NBER Reporter NATIONAL BUREAU OF ECONOMIC RESEARCH

Reporter OnLine at: www.nber.org/reporter 2009 Number 1

Program Report

IN THIS ISSUE

Program Report Health Economics Health Economics 1

Research Summaries Michael Grossman * Electricity Pricing … 9 Re-evaluating Learning 12 Teen and Non-Marital Childbearing 16 Organizations in the Aggregate Economy 19 The NBER’s Program in Health Economics focuses on the deter- minants of health. Two areas of particular interest are the economics NBER Profiles 22 of obesity and the economics of substance use. The program members’ Conferences 24 research has been widely supported by federal research grants and by NBER News 27 private foundations. Program and Working Group Meetings 31 Bureau Books 35 The Economics of Obesity Genetic factors cannot account for the rapid increase in obesity since 1980 — these factors change slowly over long periods of time. Therefore, economists have a role to play in examining the determi- Changing Content for nants and consequences of this trend, even though the factors at work NBER Reporter are complex, and the policy prescriptions are by no means straightfor- ward. Childhood obesity is especially detrimental, because its effects This issue of the NBERReporter breaks from the past carry over into adulthood. Shin Yi-Chou, Inas Rashad, and I estimate with regard to coverage of NBER Program meetings, the effects of fast-food restaurant advertising on television on obesity Working Group meetings, and other Conferences. among children and adolescents.1 Our results suggest that a ban on Historically, the Reporter included a list of the papers these advertisements would reduce the number of obese children ages presented at each of these NBER gatherings, along 3–11 in a fixed population by 18 percent and would reduce the num- with a summary of each paper. Beginning with this ber of obese adolescents ages 12–18 by 14 percent. Eliminating the tax issue, the Reporter lists the papers presented and deductibility of this type of advertising would produce smaller declines their authors, and provides readers with a link to a of between 5 and 7 percent in these outcomes, but would impose lower page on the NBER website that includes the paper’s costs on children and adults who consume fast food in moderation abstract, the name of the discussant if there was one because positive information about restaurants that supply this type of at the NBER meeting, and the research paper. This food would not be completely banned from television. new format for the Reporter makes it easier for read- Robert Kaestner and Xin Xu examine the association between ers to access the full-length research papers presented girls’ participation in high school sports and the physical activity, at NBER meetings and it recognizes the increasingly weight, and body mass and body composition of adolescent females prominent role of electronic archives for research publications of all types. We hope that you enjoy the * Grossman directs the NBER’s Program in Health Economics and is new Reporter format.. Distinguished Professor of Economics at the City University of New York Graduate Center.

NBER Reporter • 2009 Number 1 during the 1970s when girls’ sports partici- pation was dramatically increasing as a result NBER Reporter of Title IX of the Educational Amendments of 1972.2 Title IX requires that programs and activities that receive funds from the Department of Education must operate in a non-discriminatory manner. Kaestner and Xu find that increases in girls’ participation in The National Bureau of Economic Research is a private, nonprofit research orga- high school sports, a proxy for expanded ath- nization founded in 1920 and devoted to objective quantitative analysis of the letic opportunities for adolescent females, are American economy. Its officers and board of directors are: associated with an increase in physical activ- President and Chief Executive Officer — James M. Poterba ity and an improvement in weight and body Controller — Kelly Horak mass among girls. In contrast, adolescent boys BOARD OF DIRECTORS experienced a decline in physical activity and Chairman — John S. Clarkeson an increase in weight and body mass during Vice Chairman — Kathleen B. Cooper the period when girls’ athletic opportunities Treasurer — Robert Mednick were expanding. Taken together, these results DIRECTORS AT LARGE strongly suggest that Title IX and the increase Peter Aldrich Jessica P. Einhorn Alicia H. Munnell in athletic opportunities among adolescent Elizabeth E. Bailey Mohamed El-Erian Rudolph A. Oswald females it engendered had a beneficial effect Richard Berner Jacob A. Frenkel Robert T. Parry John Herron Biggs Judith M. Gueron James M. Poterba on the health of adolescent girls. John S. Clarkeson Robert S. Hamada John S. Reed Rusty Tchernis, Daniel Millimet, and Don R. Conlan Karen N. Horn Marina v. N. Whitman Muna Husain provide conflicting evidence Kathleen B. Cooper John Lipsky Martin B. Zimmerman Charles H. Dallara Laurence H. Meyer with regard to the effectiveness of school George C. Eads Michael H. Moskow nutrition programs in combating childhood obesity.3 They find that the School Breakfast DIRECTORS BY UNIVERSITY APPOINTMENT Program is a valuable tool in the current bat- George Akerlof, California, Berkeley Joel Mokyr, Northwestern tle against obesity. On the other hand, the Jagdish W. Bhagwati, Columbia Andrew Postlewaite, Pennsylvania National School Lunch Program exacerbates Glen G. Cain, Wisconsin Uwe E. Reinhardt, Princeton Ray C. Fair, Yale Nathan Rosenberg, Stanford the current epidemic. Franklin Fisher, MIT Craig Swan, Minnesota Turning to one consequence of obesity in Mark Grinblatt, California, Los Angeles David B. Yoffie, Harvard adulthood, Erdal Tekin and Roy Wada con- Saul H. Hymans, Michigan Arnold Zellner, Chicago Marjorie B. McElroy, Duke sider whether the obese pay a penalty in terms of lower wage rates.4 They point out that DIRECTORS BY APPOINTMENT OF OTHER ORGANIZATIONS previous research in this area relied on body Jean Paul Chavas, Agricultural and Applied Economics Association Gail D. Fosler, The Conference Board weight or body mass index (BMI, defined Martin Gruber, American Finance Association as weight in kilograms divided by height Timothy W. Guinnane, Economic History Association in meters squared5) for measuring obesity Arthur B. Kennickell, American Statistical Association Thea Lee, American Federation of Labor and despite the growing agreement in the medi- Congress of Industrial Organizations cal literature that they represent misleading William W. Lewis, Committee for Economic Development measures of obesity because of their inabil- Robert Mednick, American Institute of Certified Public Accountants Angelo Melino, Canadian Economics Association ity to distinguish between body fat and fat- Harvey Rosenblum, National Association for Business Economics free body mass. Using these two variables, John J. Siegfried, American Economic Association they find that increased body fat is unambig- The NBER depends on funding from individuals, corporations, and private foun- uously associated with decreased wages for dations to maintain its independence and its flexibility in choosing its research both males and females. This result is in con- activities. Inquiries concerning contributions may be addressed to James M. Poterba, President & CEO, NBER 1050 Massachusetts Avenue, Cambridge, MA trast to the mixed and sometimes inconsistent 02138-5398. All contributions to the NBER are tax deductible. results from the previous research using BMI. They also find new evidence indicating that The Reporter is issued for informational purposes and has not been reviewed by the Board of Directors of the NBER. It is not copyrighted and can be freely repro- a higher level of fat-free body mass is consis- duced with appropriate attribution of source. Please provide the NBER’s Public tently associated with increased hourly wages. Information Department with copies of anything reproduced. The body composition measures they employ Requests for subscriptions, changes of address, and cancellations should be sent represent significant improvements over the to Reporter, National Bureau of Economic Research, Inc., 1050 Massachusetts previously used measures because they allow Avenue, Cambridge, MA 02138-5398. Please include the current mailing label.

