disappearing Water. Rev. ed. New York: Penguin Books. Reuss, Martin. 1988. “Introduction.” In The Flood Control Challenge: Past, Present, and Future, Proceedings of a National Symposium, New Orleans, Louisiana, September 26, 1986, by Howard Rosen and Martin Reuss. Chicago: Public Works Historical Society. ———. 1989, Water Resources People and Issues: Interview with Professor Arthur Maass. Fort Belvoir, VA: U.S. Army Corps of Engineers, Office of History. ———. 1991. Reshaping National Water Politics: The Emergence of the Water Resources Development Act of 1986. IWR Policy Study 91-PS-1. U.S. Army Institute for Water Resources. October.

161

———. 1992. “Coping with Uncertainty: Social Scientists, Engineers, and Federal Water Resources Planning.” Natural Resources Journal 32 (1) (Winter): 101–35. ———. 1993. Water Resources People and Issues: Interview with Gilbert F. White. Fort Belvoir, VA: U.S. Army Corps of Engineers, Office of History. Roosevelt, President Theodore. 1901. First Annual Message (December 3). Rosier, Paul C. 1995. “Dam Building and Treaty Breaking: The Kinzua Dam Controversy, 1936–1958.” Pennsylvania Magazine of History & Biography 119 (4) (October): 345–68. Rowley, William D. 2006. The Bureau of Reclamation: Origins and Growth to 1945. Denver: Bureau of Reclamation. S. Doc. 87-97. See U.S. President’s Water Resources Council, 1962. Schaake, John C., Jr., and David C. Major. 1972. “A Model for Estimating Regional Water Needs.” Water Resources Research 8 (June): 755–59. Schad, Theodore M. 1988. “Evolution and Future of Flood Control in the United States.” In The Flood Control Challenge: Past, Present, and Future, Proceedings of a National Symposium, New Orleans, Louisiana, September 26, 1986, by Howard Rosen and Martin Reuss, 28–45. Chicago: Public Works Historical Society. Schilling, Kyle E. n.d. “1970 NAR Demand Model as Adapted for Use in the 1975 National Assessment.” Fort Belvoir, VA: U.S. Army Corps of Engineers Institute for Water Resources. ———. 2002. “Reflections on Water Management, Academia and Disciplinary Perspectives in the United States.” Universities Council on Water Resources, Water Resources Update, no. 123 (June): 7–12. Schwarz, Harry E. 1972a. “The NAR Study: A Case Study in Systems Analysis.” Water Resources Research 8 (June): 751–54. ———. 1972b. “Water Resources Systems Symposium: Introduction.” Water Resources Research 8 (June): 750. Schwarz, Harry E., and David C. Major. 1971. “An Experience in Planning: The Systems Approach.” Water Spectrum 3 (3) (Fall): 29–34.

162

———. 1993. “On the Origins of Innovative Planning.” Water International 18 (1) (March): 71–72. Schwarz, Harry E., David C. Major, and John E. Frost Jr. 1975. “The North Atlantic Regional Water Resources Study.” In Proceedings of the Conference on Interdisciplinary Analysis of Water Resource Systems, edited by J. Ernest Flack, 245–71. New York: American Society of Civil Engineers. Shabman, Leonard. 1988. “The Benefits and Costs of Flood Control: Reflections on the Flood Control Act of 1936.” In The Flood Control Challenge: Past, Present, and Future, Proceedings of a National Symposium, New Orleans, Louisiana, September 26, 1986, by Howard Rosen and Martin Reuss, 109–23. Chicago: Public Works Historical Society. Souris-Red-Rainy River Basins Commission. 1972. Souris-Red- Rainy River Basins Comprehensive Study: Recreation and Preservation, vol. 5, app. 1. Stakhiv, Eugene Z. 1986. “Achieving Social and Environmental Objectives in Water Resources Planning: Theory and Practice.” In Social and Environmental Objectives in Water Resources Planning and Management: Proceedings of an Engineering Foundation Conference, edited by W. Viessman, Jr., and K. E. Schilling. New York: American Society of Civil Engineers. Steiner, Peter O. 1969. Public Expenditure Budgeting. Studies of Government Finance. Washington, DC: Brookings Institution. Stern, Nicholas. 2007. The Economics of Climate Change: The Stern Review. New York: Cambridge University Press. ———. 2009. The Global Deal: Climate Change and the Creation of a New Era of Progress and Prosperity. New York: Public Affairs. U.S. Army Corps of Engineers. 1959. Laws and Procedures Governing Conduct of the Civil Works Program, EM 1120-2-101. Washington, DC. U.S. Army Corps of Engineers, Baltimore District. 1963. Potomac River Basin Report. 9 vols. February. ———. 1970. Potomac River Basin Report. House Doc. 91-343. U.S. Army Corps of Engineers, Civil Works Directorate. 2010. Building Strong Collaborative Relationships for a

