Reporter NATIONAL BUREAU of ECONOMIC RESEARCH

NBER Reporter NATIONAL BUREAU OF ECONOMIC RESEARCH

A quarterly summary of NBER research 2016 Number 3

The 2016 Martin Feldstein Lecture

The Dramatic Economics of the U.S. Market for Higher Education

Caroline M. Hoxby*

We have in the United States what is arguably the world’s only true market for higher education, as opposed to systems that are largely centrally controlled or financed. This market exhibits a strong positive correlation between students’ college readiness (hereafter “CR”) and the educational resources of the institution they attend. Moreover, my research shows, the more powerful the market forces, the stronger the correlation. From my latest research, which breaks new ground with both data Caroline M. Hoxby and methods, I show the productivity of institutions across this market. Strikingly, among institutions that experience strong market forces, the productivity of a dollar of educational resources is fairly similar, even if the schools serve students with substantially different CR. On the other ALSO IN THIS ISSUE hand, among institutions that experience weak market forces, productivity is lower and more dispersed. These facts suggest that market forces are Business Cycle Impacts on Health Behaviors 7 needed to keep schools productive and to allocate resources across schools China’s Trade Policies 11 in a way that assures that the marginal return to additional resources at dif- Multinational Firms, Value Chains, ferent institutions is roughly comparable. and Vertical Integration 14 If we take the productivity results and the resources-CR correlation as Regional Evidence on Business Cycle manifestations of market forces, then it follows that a student with higher Fluctuations and Countercyclical Policy 20 CR must make more productive use of any marginal dollar of educational resources than a student who is less prepared for college. This property, NBER News 24 which economists call “single-crossing,” has long been hypothesized to be Conferences 28 a law of nature, at least in tertiary education. This is the first compelling Program and Working Group Meetings 34 evidence. Single-crossing has profound consequences for the role of higher NBER Books 35 education in income growth, a point I clarify when concluding. The U.S. market for higher education includes about 7,500 institu-

* This is an edited and annotated version of the Martin Feldstein Lecture delivered on July 27, 2016. Caroline M. Hoxby is the Scott and Donya Bommer Professor in Economics at Stanford University. She is an NBER research associate and director of the NBER Economics of Education Program.

Reporter OnLine at: www.nber.org/reporter tions. Some are publicly controlled; others are private non-profits or for-profits. Institutions NBER Reporter are largely free to decide on pricing (tuition, fees, grants), CR requirements, students, faculty, curriculum, salaries, financial aid, and how to raise money from donors if non-profit The National Bureau of Economic Research is a private, nonprofit research organization founded in 1920 and devoted to objective quantitative analysis of the or investors if for-profit. Although public American economy. Its officers and board of directors are: schools have less discretion, they still have President and Chief Executive Officer — James M. Poterba enormous autonomy by world standards, partly Controller — Kelly Horak because they are controlled by local and state Corporate Secretary — Alterra Milone governments, not a national one, and partly BOARD OF DIRECTORS because those governments recognize that pub- Chairman — Martin B. Zimmerman lic schools must be given latitude if they are Vice Chairman — Karen N. Horn to compete with local private schools. Federal Treasurer — Robert Mednick intervention is mainly in forms that students DIRECTORS AT LARGE can receive regardless of the school they attend: Peter Aldrich Mohamed El-Erian Michael H. Moskow means-tested grants, tax credits and deduc- Elizabeth E. Bailey Jacob A. Frenkel Alicia H. Munnell tions, subsidized loans. On the whole, it is best John H. Biggs Judith M. Gueron Robert T. Parry John S. Clarkeson Robert S. Hamada James M. Poterba to think of the U.S. tertiary sector as a market Don R. Conlan Peter Blair Henry John S. Reed with numerous price distortions relative to lais- Kathleen B. Cooper Karen N. Horn Marina v. N. Whitman sez-faire, but without central control.1 Charles H. Dallara Lisa Jordan Martin B. Zimmerman George C. Eads John Lipsky U.S. institutions vary enormously in selec- Jessica P. Einhorn Laurence H. Meyer tivity — that is, in the CR of their students. Selectivity is holistic but, roughly speaking, the DIRECTORS BY UNIVERSITY APPOINTMENT “most” selective institutions’ average student has Timothy Bresnahan, Stanford Benjamin Hermalin, California, Berkeley a combined (math plus verbal) SAT (or trans- Pierre-André Chiappori, Columbia Marjorie B. McElroy, Duke Alan V. Deardorff, Michigan Joel Mokyr, Northwestern lated ACT) score above 1300, the 90th per- Ray C. Fair, Yale Andrew Postlewaite, Pennsylvania centile among test-takers. (Since some students Edward Foster, Minnesota Cecilia Elena Rouse, Princeton do not take the tests, this corresponds to the John P. Gould, Chicago Richard L. Schmalensee, MIT Mark Grinblatt, California, Los Angeles David B. Yoffie, Harvard 96th percentile among all students.) “Highly” Bruce Hansen, Wisconsin-Madison or “very” selective institutions have an average DIRECTORS BY APPOINTMENT OF OTHER ORGANIZATIONS student with combined scores above the 75th Jean-Paul Chavas, Agricultural and Applied Economics Association percentile (about 1170). “Selective” (without Martin Gruber, American Finance Association a modifier) institutions ask students to sub- Arthur Kennickell, American Statistical Association mit scores, grades, and other materials and turn Jack Kleinhenz, National Association for Business Economics Robert Mednick, American Institute of Certified Public Accountants down those judged to be inadequately prepared. Alan L. Olmstead, Economic History Association Schools with combined scores above 1000 (the Peter L. Rousseau, American Economic Association 47th percentile) are at least modestly selective. Gregor W. Smith, Canadian Economics Association William Spriggs, American Federation of Labor and Non-selective schools usually only require that a Congress of Industrial Organizations student have a high school diploma or the equiv- Bart van Ark, The Conference Board alent and often have average combined scores of The NBER depends on funding from individuals, corporations, and private 800 (the 15th percentile) or below. The divide foundations to maintain its independence and its flexibility in choosing its between non-selective and modestly selective research activities. Inquiries concerning contributions may be addressed to James schools is rough but somewhere between 800 M. Poterba, President & CEO, NBER, 1050 Massachusetts Avenue, Cambridge, MA 02138-5398. All contributions to the NBER are tax deductible. and 1000. At its highly selective end, the market is The Reporter is issued for informational purposes and has not been reviewed by well integrated across geography. Schools com- the Board of Directors of the NBER. It is not copyrighted and can be freely repro- pete for students and faculty. Schools are highly duced with appropriate attribution of source. Please provide the NBER’s Public Information Department with copies of anything reproduced. informed about their applicants, and students are fairly well informed about schools. High Requests for subscriptions, changes of address, and cancellations should be sent CR students are so valued that they are admit- to Reporter, National Bureau of Economic Research, Inc., 1050 Massachusetts Avenue, Cambridge, MA 02138-5398 (please include the current mailing label), ted without regard to their ability to pay, and or by email to [email protected]. Print copies of the Reporter are only mailed to alumni-donated funds fill the gap between what subscribers in the U.S. and Canada; those in other nations may request electronic a student pays and what his education costs.2 subscriptions at www.nber.org/drsubscribe/. 2 NBER Reporter • 2016 Number 3 The high CR students who populate this part of the market make college choices that are elastic with respect to schools’ academic and other resources. Such students have strong incentives to internalize the effects of their tertiary education on their future outcomes because they pay for most of their education themselves using family funds or loans they can expect to repay with near cer- tainty. Even alumni-supported grants, which allow lower-income students to attend highly selective schools, generate strong incentives since the schools have every incentive to internalize the effects of educating one generation on their ability to finance the next. (It is worth noting that the market has not always been like this. Rather, the aforementioned features have arisen as information and mobility costs have fallen. I describe the market’s evolution in the full-length lecture.3) At the non-selective end of the market, fairly opposite conditions prevail. Geographic integration, competition, and information are poor. Students, who typically but not universally Figure 1 have low CR, seem not to look outside a set of local schools. Even within this set, they appear insensitive to differences in schools’ resources. Because many of these students rely heavily on government funds — including veterans’ benefits and loans that will predictably remain unrepaid — their incentives to internalize the effects of their education on their future outcomes is somewhat weak. I provide evidence in Figures 1 through 3. Figure 1 shows that the educational resources a student experiences tend to rise monotonically with her CR and that the most selective schools’ per-student resources are an order of magnitude greater than those of non-selective schools. Figure 2 shows that nearly all students who apply to a very selective school send most of their (approximately 10) “competing applications” to schools located in a community other than that of the first school. Hardly any applicants to a non-selective school do this, partly because they usually apply to only one school and partly because those who do send multiple applications do so within a small geographic radius. Figure 2 Figure 3 shows that, at non-selective schools, students themselves account for only about 40 percent of undergraduate-related revenue. In contrast, they account for 80 to 95 percent of such revenue at highly selective schools. Information about schools is easily characterized: 100 percent of highly selective schools and about 0 percent of non-selective schools provide comparable information to the common dataset that is used to construct college guides. The availability of information jumps dramatically between non- selective and selective schools. In short, every indicator — integration, competition, information, financing that generates incentives to internalize consequences — points to market forces being far stronger among highly selective than among non-selective schools. Measuring the productivity of institutions is crucial if we are to gain a deeper understanding of the market. But producing reli- able measures has traditionally been extremely challenging, prin- cipally because the strong positive correlation between CR and educational resources generates a formidable selection problem. Do Harvard’s graduates have such high lifetime earnings because Figure 3

NBER Reporter • 2016 Number 3 3 the school spends so much on their edu- schools of equal selectivity but can choose would otherwise be underrepresented cation or because they had such high CR only one. (From a statistical point of view, in the class being admitted.) My proce- when admitted that they would have had it is best if they make such choices “quasi- dure combines all of these student-stu- high earnings regardless of their college? randomly” — that is, based on some arbi- dent matchups using paired comparison To measure productivity, I need data and trary factor that matters to them but does statistical methods which ensure that the methods that allow me to deal with selec- not affect long-term outcomes in a sig- resulting measures rely solely on “apples- tion and estimate a school’s value-added, nificant, consistent way. Examples would to-apples” comparisons (common sup- as opposed to the raw outcomes port, in the language of econo- of its graduates. Value-added is metrics). The procedure generates the numerator of productivity. value-added measures for nearly A lesser but still-serious all institutions. problem plagues the denomina- Because there is no broadly tor of productivity: the resources applicable way to account for the devoted to students’ education, decision to attend a non-selec- not only what they spend person- tive post-secondary institution as ally but what society spends in opposed to common alternatives total, including the government such as the military and on-the- and philanthropic funds. I call job training, I do not attempt this “social investment” and it to compute the value-added of includes not only investment in the lowest selectivity schools rela- initial undergraduate schooling tive to such alternatives. Instead, but all follow-on education that I normalize their value-added to students are induced to take up. zero; it is plausibly greater than For instance, a Harvard educa- Figure 4 or less than zero. It is important tion not only uses more resources to keep this normalization in mind (funded by students, donors, and taxpay- be the school’s architectural style or the when assessing the figures that follow. ers) per student each year. Its graduates weather on the day they visited.) Other Figure 4 shows value-added versus are more likely to persist as undergradu- comparisons are between two students raw lifetime earnings for nearly all schools, ates and more likely to go to graduate who are both “on the bubble” for admis- from the non-selective to the most selec- school. Thus, its graduates’ lifetime earn- sion. Admissions staff admit one, quasi- tive. Value-added rises with selectivity, ings reflect more years, as well as more randomly; the other student ends up though not nearly as fast as raw earn- expensive years, of education. Therefore, attending a slightly less selective school. ings. This shows that much of the earn- individuals’ longitudinal educa- ings increase is due to students’ tional histories are needed. higher CR. However, Figure 5 My productivity measure- shows that the denominator of ments use such histories for vir- productivity, the lifetime educa- tually all individuals who were tional resources students experi- in the U.S. during the prime ages ence, rises greatly with selectivity. for tertiary education — 18 to Notice that these resources rise 25 — and who were age 32 in more steeply with selectivity than 2014. Measurements based on a single year of resources, shown adjacent cohorts are very similar; in Figure 1. This indicates that I use age 32 because it is the ear- higher CR students attend more liest age at which one can predict years of higher education. earnings through age 65 well.4 Thus, productivity may rise I address the selection prob- or fall with selectivity depending lem by comparing students who on the “race” between its rising attend different schools but who numerator and rising denomina- have identical college assess- Figure 5 tor. In fact, as shown by Figure ment scores — indicating extremely sim- (Admissions staff often make quasi-ran- 6, the average productivity of a dollar is ilar CR — and who apply to the same dom decisions. For instance, two stu- fairly flat among schools that are selective. schools, thus demonstrating similar taste dents may have equal academic quali- There is, however, a notable increase in the and motivation. Some of the comparisons fications, but one may be from an area level of productivity as we move from the are between students who get into two or have an extracurricular interest that non-selective to selective schools. Figure