2 NBER Reporter • 2009 Number 1 for the effects of fat and fat free com- the effect of health insurance on weight Cigarettes ponents of body composition to be sep- outcomes. Bhattacharya, Bundorf, arately identified. Noemi Pace, and Sood provide this In two related studies, Donald Clearly, obesity carries a high per- missing piece by showing that both pri- Kenkel, Philip DeCicca, Alan Mathios, sonal cost. But does it carry a high vately insured individuals and those and colleagues question the consensus enough social cost to make it a con- with Medicaid coverage have a larger in the literature concerning the inverse cern of public policy? The case for body mass index and a higher prob- relationship between the price of ciga- government intervention in the food ability of being obese than persons rettes and various measures of cigarette choices of its citizens is weakened if with no health insurance.11 Rashad and consumption by teenagers and young fully informed consumers are taking Sara Markowitz report similar results adults. Controlling for a direct mea- account of all the costs of their food for BMI but not for the probability sure of state- and time-specific anti- choices, and strengthened if the obese of being obese.12 Both studies take smoking attitudes of adults, DeCicca, do not pay for their higher medical account of the potential endogeneity of Kenkel, Mathios, Yoon-Jeong Shin, expenditures through differential pay- health insurance. and Jae-Young Lim show the effect of ments for health care and health insur- price on youth smoking participation.14 ance, and if body weight decisions are The Economics of Cigarette consumption, conditional on responsive to the incidence of the medi- Substance Use positive participation, continues to be cal care costs associated with obesity. inversely related to price. In the second Several program members have Program members have been study- study, DeCicca, Kenkel, and Mathios examined the effects of weight on med- ing the determinants and consequences find no evidence that higher cigarette ical care costs and the impacts of insur- of cigarette smoking, excessive alcohol taxes prevent smoking initiation but ance on weight. Focusing on adolescents, use, and consumption of such illegal some evidence that higher taxes are Alan C. Monheit, Jessica P. Vistnes, and drugs as marijuana, cocaine, and heroin associated with increased cessation.15 Jeannette A. Rogowski report that in for nearly three decades. Much of this On the other hand, using repeat private group health plans, obese girls research has focused on their respon- cross-sections for the period from 1991 have expected health plan payouts that siveness to price. My time-series study through 2005 — a much longer period are approximately $1,000 greater than of trends in cigarette smoking, binge than those that Kenkel and colleagues females of normal weight.6 They find alcohol drinking (consumption of five considered — Christopher Carpenter no differences for obese boys in these or more drinks in a row on at least one and Philip J. Cook report that the large plans or for obese girls or boys with day in the past two weeks), and mari- state tobacco tax increases of the past public (Medicaid or the State Child juana use by high school seniors sets the 15 years were associated with signifi- Insurance Program) coverage. stage for the studies to be discussed.13 I cant reductions in smoking participa- Jay Bhattacharya and colleagues show that changes in price can explain tion by youths.16 This result emerges consider in detail the health care cost a good deal of the observed changes even after the anti-smoking sentiment externality associated with adult obe- in these behaviors for the period from measure used by Kenkel and colleagues sity.7 They estimate that the obese 1975 through 2003. For example, the is held constant. This appears to be an impose an external cost of approximately 70 percent increase in the real price of area in which a good deal of additional $150 on the non-obese.8 Bhattacharya cigarettes since 1997 attributable to the research would be fruitful. and M. Kate Bundorf find, however, Medicaid Master Settlement Agreement Turning to other determinants of that the incremental healthcare costs explains almost all of the 12 percentage cigarette smoking and determinants associated with obesity are passed on point reduction in the cigarette smok- of outcomes related to that behavior, to obese workers with employer-spon- ing participation rate since that year. Henry Saffer, Melanie Wakefield, and sored health insurance in the form of The 7 percent increase in the real price Yvonne Terry-McElrath examine the lower cash wages.9 Obese workers in of beer between 1990 and 1992 due effect of nicotine replacement therapy firms without employer-sponsored to the Federal excise tax hike on that (NRT) advertising on youth smoking.17 insurance do not have a wage offset beverage in 1991 accounts for almost They find that an increase in this type relative to their non-obese counter- 90 percent of the 4 percentage point of advertising has no impact on youth parts. Their estimate of the wage off- decline in binge drinking in the period smoking participation but causes the set exceeds estimates of the expected at issue. The wide swings in the real number of cigarettes smoked per day incremental health care costs of these price of marijuana explain 70 percent by youths who smoke to increase. They individuals for obese women, but not of the reduction in participation from provide a moral hazard explanation of for men.10 1975 to 1992, 60 percent of the subse- this result: NRT advertising increases None of the studies just summa- quent growth to 1997, and almost 60 the expectation that cessation is rela- rized contains an empirical estimate of percent of the decline since that year. tively easy. They estimate that a ban