163 Sustainable Water Resources Future: National Report Responding to National Water Resources Challenges. August. Washington, DC. U.S. Army Corps of Engineers, New England Division. 1934. Connecticut River 308 Report Index. U.S. Army Corps of Engineers, North Atlantic Division. 1975. Northeastern United States Water Supply (NEWS) Study, Washington Metropolitan Area Water Supply Study: Report. November. New York. U.S. Army Corps of Engineers, Philadelphia District. n.d. “District History on Dams.” http://www.nap.usace.army.mil/Portals/39/docs/History/ ch02.pdf. ———. 1960. Delaware River Basin Report. December; rev. May 1961. Vol. 11 printed as H. Doc. 87-522, 87th Congress, 2nd Session, August 16, 1962. All DRBR reports are available online from ProQuest. U.S. Bureau of Reclamation. 1922. Colorado River Compact. http://www.usbr.gov/lc/region/pao/pdfiles/crcompct.pdf. ———. n.d. Grand Coulee Dam. http://www.usbr.gov/pn/grandcoulee/. ———. n.d. Hoover Dam. http://www.usbr.gov/lc/hooverdam/History/essays/bigge st.html. U.S. Bureau of the Budget. 1952. Reports and Budget Estimates Relating to Federal Programs and Projects for Conservation, Development, or Use of Water and Related Land Resources. Circular A-47. December 31. U. S. Congress. 1862a. An Act to Secure Homesteads to Actual Settlers on the Public Domain (The Homestead Act). May 20. ———. 1862b. An Act Donating Public Lands to the Several States and Territories Which May Provide Colleges for the Benefit of Agriculture and Mechanic Arts (The Morrill Act). July 2. ———. 1902a. An Act Appropriating the Receipts from the Sale and Disposal of Public Lands in Certain States and Territories to the Construction of Irrigation Works for the Reclamation of Arid Lands, P.L. 57-161, 57th Congress, 1st Session. ———. 1902b. River and Harbor Act, P.L. 57-154.

164

———. 1906. An Act to Extend the Irrigation Act to the State of Texas. June 12. ———. 1908. An Act Making Appropriations for the Current and Contingent Expenses of the Indian Department, for Fulfilling Treaty Stipulations with Various Indian Tribes, and for Other Purposes, for the Fiscal Year Ending June 30, 1909. April 30. ———. 1910. An Act to Authorize Advances to the “Reclamation Fund,” and for the Issue and Disposal of Certificates of Indebtedness in Reimbursement Therefor, and for Other Purposes. June 25. ———. 1914. An Act Extending the Period of Payment under Reclamation Projects, and for Other Purposes. August 13. ———. 1917. An Act to provide for the control of the floods of the Mississippi River and of the Sacramento River, and for other purposes. March 1. ———. 1920a. River and Harbor Act, P.L. 66-263. June 5. ———. 1920b. Federal Water Power Act, P.L. 66-280. June 10. ———. 1925. An Act Authorizing the Construction, Repair, and Preservation of Certain Public Works on Rivers and Harbors, and for Other Purposes. March 3. ———. 1926. H. Doc. 69-308, Estimate of Cost of Examinations, etc., of Streams Where Power Development Appears Feasible. April 13. ———. 1927. An Act Authorizing the Construction, Repair, and Preservation of Certain Public Works on Rivers and Harbors, and for Other Purposes, P.L. 69-560. January 21. ———. 1928. An Act to Provide for the Construction of Works for the Protection and Development of the Colorado River Basin, for the Approval of the Colorado River Compact, and for Other Purposes. ———. 1930a. H. Doc. 71-328, Tennessee River and Tributaries North Carolina, Tennessee, Alabama and Kentucky. March 24. ———. 1930b. H. Doc. 71-646, Androscoggin River, ME and NH. December 2. ———. 1933a. The Tennessee Valley Authority Act of 1933, P.L. 73-17. ———. 1933b. H. Doc. 73-103, Columbia River and Minor Tributaries.