4 NBER Reporter • 2016 Number 3 7 shows, in addition, that very selective schools grow both with the population resources; students with higher CR make schools whose students have the same CR and by enrolling students of lower CR — more productive use of any marginal tend to have similar productivity. In con- those who would not have been candi- dollar of resources. Suppose also that trast, non-selective schools differ widely dates for tertiary education in past years. (ii) students maximize their return on in productivity. education; (iii) college choices The fact that productivity is are not based on geography but, fairly flat across institutions that instead, are elastic with respect range from modestly to highly to schools’ resources and out- selective is striking because the comes; and (iv) students are least selective schools in this fully informed and not liquid- range have much lower educa- ity constrained. In this world, tional resources than the most market forces would generate an selective schools. Thus, the flat- assortatively matched allocation ness indicates that resources in which higher CR students somehow scale up with students’ would be paired with greater CR so that there would be no educational resources. Crucially, easy gains from reallocating dol- in this world, market forces lars among the modestly selec- would require that each dollar of tive and most selective schools. resources be equally productive. This is undoubtedly the most This model, although overly important result in this lecture Figure 6 simple, aligns fairly well with because, when combined with the evi- As indicated above, it is hard to say what we see in the selective tertiary sec- dence on market forces, it has profound whether the average non-selective school tor where assumptions (i) through (iv) implications. is a good investment relative to alterna- are not grossly wrong. (It is surprising The evidence in Figure 6 also indi- tives like the military or on-the-job train- that the simple model fits as well as it cates that the productivity of a dollar at ing. However, Figure 7 tells us that the does given that even the selective sector selective institutions is sufficiently posi- average is not of first-order importance abounds in price distortions, information tive that they are a good investment for anyway. The striking fact is that non-selec- gaps, and financing problems.) The simple the students who attend them. (For this tive schools differ greatly in productivity. model does not align at all with the non- not to be true, the least selective selective sector, and we should schools would need to have nota- not expect that it would. We have bly negative value-added instead seen evidence that the non-selec- of the zero to which I have nor- tive sector has little integration, malized it.) Note that Figure weak competition, poor infor- 6 does not imply that selective mation, and blunted incentives schools make maximally produc- for participants to internalize tive use of resources, just that the consequences of their educa- they make similarly efficient use tional choices. of resources. What are the broader impli- Figure 6 indicates that the cations of the evidence and average non-selective institution model? If the education pro- is less productive than selective duction function for tertiary ones. This result is concerning, education does indeed exhibit especially because enrollment in strong single-crossing, they are non-selective schools has grown profound. To make its maxi- Figure 7 substantially faster than enroll- mum contribution to economic ment in selective schools since at least This means that students choosing among growth, the higher education sector must 1970. The proximate causes of the higher non-selective schools can make mistakes allow educational resources to scale up growth rate are fairly clear. The distribu- that have very serious consequences for with CR. Yet, the extent of scaling up that tion of CR among U.S. secondary school their life outcomes. occurs in the U.S. is unique. Most coun- students is largely unchanged since 1970. Can economics make sense of all tries allocate more, but only modestly Thus, selective schools that maintain their the evidence reviewed so far? Consider more, resources to higher CR students. CR standards can only grow as fast as a simple world in which (i) there is Moreover, single-crossing implies the population grows. But non-selective single-crossing in CR and educational that if a country can make all its students

NBER Reporter • 2016 Number 3 5 attain high CR, it would be growth-max- integration, stronger competition, and the University,” in J. Brown and C. Hoxby imizing to invest more in the tertiary financing that makes students and schools eds., How the Financial Crisis and Great education of all of them. Educational internalize the consequences of their Recession Affected Higher Education, resources are investments, not consump- choices. Such policies appear to drive Chicago, Illinois: University of Chicago tion, so that, to a first-order, there is educational resources toward a relatively Press, 2015, pp. 15–41. nothing zero-sum about them. This key efficient allocation. While we have not Return to text point is often misunderstood: One might seen evidence here that all tertiary edu- 3 See the video of this year’s Feldstein incorrectly think that there is a fixed pie cation investments earn higher returns Lecture: http://nber.org/feldstein_lec- of resources so that some students must than competing uses, such as investments ture_2016/feldsteinlecture_2016.html. necessarily get fewer resources if others in physical capital, policies that generate Return to text get more. greater efficiency within higher education 4 For more on the data and methods Nevertheless, single-crossing in ter- are almost certainly pro-growth. described in this and the next few para- tiary education puts great pressure on graphs, see C. Hoxby, “Computing the the primary and secondary systems to Value-Added of American Postsecondary ensure that all students, regardless of 1 For a succinct history of the market’s Institutions,” Internal Revenue Service background, are able to attain CR. If evolution, see C. Hoxby, “The Changing Statistics of Income Division Working educational opportunities are restricted Selectivity of American Colleges,” Journal Paper, July 2015; see also C. Hoxby, to a subset of students, single-crossing of Economic Perspectives, 23(4), 2009, “The Productivity of U.S. Postsecondary is likely to reduce income mobility and pp. 95–118. This article also describes Institutions,” in C. Hoxby and K. Stange equality. It is important to note that pre- some of the market’s key internal logic. eds., Productivity in Higher Education, tertiary education may not exhibit single- Some of the historical evidence is reviewed forthcoming, Chicago, Illinois: University crossing just because tertiary education in the full-length version of this lecture. See of Chicago Press. does. Indeed, some economists hypoth- http://nber.org/feldstein_lecture_2016/ Return to text esize that, in early childhood education, feldsteinlecture_2016.html. 5 Such a phenomenon could arise simply every marginal dollar of resources is more Return to text because more advantaged parents are bet- productive for children from more disad- 2 This logic is described in detail ter substitutes for early childhood educa- vantaged backgrounds.5 in C. Hoxby (2009) and C. Hoxby, tion. In other words, the phenomenon The other key implication of the “Endowment Management Based on a would not require that a fundamental evidence and model is that economic Positive Model of the University,” NBER complementarity between aptitude and growth is likely to increase with policies Working Paper No. 18626, December educational resources reverse itself as chil- that facilitate market forces in higher 2012, and published as “Endowment dren age. education — better information, greater Management Based on a Positive Model of Return to text

6 NBER Reporter • 2016 Number 3 Research Summaries

Business Cycle Impacts on Health Behaviors

Dhaval Dave

The unemployment rate more than the economic cycle. Examining these doubled in the United States during the proximate pathways also is important for Great Recession, from 4.4 percent to 10 judging the validity of the prior, at times percent, imposing a heavy financial bur- contradictory, evidence on health. den on households. However, whether Consider, for instance, the various Dhaval Dave is the Stanton Research such economic downturns also impose studies that assess whether area-level Professor in Economics at Bentley University. a health burden is a subject of much unemployment affects obesity. It is pre- He is also a research fellow at the Institute for debate. Exploiting area-level variation in sumed that unemployment leads to a the Study of Labor and a research associate in measures of labor demand, a large litera- change in energy expenditure and/or the NBER Program on Health Economics. ture, starting with Chris Ruhm’s seminal energy intake, which in turn affects body- He holds a Ph.D. from the Graduate Center work, has explored how the business cycle weight. While some studies find that obe- of the City University of New York, and B.S. affects population health.1 sity decreases during recessions, others and M.A. degrees from Rutgers University. While it may be intuitive to suppose find the opposite, and still others find no Before joining the Bentley faculty, Dave was that population health would improve consistent effects. Many of these studies a John A. Olin postdoctoral research fellow with the macroeconomy, the evidence is use similar methods and data sets. Thus, at the Wharton School of the University of surprisingly murky. Some adverse health additional evidence bearing on the sepa- Pennsylvania. effects of economic downturns are direct rate links in the causal chain would help Dave is an applied microeconomist with and undisputed, such as increases in psy- assess the credibility of the link between primary research areas at the intersection chological stress, depression, and related labor demand and obesity. Similarly, of health and labor economics. His current illnesses, while others are indirect and less research has examined the effects of labor research examines the demand for electronic clear. Some studies indicate that health is demand on heart disease, with one pre- cigarettes, the link between welfare policy and countercyclical, with various measures of sumed causal pathway being that unem- longer-term effects on behavioral outcomes mortality, including those from cardio- ployment leads to less physical exertion for parents and their children, broader non- vascular disease and motor vehicle fatali- which leads to fewer heart attacks. economic effects of the minimum wage, and ties, declining with reduced economic In a series of papers with Gregory labor market effects of the Affordable Care activity, while others find the opposite. Colman and Inas Kelly, I examine how Act. He is also interested in the economics of None of these studies of the link labor demand affects health behaviors, in crime, and is analyzing interventions in the between labor demand and health out- order to shed light on the effects of the juvenile justice system and how they impact comes presume a direct effect. Rather, the economic cycle on health. youth recidivism and educational outcomes. presumption is that labor demand affects Dave’s research has been supported by the workers’ environment (for instance, pol- Energy Expenditure National Institutes of Health, the Agency for lution or crowding) or their behavior (for and Time Use Healthcare Research and Quality, and various instance, physical activity, diet, tobacco research foundations. He is currently serving as and alcohol use), which then affects Prior evidence on the effects of an associate editor of Economics and Human health. Health effects may take time to unemployment on energy expenditure Biology. materialize, making it challenging to iden- has been confined to recreational exercise, A native of India, Dave grew up in north- tify them empirically in the short term. and has been inconsistent. While recre- ern New Jersey. He splits his residence between Thus, it is important to examine the inter- ational exercise is certainly an important the New York metro area and Boston. In his mediate links, that is, effects on health behavioral outcome, it constitutes only spare time, he enjoys traveling and hiking, behaviors, which may respond more read- about 4 percent of total physical activ- improving his chess, and reading British mys- ily than health itself to changes in house- ity. Furthermore, in a study with Henry tery novels and books on cosmology. holds’ time and income constraints over Saffer, Michael Grossman, and Leigh

NBER Reporter • 2016 Number 3 7 Ann Leung, I find significant substitution declines during a recession, as the aver- in subgroup-specific unemployment and across recreational exercise, work-related age individual’s loss in work activity is employment rates, we find that individu- physical activity, and other modes of activ- not offset by the increases in exercise and als’ food consumption choices systemati- ity.2 Thus, it cannot be presumed that, other home-based, mostly low-MET lei- cally vary over the economic cycle, though because exercise improves health, if unem- sure activities. in ways that defy simple characterization. ployment increases exercise it must also As a validation check, we find that Specifically, we find consistent evi- improve health. It is total physical activity, these effects are concentrated among dence that a higher unemployment rate not just recreational exercise per se, which groups — particularly males who are is associated with reduced frequency of is the salient input into the individual’s low-educated or employed in physically fruit and vegetable consumption, and health production function. demanding occupations — whose employ- weak evidence of an increased frequency Colman and I study whether shifts ment was most adversely affected by the of consuming snacks and foods relatively in labor demand induce individuals to recent economic collapse. The decrease in dense in calories and fat, such as ham- become more or less physically active.3 We physical activity and exertion during an burgers and fried chicken. Together with exploit within-state variation in gender- economic downturn may partly explain the ATUS data, the results indicate that specific employment ratios matched with the positive association often found reduced employment is associated with an detailed time diary information from the between unemployment and depression, increase in time spent eating and drinking. American Time Use Survey (ATUS) over and also lends some credibility to studies While this may not necessarily reflect calo- 2003–10, a period which included the that uncover a procyclical relationship in ries consumed, it may reflect an increase in Great Recession. The ATUS is based on a mortality from cardiovascular causes. “secondary eating,” that is, snacking while national sample drawn from the Current watching television — both of which are Population Survey (CPS) and tracks all Diet and Food Intake activities our studies show tend to increase activities undertaken by the respondent during a recession. in the past 24 hours. For each activity, in The flip side to energy expenditure and One issue with the BRFSS measures of addition to duration, we measure intensity physical exertion is how a recession affects consumption of foods such as hamburgers using the Metabolic Equivalent of Task food intake, a question that I address in a and fried chicken is that they conflate con- (MET). A unit of MET is defined as the study with Kelly.5 We utilize individual- sumption of such foods prepared at home ratio of a person’s working metabolic rate level data from the Behavioral Risk Factor with those consumed in fast food restau- relative to his resting metabolic rants and other establishments. rate.4 By combining informa- Using data from the National tion on the duration of each Longitudinal Survey of Youth activity with its MET value, we (NLSY), Colman and I specif- are able to group activities and ically assess effects on fast food also to construct a standard- consumption and find that ized and consistent measure of unemployment reduces the total physical activity or exer- number of fast food meals that tion during the day. respondents consume weekly.6 Figure 1, which compares There is considerable heteroge- unadjusted means before and neity in these effects. As with after the recession began in late the results for exercise and 2007, summarizes our main physical activity, the reduc- results. We find that a reduc- tions are larger among males tion in employment increases Figure 1 and lower-educated individu- exercise, and specifically exercise als — groups which tend to be activities which are relatively less vigorous, Surveillance System (BRFSS) spanning the concentrated in boom-and-bust industries with a MET value of 4 or lower, such as 20 years of 1990–2009 and including the such as manufacturing and construction walking or golfing. The increase in exercise comparatively mild 1990–91 and 2001 and thus relatively more vulnerable to the during a recession is consistent with a reces- recessions and the severe 2007–09 down- adverse employment effects of a recession. sion-induced easing of time constraints. turn. While self-reported measures of types We also find that part of the time freed of foods consumed and frequency of con- Mechanisms from a decrease in working hours over the sumption in the BRFSS are subject to mea- and Intra-Household Spillovers recession flows into other time-intensive surement error and less than ideal, the long activities such as housework, childcare, eat- time span and the large sample sizes allow In these studies, we assess both ing and drinking, watching television, and us to provide some of the first evidence on directly and indirectly the role of vari- sleeping. Total physical exertion, however, this issue. Exploiting within-state variation ous mechanisms that may underlie the