NBER Reporter • 2009 Number 1 3 on NRT advertising is equivalent to cate that higher state excise tax rates on of Justice.22 The household surveys con- a 10 percent increase in the price of beer (the most popular alcoholic bever- tain imperfect measures of chronic drug cigarettes. age among youths and young adults) are use and obviously exclude certain groups Sudden Infant Death Syndrome associated with lower gonorrhea inci- of heavy users such as the homeless and (SIDS) is a leading cause of mortality dence rates for males ages 15–19 and criminals. Therefore, Dhaval Dave among infants between the ages of one 20–24. These higher taxes also are asso- employs rates of hospital emergency and twelve months. Prenatal maternal ciated with lower AIDs rates for males room mentions for cocaine and heroin smoking and postnatal environmental ages 20–29. Zero tolerance laws, which and the percentage of arrestees testing smoke have been identified as strong typically set the maximum blood alco- positive for each substance based on risk factors for SIDS. Given these links, hol percentage at 0.02 for underage urine tests to fit demand functions for Markowitz examines the relationship drinkers, reduce gonorrhea rates among heavy users. In the emergency room between cigarette prices, taxes, and 15–19 year-old boys. study, he finds that the elasticity of the clean indoor air laws and the incidence Carpenter and Carlos Dobkin esti- probability of a cocaine mention with of SIDS.18 She finds that a 10 percent mate the effect of alcohol consump- respect to own-price is a negative 0.27, increase in the price of cigarettes low- tion on mortality using the minimum and the corresponding elasticity of the ers SIDS deaths by approximately 7 per- drinking age in a regression disconti- probability of a heroin mention is a cent. Stronger restrictions on smoking nuity design.21 They find that granting negative 0.1.23 The probability of any in restaurants and child care centers are legal access to alcohol at age 21 leads to drug related episode, which captures also effective in reducing SIDS deaths. large and immediate increases in sev- polydrug usage, is also significantly neg- eral measures of alcohol consumption, atively related to both cocaine and her- Alcohol including a 21 percent increase in the oin prices. Cross-price effects are number of days on which people drink. consistent with a complementary rela- Program members have focused This increase in alcohol consumption tionship between cocaine and heroin. on the determinants of excessive con- results in a discrete 9 percent increase The arrestee study supports these results sumption and on the effects of alco- in the mortality rate at age 21. The and contains own-price elasticities of a hol taxes or prices and other regula- overall increase in deaths is attributable negative 0.3 for cocaine participation tions on violent behavior and on risky primarily to a 14 percent increase in and a negative 0.2 for heroin participa- sexual behavior by teenagers and young deaths due to motor vehicle accidents, tion.24 These results imply that higher adults. Dhaval Dave and Saffer consider a 30 percent increase in alcohol over- penalties, more stringent enforcement, the effects of alcohol taxes on chronic doses and alcohol-related deaths, and a and supply reduction, all of which raise alcohol consumption (consumption of 15 percent increase in suicides. A com- illegal drug prices, can discourage par- more than two drinks a day on average) bination of the reduced-form estimates ticipation by heavy users. among older adults ages 55 and over.19 reveals that a 1 percent increase in the Illegal drug use by pregnant women They find that the elasticity of this out- number of days a young adult drinks can have serious consequences for the come with respect to the real beer tax is or drinks heavily results in a .4 percent health of their infants. Hope Corman, approximately -0.3. Their study is the increase in total mortality. Given that Kelly Noonan, Nancy E. Reichman, first to include a measure of risk prefer- mortality due to external causes peaks and Dave shed a considerable amount ence in the demand function for alco- at about age 21, and that young adults of light on the magnitude of this effect hol and to allow this measure to interact report very high levels of alcohol con- in a large urban sample that over-rep- with the tax effect. Since the tax elastic- sumption, their results suggest that pub- resents unmarried, young, minority ity is similar across both risk-averse and lic policy interventions to reduce youth women.25 They estimate the effect of risk-tolerant individuals, tax policies drinking can have substantial public prenatal drug use both on the prob- are equally effective deterrents among health benefits. ability of low birth weight (less than those who have a higher (the risk toler- 2,500 grams) — a marker for poor ant) versus a lower (the risk averse) pro- Illegal Drugs health — and on a direct measure of pensity for excessive consumption. infant health. Given the link between excessive Most estimates of demand func- Corman, Noonan, Reichman, alcohol consumption and risky sexual tions for illegal drugs combine house- and Dave find that prenatal drug use practices, Kaestner, Markowitz, and I hold surveys with year- and city-specific increases the probability of low birth- explore the effects of alcohol taxes and cocaine and heroin prices contained in weight by between 4 and 6 percentage statutes pertaining to drunk driving on the System to Retrieve Information points and that it increases the proba- a direct consequence of these practices: from Drug Evidence (STRIDE) main- bility of an abnormal infant health con- the incidence of sexually transmitted tained by the Drug Enforcement dition by between 7 and 12 percentage diseases (STDs).20 Our results indi- Administration of the U.S. Department points. The effect of maternal cigarette