165 ———. 1935. An Act Authorizing the Construction, Repair, and Preservation of Certain Public Works on Rivers and Harbors, and for Other Purposes. August 30. ———. 1936. Flood Control Act of 1936, P.L. 74-738. ———. 1944. Flood Control Act of 1944, P.L. 78-534. ———. 1961. Delaware River Basin Compact, P.L. 87-328. September 27. ———. 1965. Water Resources Planning Act, P.L. 89-80, 89th Congress, 1st Session, July 12. ———. 1969. National Environmental Policy Act, P.L. 91-190. ———. 1972. Federal Water Pollution Control Act Amendments of 1972, P.L. 92-500. ———. 1974. Water Resources Development Act of 1974, P.L. 93-251, 93rd Congress, 2nd Session. U.S. Environmental Protection Agency. n.d. Surf Your Watershed. https://cfpub.epa.gov/surf/huc.cfm?huc_code=01040002. U.S. Geological Survey. n.d. Office of the Delaware River Master. http://water.usgs.gov/osw/odrm/index.html. U.S. Inter-Agency River Basin Committee, Subcommittee on Benefits and Costs. 1958. Proposed Practices for Economic Analysis of River Basin Projects. Report to the Federal Inter-Agency River Basin Committee. Rev. ed. Washington, DC, originally published 1950. U.S. North Atlantic Regional Water Resources Study Coordinating Committee. 1972. North Atlantic Regional Water Resources Study, Report; Annexes and Appendices. United States Army Corps of Engineers, North Atlantic Division. U.S. President’s Water Resources Council. 1962. Policies, Standards, and Procedures in the Formulation, Evaluation, and Review of Plans for Use and Development of Water and Related Land Resources. Senate Doc. 87-97, 87th Congress, 2nd Session, May 19. U.S. Senate, Committee on Public Works. 1971. “Procedures for Evaluation of Water and Related Land Resources: Findings and Recommendations of the Special Task Force of the United States Water Resources Council,” Ser. 92-20. Washington, DC. September.

166

U.S. Water Resources Council, 1968. The Nation’s Water Resources: The First National Assessment of the Water Resources Council, parts 1–7. Washington, DC. ———. 1969. Report to the Water Resources Council by the Special Task Force: Procedures for Evaluation of Water and Related Land Resource Projects. Washington, DC. ———. 1970a. Report to the Water Resources Council by the Special Task Force: Principles for Planning Water and Land Resources. Washington, DC. (Also published with U.S. Water Resources Council [1970b, 1970c] as U.S. Senate, Committee on Public Works, Procedures for Evaluation of Water and Related Land Resource Projects: Findings and Recommendations of the Special Task Force of the United States Water Resources Council, committee print, Ser. 92-20, 92nd Congress, 1st Session, September 1971.) ———. 1970b. Report to the Water Resources Council by the Special Task Force: Standards for Planning Water and Land Resources. Washington, DC. ———. 1970c. Report to the Water Resources Council by the Special Task Force: Findings and Recommendations. Washington, DC. ———. 1970d. Report to the Water Resources Council by the Special Task Force: A Summary Analysis of Nineteen Tests of Proposed Evaluation Procedures on Selected Water and Land Resource Projects. Washington, DC. ———. 1971. “Proposed Principles and Standards for Planning Water and Related Land Resources.” Federal Register 36 (245) (December 21): 24,144–194. ———. 1973. “Water and Related Land Resources: Establishment of Principles and Standards for Planning.” Federal Register 38 (174) (September 10): 24,778–869. ———. 1979. The Nation’s Water Resources: 1975–2000: The Second National Water Assessment. Washington, DC. ———. 1983. Economic and Environmental Principles and Guidelines for Water and Related Land Resources Implementation Studies. Washington, DC. Vogel, Eve, and Alexandra Lacy. 2012. “The New Deal versus Yankee Independence: The Failure of Comprehensive Development on the Connecticut River, and Its Long- Term Consequences.” Northeastern Geographer 4 (2).

167

Wahl, Richard W. 1989. Markets for Federal Water: Subsidies, Property Rights, and the Bureau of Reclamation. Washington, DC: Resources for the Future. White, Gilbert F. 1936. “The Limit of Economic Justification for Flood Protection.” Journal of Land and Public Utility Economics 12:133–48. ———. 1945. Human Adjustment to Floods. Department of Geography Research Paper no. 29. Chicago: University of Chicago. Willingham, William F. 2006. Northwest Passages: A History of the Seattle District U.S. Army Corps of Engineers, vol. 2: 1920–1970. Seattle: U.S. Army Corps of Engineers. WRC. See U.S. Water Resources Council. Wright, James M. 2000. “The Nation’s Response to Flood Disasters: A Historical Account.” Report by the Association of State Floodplain Managers. April. https://www.floods.org/PDF/hist_fpm.pdf.