8 NBER Reporter • 2016 Number 3 observed changes in behaviors. Own job- overs. For married or cohabiting couples, a challenge in this literature, as very large loss can affect exercise and diet by easing for instance, a spouse’s job-loss can affect a sample sizes are required to reliably detect time endowment constraints as well as respondent’s behavior due to joint house- it. If we assume that the effect is being real- through a negative income shock. Further, hold production even if their own labor ized only for individuals who lose their jobs it may lead to loss of health insurance and supply remains unchanged. Using the during a recession, then this APE translates reduced access to care, which may also ATUS, Colman and I assess the impor- into a treatment-on-the-treated (TOT) impact health behaviors. In prior work, tance of such spousal spillover effects. effect of a 27-minute increase in time spent Robert Kaestner and I find evidence ofex Due to the segregation of gen- exercising, a meaningful effect size. ante moral hazard whereby loss of cover- ders across industries and sectors and to Consider the effect on total physi- age may actually lead individuals to behave the much stronger adverse employment cal exertion for low-educated males, the more healthily, though there is also a coun- effects on male-dominated sectors dur- group most affected by the recent eco- teracting effect from reduced contact with ing the recent recession, there is substan- nomic downturn. We find that total physi- physicians due to loss of health care cov- tial within-state variation in each gender’s cal activity declines by between 5.1 and 6.3 erage, which can lead to an increase in employment ratio independent from the MET-adjusted minutes for every one per- unhealthy behaviors.7 While these are other. Exploiting this variation, we find centage point decrease in the employment- direct “internal” effects from recession- some evidence of spousal spillovers. Where to-population ratio. Again, if the effect is induced job loss, an economic downturn the husband’s and wife’s time are substitute the result of changes in behavior only of may, in addition, have external spillovers inputs — for instance, housework, child- those who become unemployed, this trans- on health behaviors, conditional on own care, and shopping — one spouse’s job loss lates into a decline in total daily physi- labor supply. Inability to find work, risk reduces the other spouse’s time use in these cal exertion of about 21 to 24 percent for of job loss, and expectations may affect activities. Thus, spousal job-loss allows the the average laid-off individual. If there are mental health and perceived health status, spouse to take over some of these activities, external spillovers of the depressed labor which may affect behaviors. and frees up the other spouse’s time which demand on other individuals, then the We assess the role of some of these then appears to be spent on personal care, TOT will be smaller. For instance, if we pathways in explaining the changes in socializing and relaxing, and sleeping. The assume that the external effects are as large observed food consumption choices. We presence of these and other external effects as the “internal” effects — so for instance, find that, to varying degrees, shifts in also underscores why it is not appropriate the recession affects as many other indi- household income, time constraints, and to use area-specific labor demand shocks as viduals as those who lose their jobs — then mental health status play important roles. instrumental variables for own labor sup- this implies a reduction in total physical With respect to reduced fast food con- ply to identify effects on health behaviors exertion of 10 to 12 percent a day. sumption associated with unemployment, and outcomes. When studying individuals’ food con- we find that this mostly reflects the greater sumption choices, we find indirect evi- availability of time for cooking rather than Average Population Effect versus dence of these external effects. That is, a less income available to purchase fast food. ‘Treatment-on-the-Treated’ higher rate of unemployment does not This is supported by data from the ATUS, just affect food consumption among those which show that the time spent on meal An important issue that arises in this who actually lose their jobs, but also among preparation is positively associated with literature relates to the interpretation of those “at risk” of becoming unemployed the unemployment rate. For these behav- effect sizes, and whether they are econom- during a recession based on their socio- iors, we do not find insurance coverage to ically significant. In most of these stud- economic characteristics. Specifically, we be an important mediator, possibly due to ies, including some of our own, area-level find a 3 to 6 percent reduction in the fre- the counteracting incentives noted above, measures of labor demand are linked to quency of consuming fruits and vegetables and partly due to the increase in pub- person-level data. What is being estimated among “at risk” individuals. These effect lic coverage buffering the drop in private is a reduced-form or average population sizes are 6 to 10 times larger than what we coverage. We also assess whether shifts in effect (APE), which conflates those who find for the average person. With respect the relative prices of food over the busi- are affected and those who are not affected to the frequency of fast food consumption, ness cycle can explain any substantial part by the recession. For instance, we find that using longitudinal data and a different of the link between unemployment and a one percentage point decrease in the identification strategy, described below, we food consumption, and generally do not employment-to-population ratio increases estimate the effect of own unemployment, find this to be the case, with the caveat that time spent exercising by 0.27 minutes per and thus a direct TOT effect for those laid measuring the relative prices of food is sub- day, an effect which is precisely estimated off. Here, we find that own unemployment ject to multiple challenges. but appears to be very small. This APE is reduces the number of fast food meals One point generally overlooked in expected to be small, however, since most respondents consume by about half a meal the literature is the possibility of exter- individuals are not affected and do not lose per week — a sizeable 29 percent decrease nal effects due to intra-household spill- their jobs during a recession. This also poses relative to the baseline mean.

NBER Reporter • 2016 Number 3 9 Longitudinal Evidence mer smokers due to stress. However, both of the recent economic downturn, interest males and females who were heavy smok- in these questions has reemerged among Colman and I provide some of the first ers at baseline tend to somewhat reduce economists and research along these lines longitudinal evidence on these questions.8 their cigarette consumption, consistent will help inform efforts to determine the We specifically consider the effects of indi- with an income effect. A longer unem- true economic costs of recessions and the viduals’ job loss on their health behav- ployment duration is also associated with appropriate policy responses. iors, using alternate measures — becoming a greater likelihood of delaying a doctor unemployed during a recession, becom- visit, which may reflect individuals delay- 1 See for instance C. Ruhm, “Health ing unemployed because of being laid off, ing or postponing utilization until they Effects of Economic Crises,” NBER becoming unemployed due to plant or have a job and health care coverage. Working Paper No. 21604, October 2015. business closure — that are plausibly exog- Prior research on the effects of unem- Return to text enous, based on data from the 1979 NLSY ployment on the body mass indes (BMI) 2 H. Saffer, D. Dave, M. Grossman, and Cohort and the Panel Study of Income has either found small effects on both L. Leung, “Racial, Ethnic, and Gender Dynamics (PSID). The use of longitudinal sides or no effects. This may reflect that Differences in Physical Activity,” Journal of information allows us to address several the true effect, if it exists, is simply too Human Capital, 7(4), 2013, pp. 378–410. lingering questions in the literature. small to measure in a population-based Return to text For instance, if recessions reduce sample. Thus, there is also some value in 3 G. Colman and D. Dave, “Exercise, smoking, cross-sectional data have a dif- being able to measure energy expenditure Physical Activity, and Exertion over ficult time determining whether this (proxied by exercise and physical activity), the Business Cycle,” Social Science & reflects light smokers quitting or heavy energy intake (proxied by consumption of Medicine, 93(c), 2013, pp. 11–20. smokers cutting back. Responses may also fast food, snacks, and other food), and the Return to text vary based on the duration of unemploy- net effect (BMI) for the same individual 4 One MET represents the energy it takes ment. Recent job losers will change their over time. Our interpretation of the joint to sit quietly, which for the average adult behavior little if they expect to be re- results of physical activity, fast food con- represents about one calorie per every 2.2 employed, whereas if they expect jobless- sumption, and BMI is that both energy pounds of body weight per hour; walking, ness to last, they will adjust to a possibly expenditure and energy intake tend to for instance, has a MET value of 2. prolonged decline in income and increase decline after a job loss, leaving observed Return to text in non-work time. Longitudinal data also BMI unchanged or only slightly higher, 5 D. Dave and I. Kelly, “How Does the allow us to control for potential compo- mostly among previously obese individu- Business Cycle Affect Eating Habits?” sitional selection arising from interstate als, even with prolonged unemployment. Social Science & Medicine, 74(2), 2012, migration that may be correlated with job pp. 254–62. prospects and health. Conclusion Return to text Consistent with our work with the 6 G. Colman and D. Dave, ATUS, we find that becoming unem- The research presented here shows “Unemployment and Health Behaviors ployed is associated with a small increase that the effects of unemployment and risk over the Business Cycle: A Longitudinal in recreational exercise but a substan- of job loss on health behaviors are complex View,” NBER Working Paper No. 20748, tial drop in total physical activity. These and multi-faceted, and cannot necessarily December 2014. effects are more pronounced with lon- be reduced to broad generalizations along Return to text ger unemployment duration. We also find the form of recessions leading individuals 7 D. Dave and R. Kaestner, “Health some suggestive evidence for other health to engage in more or less healthy lifestyles. Insurance and Ex Ante Moral Hazard: behaviors including a moderate decrease Different behaviors vary in terms of their Evidence from Medicare,” International in smoking. Prior evidence on the effect relative intensity of time- versus market- Journal of Health Care Finance and of unemployment on smoking has been purchased inputs, and thus respond dif- Economics, 9(4), 2009, pp. 367–90. mixed, and our longitudinal evidence sug- ferently to shifts in resource constraints Return to text gests that this may be due to heterogeneity over the economic cycle. While our work 8 G. Colman and D. Dave, across various margins. Among females, yields some insights on these relation- “Unemployment and Health Behaviors job loss is associated with an increase in ships, it only touches on a few behavioral over the Business Cycle: A Longitudinal the probability of being a current smoker, outcomes and processes at play linking View,” NBER Working Paper No. 20748, consistent with a decline in smoking ces- the broader macroeconomy to micro-level December 2014. sation or relapse into smoking among for- choices. In light of the far-reaching effects Return to text

10 NBER Reporter • 2016 Number 3 China’s Trade Policies

John Whalley

Recent developments in China’s trade We also find that total world produc- policy include discussions of the possibil- tion and welfare will increase under a TPP ity of joining the Trans-Pacific Partnership, regional free trade initiative and TPP will exploration of mega trade deals with a benefit member countries significantly. number of trade partners, and enactment Smaller TPP countries gain proportionally of a China-Korea free trade agreement. more than the U.S. because of their sub- My research program applies numerical stantial intra-Pacific trade. These results simulation methods to various economic appear to be reasonably robust to changes models of China and its trading partners in key model parameters, such as price John Whalley is a to analyze the potential impacts of such elasticities of demand. research associate in the changes. The work draws on the output We use our model to simulate the NBER International Trade of two research efforts by young Chinese effects of Japan joining the TPP and find and Investment program and scholars that intensively examined a broad that this would be a beneficial step for a professor of economics and range of Chinese economic topics.1 Japan and all other TPP countries, but that co-director of the Centre for this action would have negative effects on the Study of International China and the TPP China and the rest of the world. We eval- Economic Relations at Western uate the effect of China joining the TPP, University in Canada. He is The Trans-Pacific Partnership (TPP) and find that China and other TPP coun- a Distinguished Fellow at is a proposed regional arrangement among tries would all gain, while non-TPP coun- the Centre for International 13 countries; China is not a participant. tries would be hurt. In our model, the Governance Innovation (CIGI) Chunding Li and I assess the potential effects of TPP are different from those of in Canada. effects of the TPP on China and other global free trade. Global free trade ben- Whalley’s major research countries.2 We use a numerical five-coun- efits all countries, but TPP benefits only interests include computable try global general equilibrium model member countries. Moreover, the positive general equilibrium, interna- which incorporates trade costs and a mon- effects of global free trade are considerably tional trade, public finance, etary structure that incorporates inside higher than those of TPP. development economics and money and thereby allows for impacts transition economies, global- China and Mega Trade Deals ization, and climate change economics. Li, Jing Wang, and I explore potential He holds a B.A. in eco- impacts on China and other major coun- nomics from Essex University tries of mega trade deals beyond TPP.3 (1968), and an M.A. (1970), These include the Regional Comprehensive M.Phil. (1971) and a Ph.D. Economic Partnership (RCEP), China- (1973) from Yale University. Japan-South Korea Free Trade Agreement, He won the Hellmuth Prize China-TPP, and possible China-U.S. and for Achievement in Research China-India free trade agreements. We also in 2009, and was also a 2012 use numerical general equilibrium simula- Killam Prize Winner. tion methods, but introduce two important novelties. First, we divide trade costs into tariff and non-tariff barriers and again Figure1 calculate trade costs between countries empirically using gravity-model methodology. This allows exploration of free-trade agreement effects from both on trade imbalances. Trade costs are cal- tariff and non-tariff reduction. Secondly, we use an inside money structure culated using a method based on gravity to form an endogenous trade imbalance model that captures important reali- equations. Our simulation results show ties in China’s large trade imbalances. Using a 13-country Armington-type small negative effects of the TPP on China global general equilibrium model, we endogenously determine trade imbalance and other non-TPP countries. effects from the trade initiative and calibrate our model to a base case captur-

NBER Reporter • 2016 Number 3 11 ing China’s large trade surplus. We cali- This yields a measure of both absolute rable statistic for the OECD is about 6 brate the model to 2011 data and use and relative gain to China from global percent, higher than its GDP growth counterfactual simulations to explore trade policy bargaining. We calibrate rate. When we use an alternative Kalai- the effects. the model to base case data for 2008 Smorodinsky (KS) solution concept, Our simulation results show that and use a model formulation in which things are different. China’s share of almost all mega deal member countries there are heterogeneous goods across global gains is initially smaller — only will gain and nearly all mega deal non- countries. The gains from trade bar- 10.6 percent in 2010 — but grows member countries will lose. The more gaining accrue more heavily to coun- much more rapidly to 70.9 percent in non-tariff barriers are eliminated, the tries other than China when we focus 2050 and to 99.1 percent in 2100. We more significant the impacts the mega on the economic circumstances in 2008 get these results under the assumption deal will have on all countries. All than when we use data from a later year. that China maintains its growth rate at mega deals will benefit China in terms We consider the impact of differing 10.47 percent, its average in 2001-10, of welfare, trade, exports, and imports. prospective national growth estimates, and the OECD stays at a rate of 1.66 Comparatively, the RCEP and China which sharply increase China’s size rel- percent, its average in the same period. in the TPP generate the highest wel- ative to its trading partners. Our objec- These findings imply important fare outcomes in the model. The next tive is to assess how China’s gains from differences when using Nash and KS highest is a China-Japan- bargaining solution con- Korea free trade agreement cepts for numerical policy- (FTA), and then a China- based work. With asym- U.S. agreement. For the U.S., metric shifts in the utility China in the TPP gener- possibility frontier due to ates the highest welfare gain. growth, the Nash bargain- The next highest is a China- ing approach uses tangencies U.S. FTA. For the European between an implicit Cobb- Union, all China-involved Douglas preference func- mega deals except a China- tion and the utility possi- U.S. FTA generate negative bility frontier, while the KS welfare outcomes. For Japan, uses a utopia point propor- RCEP generates the high- tional to intersections with est welfare gain; the next axes. The two equilibrium highest is China in the TPP. concepts behave differently. For Korea, RCEP generates Additionally, if China joins the highest welfare gain, fol- with India and Brazil to bar- lowed by a China-Japan- Figure 2 gain jointly with the OECD, Korea FTA. For India, RCEP China’s welfare gain from bar- generates the highest welfare gain, fol- bargaining change over time, and in gaining increases by 40 percent com- lowed by a China-India FTA. particular whether they grow at a faster pared to the Nash bargaining China- rate than GDP. OECD case. We also find that if we China and Trade Our simulation results show that take account of the relative size of Policy Bargaining China’s welfare gains from bargain- China’s economy by making a purchas- ing with the OECD increase over ing power parity correction to our ini- Timing is an issue in China’s trade time if all countries keep their pres- tial calibration, China’s welfare gain bargaining, since the country is grow- ent GDP growth rates. Using the Nash would be even larger. ing faster than its OECD partners. Li Equilibrium solution concept, China’s and I use a multi-country, single-period share of global bargaining gains in China’s Service Trade numerical general equilibrium model the simulation is 41 percent in 2010, which describes the economies of 67.7 percent in 2050, and 88.7 per- Services are an increasingly impor- China and its major trading partners to cent in 2100. [See Figure 2.] This tant part of China’s trade. Chen and I examine the outcomes of trade policy shows growth in bargaining gains at discuss the country’s service trade per- bargaining solutions — bargaining over roughly the rate of increase in relative formance from 1980 to 2010, focus- tariffs and financial transfers — over GDP. China’s annual average growth ing on service subsectors in both the time.4 We compute gains relative to rate in its trade bargaining welfare gain Chinese and the world economies.5 We non-cooperative Nash equilibria (NE) is about 11 percent, just a little higher summarize and present data on the size for a range of model parameterizations. than its GDP growth rate. The compa- of China’s service trade, its growth rate,