 NBER Reporter • 2009 Number 1 smoking during pregnancy on low birth- Lleras-Muney provide evidence of estimate the health impacts of the weight is slightly larger than that of drug mechanisms via which schooling affects change. The experience of Taiwan, how- use. In contrast, smoking is not signifi- health.28 The obvious economic expla- ever, provides a natural experiment that cantly related to abnormal infant condi- nations — education is related to income Shin-Yi Chou, Jin-Tan Liu, and I tions. These results may reflect that low or occupational choice — explain only a exploit.30 Prior to the introduction of birth weight is a marker for poor infant part of the education effect. In terms NHI in March 1995, government work- health, whereas abnormal conditions of the relation between education and ers possessed health insurance policies are a direct measure. In a companion various health risk factors — smok- that covered prenatal medical care, new- study, the authors find that the demand ing drinking, diet, exercise, use of ille- born deliveries, neonatal care, and med- for illicit drugs among pregnant women gal drugs, household safety, and care ical care services received by their chil- is fairly elastic with respect to the price for hypertension and diabetes — dren beyond the first month of life. of cocaine.26 Taken together, the two Cutler and Lleras-Muney show that the Private sector industrial workers and studies suggest that drug enforcement is better educated have healthier behaviors farmers lacked this coverage. All house- a potentially promising tool for improv- along virtually every margin. They also holds received coverage for the services ing birth outcomes. suggest and provide tentative evidence just mentioned as of March 1995. This that increasing levels of education lead creates treatment and control groups. Other Determinants of Health to different thinking and decision-mak- The former group consists of non-gov- ing patterns. The monetary value of the ernment employed households, while Schooling return to education in terms of health is the latter group consists of government- perhaps half of the return to education employed households. Many studies suggest that years of on earnings, so policies that affect edu- We focus on postneonatal mortal- formal schooling completed is the most cational attainment could have a large ity. We do not observe negative and sig- important correlate of a variety of mea- effect on population health. nificant effects for private workers, but sures of good health. The causal inter- In a study with Seema Jayachandran, we do observe negative and significant pretation of this finding has been diffi- Lleras-Muney exploits a sudden drop in effects for farmers. In the sample as a cult, however, on the grounds that there maternal mortality risk in Sri Lanka whole, we find that the introduction may be omitted “third variables” or between 1946 and 1953, which created of NHI lowered the postneonatal mor- reverse causality. Shin-Yi Chou, Jin-Tan a sharp increase in life expectancy for tality rate of infants born to the wives Liu, Ted Joyce, and I exploit a natural school-age girls, to obtain consistent of farmers by 0.48 deaths per thou- experiment to estimate the causal impact estimates of the effects of an increase sand infants who survived the neonatal of parental education on child health in in life expectancy on schooling.29 This period. This is a reduction of 11 percent Taiwan.27 In 1968, the Taiwanese gov- development allows them to use boys as relative to the mean in the pre-NHI ernment extended compulsory educa- a control group. They find that the 70 period of 4.26 deaths per thousand sur- tion from six to nine years. From that percent reduction in maternal mortal- vivors. The impacts of NHI on farm year through 1973, the government ity risk over the sample period increased households are larger for less educated opened 254 new junior high schools, an female life expectancy at age 15 by mothers, for farmers who live in rural 80 percent increase, at a differential rate 4.1 percent, female literacy by 2.5 per- areas, and for farm households with a among regions. Within each region, we cent, and female years of education by premature or low-weight birth. In the exploit variations across cohorts in new 4.0 percent. While their results suggest case of prematurity, the postneonatal junior high school openings to con- reverse causality from life expectancy to mortality rate is lowered by six deaths struct an instrument for schooling. We schooling in the developing world, they per thousand survivors or by 36 percent use this instrument to estimate the probably do not translate to the United relative to the pre-NHI mean of 16.71. causal effects of mother’s or father’s States and other developed countries in Our results imply that lack of health schooling on the incidence of low birth- which maternal mortality is extremely insurance may be a major contributor weight and mortality of infants born to rare. to poor infant health outcomes in the women in the treatment and control rural sector of developing economies. groups, or to the wives of men in these National Health Insurance They also suggest that the provision groups. Parents’ schooling, especially of health insurance is a more effective mother’s schooling, does indeed cause There is enormous interest in the policy tool if it is accompanied by the favorable infant health outcomes. The impacts of the introduction of National introduction and use of advanced medi- increase in schooling associated with Health Insurance (NHI) on health out- cal technologies. the reform resulted in a decline in infant comes, but the very nature of this inter- June E. O’Neill and Dave M. mortality of approximately 11 percent. vention, whereby entire nations are cov- O’Neill address the NHI issue by com- David M. Cutler and Adriana ered universally, makes it difficult to paring Canada’s publicly funded, single-

NBER Reporter • 2009 Number 1 5 payer health care system to the multi- takes the form of a reduction in preven- variety of health risks, and extra hours payer heavily private U.S. system.31 tion activities. Dave and Kaestner assess could reduce sleep. Job stress may rise They argue that differences between the importance of this phenomenon during economic expansions and may the United States and Canada in infant in the context of an estimation strat- be exacerbated by production speed- mortality and life expectancy — the egy that allows for the possibility that ups and inexperienced workers. Ruhm two indicators most commonly used health insurance has both a direct (ex emphasizes that the findings do not as evidence of better health outcomes ante moral hazard) and indirect effect imply that recessions should be encour- in Canada — cannot be attributed to on health behaviors.33 The indirect aged. Instead, they highlight that the differences in the effectiveness of the effect works through changes in health effects of economic growth are not uni- two health care systems because they promotion information and the proba- formly beneficial and that physicians are strongly influenced by differences bility of illness that may be a byproduct may need to identify patients at higher in cultural and behavioral factors, such of insurance-induced greater contact risk when the economy strengthens. as the relatively high U.S. incidence of with medical professionals. They iden- obesity and of accidents and homicides. tify these two effects and in doing so Reproductive Behavior, Maternal Direct measures of the effectiveness of identify the pure ex ante moral hazard Nutrition, and Infant Health medical care show that five-year relative effect. They find limited evidence that Outcomes survival rates for individuals diagnosed obtaining health insurance reduces pre- with various types of cancer are higher vention and increases unhealthy behav- The program has had a long-stand- in the United States than in Canada, as iors among elderly persons. There is ing interest in the impacts of a variety of are infant survival rates of low-birth- more robust evidence that physician determinants of infant health outcomes. weight babies. These successes are con- counseling is successful in changing Joyce and his colleagues have made very sistent with the greater U.S. availability health behaviors. important contributions in this area of high level technology, higher rates of over a long period of time and have con- screening for cancers, and higher treat- Unemployment tinued their efforts since my last pro- ment rates of the chronically ill. The gram report in the spring of 2004. In need to ration when care is delivered U.S. citizens are experiencing a one set of studies, Joyce, Kaestner, and “free” ultimately leads to long waits. number of negative consequences of Silvie Colman focus on the reproduc- The health-income gradient is at least the current recession, but an increase tive behavior of minors.35 Clearly, this as prominent in Canada as it is in the in the risk of death from acute myo- is a very important group to consider United States. cardial infarction (AMI) may not be because their infants have worse health Focusing on the United States, one of them. Christopher Ruhm finds outcomes than those of other groups. Cutler, Dobkin, and Nicole Maestas that a 1 percentage point reduction Joyce and colleagues are particularly exploit the sharp change in health insur- in unemployment is predicted to raise concerned with the effects of paren- ance characteristics of the population AMI mortality by 1.3 percent, with a tal involvement laws, which require that occurs at age 65, because most larger increase in relative risk for 20–44 parental involvement in a minor’s deci- people become eligible for Medicare, year olds than older adults, particularly sion to terminate a pregnancy. Previous to investigate whether this change mat- if the economic upturn is sustained.34 research has found that minors’ abor- ters for health.32 They address this issue Nevertheless, the much higher abso- tion rates fall following the enactment by examining differences in mortality lute AMI fatality rate of senior citizens of a notification law and that birth rates for severely ill people who were admit- implies that they account for most of do not rise — a “win-win” situation. ted to California hospitals just before the additional deaths. Joyce and colleagues point out that and just after their 65th birthday. They These results suggest the impor- this research has serious methodologi- estimate a nearly 1 percentage point tance of factors like air pollution and cal limitations. It is not able to mea- drop in 7-day mortality for patients traffic congestion that increase with sure cross-state travel, and it misclassi- at age 65, implying that Medicare eli- economic activity, are linked to coro- fies exposure. With regard to the latter gibility reduces the death rate of this nary heart disease, and may have par- issue, three-quarters of minors who con- severely ill patient group by 20 percent. ticularly strong effects on vulnerable ceive at age 17 give birth at age 18. This The mortality gap persists for at least segments of the population, such as creates a bias toward finding no impact two years following the initial hospital the frail elderly. For the younger age on births. In addition, minors can delay admission. group, the longer working hours that an abortion until they reach age 18. A potential unintended conse- accompany an expansion could make Joyce and colleagues remedy these quence of the acquisition of Medicare it more difficult for individuals to take deficiencies by using data for Texas with by the previously uninsured is that it the time to exercise or eat properly. exact dates (month, day, and year) of may induce ex ante moral hazard that Inadequate sleep is associated with a conception, abortion, and birth before