168

Index

Note: page references to photographs Green Book, see United States Inter- and maps are in italic type. Agency River Basin Committee, Subcommittee on Benefits and Costs 308 Reports, 5, 19-29, 153-4; authorization and reports, 19-22; Harvard Water Program, 111-2; see plans and planning methods, 22-8; also Maass, Arthur perspectives, 28-9; Hoover Dam, 12-3, 14 Acknowledgements, xi Human Adjustment to Floods, see Androscoggin River, 308 report, 25- White, Gilbert F. 6, 27 Interest rate, in the Green Book, 68; Caulfield, Henry, 104, 104 in Krutilla and Eckstein, 72-3; in the DRBR, 88; in the U.S. President’s Clay, Lucius D., Captain (later Water Resources Council Report, General), and the Flood Control Act 108; in Design of Water-Resource of 1936, 31 Systems, 114-5; in the Principles and Standards, 147-8 Connecticut River, controversy over proposed projects, 24-5 Kings River case study, in Muddy Copeland, Dr. Royal S., Senator, and Waters, 59 and map, 59 the Flood Control Act of 19367, 31, 33 Krutilla, John V., Multiple Purpose

River Development, 40, 71-82, 154-5; Delaware River Basin Report , 40, economic analysis and the economics 83-98, 154; planning methods, 84-91; of river basin development, 72; cost recommendations and results, 92-97; of capital, 72-3; case studies: perspectives, 97-98 economics, 73-7; case studies: Design of Water-Resource Systems, distribution of costs and gains, 77-81; see Maass, Arthur perspectives, 81-82

DRBR, see Delaware River Basin Markham, Edward M. Major Report General, and the Flood Control Act of 1936, 31, 34 Eckstein, Otto, Multiple Purpose River Development, 40, 71-82, 154-5; Maass-White Library Series in Water economic analysis and the economics Planning and Management, see of river basin development, 72; cost Pietrowski, Robert A of capital, 72-3; case studies: Maass, Arthur, Muddy Waters, 39, economics, 73-7; case studies: 51-61; Ickes Foreword to, 52-4; distribution of costs and gains, 77-81; content of Muddy Waters, 54-60; perspectives, 81-82 perspectives, 60-1; and Gilbert

White, 61-2; and Design of Water- Flood Control Act of 1936, 5, 31-8; Resource Systems, 99, 111-21, 155; flood control as a federal research of the Harvard water responsibility, 31-3; benefits and program, 112; new investment costs, 33-4; local contributions, 34-5; criteria, 113-5; conventional projects, 35; perspectives, 35-8 techniques, mathematical modeling, Flooding in Pittsburgh, 1936, 32 and the design process, 115-8; political decision making, 119; Grand Coulee Dam, 24, 25 perspectives, 120-1

169

Muddy Waters, see Maass, Arthur Principles and Standards, 100, 139- 49, 155-6; Background 139-41; Multiple Objectives, in the U.S. Authority, 141-2; Objectives and President’s Water Resources Council Criteria, 142-8; Perspectives, 148-9 Report, 104-7; in Design of Water- Resource Systems, 113-4; in the NAR Reclamation Act of 1902, 5, 7-18, study, 128-30, 129, 131; in the 153; planning methods and Principles and Standards, 139-41, procedures, 9-11; development and 144-47; design and display results of reclamation programs, 11- objectives, 145-6, 146; in 16; perspectives, 16-18 government actions, 156; see also Objectives Rivers and Harbors Act, 1925, authorizing cost estimates for Multiple Purpose River Development, multipurpose development of see Krutilla, John V., and Eckstein, streams, 19-21 Otto S. Doc. 87-97, see U.S. President’s NAR see North Atlantic Regional Water Resources Council Report Water Resources Study

NEWS Study, see Northeastern Tennessee River, 308 report as basis United States Water Supply Study for the Tennessee Valley Authority, 24 North Atlantic Regional Water Resources Study (NAR), 99-100, U.S. Bureau of the Budget, Budget 123-37; institutional arrangements, Circular A-47, 102-3 125-6; regional subdivisions for planning, 126, 127; planning period U.S. President’s Water Resources and forecast years, 127-8; new Council Report, 99, 101-9, 155; methods: multiple objectives, 128-30, objectives and project formulation, 129, 131; new methods: 104-8; perspectives, 108-9 mathematical models,130-3; needs, U.S. Water Resources Council, see sources, and devices, 133-4; forecasts Principles and Standards and recommendations, 134-6; perspectives, 136-7 United States Inter-Agency River Basin Committee, Subcommittee on Benefits Northeastern United States Water and Costs, 39, 63-70, 154-5; Supply (NEWS) Study, 136 economics of project evaluation, 65-

8; multiple objectives, 68-9; Objectives, in the reclamation perspectives, 69-70 program, 10, 17-18; in the 308 program, 28-9; in the Green Book, White, Gilbert F.. Human Adjustment 68-9; in Krutilla and Eckstein, 81-2; to Floods, 39, 41-9; context, 42-4; in the DRBR, 87-88; see also content and approach, 44-7; Multiple Objectives perspectives, 47-9; Maass’s Oroville Dam, 14-5 comments on thesis, 48; and Arthur Maass, 61-2 Pathfinder Dam 12, 12

Pietrowski, Robert A., The Maass- White Library Series in Water Planning and Management, v-vii

170

IWR Maass-White Series