12 NBER Reporter • 2016 Number 3 sectoral decomposition, comparative or foreign investment, but the institu- which was supported by Western advantage, and degree of openness. The tional innovation it has generated. The University, Ontario Research Fund data suggest that despite China’s high most significant changes include imple- (ORF), CIGI, and a number of Chinese growth rate, development of service mentation of a “negative list” model for universities and research institutions. trade lags behind merchandise trade. foreign investment management, more Return to text The openness index for China’s service efficient operation of new trade supervi- 2 C. Li and J. Whalley, “China and trade differs across subsectors, and the sory institutions for trade execution, the the TPP: A Numerical Simulation international competitiveness of major launch of financial reform experiments Assessment of the Effects Involved,” service subsectors remains low. on capital account convertibility and in NBER Working Paper No. 18090, May We examine China’s service trade financial services, and the cutting of red 2012, and published as “China and the in light of prospective development tape in administration. Trans-Pacific Partnership: A Numerical strategies and assess potential effects The Shanghai zone is a trial intro- Simulation Assessment of the Effects on the Chinese and global economies. duction of both floating exchange rates Involved,” The World Economy, 37(2), China has adjusted its long-term policy and capital account liberalization into 2014, pp.169–92. bias in favor of merchandise manufac- China’s macro policy mix. We examine Return to text turing and heavy industries to encour- three measures to evaluate its effects: the 3 C. Li, J. Wang, and J. Whalley, age high-tech manufacturing and ser- price spread between the Chinese yuan “Numerical General Equilibrium vices in its far-reaching 12th Five-Year in Hong Kong and mainland China, the Analysis of China’s Impacts from Plan. A series of facilitating policies on yield gaps between Renminbi accounts Possible Mega Trade Deals,” NBER taxation, finance, land use, and other in onshore and offshore markets, and Working Paper No. 20425, August elements has been launched to boost the extent to which changes in China’s 2014, and published as “Impact the service trade, which already has money supply lead to changes in foreign of Mega Trade Deals on China: A had large impacts on the country’s eco- interest rates. We find that the yield gap Computational General Equilibrium nomic growth, employment, and tech- between three-month notes onshore Analysis,” Economic Modelling, 57, nology diffusion. The potential global and offshore declined after the found- 2016, pp. 13–25. impacts of China’s service trade devel- ing of the Shanghai zone. Our results Return to text opment include changes in China’s more generally suggest that China’s cap- 4 C. Li and J. Whalley, “China’s competitiveness in offshore service out- ital controls have weakened since initia- Potential Future Growth and Gains sourcing, shifts in global FDI patterns tion of the zone. from Trade Policy Bargaining,” NBER and flows, and international migration The zone incorporates many policy Working Paper No.17826, February of educated labor. innovations such as free trade accounts 2012, and published as “China’s and a negative list for foreign invest- Potential Future Growth and Gains The Shanghai Pilot ment, as well as new trade facilities. from Trade Policy Bargaining: Some Free Trade Zone These reforms enable funds to flow in Numerical Simulation Results,” and out of China more freely, and inte- Economic Modelling, 37(C), 2014, China still maintains relatively rig- grate the Chinese financial market more pp. 65–78. orous capital controls for both state secu- into the international market. Return to text rity and policy independence reasons. 5 J. Whalley and H. Chen, “China’s The adoption of the China (Shanghai) 1 One was a “Young China Scholars Service Trade,” Journal of Economic Pilot Free Trade Zone in September Network on China’s Policy Options Surveys, 28(4), 2014, pp. 746–74. 2013 was part of an ambitious new in a Post Crisis World” sponsored Return to text round of reform, designed to liberalize by Western University, the Centre 6 D. Yao and J. Whalley, “The the capital account and facilitate trade for International Governance China (Shanghai) Pilot Free Trade in the small area of Shanghai to which Innovation (CIGI, Waterloo), and the Zone: Background, Developments the zone’s special policies apply. Daqing International Development Research and Preliminary Assessment of Initial Yao and I discuss the reasons for and Centre (IDRC, Ottawa) between 2010 Impacts,” NBER Working Paper No. objectives of China’s adoption of such and 2013. Another was a six-year 20924, February 2015, and The a zone and review its first year of opera- (October 1, 2009, to September 31, World Economy, 39(1), 2016, pp. tion.6 We find that the main impacts of 2015) project titled “The Western- 2–15. the zone has not been its trade volume CIGI-China (BRIC)-Ontario Project” Return to text

NBER Reporter • 2016 Number 3 13 Multinational Firms, Value Chains, and Vertical Integration

Laura Alfaro

In recent decades there has been a and compared that concentration with very rapid increase in flows of goods and the analogous pattern for domestic capital between countries and between firms. Maggie Chen and I find evi- Laura Alfaro is a research asso- firms, driven by technological progress dence of MNC clusters, which we label ciate in the NBER’s International and falling cross-border restrictions. The agglomeration.1 MNCs’ offshore sub- Finance and Macroeconomics rising ability to retain or outsource vari- sidiaries’ higher productivity, verti- (IFM) Program and the Warren ous production stages within firms and cally integrated production, and higher Alpert Professor of Business across country boundaries has fueled frag- knowledge- and capital-intensities all Administration in the business, mentation of production and the emer- suggest that their motives for agglomer- government, and international gence of global value chains. Cross-border ation are different from those of domes- economy unit at Harvard Business production, investment, and trade in final tic firms. We quantify patterns of spa- School, the faculty of which she and intermediate goods by multinational tial location by constructing an index of joined in 1999. For several years she corporations (MNCs) are key drivers of agglomeration that compares establish- has co-organized the NBER’s IFM this phenomenon. ments at both the industry and plant spring meeting. In a series of papers combining new levels.2 The index quantifies the extent In 2010–12, Alfaro served firm-level datasets and novel insights to which MNC establishments are as Minister of National Planning from trade and organizational econom- more or less likely to agglomerate than and Economic Policy in her native ics, my colleagues and I have examined their domestic counterparts. Dun and Costa Rica during a leave of absence the characteristics and determinants of Bradstreet’s WorldBase data enables us from Harvard. MNCs, value chains, and vertical produc- to compute this index based on plant- Alfaro’s main research inter- tion. We have found new patterns of for- level observations. The dataset includes ests are in international economics, eign direct investment (FDI), and inves- primary and secondary industries, own- in particular international capital tigated the relationships among market ership information, and plant-level flows, sovereign debt, foreign direct conditions, vertical integration, and the physical location, which can be used to investment, and multinational effects of foreign capital. calculate the distance between pairs of activity. Her recent research focuses We document the emergence of new establishments. on multinational activity, value MNC industrial clusters and their dis- Our comparative analysis generates chains, and vertical integration. tinct agglomeration patterns. The orga- a rich array of new findings. MNC Alfaro earned her Ph.D. in eco- nizational choices that firms make in headquarters are, on average, the most nomics in 1999 from the University structuring their value chains suggest that agglomerative, meaning that they are of California, Los Angeles, from complex production and process decisions most concentrated geographically. which she received a dissertation involving multiple stages explain intra- Headquarters facilities are followed by fellowship award. She received a firm activity. Our work enhances under- MNC foreign subsidiaries and domes- Licenciatura from the Pontificia standing of the sources of productivity tic plants in their degree of concentra- Universidad Católica de Chile in gains and resilience to external shocks tion. The differences in the degree of 1994, from which she graduated afforded to host countries by MNC activ- agglomeration of these three different with highest honors, and a B.A. in ity and cross-border vertical relations. types of facilities suggest that MNC economics with honors from the offshore clusters are not simply a reflec- Universidad de Costa Rica in 1992. Global Patterns of tion of domestic industrial ones. She lives in Brookline, MNC Activity Figure 1, on the following page, Massachusetts, with her husband plots the distributions of pairwise and daughter. They spend a lot of Agglomeration industries’ agglomeration densities, time in Costa Rica and in Brazil, computed using a distance of 50 km where her husband was born. One strand of my research has to define “close” establishments for examined the geographic concentra- MNC foreign subsidiaries and domestic tion of the plants operated by MNCs, plants, respectively. MNC foreign sub-

14 NBER Reporter • 2016 Number 3 Figure 1 sidiaries are more agglomerative than ditional classification of horizontal cerns about transfer pricing affect- domestic plants in capital-, skilled FDI.3 Although patterns of foreign ing values. labor-, and R&D-intensive indus- investment are recognized as com- We find that the bulk of MNC tries. In industries with greater than plex, the literature has tradition- activity occurs between rich nations, median levels of capital intensity, the ally, for analytical simplicity, dis- but some of our plant-level findings distribution of agglomeration indices tinguished between two forms of, provide a new perspective that goes is rightward-shifted for MNC for- and motivations for, locating activi- beyond this traditional wisdom. Many eign subsidiaries compared to domes- ties abroad: horizontal — replicating vertical subsidiaries, which we find tic plants. This pattern is similarly a subset of activities or processes in are larger than commonly thought, observed for industries with greater another country, and vertical — frag- are located in sectors related to higher than median levels of skilled labor menting production by function. In skill input in high-skill countries. and R&D intensities. We also evaluate general, market access models are These subsidiaries have been assumed how agglomeration economies, par- favored empirically over comparative to be market seeking.4 We term such ticularly input-output linkages, labor advantage models. Our results sug- subsidiaries intra-industry vertical and capital goods market externali- gest that data limitations have led the FDI and show them to be qualita- ties, and technology diffusion, affect prior studies to underestimate verti- tively different from vertical subsid- MNCs relative to domestic firms. We cal FDI systematically. iaries that cross two-digit industry find that MNCs’ choice of location is We use a combination of four- codes, that is, inter-industry vertical significantly influenced by technol- digit, sector-level information from FDI. Although both are vertical, intra- ogy diffusion and capital-good mar- the WorldBase data together with industry vertical FDI is more diffi- ket externalities. input-output tables to distinguish cult to explain via standard theories These findings are largely consis- between horizontal and vertical FDI. that emphasize factor cost differences tent with the MNCs’ vertically inte- We classify a horizontal subsidiary as the primary motivation for frag- grated organizational form and sub- as a plant in the same sector as its mentation. We argue that the pat- stantial investment in technology and foreign parent owner, and a verti- terns of vertical FDI and the motiva- capital goods, as well as with the cal subsidiary as a plant in sectors tion for sourcing an input within firm increasing segmentation of activities that are inputs to the foreign par- boundaries also involve the subsid- within firm boundaries and increas- ent’s product. As we do not observe iary’s intended position in the produc- ingly complex sourcing strategies. interplant trade, this approach infers tion chain. We define a variable that vertical relations from information captures the proximity of two four- Intra-Industry FDI about the goods produced in each digit sectors in a vertical production establishment and their input-output chain as the proportion of the inter- Andrew Charlton and I show relationships. While we acknowledge mediate product used directly in the that large FDI flows across rich coun- its limitations, this method yields a final good; for example, less-processed tries associated with these more com- large amount of data for many coun- materials have low proximity variables. plex strategies do not fit the tra- tries and industries and avoids con- We find proximity to be significantly

NBER Reporter • 2016 Number 3 15 higher, on average, between two verti- We find that a firm’s propensity tional arrangements to elicit the right cally related plants than between two to integrate upstream inputs depends incentives from suppliers positioned randomly selected ones. critically on the elasticity of demand at early stages. We construct a mea- for its final good and the elasticity sure of input contractibility for each Firm Boundaries and of substitution across its production industry and find that a greater degree Organizational Choices stages. When demand is elastic or of contractibility of upstream inputs inputs are not particularly substitut- increases the likelihood that a firm Value Chains able, input investments are sequen- facing high elasticity of demand will tial complements; the greater the integrate upstream inputs. These Pol Antràs, Davin Chor, Paola upstream supplier’s investments, the empirical patterns provide strong evi- Conconi, and I examine firms’ orga- greater that supplier’s marginal incen- dence that the position of inputs in nizational choices along value chains tive to undertake relationship-specific the production process and contrac- and their key decisions regarding investments. In this case, it is optimal tual frictions critically shape a firm’s which segments of the production to contract at arm’s length to incen- integration choices. process to own and which to out- tivize upstream suppliers’ investment source.5 We combine WorldBase data efforts and integrate the stages fur- Prices and Vertical Integration on firm activities across many coun- thest downstream to capture surplus. tries and industries with information When demand is inelastic or inputs The impact of market conditions, from standard input-output tables to are sufficiently substitutable, input in particular prices, on firms’ orga- study the differences between value investments are instead sequential nizational choices is a long-standing chains for integrated and question in organizational non-integrated inputs. economics. In a recent We construct an indus- paper, Conconi, Harald try-pair specific measure Fadinger, Andy Newman, of the position of differ- and I find strong support ent industries along the for the view that output value chain that summa- prices are a key determi- rizes the extent to which nant of vertical integra- a firm’s integrated inputs tion.7 This result stems tend to be more upstream from managers not only compared to its non-inte- having a stake in the orga- grated inputs. nizational goal, but also We develop a rich standing to derive pri- theoretical framework of vate, non-contractible firm behavior amenable benefits.8 Suppose that to estimation using firm- integration increases pro- level data.6 In an incom- ductivity, but at a cost; plete-contracts setup in improved coordination which the manufacture among suppliers, for of final goods entails a example, could engender large number of produc- administrative costs inde- tion stages performed in a pendent of output and predetermined order, sup- product price. A price- pliers engaged in different Figure 2 taking firm would choose stages undertake relation- to integrate only if the ben- ship-specific investments. The divi- substitutes. In this case, firms choose efits of increased profitability out- sion of surplus between the final-good to integrate relatively upstream stages weigh the cost of integrating. At low producer and each supplier is gov- and outsource downstream suppliers. prices, productivity gains from inte- erned by bargaining after inspection Figure 2 illustrates these patterns for gration are seldom sufficiently valu- of the completed stage. We allow for different quintiles of the parent firm’s able to justify the cost. As the mar- heterogeneity in the importance of elasticity of demand. ket price rises, the tradeoff resolves in inputs for production as well as in sup- In our model, greater upstream favor of more integration. pliers’ marginal cost of production at use of contract arrangements reduces Testing whether product prices different points along the value chain. a firm’s need to rely on organiza- affect organizational design requires