 NBER Reporter • 2009 Number 1 and after the enactment of a parental Girls’ Physical Activity and Weight,” N. Pace, and N. Sood, “Does Health notification law on January 1, 2000. NBER Working Paper No. 12113, Insurance Make You Fat?” presented They find that the abortion rate of 17- March 2006, and The Economics of at the NBER Conference on Economic year-olds at conception fell by 16 per- Obesity, Volume 17 of Advances in Aspects of Obesity, Louisiana State cent relative to those of 18-year-olds Health Economics and Health Services University, November 10–11, 2008. because of the law. In addition, the Research, K. Bolin and J. Cawley, eds., 12 I. Rashad and S. Markowitz, birth rate of 17-year-olds at concep- Amsterdam: JAI an imprint of Elsevier “Incentives in Obesity and Health tion rose by 4 percent. Finally, abor- Ltd., 2007, pp. 79–111. Insurance,” NBER Working Paper No. tions rose by approximately 30 percent 3 D.L. Millimet, R. Tchernis, and M. 13113, May 2007. among teens who did not reach the Husain, “School Nutrition Programs 13 M. Grossman, “Individual age of 18 until after the first trimester and the Incidence of Childhood Obesity,” Behaviors and Substance Use: The of pregnancy. These second-trimester NBER Working Paper No. 14297, Role of Price,” NBER Working Paper abortions involve greater health risks September 2008. No. 10948, December 2004, and than first-trimester abortions. 4 R. Wada and E. Tekin, “Body Substance Use: Individual Behavior, Recent analyses differ on how effec- Composition and Wages,” NBER Social Interactions, Markets and tive the Special Supplemental Nutrition Working No. 13595, November 2007. Politics, Volume 15 of Advances Program for Women, Infants and 5 Persons 18 years of age and older are in Health Economics and Health Children (WIC) is at improving infant classified as obese if their BMI equals Services Research, B. Lindgren and health. Joyce, Racine, and Cristina or exceeds 30. Persons under that age M. Grossman, eds., Amsterdam: JAI Yunzal-Butler use data from nine are classified as obese if their BMI is at an imprint of Elsevier Ltd., 2005, pp. states that participate in the Pregnancy or above the 95th percentile based on 15–39. Nutrition Surveillance System to age-and gender specific growth charts for 14 P. DeCicca, D.S. Kenkel, A.D. address limitations in previous work.36 children and adolescents at a fixed point Mathios, Y.-J. Sin, and J.-Y. Lim, “Youth With information on the mother’s in time. Smoking, Cigarette Prices, and Anti- timing of WIC enrollment, they test 6 A.C. Monheit, J.P. Vistnes, and J.A. Smoking Sentiment,” NBER Working whether greater exposure to WIC is Rogowski, “Overweight in Adolescents: Paper No. 12458, August 2006, and associated with less smoking, improved Implications for Health Expenditures,” Health Economics, 17 (6) (June 2008), weight gain during pregnancy, better NBER Working Paper No. 13488, pp. 733–49. birth outcomes, and greater likelihood October 2007, and Economics and 15 P. DeCicca, D.S. Kenkel, and A.D. of breastfeeding. Their results suggest Human Biology, forthcoming. Mathios, “Cigarette Smoking and that much of the often-reported asso- 7 See J. Bhattacharya, “Who Pays the Transition from Youth to Adult ciation between WIC and lower rates for Obesity?” NBER Reporter, 2008 Smoking: Smoking Initiation, Cessation, of preterm birth is likely spurious, the Number 3, pp. 4–6 for a detailed sum- and Participation,” NBER Working result of gestational age bias. They find mary of this research. Paper No. 14042, May 2008, and modest effects of WIC on fetal growth, 8 J. Bhattacharya and N. Sood, “Health Journal of Health Economics, 27 (4) inconsistent associations between WIC Insurance and the Obesity Externality,” (July 2008), pp. 904–17. and smoking, limited associations with NBER Working Paper No. 11529, 16 C. Carpenter and P.J. Cook, gestational weight gain, and some rela- July 2005, and The Economics of “Cigarette Taxes and Youth Smoking: tionship with breastfeeding. A WIC Obesity, Volume 17 of Advances in New Evidence from National, State, effect exists, but on fewer margins and Health Economics and Health Services and Local Youth Risk Behavior Surveys,” with less impact than has been claimed Research, K. Bolin and J. Cawley, eds., NBER Working Paper No. 13046, by policy analysts and advocates. Amsterdam: JAI an imprint of Elsevier April 2007, and Journal of Health Ltd., 2007. Economics, 27 (2) (March 2008), pp. 9 J. Bhattacharya and M.K. Bundorf, 287–99. 1 S.-Y. Chou, I. Rashad, and M. “The Incidence of the Healthcare Costs 17 H. Saffer, M. Wakefield, and Y. Grossman, “Fast-Food Restaurant of Obesity,” NBER Working Paper Terry-McElrath, “The Effect of Nicotine Advertising on Television and Its No. 11303, revised September 2006, Replacement Therapy Advertising on Influence on Childhood Obesity,” NBER forthcoming in Journal of Health Youth Smoking,” NBER Working Paper Working Paper No. 11879, December Economics. No. 12964, March 2007. 2005, and Journal of Law and 10 J. Bhattacharya and M. Packalen, 18 S. Markowitz, “The Effectiveness of Economics, 51 (4) (November 2008), “The Other Ex-Ante Moral Hazard Cigarette Regulations in Reducing Cases of pp. 599–618. in Health,” NBER Working Paper No. Sudden Infant Death Syndrome,” NBER 2 R. Kaestner and X. Xu, “Effects of 13863, March 2008. Working Paper No. 12527, September Title IX and Sports Participation on 11 J. Bhattacharya, M.K. Bundorf, 2006, and Journal of Health Economics,

NBER Reporter • 2009 Number 1 7 27 (1) (January 2008), pp. 106–33. (April 2007), pp. 361–84. in Health Policy Research), Vol. 19 D. Dave and H. Saffer, “Risk 26 H. Corman, K. Noonan, N.E. 10, (1), D.M. Cutler, A.M. Garber, Tolerance and Alcohol Demand Reichman, and D. Dave, “Demand D. Goldman, and T. Philipson, eds. Among Adults and Older Adults,” for Illicit Drugs by Pregnant Women,” Berkeley, CA: Berkeley Electronic Press, NBER Working Paper No. 13482, NBER Working Paper No. 10688, 2007, pp. 1–45. October 2007, published as “Alcohol August 2004, published as “Demand 32 D.M. Cutler, C. Dobkin, and N. Demand and Risk Preference,” Journal for Illicit Drugs Among Pregnant Maestas, “Does Medicare Save Lives?” of Economic Psychology, 29 (6) Women,” in Substance Use: Individual NBER Working Paper No. 13668, (December 2008), pp. 810–31. Behavior, Social Interactions, Markets November 2007, and Quarterly Journal 20 M. Grossman, R. Kaestner, and S. and Politics, Volume 15 of Advances of Economics, forthcoming. Markowitz, “An Investigation of the in Health Economics and Health 33 D. Dave and R. Kaestner, “Health Effects of Alcohol Control Policies on Services Research, B. Lindgren and Insurance and Ex Ante Moral Hazard: Youth STDs,” NBER Working Paper M. Grossman, eds., Amsterdam: JAI Evidence from Medicare,” NBER No. 10949, December 2004, and an imprint of Elsevier Ltd., 2005, pp. Working Paper No. 12764, December Substance Use: Individual Behavior, 41–60. 2006. Social Interactions, Markets and 27 S.-Y. Chou, J.-T. Liu, M. Grossman, 34 C.J. Ruhm, “A Healthy Economy Can Politics, Volume 15 of Advances and T.J. Joyce, “Parental Education and Break Your Heart,” NBER Working in Health Economics and Health Child Health: Evidence from a Natural Paper No. 12102, March 2006, and Services Research, B. Lindgren and Experiment in Taiwan, NBER Working Demography, 44 (4) (November 2007), M. Grossman, eds., Amsterdam: JAI Paper No. 13466, October 2007. pp. 829-48. an imprint of Elsevier Ltd., 2005, pp. 28 D.M. Cutler and A. Lleras-Muney, 35 S. Colman, T.J. Joyce, and R. 229–56. “Education and Health: Evaluating Kaestner, “ Methodological Issues in the 21 C. Carpenter and C. Dobkin, “The Theories and Evidence,” NBER Working Evaluation of Parental Involvement Effect of Alcohol Consumption on Paper No. 12352, July 2006, and Laws: Evidence from Texas,” NBER Mortality: Regression Discontinuity Making Americans Healthier: Social Working Paper No. 12608, October Evidence from the Minimum Drinking and Economic Policy as Health Policy, 2006, published as “Changes in Age,” NBER Working Paper No. 13374, J. House, R. Schoeni, G. Kaplan, and Abortions and Births and the Texas September 2007, and American H. Pollak, eds., New York: Russell Sage Parental Notification Law,” New Economic Journal: Applied Economics, Foundation,2008, pp. 29–60. England Journal of Medicine, 354 (10) forthcoming. 29 S. Jayachandran and A. Lleras- (March 9, 2006), pp. 1031–8. See also, 22 For a review of these studies, see M. Muney, “Life Expectancy and Human S. Colman, T.J. Joyce, and R. Kaestner, Grossman, “Individual Behaviors and Capital Investment: Evidence from “Misclassification Bias in the Evaluation Substance Use: The Role of Price.” Maternal Mortality Declines,” NBER of Parental Involvement Laws: A 23 D. Dave, “The Effects of Cocaine Working Paper No. 13947, April 2008, Minor Oversight with a Major Impact,” and Heroin Prices on Drug-Related and Quarterly Journal of Economic, American Journal of Public Health, 98 Emergency Department Visits,” NBER forthcoming. (10) (October 2008), pp. 1881–5, and Working Paper No. 10619, July 2004, 30 S.-Y. Chou, M. Grossman, and J.T. S. Colman and T.J. Joyce, “Behavioral and Journal of Health Economics, 25 Liu, “The Impact of National Health Responses to Parental Involvement (2) (March 2006), pp. 311–33. Insurance on Infant Health Outcomes: Laws: The Case of Delay in the Timing 24 D. Dave, “Illicit Drug Use Among A Natural Experiment in Taiwan,” of Abortions Until Age 18,” Perspectives Arrestees and Drug Prices,” NBER NBER Working Paper, forthcoming, on Sex and Reproductive Health, forth- Working Paper No. 10648, July 2004, presented at the Fifth World Congress coming. published as “Illicit Drug Use Among of the International Health Economics 36 T.J. Joyce, A.D. Racine, and C. Arrestees, Prices and Policy, Journal of Association, Barcelona, Spain, July Yunzal-Butler, “Reassessing the WIC Urban Economics, 63 (3) (May 2008), 10–13, 2005. Effect: Evidence from the Pregnancy pp. 694–714. 31 J.E. O’Neill and D.M. O’Neill, Nutrition Surveillance System,” NBER 25 K. Noonan, N.E. Reichman, H. “Health Status, Health Care and Working Paper No. 13441, September Corman, and D. Dave, “Prenatal Drug Inequality: Canada vs. the U.S.,” 2007, and Journal of Policy Analysis Use and the Production of Infant Health,” NBER Working Paper No. 13429, and Management, 27 (2) (Spring NBER Working Paper No. 11433, June September 2007, and Forum for Health 2008), pp. 277–303. 2005, and Health Economics, 16 (4) Economics and Policy (Frontiers