16 NBER Reporter • 2016 Number 3 an exogenous source of price varia- Effects of Multinational panel compilation of financial, oper- tion. Trade policy provides one such Firms Productivity, ating, and ownership information for source, since the degree of trade pro- Selection, and Reallocation companies. tection obviously affects equilibrium We find within-firm productivity prices, but it is unlikely to be influ- The impact of MNCs on their improvement and between-firm selec- enced by firms’ vertical integration host countries has been widely stud- tion to be significant but distinctly decisions. Under the most favored ied.9 Positive gains from MNC activ- different sources of gains from MNC nation (MFN) principle set out in ity are often attributed to within-firm production. We also explore the pos- the General Agreement on Tariffs productivity improvements resulting sibility of between-industry produc- and Trade (GATT), member coun- from productivity spillover from for- tivity spillover through vertical pro- tries agree not to discriminate among eign MNCs to domestic firms, or from duction linkages and find linkages to trading partners, with some excep- self-upgrading by domestic firms. But affect less and more productive firms tions. Long-term multilateral trade MNC production can also precipitate differently. The data are consistent negotiations render MFN tariffs less more intense competition in product with both between-firm selection and responsive to domestic political pres- and factor markets, as well as reallo- market reallocation. Ignoring them sure. Reverse causality is also unlikely cation of resources from domestic to could bias estimates of the origin and to be a concern in our analysis as the multinational firms and from less pro- magnitude of productivity gains from MFN tariffs that firms faced in 2004, ductive to more productive domes- MNC production.11 the year we examine, were determined tic firms. Although both channels during the Uruguay Round of multi- imply aggregate productivity gains, Foreign Ownership, Vertical lateral trade negotiations (1986–94). they represent two distinct margins. Linkages, and Resilience Combining information on firms’ Within-firm productivity improve- production activities drawn from ment operates through an “intensive Firms’ integration choices across WorldBase with input-output tables, margin” where foreign production borders can also affect a host coun- we construct firm-level vertical inte- increases the productivity of domes- try’s performance. MNCs’ ability to gration indices that measure the frac- tic firms that persist, while between- shift production back home likely tion of inputs used in the production firm selection and market realloca- results in more volatile performance of a final good that can be produced tion operates at an “extensive margin” for horizontal subsidiaries while in-house. where foreign competition induces intra-firm demand may help absorb We find that the higher the tar- the exit of the least productive firms. negative demand shocks in the host iff on imports of a given product, and The implications for domestic econo- country, resulting in more resilient thus the higher the domestic price, the mies are also sharply different: growth responses to crises. more vertically integrated are the firms or contraction of domestic industries. Chen and I examine the differen- that produce the product in that coun- My recent work with Chen dis- tial performance of establishments, try. The effect is larger precisely where entangles the two channels in deter- with particular emphasis on the role organizational decisions ought to be mining aggregate productivity gains of foreign ownership during the more responsive to import tariffs — for from MNC production.10 We inves- 2008–09 global financial crisis. This firms that serve only the domestic mar- tigate the ways market reallocation crisis was notable for its speed, sever- ket and in sectors in which tariffs have and knowledge spillovers influence ity, and international span.12 We pro - a greater impact on domestic prices. potential gains from MNC competi- vide micro-evidence on the role of We rule out several alternative mech- tion, and their relative importance, production and financial linkages in anisms that could generate a positive using a general analytical frame- influencing how foreign ownership correlation between tariffs and verti- work based on a standard model of affects an establishment’s resilience cal integration, such as competition MNC production and heterogeneous to economic crisis. We construct a and credit constraints. Our estimates firms, accounting for self-selection direct measure of production linkages imply that price changes can have large of MNCs. Our predictions of how by examining the input-output rela- effects on firm boundaries. Contrary variation in these channels influences tionship between the primary prod- to the direction of causality suggested the distribution of domestic firms ucts of subsidiaries and parent firms. by foreclosure theories, whereby ver- along the dimensions of productivity, We also consider how MNCs’ inter- tical integration raises prices as firms revenue, employment, and survival nal capital markets lower subsidiaries’ integrate with their suppliers to reduce enable us to distinguish between the dependence on host country credit competition, our analysis suggests that two channels. We empirically evalu- conditions, an advantage particularly higher prices may induce more vertical ate these predictions using Bureau van important during credit crunches. In integration. Dijk’s Orbis, a large, cross-country- order to disentangle the effect of for-

NBER Reporter • 2016 Number 3 17 eign ownership from the effects of Return to text 8 For theoretical work, see P. Legros other observable and unobservable 4 That firm-level trade data for the and A. Newman, “A Price Theory establishment and macroeconomic United States, for example, shows a of Vertical and Lateral Integration,” factors, we match MNC subsidiaries high proportion of intra-firm trade Quarterly Journal of Economics, with local plants in the same country between developed countries is further 128(2), 2013, pp. 725–70. and industry on the basis of similarity evidence of the importance of rich Return to text in characteristics, using WorldBase’s countries’ MNC vertical activity. For 9 The literature has identified data. We infer the effect of foreign further analysis of flows between rich important roles for the institutional ownership from divergences in the and poor countries, see L. Alfaro, S. environment and policy in terms of performance paths of MNC subsidiar- Kalemli-Ozcan, and V. Volosovych, FDI and capital flowing into, in par- ies and their local matches. We com- “Why Doesn’t Capital Flow from ticular, poor countries, see L. Alfaro, pare the effect of foreign ownership Rich to Poor Countries? An Empirical S. Kalemli-Ozcan, and V. Volosovych, between the non-crisis years 2005–06 Investigation,” NBER Working Paper “Why Doesn’t Capital Flow from and the crisis period, 2007–08. No. 11901, December 2005, and The Rich to Poor Countries? An Empirical Our results shed light on why for- Review of Economics and Statistics, Investigation,” The Review of eign ownership could lead to diver- 90(2), 2008, pp. 347–68; L. Alfaro, Economics and Statistics, 90(2), gent performance. On average, foreign S. Kalemli-Ozcan, and V. Volosovych, 2008, pp. 347–68; and L. Alfaro, S. subsidiaries were more resilient than “Capital Flows in a Globalized Kalemli-Ozcan, and V. Volosovych, their domestic counterparts through World: The Role of Policies and “Capital Flows in a Globalized the crisis. Establishments with stron- Institutions,” NBER Working Paper World: The Role of Policies and ger vertical production linkages No. 11696, October 2005, and in Institutions,” in S. Edwards, ed., exhibited more resilience, especially S. Edwards, ed., Capital Controls Capital Controls and Capital Flows in host countries with greater neg- and Capital Flows in Emerging in Emerging Economies: Policies, ative demand shocks. Horizontally Economies: Policies, Practices and Practices and Consequences, linked establishments, in contrast, Consequences, Chicago, Illinois: Chicago, Illinois: University of performed no better than the control University of Chicago Press, 2007, pp. Chicago Press, 2007, pp. 19–72. establishments. The role of vertical 19–72; L. Alfaro, S. Kalemli-Ozcan, Evidence has shown positive effects of production linkages and the role of and V. Volosovych, “Sovereigns, FDI conditional on local conditions, financial linkages, especially in host Upstream Capital Flows and Global in particular local financial markets, countries with worsening credit con- Imbalances,” NBER Working Paper see L. Alfaro, A. Chanda, S. Kalemli- ditions, also were related to perfor- No. 17396, September 2011, and Ozcan, and S. Sayek, “How Does mance only during the crisis period. Journal of the European Economic Foreign Direct Investment Promote Association, 12(5), 2014, pp. 1240– Economic Growth? Exploring the 84. Effects of Financial Markets on 1 L. Alfaro and M. Chen, “The Return to text Linkages,” NBER Working Paper No. Global Agglomeration of Firms,” 5 L. Alfaro, P. Antràs, D. Chor, 12522, September 2006, and Journal NBER Working Paper No. 15576, and P. Conconi, “Internalizing the of Development Economics, 91(2), December 2009, and Journal of Global Value Chains: A Firm-Level 2010, pp. 242–56; and L. Alfaro, A. International Economics, 94(2), Analysis,” NBER Working Paper No. Chanda, S. Kalemli-Ozcan, and S. 2014, pp. 263–76. 21582, September 2015. Sayek, “FDI and Economic Growth: Return to text Return to text The Role of Local Financial Markets,” 2 We extend the methodology of 6 P. Antràs and D. Chor, Journal of International Economics, G. Duranton and H. Overman, “Organizing the Global Value Chain,” 64(1), 2004, pp. 89–112. “Testing for Localization Using NBER Working Paper No. 18163, Return to text Micro Geographic Data,” Review of June 2012, and Econometrica, 10 L. Alfaro and M. Chen, Economic Studies, 2(4), 2005, pp. 81(6), 2013, pp. 2127–204. “Selection and Market Reallocation: 1077–106. Return to text Productivity Gains from Return to text 7 L. Alfaro, P. Conconi, H. Fadinger, Multinational Production,” NBER 3 L. Alfaro and A. Charlton, “Intra- and A. F. Newman, “Do Prices Working Paper No. 18207, July 2012. Industry Foreign Direct Investment,” Determine Vertical Integration?” Return to text NBER Working Paper No. 13447, NBER Working Paper No. 16118, 11 These results echo the growing September 2007, and American June 2010, and Review of Economic literature that emphasizes effects of Economic Review, 99(5), 2009, pp. Studies, 83(3), 2016, pp. 855–88. resource misallocation across estab- 2096–119. Return to text lishments. See also L. Alfaro, A.

18 NBER Reporter • 2016 Number 3 Charlton, and F. Kanczuk, “Plant- Press, 2009, pp. 243–72; and L. 12 L. Alfaro and M. Chen, “Surviving Size Distribution and Cross-Country Alfaro and A. Chari, “Deregulation, the Global Financial Crisis: Foreign Income Differences,” NBER Working Misallocation, and Size: Evidence Ownership and Establishment Paper No. 14060, June 2008, and from India,” NBER Working Paper Performance,” NBER Working Paper in J. Frankel and C. Pissarides, eds., No. 18650, December 2012, and No. 17141, June 2011, and American NBER International Seminar on Journal of Law and Economics, Economic Journal: Economic Policy, Macroeconomics 2008, Chicago, 57(4), 2014, pp. 897–936. 4(3), 2012, pp. 30–55. Illinois: University of Chicago Return to text Return to text

NBER Reporter • 2016 Number 3 19 Regional Evidence on Business Cycle Fluctuations and Countercyclical Policy

Erik Hurst

Studying cross-region variation in eco- Reserve policy may respond to aggregate nomic outcomes within a country is an shocks but not to local shocks. Ignoring increasingly common empirical research such general equilibrium factors can yield strategy in labor, public, and urban eco- estimates of local employment elasticities nomics.1 For example, local exposure to a given shock that are two to three times Erik Hurst is a to trade with China has been shown to larger than the aggregate employment elas- research associate in reduce local employment rates and wages.2 ticity to the same shock.7 Second, regional NBER’s Economic Fluctu Cross-state comparisons ations and Growth, Public variation has cannot shed Economics, and Aging been used to any light on Programs. He is the V. estimate the shocks that Duane Rath Professor incidence of affect the of Economics at the local taxes.3 entire econ- University of Chicago’s Large local omy in the Booth School of Business public work same way. and is a co-editor of programs Such shocks the Journal of Political have been are “differ- Economy. used to doc- enced away” His research spans ument the when the dif- many fields, including existence of ferent expe- empirical macroeconom- agglomera- riences of ics, labor markets, hous- tion econo- different ing markets, and consumer mies in man- regions are 4 behavior. He received ufacturing. Figure 1 compared. his Ph.D. in economics Cross- Note: Estimated wages are for white men, aged 40-44 who attended Much from the University of region variation some college, working 40 hours per week of my current Michigan in 1999 and a provides macro- research uses a B.A. in economics from economic researchers with richer informa- combination of local and aggregate data to Clarkson University in tion on economic fluctuations than aggre- learn about the drivers of aggregate busi- 1993. In 2007, he received gate time series data at the national level. ness cycles and to explore the regional con- the TIAA-CREF Paul For example, during the Great Recession sequences of aggregate government poli- Samuelson Award for in the United States, some metropolitan cies. A combination of local and aggregate the best paper on life- areas — Las Vegas, for example — experi- data, along with a structural economic long financial security enced larger declines in employment than model, often is needed to use local varia- for “Consumption vs. many others. Explaining this variation can tion to address macro questions. Expenditure” (with Mark shed light on potential causes of the aggre- Aguiar). In 2012, he was gate recession.5 Understanding the Causes awarded the Ewing Marion Drawing inferences about the aggre- of the Great Recession Kauffman Prize Medal for gate economy from data on regional vari- Distinguished Research in ation is complicated by two issues, how- Despite aggregate employment rates Entrepreneurship. ever. First, the way a regional economy falling substantially during the Great Erik currently lives in responds to a given economic shock could Recession, aggregate real wage growth dur- Chicago with his wife and be substantively different from the way ing the 2008-2010 period remained on two children. a national economy responds because of its pre-recession trend. If employment fell both factor mobility and general equi- because of a labor demand decline, the librium forces.6 Factors such as Federal employment decline during the recession