 NBER Reporter • 2009 Number 1 Research Summaries

Electricity Pricing that Reflects Its Real-Time Cost

Severin Borenstein*

Long before U.S. electricity restruc- tion of RTP. Under regulation, the util- power market. The underlying economic turing began in the 1990s there was a ity nearly always charges prices that are model for this market, however, required recognition that the marginal cost of based on some notion of average cost, that prices occasionally rise to well above producing electricity could change sig- including the accounting amortization the marginal cost of producing most nificantly hour to hour. Combined with of long-term capital expenditures. Such units of output in order for firms to earn the high cost of storing electricity, this an approach is targeted at cost recov- operating profits on infra-marginal units, meant that the true opportunity cost of ery, not efficient pricing. Also, regulated operating profits that allowed the firm to consuming electricity also would vary utilities may be less likely to appreciate cover its capital cost, at least in expecta- constantly. For many decades economists one of the main attractions of RTP, the tion. In the simple framework of a con- have argued that retail electricity prices effect it has in shaving demand peaks and stant marginal cost of each generator up should fluctuate accordingly — this is reducing the need for capital investment. to its capacity, this meant that the market known as real-time pricing (RTP) — but If regulators allow utilities to earn gener- had to sometimes clear “on the demand the technology to meter hourly con- ous returns on investment, or if the util- side.” That is, high prices would occur at sumption and to communicate fluctuat- ity management simply wants to grow times of high demand or reduced sup- ing prices was quite costly. the company, a pricing strategy that con- ply, and those high prices would cause In the last half of the twentieth cen- strains new capital investment is unlikely quantities demanded to decline until tury, the industry created a system meant to be popular with managers. they were in line with system capac- to approximate RTP with standard tech- It is not that utilities did not under- ity. Such price-responsive demand would nology: “time-of-use” prices that varied stand or calculate their marginal cost. In constrain prices from jumping too high, systematically by time of day and day fact, engineers tasked with minimizing whether the tight market was caused by of the week, usually with a higher price production costs were constantly cal- a true supply shortage or an artificial Monday through Friday during business culating “system lambda,” the value of shortage caused by some firms exercis- hours and a lower price at all other times. the production constraint, which cor- ing market power. What went largely The two prices (or sometimes three, with responds directly to economic marginal unnoticed at the time was that the tech- an added “shoulder” pricing period) cost. They needed this information in nology and market organization to were set months in advance, however, order to choose among different produc- enable RTP was not in place in any and did not change to reflect system tion resources. The information was just of the markets headed towards deregu- demand/supply balance on a daily basis. not used on the consumption side. lation. My own pre-deregulation work Because of the cost of even this simple As metering technology improved, with Jim Bushnell, which forecast mar- pricing and metering scheme, it was used a few utilities began to experiment with ket power problems in deregulated elec- only for large commercial and industrial RTP. The pioneer and still a leader in tricity markets, just assumed that there customers. this regard is Georgia Power, a company would be some degree of real-time price In a regulatory environment, two that was, and remains, a traditional regu- response.1 additional factors worked against adop- lated utility. GP introduced its first RTP When deregulated markets launched program in 1991 for large industrial cus- in California, Pennsylvania-New Jersey- * Severin Borenstein is a Research Associate tomers. By 2000, nearly one-third of its Maryland, and New England in the late in the NBER’s Industrial Organization entire electricity demand was on RTP. 1990s, retail customers could choose and Environmental and Energy Economics Wholesale electricity markets were among retail providers who were buying Programs. He is also Director of the deregulated in many parts of the United power out of the wholesale power mar- University of California Energy Institute and the E.T. Grether Professor of Business States in the late 1990s. The idea was that ket. Since electricity is a homogeneous Economics and Public Policy, Haas School of electricity generation could be a com- good delivered over a common-carrier Business, University of California, Berkeley. petitive industry with many generators infrastructure of transmission and distri- His profile appears later in this issue. vying to sell their output into a common bution wires, there was no ability to dif-

NBER Reporter • 2009 Number 1 9 ferentiate the product sold. In nearly all pricing because prices would not move dollars in California, but it wasn’t going cases, the final delivery and metering of with marginal cost, it would not even to fundamentally change the cost of sup- usage was also left to the still-regulated result in the second-best (that is, least plying electricity. The reason for this is utility that was providing transmission inefficient) non-varying retail price. The worth highlighting: in an electric sys- and distribution services. The centraliza- result would be inefficient investment tem that must always stand ready to tion of the metering service meant that levels, even given the constant-price con- meet all demand at the retail price, the even if a retailer wanted to offer new straint. We also showed that incentives cost of a constant-price structure is the time-varying retail pricing structures it for adoption of RTP, if there is some cost need to hold substantial capacity that is was difficult to actually do so. to adopt, such as metering, could be too hardly ever used. But utilities optimize Still, because these markets were weak or too strong from a societal point by building “peaker plants” for this pur- in a period of excess capacity, prices of view. Essentially, this is because the pose, capacity that has low capital cost remained low and steady at first, with the RTP adopters change the price and lev- and high operating cost. The social cost primary complaint coming from produc- els of investment for the non-adopters as of holding idle capacity of this form ers who argued that prices were too low well, an externality that can be positive turns out to be not as great as one might to justify new investment. That changed or negative. think. The analysis, however, does not dramatically in summer 2000 with the In the same paper, we used simula- capture some other potential benefits of onset of the California electricity crisis tions to examine how large the societal RTP, including reduced vulnerability to that brought extremely high prices and gains from switching to RTP are likely to supplier market power and greater resil- a few isolated blackouts when total sup- be.4 These simulations used realistic pro- iency in emergency situations, such as ply from generation didn’t keep up with duction cost parameters to analyze how transmission outages, so the simulation demand. the long-run equilibrium investment and estimates are only a piece of the gains. The view of most economists who pricing would change as more custom- While RTP holds potential for real have studied the electricity crisis is that ers moved from a time-invariant pricing efficiency gains, it is unfortunately often it resulted from a true scarcity in the plan to RTP. The result was significant confused with energy efficiency pro- wholesale market greatly exacerbated by and at the same time sobering. grams that are designed to reduce overall the ability of a few sellers to exercise sig- It was significant in that the poten- consumption. RTP is even occasionally nificant market power. This is supported tial gains from RTP were almost cer- touted for having environmental ben- by my own work with Jim Bushnell and tainly many times greater than the esti- efits. While it might cause decreased Frank Wolak, and a paper by Paul Joskow mated costs of implementing such a consumption in some cases, there is no and Ed Kahn.2 Virtually all economists program. In addition, the gains were evidence that the effect on net would agree that the outcome was exacerbated largest for the first tranche of customers generally be in that direction. The effect by the inability of the demand side of moved to RTP. In fact, with reasonable of redistributing consumption from peak the market to respond to real or artifi- elasticity assumptions, it is likely that to off-peak periods could have positive or cial supply shortages. This realization one-half of the possible total surplus gain negative environmental effects. Stephen prompted my research stream on real- could result from putting only one-third Holland and Erin Mansur’s work on time electricity pricing. of all demand on RTP. This was impor- the environmental effect of RTP shows It was recognized during the crisis tant because the cost of implementing that in many parts of the country where that RTP would lower prices during a RTP at the residential level may be sub- coal provides baseload power, smooth- time of supply shortage and would reduce stantially higher — because each house- ing demand is likely to increase most the incentive of sellers to exercise market hold consumes fairly little yet has nearly pollutants, including greenhouse gases, power by making demand more elastic, the same metering and billing costs as a by redistributing more production to thus greatly reducing the wealth trans- large industrial customer — so an RTP coal-fired generation.5 They find, how- fer from consumers to producers. The program is likely to start with large indus- ever, that California, which relies less efficiency effects, however, were much trial and commercial customers. on coal, is an exception where demand less well understood. In a 2003 paper, The results were sobering because smoothing from RTP is likely to benefit Stephen Holland and I explored the as exciting as the prospect of “getting the environment. short-run pricing and long-run invest- prices right” may be to economists, the The industry and public policy ment inefficiencies that result when potential gains were likely to be only 5 debate about RTP in the years following some or all customers face retail prices percent or less of the energy bill. And the California electricity crisis brought that do not vary with the wholesale mar- energy is generally only about half of out a broad range of producer and cus- ket.3 The theoretical analysis showed that the entire electricity bill, the remain- tomer concerns about RTP. Many were the competitive equilibrium in a market der being transmission, distribution, and easy to address. For instance, manag- without RTP could be quite inefficient. customer administration costs. It still ers at regulated utilities worry that RTP Not only would it fail to attain first-best amounted to hundreds of millions of would make it more difficult for them