20 NBER Reporter • 2016 Number 3 should have been accompanied by a the shocks that drove both aggre- shock during the Great Recession? A decline in real wages. Many people who gate and local business cycles dur- rising degree of skill mismatch could believe that the Great Recession was ing the Great Recession period. We generate empirical patterns like those primarily caused by a lack of demand estimate the amount of wage sticki- from a labor supply shock. For exam- appeal to wages being “sticky” as the ness using local data throughout the ple, manufacturing workers may not reason a decline in real wages during analysis. Our main finding is that be able to fill jobs in the computer the Great Recession did not accompany the shock to household intertempo- sector regardless of the wage being the sharp decline in employment.8 [See ral consumption decisions — some- offered for the computer sector jobs. Figure 1.] thing akin to a traditional Keynesian In my work with Kerwin Charles In new work with Martin Beraja demand shock — explains at best only and Matt Notowidigdo, we document and Juan Ospina, I estimate the 40 percent of the employment decline the extent to which the secular decline amount of wage stickiness using cross- during 2008–10 and essentially none in manufacturing employment during state variation.9 Using a variety of of the persistently low employment the 2000s contributed to the sharp data sources, we show that decline in employment rates states with the largest rela- experienced in the U.S. tive employment declines economy between 2000 had the smallest relative and 2015.10 Employment wage increases. We con- rates for men and women struct state-level measures between the ages of 21 and of real wages by combining 54 with less than a four- state-level nominal wage year college degree fell by data from the American roughly 2 percentage points Community Survey with between 2000 and 2007 and state-level price indices then fell by an additional 7 constructed from scanner percentage points between data. From this analysis, 2007 and 2010. In 2015, we estimate that wages are employment rates for this fairly flexible. While there group were still roughly 7 is some stickiness to wages percentage points below the at the local level, real wages 2000 level. do respond to contempora- The U.S. economy lost neous local labor demand roughly 3.5 million man- shocks. [See Figure 2.] Figure 2 ufacturing jobs during To understand the the 2000–07 period and broad causes of the Great Recession, rate between 2010 and 2012. The another 2 million manufacturing jobs we construct a model of local econo- degree of wage stickiness necessary during the 2007–10 period. In 2015, mies that can aggregate to the national for demand shocks to be the pri- U.S. manufacturing employment was economy. The model allows for the mary cause of aggregate employment still roughly 5.5 million jobs below trade of goods across local econo- decline during the Great Recession the 2000 level. The decline in manu- mies, and for a national monetary is inconsistent with the flexibility of facturing employment in the U.S. dur- authority that sets a common inter- wages that we estimate from crossing the 2000s was almost three times est rate across local economies. We state variation. Explaining aggregate as large as the decline during the embed within the model four shocks: wage patterns requires an aggregate 1980s and 1990s. a shock to households’ intertempo- labor supply shock. In our analysis, A common question for struc- ral consumption decisions, a shock to regional data are essential for identi- tural explanations of the employment firms’ marginal products of labor, a fying a key parameter that helps dis- declines since early 2000 is why so shock to households’ current choices tinguish between various shocks driv- much of the decline was concentrated between market consumption and lei- ing aggregate business cycles. during the 2007–10 period. It is often sure, and a monetary policy shock. All argued that structural forces result but the monetary policy shock have Understanding the Decline in in more gradual changes while cycli- both aggregate and local components. Employment Rates, 2000–15 cal factors can result in more abrupt Using data on both aggregate changes. If manufacturing employ- and local employment rates, prices, What factors could have contrib- ment declines are contributing to low and nominal wages, we estimate uted to an aggregate labor supply employment rates in the U.S. econ-

NBER Reporter • 2016 Number 3 21 omy, why is it that the employment region variation, we show that employ- Vegas were underwater as of November rate did not fall more during the ment rates in manufacturing areas plum- 2008 while most homeowners in Dallas early 2000s, when the manufacturing meted well before the start of the Great were not. Consumption increased much declines were larger? Recession. Additionally, we show that more after QE in Dallas than in Las Charles, Notowidigdo, and I docu- employment rates were well above long- Vegas. Local refinancing booms were ment that the housing boom masked run trends in housing boom areas during correlated with local spending booms. the labor market effects of the declin- the early 2000s. At the national level, Collectively, the results show that effects ing manufacturing sector. As manufac- these two effects roughly offset. Part of monetary policy most helped regions turing declined, lower skilled men and of the masking we document occurred that needed help least. women were displaced, but the early because local areas were exposed differ- While the Federal Reserve may not 2000s housing boom actually increased entially to manufacturing declines and be independently interested in the dis- demand for construction workers, housing booms. However, we also docu- tributional aspects of monetary pol- mortgage brokers, real estate agents, ment that masking occurred at the indi- icy decisions, our paper highlights that and local services. This latter effect vidual level. If an individual who was the effectiveness of monetary policy in was due to a wealth or liquidity effect displaced from manufacturing in the stimulating consumption through home from rising house prices. As housing early 2000s lived in an area experiencing equity borrowing can be time-varying. prices increased, households increased a housing boom, that person was more Monetary policy has a lower stimulus their spending on local goods.11 Male likely to be re-employed than a displaced effect through this channel when the workers with less than a four-year col- manufacturing worker in an area with- economy is experiencing a housing bust lege degree moved sharply into con- out a housing boom. and many homeowners are underwater. struction during the early 2000s. By exploiting cross-region varia- Simultaneously, female workers with tion, we find that the labor market was U.S. Mortgage Markets and less than a four-year college degree structurally weaker prior to the Great Cross-Region Redistribution moved sharply into real estate sales, Recession than had previously been rec- mortgage brokerage, and local services. ognized, and that the housing boom The extent to which households can Using aggregate data, we docu- temporarily propped up labor market borrow to self-insure against regional ment that the decline in manufactur- statistics in aggregate data. shocks depends crucially on how the ing employment for both men and interest rate varies with regional eco- women with less than a four-year col- The Regional Effects of nomic conditions. Most economic lege degree was almost completely U.S. Monetary Policy models assume that regions within a offset by an increase in employment monetary union share a common risk- in sectors spurred on by the housing In addition to my work using adjusted interest rate. For most house- boom. In 2007, the housing market regional variation to learn about the holds, mortgages are the primary instru- collapsed and employment spurred by drivers of the aggregate macroecon- ment of borrowing. In a recent paper the housing boom collapsed with it. omy, I also have studied how aggre- with Ben Keys, Amit Seru, and Vavra, Manufacturing employment continued gate U.S. macro policies differentially I show that the assumption of constant to decline throughout the recession. affect various regions of the country. risk-adjusted mortgage rates is well sup- Thus, a sharp downward trend in man- In recent work with Beraja, Andreas ported by data for mortgages securitized ufacturing employment coupled with Fuster, and Joe Vavra, we show that the by Government Sponsored Enterprises a boom and bust in housing-related Federal Reserve’s policy of quantitative (GSEs).13 Using loan-level data, we doc- employment caused aggregate employ- easing (QE) beginning in November ument that there is no spatial variation ment to remain relatively flat during 2008 disproportionately helped regions at all in mortgage rates and fees across the early 2000s and then to fall sharply of the country that were doing relatively U.S. metropolitan areas for loans secu- at the start of the recession. well.12 We document that right after ritized by the GSEs despite there being While the patterns are present in QE began, mortgage rates fell sharply. large ex ante predictable differences in aggregate data, by exploiting regional As rates fell, a mini refinancing boom default risk across those regions. All of variation the “masking hypothesis” can occurred. The increase in refinancing our results control for loan-level observ- be illustrated clearly. Certain metro- was associated with individuals tapping ables like FICO score and loan-to-value politan areas — like Detroit — expe- into their home equity. However, the ratios. For example, GSE-secured loans rienced large manufacturing declines increase in refinancing was concentrated originated in 2007 in Las Vegas, con- and no housing boom. Other metro- in locations that had lower unemploy- ditional on borrower and loan observ- politan areas — like Las Vegas — experiment rates. The boom was much larger ables, faced the same mortgage rate as enced little manufacturing decline and in Dallas than it was in Las Vegas. The GSE-secured loans originated in 2007 a large housing boom. Exploiting cross- reason is that most homeowners in Las in Dallas, despite the Las Vegas loans,

22 NBER Reporter • 2016 Number 3 conditional on observables, having a 2 D. Autor, D. Dorn, and G. Hanson, No. 21956, February 2016. much higher predicted default probabil- “The China Syndrome: Local Labor Return to text ity, given that housing prices were falling Market Effects of Import Competition 8 See, for example, P. Krugman, to a greater extent there. in the United States,” NBER Working “Yellen, Wages, and Intellectual The patterns found within a sample Paper No. 18054, May 2012, and The Honesty,” blog post from August 25, of GSE loans differ markedly from the American Economic Review, 103(6), 2014. patterns in a sample of otherwise similar 2013, pp. 2121–68. Return to text mortgages not securitized by the GSEs. Return to text 9 M. Beraja, E. Hurst, and J. Ospina, These loans were similar in all dimen- 3 M. Harding, E. Leibtag, and M. “The Aggregate Implications of Regional sions aside from size. Loans above a cer- Lovenheim, “The Heterogeneous Business Cycles,” NBER Working Paper tain threshold are ineligible for securi- Geographic and Socioeconomic Incidence 21956, February 2016. tization by the GSEs. Mortgage rates of Cigarette Taxes: Evidence from Return to text on these jumbo loans were higher in Nielsen Homescan Data,” American 10 K. Charles, E. Hurst, and M. areas where predicted mortgage default Economic Journal: Economic Policy, Notowidigdo, “The Masking of the rates were higher. We show evidence 2012, 4(4), 2012, pp. 169–98. Decline in Manufacturing Employment that the GSEs are bound by political Return to text by the Housing Bubble,” Journal of constraints that prevent them from 4 P. Kline and E. Moretti, “Local Economic Perspectives, 30(2), 2016, charging spatially different mortgage Economic Development, Agglomeration pp. 179–200; K. Charles, E. Hurst, rates conditional on borrower and loan Economies and the Big Push: 100 and M. Notowidigdo, “Housing Booms, characteristics. Years of Evidence from the Tennessee Manufacturing Decline, and Labor If mortgage rates do not respond Valley Authority,” NBER Working Market Outcomes,” University of to local economic shocks that increase Paper No. 19293, August 2013, and Chicago Working Paper, 2016, http:// ex ante default risk, then households in The Quarterly Journal of Economics, faculty.wcas.northwestern.edu/noto/ these regions face lower borrowing costs 129(1), 2014, pp. 275–331. research/CHN_manuf_decline_hous- than they would if default risk were Return to text ing_booms_mar2016.pdf. The prior priced into interest rates. This reduc- 5 A. Mian and A. Sufi, “What two papers are based on K. Charles, tion in borrowing costs may in turn off- Explains the 2007–09 Drop in E. Hurst, and M. Notowidigdo, set some of the negative local economic Employment?” Econometrica, 82(6), “Manufacturing Decline, Housing shocks that increased default risk in the 2014, pp. 2197–223. An earlier ver- Booms and Non-Employment,” NBER first place. Conversely, those in regions sion of this paper circulated as “What Working Paper No. 18949, April 2013. with low default risk will face higher Explains High Unemployment? The Return to text borrowing costs than they would if this Aggregate Demand Channel,” NBER 11 A. Mian and A. Sufi, “House Price low default risk was priced into inter- Working Paper No. 17830, February Gains and U.S. Household Spending est rates. Thus, the constant interest rate 2012. from 2002 to 2006,” NBER Working “policy” followed by the GSEs results in Return to text Paper No. 20152, May 2014. state-contingent regional transfers. We 6 E. Nakamura and J. Steinsson, Return to text estimate that during the Great Recession “Fiscal Stimulus in a Monetary Union: 12 M. Beraja, A. Fuster, E. Hurst, and about $47 billion was transferred via Evidence from U.S. Regions,” NBER J. Vavra, “Regional Heterogeneity and the mortgage market from regions with Working Paper No. 17391, September Monetary Policy,” Federal Reserve Bank smaller employment declines (above the 2011, and American Economic of New York Staff Reports No. 731, June median) to regions with larger employ- Review, 104(3), 2014, pp. 753–92; 2015. ment declines (below the median). M. Beraja, E. Hurst, and J. Ospina, Return to text “The Aggregate Implications of Regional 13 E. Hurst, B. Keys, A. Seru, and Business Cycles,” NBER Working Paper J. Vavra, “Regional Redistribution 1 An influential example of this includes No. 21956, February 2016. Through the U.S. Mortgage Market,” O. Blachard and L. Katz, “Regional Return to text NBER Working Paper No. 21007, Evolutions,” Brookings Papers on Economic 7 M. Beraja, E. Hurst, and J. Ospina, March 2015, and forthcoming in Activity, 23(1), 1992, pp. 1–76. “The Aggregate Implications of Regional American Economic Review. Return to text Business Cycles,” NBER Working Paper Return to text

NBER Reporter • 2016 Number 3 23 NBER News

Lisa Jordan Elected to NBER Board of Directors

At its September 2016 meeting, the NBER nomics, industrial relations, and race and gender eco- Board of Directors elected Dr. Lisa Jordan to a five- nomics. She currently serves as an adviser to the AFL- year membership term as an at-large director. Jordan CIO’s Commission on Race and is a past board member is director of education and membership develop- of the Labor and Employment Relations Association. ment for the United Steelworkers of America. She Jordan graduated from the University of Notre and her staff create and coordinate the union’s educa- Dame with a Ph.D. in labor economics. She served tion program throughout North America and in the as a labor educator and assistant professor at the Jordan United Kingdom. University of Illinois and at the University of Jordan is engaged in a range of applied economic Minnesota before becoming the director of the research. Her primary areas of interest are labor eco- School of Business at Brevard College.

39th Annual NBER Summer Institute

The NBER hosted its 39th annual Summer Institute during ilar. An edited text of the lecture appears earlier in this issue of a three-week period in July 2016. There were 2,763 registered The NBER Reporter. participants, taking part in 54 distinct meetings led by 124 orga- A panel discussion on the Brexit vote in the United Kingdom nizers. About one in five participants — 554 researchers — were and its implications for global financial markets and economic attending their first Summer Institute. There were 169 gradu- growth drew a large turnout and provided participants with new ate student participants. Two thirds of the participants were not insights on the consequences of this important change in inter- NBER affiliates. national economic relations. Caroline Hoxby, who is the Scott and Donya Bommer A group of five researchers — Al Roth of Stanford, Parag Professor of Economics at Stanford University and the director A. Pathak of MIT, Atila Abdulkadiroglu of Duke University, of the NBER’s Education Program, delivered the 2016 Martin Nikhil Agarwal of MIT, and Itai Ashlagi of Stanford — pre- Feldstein Lecture on “The Dramatic Economics of the U.S. sented the 2016 Methods Lectures on “The Theory and Practice Market for Higher Education.” Her talk described the hetero- of Market Design.” They described the deferred acceptance algo- geneity in the cost of delivering undergraduate education at dif- rithm that is widely used in designing rules for matching markets ferent institutions, the differences in student test scores across and explained how a number of such markets operate in practice. schools, and the correlation between educational resources and All of the presentations — the Feldstein Lecture, the Brexit student scores. Her analysis suggested that for a broad range of Panel, and the Methods Lectures — have been videotaped and selective colleges and universities, the marginal productivity of can be accessed on the NBER website under the NBER Videos additional resources devoted to undergraduate education is sim- tab on the left side of the homepage.

24 NBER Reporter • 2016 Number 3 Jonathan Skinner Is New Director of Economics of Aging Program

Jonathan Skinner, the James O. Freedman Program: health status, saving behavior, the role of Presidential Professor of Economics at Dartmouth the public and private sectors in providing old age College and a professor of family and community security, the operation of insurance markets, and medicine at the Geisel School of Medicine, is the intergenerational linkages that affect older house- new director of the Economics of Aging Program. A holds. Skinner chairs the American Economic member of the Institute of Medicine and an NBER Association’s Committee on Government Relations, research associate since 1989, Skinner is one of the is on the steering committee of the Health and Skinner most active and influential researchers in the fields Retirement Study, and has been a member of the of health economics and the economics of aging. editorial boards of journals in both economics and Skinner received his B.A. from the University medicine. of Rochester and his Ph.D. from UCLA. He was a He is well known for his work on precaution- faculty member at the University of Virginia for 14 ary saving and social insurance, and for his contribu- years before joining the Dartmouth faculty in 1995. tions to the “Dartmouth Atlas” research program on Skinner’s research spans the many different top- the explanation of regional variations in health care ics that come together in the Economics of Aging costs in the United States.

Catherine Wolfram Is New Director of Environment and Energy Economics Program

Catherine Wolfram, the Cora Jane Flood since its inception. Professor of Business Administration at the Wolfram’s research interests span a range of University of California, Berkeley, Haas School energy and environment topics. She has studied elec- of Business, is now director of the Environment tricity markets, climate policy, U.S. air pollution reg- and Energy Economics Program. A member of the ulations, and energy demand in the developing world. Berkeley faculty since 2000, Wolfram received her A faculty director of Berkeley’s Energy Institute at A.B. from Harvard and her Ph.D. from MIT. She Haas, she is also the principal investigator on sev- Wolfram was an assistant professor at Harvard for several eral large grants, including one from the Alfred P. years before moving to Berkeley. Sloan Foundation to support research on energy An NBER research associate since 2006, efficiency and one from the U.K.’s Department for she has been a member of the Environment and International Development to fund research on Energy Economics Program steering committee energy in the developing world.

China Working Group Director Shang-Jin Wei Returns

After spending over two years based in Manila Wei is the N.T. Wang Professor of Chinese as chief economist of the Asian Development Bank, Business and Economy and professor of finance and Shang-Jin Wei has returned to Columbia University economics at the Columbia Graduate School of and resumed his role as director of the NBER’s China Business. He is a research associate in the NBER’s Working Group. While Wei was on leave, Hanming Development Economics, International Finance Fang of the University of Pennsylvania served as and Macroeconomics, and International Trade and Wei interim director and provided outstanding leadership. Investment programs.

NBER Reporter • 2016 Number 3 25 Liran Einav Is New Director of Industrial Organization Program

Liran Einav is a professor of economics at credit and health insurance markets. He has devel- Stanford, where he has taught since 2002. He oped empirical models of both insurance demand received his B.A. from Tel Aviv University and his and insurance pricing, and explored the implications Ph.D. from Harvard. of both adverse selection and moral hazard. He has An NBER research associate since 2008, he has also studied the economics of online markets, such as been a co-director of the Bureau’s Working Group on peer-to-peer internet markets, and the economic per- Insurance Economics since 2012. formance of the motion picture industry. Einav Einav has made important contributions to a He is widely sought-after for his editorial exper- range of topics in industrial organization and applied tise: In June 2017 he will complete a term as co-edi- microeconomics more generally, including the role of tor of Econometrica, and he will become a co-editor competition and imperfect information in consumer of the American Economic Review in 2018.

Ben Handel and Motohiro Yogo Are New CoDirectors of Insurance Working Group

Benjamin Handel, an associate professor of eco- until 2011 and has been a research associate since nomics at the University of California, Berkeley, 2015. and Motohiro Yogo, a professor of economics at Handel’s work is primarily concerned with the Princeton, have become co-directors of the Insurance economic analysis of health insurance markets. He Working Group. has studied the role of adverse selection, the nature Handel, who has taught at Berkeley since 2010, of competition between insurance providers, and the received his A.B. from Princeton and his Ph.D. from role of behavioral economics in explaining insur- Handel Northwestern. He was named an NBER faculty ance plan choice. His 2015 Econometrica paper with research fellow in industrial organization in 2011 and Igal Hendel and Michael Whinston on “Equilibria became a research associate in 2016. in Health Exchanges” was awarded the Econometric Yogo received his A.B. from Princeton in 2000 Society’s Frisch Medal. Handel is a recent recipient of and his Ph.D. from Harvard in 2004. He became a an NSF CAREER Award. junior faculty member at the Wharton School at the Yogo’s research focuses on insurance markets, University of Pennsylvania and, after a period as a financial economics, and econometrics. He has been research economist at the Federal Reserve Bank of particularly interested in how regulatory policy affects Minneapolis, joined the Princeton faculty in 2015. insurance companies and in the supply of and demand Yogo was an NBER faculty research fellow from 2006 for insurance products. Yogo

Stephen Redding Is New Director of International Trade and Investment Program

Stephen Redding is the Harold T. Shapiro ’64 demic career. He is an associate editor of Econometrica. Professor of Economics in the department of econom- Redding has been an active member of the ics and the Woodrow Wilson School of Public and International Trade and Investment Program since International Affairs at Princeton University, where his appointment as an NBER research associate in he has taught since 2010. He was a member of the fac- 2011. He has made important research contribu- ulty at the London School of Economics and at Yale tions to a wide range of questions in international University before moving to Princeton. trade, economic geography, and growth and produc- Redding Redding received his undergraduate and graduate tivity analysis. His work spans traditional questions training at Oxford University and was a research econ- in international trade as well as emerging issues in omist at the Bank of England before beginning his aca- spatial economics.

26 NBER Reporter • 2016 Number 3 Jeffrey Brown, Courtney Coile, and James Choi Lead Retirement Research Center

Jeffrey Brown, the Josef and Margot Science Center at Wellesley College. Lakonishok Professor and dean of the busi- Choi has done pioneering work on how ness school at the University of Illinois, defaults and other plan design features Urbana-Champaign, is the new direc- affect retirement saving. Coile has made tor of the NBER Retirement Research important contributions to understand- Center. Brown’s work has touched on ing the interplay between labor market many issues related to retirement secu- conditions, Social Security, and retire- Brown Choi rity, saving, and the provision of lifetime ment well-being. She is the editor of The income. NBER Bulletin on Aging and Health. Assistant directors of the center are Jeffrey Liebman, the Malcolm Wiener James Choi, professor of finance at Yale Professor of Public Policy at the Harvard University’s School of Management, and Kennedy School, is senior adviser to the Courtney Coile, the professor of eco- Center. He previously served as associate nomics and director of the Knapp Social director.

Coile Liebman

David Autor and Nicole Maestas Lead Disability Research Center

David Autor, the Ford Professor of associate professor of health care policy Economics and associate economics depart- at Harvard Medical School, is now asso- ment head at MIT, is the new director of the ciate director. She previously was a senior NBER Disability Research Center. Autor economist at the RAND Corporation, and has written seminal papers on disability has done important work on labor mar- policy and previously served as the associ- ket effects of disability insurance. Jeffery ate director of the Center. Nicole Maestas, Liebman is senior adviser to the Center. Autor Maestas

NBER Reporter • 2016 Number 3 27 Conferences

Trans-Atlantic Public Economics Seminar on Social Insurance

The NBER’s bi-annual Trans-Atlantic Public Economics Seminar met in Mannheim, Germany, on June 13–15. The meeting was organized by Research Associate Roger Gordon of the University of California, San Diego, Andreas Peichl of the University of Mannheim, and NBER President James Poterba of MIT. These researchers’ papers were presented and discussed:

• Anna Raute, University of Mannheim, “Can Financial Incentives Reduce the Baby Gap? Evidence from a Reform in Maternity Leave Benefits”

• Itzik Fadlon, University of California, San Diego, and NBER, and Torben Heien Nielsen, University of Copenhagen, “Household Labor Supply Responses to Severe Health Shocks and the Gains from Social Insurance”

• Stefan Pichler, ETH Zurich, and Nicolas R. Ziebarth, Cornell University, “The Pros and Cons of Sick Pay Schemes: Testing for Contagious Presenteeism and Shirking Behavior”

• Mathias Dolls, Philipp Doerrenberg, Andreas Peichl, and Holger Stichnoth, ZEW (Mannheim), “Labor Market and Savings Responses to Pension Reforms — Quasi-Experimental Evidence from Germany”

• Stuart Adam, David Phillips, and Barra Roantree, Institute for Fiscal Studies (London), “35 Years of Reforms: A Panel Analysis of the Incidence of, and Employee and Employer Responses to, Social Security Contributions in the U.K.”

• Bruce D. Meyer, University of Chicago and NBER, and Wallace K.C. Mok, Chinese University of Hong Kong, “Disability, Earnings, Income, and Consumption” (NBER Working Paper No. 18869)

• Liran Einav, Stanford University and NBER; Amy Finkelstein, MIT and NBER; and Paul Schrimpf, University of British Columbia, “Bunching at the Kink: Implications for Spending Responses to Health Insurance Contracts” (NBER Working Paper No. 22369)

• Jeffrey Clemens and Michael Wither, University of California, San Diego, “Is Tinkering with Safety Net Programs Harmful to Beneficiaries? Evidence from the Medicaid Notch and the Minimum Wage”

• Torben Fischer, Markus Frölich, and Andreas Landmann, University of Mannheim, “Adverse Selection in Micro- Health Insurance Markets: Evidence from a Randomized Control Trial in Pakistan”

• Andreas Lichter, IZA Institute for the Study of Labor (Bonn), “Benefit Duration and Job Search Effort: Evidence from a Natural Experiment”

• Thomas Le Barbanchon, Bocconi University (Milan), “Optimal Partial Unemployment Insurance: Evidence from Bunching in the U.S.”

• Ioana Marinescu, University of Chicago and NBER, “The General Equilibrium Impacts of Unemployment Insurance: Evidence from a Large Online Job Board” (NBER Working Paper No. 22447)

Summaries of these papers are at: http://www.nber.org/confer/2016/TAPES16/summary.html

28 NBER Reporter • 2016 Number 3 East Asian Seminar on Economics

The NBER, the Australian National University, the Peking University China Center for Economic Research, the Chung- Hua Institution for Economic Research (Taipei), the Hong Kong University of Science and Technology, the Korea Development Institute, the National University of Singapore, the Tokyo Center for Economic Research, and Tsinghua University (Beijing) jointly sponsored the NBER’s 27th Annual East Asian Seminar on Economics. It took place in Singapore on June 23–24. Research Associates Takatoshi Ito of Columbia University and Andrew K. Rose of the University of California, Berkeley, organized the conference. These researchers’ papers were presented and discussed:

• Benjamin J. Keys and Amit Seru, University of Chicago and NBER; Tomasz Piskorski, Columbia University and NBER; and Vincent Yao, Georgia State University, “Mortgage Rates, Household Balance Sheets, and the Real Economy” (NBER Working Paper No. 20561)

• Mathias Hoffmann, University of Zurich, and Iryna S. Stewen, University of Mainz, “Holes in the Dike: The Global Savings Glut, U.S. House Prices and the Long Shadow of Banking Deregulation”

• Sumit Agarwal, Georgetown University, and Cristian Badarinza and Wenlan Qian, National University of Singapore, “The Effectiveness of Housing Collateral Tightening Policy”

• Sumit Agarwal; Gene Amromin, Federal Reserve Bank of Chicago; Souphala Chomsisengphet, Office of the Comptroller of the Currency; Tomasz Piskorski; Amit Seru; and Vincent Yao, “Mortgage Refinancing, Consumer Spending, and Competition: Evidence from the Home Affordable Refinancing Program” (NBER Working Paper No. 21512)

• Sumit Agarwal; Changcheng Song, National University of Singapore; and Vincent Yao, “Banking Competition and Shrouded Attributes: Evidence from the U.S. Mortgage Market”

• Brent Ambrose and Jiro Yoshida, Pennsylvania State University, and N. Edward Coulson, University of Nevada, Las Vegas, “Inflation Rates Are Very Different When Housing Rents Are Accurately Measured”

• Hanming Fang, University of Pennsylvania and NBER; You Suk Kim, Federal Reserve Board; and Wenli Li, Federal Reserve Bank of Philadelphia, “The Dynamics of Adjustable-Rate Subprime Mortgage Default: A Structural Estimation” (NBER Working Paper No. 21810)

• Daisuke Miyakawa and Iichiro Uesugi, Hitotsubashi University (Tokyo), and Chihiro Shimizu, National University of Singapore, “Geography and Realty Prices: Evidence from International Transaction-Level Data”

• Peter Chinloy, American University, and Man Cho and Inho Song, Korea Development Institute, “The Asset Price of a House”

• Bo Zhao, Peking University (Beijing), “Too Poor to Retire? Housing Prices and Retirement”

• Wen-Chieh Wu, National Chengchi University (Taipei), and Yu-Chun Ma and Jiann-Chyuan Wang, Chung-Hua Institution for Economic Research (Taipei), “Childhood Housing Environment and Young Adulthood Health Status”

Summaries of these papers are at: http://www.nber.org/confer/2016/EASE16/summary.html

NBER Reporter • 2016 Number 3 29 International Seminar on Macroeconomics

The NBER’s 39th International Seminar on Macroeconomics took place in Sofia, Bulgaria on June 24–25. The seminar was organized by Research Associates Richard H. Clarida of Columbia University, Jeffrey Frankel of Harvard University, and Hélène Rey and Lucrezia Reichlin of London Business School. These researchers’ papers were presented and discussed:

• Enrique Alberola, Bank for International Settlements, and Gianluca Benigno, London School of Economics, “Revisiting the Commodity Curse: A Financial Perspective”

• Yusuf Soner Baskaya and Mehmet Fatih Ulu, Central Bank of the Republic of Turkey; Julian di Giovanni and José- Luis Peydró, Pompeu Fabra University (Barcelona); and Şebnem Kalemli-Özcam, University of Maryland and NBER, “International Spillovers and Local Credit Cycles”

• Luca Dedola and Livio Stracca, European Central Bank, and Giulia Rivolta, University of Brescia, “If the Fed Sneezes, Who Catches a Cold?”

• Vania Stavrakeva, London Business School, “Optimal Capital Flows in a Model with Financial Frictions and Imperfectly Competitive Banking Sector”

• Luis Céspedes, Adolfo Ibáñez University (Chile); Roberto Chang, Rutgers University and NBER; and Andrés Velasco, Columbia University and NBER, “Financial Intermediation, Exchange Rates, and Unconventional Policy in an Open Economy” (NBER Working Paper No. 18431)

• Kathryn Holston and Thomas Laubach, Federal Reserve Board, and John Williams, Federal Reserve Bank of San Francisco, “Measuring the Natural Rate of Interest: International Trends and Determinants”

• Andrés Fernández, Inter-American Development Bank, and Stephanie Schmitt-Grohé and Martín Uribe, Columbia University and NBER, “World Shocks, World Prices, and Business Cycles: An Empirical Investigation”

• Karen K. Lewis, University of Pennsylvania and NBER, and Edith Liu, Federal Reserve Board, “Disaster Risk and Asset Returns: An International Perspective”

Summaries of these papers are at: http://www.nber.org/confer/2016/ISOM16/summary.html

The Economics of Asset Accumulation and Poverty Traps

An NBER conference, “The Economics of Asset Accumulation and Poverty Traps,” took place in Washington, D.C., on June 28–29. The meeting was organized by Christopher B. Barrett of Cornell University, Research Associate Michael Carter of the University of California, Davis, and NBERR Board Member Jean-Paul Chavas of the University of Wisconsin-Madison. These researchers’ papers were presented and discussed:

• Duncan Thomasand Elizabeth Frankenberg, Duke University and NBER, “Shocks and Nutrition, Health, and Human Capital”

• Emma Boswell Dean and Heather Schofield, University of Pennsylvania, and Frank Schilbach, MIT and NBER, “Poverty and Cognitive Function”

• Jonathan de Quidt, Institute for International Economic Studies (Stockholm), and Johannes Haushofer, Princeton University, “Depression for Economists”

30 NBER Reporter • 2016 Number 3 • Tr av is Ly bb er t, University of California, Davis, and Bruce Wydick, University of San Francisco, “Poverty, Aspirations, and the Economics of Hope: A Framework for Study with Preliminary Results from the Oaxaca Hope Project”

• Francisco J. Buera, Federal Reserve Bank of Chicago; Joseph P. Kaboski, University of Notre Dame and NBER; and Yongseok Shin, Washington University in St. Louis and NBER, “Taking Stock of the Evidence on Micro-Financial Interventions” (NBER Working Paper No. 22674)

• Michael Carter; Munenobu Ikegami, International Livestock Research Institute (Nairobi); and Christopher B. Barrett, “Poverty Traps and the Social Protection Paradox”

• Paulo Santos, Monash University (Melbourne), and Christopher B. Barrett, “Heterogeneous Wealth Dynamics: On the Roles of Risk and Ability” (NBER Working Paper No. 22626)

• Jean-Paul Chavas, “Agro-Ecosystem Productivity and the Dynamic Response to Shocks” (NBER Working Paper No. 22624)

• Karen Macours, Paris School of Economics, and Renos Vakis, The World Bank, “Medium-Term Impacts of a Productive Safety Net on Aspirations and Human Capital Investments”

• Norbert Schady, Caridade Araujo, and Mariano Bosch, Inter-American Development Bank, “Cash Transfers and Poverty Traps: A Tale of Two Generations”

• Oriana Bandiera, Robin Burgess, and Munshi Sulaiman, London School of Economics; Narayan Das, BRAC University (Bangladesh); Selim Gulesci, Bocconi University (Milan); and Imran Rasul, University College London, “Labor Markets and Poverty in Village Economies”

Summaries of these papers are at: http://www.nber.org/confer/2016/PTs16/summary.html

Japan Project

The NBER held a meeting on the Japanese economy in Tokyo on August 1. The seminar was organized by Shiro Armstrong of the Australian National University, Research Associate Charles Horioka of the Asian Growth Research Institute (Kitakyushu), Research Associate Takeo Hoshi of Stanford University, Tsutomu Watanabe of the University of Tokyo, and Research Associate David Weinstein of Columbia University. These researchers’ papers were presented and discussed:

• Robert Dekle, University of Southern California; Atsushi Kawakami, Teikyo University (Tokyo); Nobuhiro Kiyotaki, Princeton University and NBER; and Tsutomu Miyagawa, Gakushuin University (Tokyo), “Product Dynamics and Aggregate Shocks: Evidence from Japanese Product and Firm Level Data”

• David Cashin, Federal Reserve Board, and Takashi Unayama, Hitotsubashi University (Tokyo), “The Impact of a Permanent Income Shock on Consumption: Evidence from Japan’s 2014 VAT Increase”

• Jess Diamond, Hitotsubashi University (Tokyo), and Kota Watanabe and Tsutomu Watanabe, University of Tokyo, “The Formation of Consumer Inflation Expectations: Evidence from Japan’s Deflation Experience”

• Mark Koyama, George Mason University; Chiaki Moriguchi, Hitotsubashi University (Tokyo); and Tuan-Hwee Sng, National University of Singapore, “Geopolitics and Asia’s Little Divergence: A Comparative Analysis of State Building in China and Japan After 1850”

NBER Reporter • 2016 Number 3 31 • Gauti Eggertsson, Brown University and NBER; Neil Mehrotra and Sanjay Singh, Brown University; and Lawrence Summers, Harvard University and NBER, “A Contagious Malady? Open Economy Dimensions of Secular Stagnation” (NBER Working Paper No. 22299)

• Hiroshi Fujiki, Chuo University (Tokyo), and Hajime Tomur a, Waseda University (Tokyo), “Fiscal Cost to Exit Quantitative Easing: The Case of Japan”

Summaries of these papers are at: http://www.nber.org/confer/2016/JPMs16/summary.html

Economics of Commodity Markets

The NBER, supported by the Bank of Canada, held its annual Universities Research Conference, “The Economics of Commodity Markets,” in Cambridge on September 16–17. Jing Yang of the Bank of Canada and Research Associates Kenneth Singleton of Stanford University and Wei Xiong of Princeton University organized the meeting. These researchers’ papers were presented and discussed:

• Farid Farrokhi, Purdue University, “Global Sourcing in Oil Markets”

• Frank Wolak, Stanford University and NBER, “Assessing the Impact of the Diffusion of Shale Oil and Gas Technology on the Global Coal Market”

• Martijn Boons and Melissa Porras Prado, Nova School of Business and Economics (Lisbon), “Basis-Momentum in the Futures Curve and Volatility Risk”

• Ignacia Mercadal, University of Chicago, “Dynamic Competition and Arbitrage in Electricity Markets: The Role of Financial Players”

• Reinhard Ellwanger, Bank of Canada, “Driven by Fear? The Tail Risk Premium in the Crude Oil Futures Market”

• Daniele Bianchi, University of Warwick, and Jacopo Piana, Cass Business School (London), “Expected Spot Prices and the Dynamics of Commodity Risk Premia”

• Steffen Hitzemann, Ohio State University, “Macroeconomic Fluctuations, Oil Supply Shocks, and Equilibrium Oil Futures Prices”

• Michael Brandt, Duke University and NBER, and Lin Gao, University of Luxembourg, “Macro Fundamentals or Geopolitical Events? A Textual Analysis of News Events for Crude Oil”

• Jorge Fornero and Markus Kirchner, Central Bank of Chile, “Learning About Commodity Cycles and Saving- Investment Dynamics in a Commodity-Exporting Economy”

• Niko Jaakkola, Ifo Institute for Economic Research (Munich); Daniel Spiro, University of Oslo; and Arthur van Benthem, University of Pennsylvania and NBER, “Finders, Keepers?” (NBER Working Paper No. 22421)

Summaries of these papers are at: http://www.nber.org/confer/2016/CEMf16/summary.html

32 NBER Reporter • 2016 Number 3 Energy Policy Tradeoffs between Economic Efficiency and Distributional Equity

An NBER conference, “Energy Policy Tradeoffs between Economic Efficiency and Distributional Equity,” took place in Durham, North Carolina, on September 16–17. Faculty Research Fellow Tatyana Deryugina and Research Associate Don Fullerton, both of the University of Illinois at Urbana-Champaign, and Research Associate William A. Pizer of Duke University organized the meeting. These researchers’ papers were presented and discussed:

• Chris Bruegge, Stanford University; Tatyana Deryugina; and Erica Myers, University of Illinois at Urbana-Champaign, “The Distributional Effects of Building Codes”

• Sébastien Houde, University of Maryland, and Joseph Aldy, Harvard University and NBER, “Efficiency and Distributional Consequences of Heterogeneous Behavioral Responses to Energy Fiscal Policies”

• Mar Reguant, Northwestern University and NBER, “The Distributional Impacts of Large-Scale Renewable Policies”

• Lucas Davis, University of California, Berkeley, and NBER, and Christopher Knittel, MIT and NBER, “Are Fuel Economy Standards Regressive?”

• Stephen Holland, University of North Carolina at Greensboro and NBER; Erin Mansur, Dartmouth College and NBER; Nicholas Muller, Middlebury College and NBER; and Andrew Yates, University of North Carolina at Chapel Hill, “Distributional Effects of Air Pollution from Electric Vehicle Adoption”

• Arik Levinson, Georgetown University and NBER, “Are Energy Efficiency Standards Less Regressive Than Energy Taxes?”

• Julie Anne Cronin, U.S. Treasury Department; Don Fullerton; and Steven Sexton, Duke University, “Carbon Tax Rebates and Redistribution”

• Carolyn Fischer, Resources for the Future, and William A. Pizer, “Equity versus Efficiency in Energy Regulation”

Summaries of these papers are at: http://www.nber.org/confer/2016/EPTf16/summary.html

Tax Policy and the Economy

An NBER conference, “Tax Policy and the Economy,” took place in Washington, D.C., on September 22. Research Associate Robert Moffitt of Johns Hopkins University organized the meeting. These researchers’ papers were presented and discussed:

• Alan J. Auerbach, University of California, Berkeley, and NBER; Laurence J. Kotlikoff, Boston University and NBER; Darryl R. Koehler, Economic Security Planning; and Manni Yu, Boston University, “Is Uncle Sam Inducing the Elderly to Retire?”

• Gizem Kosar, Federal Reserve Bank of New York, and Robert Moffitt, “Trends in Cumulative Marginal Tax Rates Facing Low-Income Families, 1997–2007”

• Emmanuel Saez, University of California, Berkeley, and NBER, “Taxing the Rich More: Preliminary Evidence from the 2013 Tax Increase”

NBER Reporter • 2016 Number 3 33 • Louis Kaplow, Harvard University and NBER, “A Distribution-Neutral Perspective on Tax Expenditure Limitations”

• Conor J. Clarke, Yale University, and Wojciech Kopczuk, Columbia University and NBER, “Business Income and Business Taxation in the United States since the 1950s”

Summaries of these papers are at: http://www.nber.org/confer/2016/TPE16/summary.html

Program and Working Group Meetings

Economic Fluctuations and Growth

The NBER’s Program on Economic Fluctuations and Growth met in Cambridge on July 16. Research Associates Fernando Alvarez of the University of Chicago and Emi Nakamura of Columbia University organized the meeting. These researchers’ papers were presented and discussed:

• Anmol Bhandari, University of Minnesota; David Evans, University of Oregon; Mikhail Golosov, Princeton University and NBER; and Thomas Sargent, New York University and NBER, “Fiscal Policy and Debt Management with Incomplete Markets”

• Gabriel Chodorow-Reich, Harvard University and NBER, and Loukas Karabarbounis, Federal Reserve Bank of Minneapolis and NBER, “The Limited Macroeconomic Effects of Unemployment Benefit Extensions” (NBER Working Paper No. 22163)

• Lorenz Kueng, Northwestern University and NBER, “Explaining Consumption Excess Sensitivity with Near- Rationality: Evidence from Large Predetermined Payments” (NBER Working Paper No. 21772)

• Greg Kaplan, University of Chicago and NBER; Kurt Mitman, Institute for International Economic Studies (Stockholm); and Giovanni Violante, New York University and NBER, “Consumption and House Prices in the Great Recession”

• Xavier Gabaix, New York University and NBER, “Behavioral Macroeconomics via Sparse Dynamic Programming” (NBER Working Paper No. 21848)

• Sydney Ludvigson, New York University and NBER; Sai Ma, New York University; and Serena Ng, Columbia University and NBER, “Uncertainty and Business Cycles: Exogenous Impulse or Endogenous Response?” (NBER Working Paper No. 21803)

Summaries of these papers are at: http://www.nber.org/confer/2016/EFGs16/summary.html

34 NBER Reporter • 2016 Number 3 NBER Books

African Successes Volume I: Government and Institutions Volume II: Human Capital Volume III: Modernization and Development Volume IV: Sustainable Growth Sebastian Edwards, Simon Johnson, and David N. Weil, editors The University of Chicago Press, 2016 $110.00 each (cloth) Studies of African economic development frequently focus on the daunting challenges the continent faces. From recurrent crises to ethnic conflicts and long-standing corruption, a raft of deep-rooted problems has led many to regard the continent as facing many hurdles to raising living standards. Yet Africa has made considerable progress in the past decade, with the GDP growth rate exceeding five percent in some regions. The African Successes series looks at recent improvements in living standards and other measures of development in many African countries with an eye toward identify- ing what shaped them and the extent to which lessons learned are transferable and can guide policy in other nations and at the international level. The first volume in the series, African Successes: Government and Institutions considers the role governments and institutions have played in recent developments and identifies the factors that enable economists to predict the way institutions will function. The second volume, African Successes: Human Capital turns the focus toward Africa’s human capital deficit, measured in terms of health and schooling. It offers a close look at the continent’s biggest challenges, including tropical disease and the spread of HIV. The third volume, African Successes: Modernization and Development looks at the rise in private production in spite of difficult institutional and physical environments. The volume emphasizes the ways that technologies, including mobile phones, have made growth in some areas especially dynamic. The fourth volume, African Successes: Sustainable Growth combines informative case studies with careful empirical analysis to consider the prospects for future African growth.

For information on ordering and electronic distribution, see http://www.press.uchicago.edu/books/orders.html or to place an order you may also contact the University of Chicago Press Distribution Center, at 1-800-621-2736 (USA and Canada) 773-702-7000 (international) Email: [email protected]

NBER Reporter • 2016 Number 3 35 Nonprofit Org. NBER U.S. Postage Reporter PAID NATIONAL BUREAU OF ECONOMIC RESEARCH National Bureau of Economic Research 1050 Massachusetts Avenue Cambridge, Massachusetts 02138-5398 (617) 868-3900

Change Service Requested