10 NBER Reporter • 2009 Number 1 to be assured they could earn revenues modest: with no price-response, about Over the last decade, real-time pric- that cover their costs, but RTP can actu- 55 percent of these customers would see ing has continued to attract attention ally match revenues to utility costs bet- their bills rise under RTP (the winners and even some adoption. New programs ter than the time-invariant price model.6 are on average larger consumers than the have appeared in Chicago, New York, But two concerns, in particular, merited losers), but with a short-run elasticity of Florida, and elsewhere. For many years, further empirical study. -0.1, the share drops to 44 percent. The resistance to adoption has rested on First, some customers would be win- results made clear that extremely few implementation costs, but technology ners and others losers with a switch to customers would see disruptive changes advances have undermined those argu- RTP. Those who consume dispropor- in their electricity bills, but that there ments. Nonetheless, as with many eco- tionate quantities at the most expensive would be a significant number of small nomically attractive ideas, public pol- times are being subsidized under time- losers. The remainder of the paper dis- icy adoption requires first examining a invariant pricing and may be worse off cussed various strategies that attempt number of issues, real and imagined, that if they cannot adjust their consumption to compensate losers without distorting fall outside the strict confines of eco- substantially under RTP. Most energy their marginal consumption incentives. nomic efficiency. In my research, I have managers in industrial and commercial Using the same dataset, I studied attempted to address both the questions customers seemed to think that their bill volatility under RTP.8 Bill volatil- of economic efficiency and the broader bills on average would rise significantly ity is caused by consumption volatility, economic questions of risk and redis- under RTP (even though total system price volatility, and the covariance of the tribution that are part of the policy costs would fall with the switch). Second, two. In electricity, departures from aver- process. even customers who thought their con- age consumption quantity and average sumption pattern was no more expen- real-time price tend to be positively cor- sive to supply than the typical customer related, exacerbating the variance of bills. 1 S. Borenstein and J.B. Bushnell, “An were still worried that their bills could Using monthly billing periods, I calcu- Empirical Analysis of the Potential for be much more volatile under RTP. Some lated customer bills under time-invari- Market Power in California’s Electricity economists dismiss such concerns about ant, time-of-use, and real-time pricing Industry,” NBER Working Paper No. variance and risk management within schemes. After adjusting for seasonal 6463, March 1998, and Journal of companies, but the effect is very real variation, which should be easy to antici- Industrial Economics, 47(3), 1999, pp. on company budgets and performance pate, I found that the coefficient of vari- 285–323. reviews of the managers responsible for ation of a customer’s bill is on average 2 S. Borenstein, J.B. Bushnell, and F.A. electricity consumption. nearly five time larger under RTP than Wolak, “Diagnosing Market Power in I was able to study both of these top- under the time-of-use structure that they California’s Restructured Wholesale ics using a confidential dataset of hourly typically face. I then examine a simple Electricity Market,” NBER Working consumption for 1142 large industrial hedging scheme in which the customer Paper No. 7868, September 2000, and and commercial customers in Northern buys its expected consumption quan- American Economic Review, 92(5), California over a four-year period. In tity (seasonally adjusted) for each hour 2002, pp. 1376–1405; and P. Joskow and the paper on wealth transfers from RTP, of the month at the actuarially fair price E. Kahn, “A Quantitative Analysis of I combined the consumption patterns for that period. This hedge incorporates Pricing Behavior in California’s Wholesale of these customers with simulated and no additional information that the cus- Electricity Market During Summer 2000,” actual wholesale prices to examine how tomer is likely to have about its business NBER Working Paper No. 8157, March big the likely transfer would be.7 The activities during the month, so it is very 2001, and Energy Journal, 23(1), 2002, results showed that the transfers were likely that a customer could refine it fur- pp. 1–35. likely to be smaller than one might think. ther. Still, even this simple hedge elimi- 3 S. Borenstein and S. Holland, “On Starting from the simple time-of-use pric- nates about 90 percent of the excess bill the Efficiency of Competitive Electricity ing these customers already faced (three volatility attributable to RTP, leaving Markets with Time-Invariant Retail different preset prices for peak/shoul- bills about 30 percent more volatile than Prices,” NBER Working Paper No. 9922, der/off-peak periods), and assuming no under time-of-use pricing. I also explored August 2003, and RAND Journal of change in consumption pattern, more a somewhat more sophisticated strategy Economics, 36(3), 2005, pp. 469–93. than 95 percent of customers would that utilizes over-hedging (buying for- 4 A more extensive version of the simu- be likely to see their bills rise or fall by ward more than 100 percent of expected lation analysis was later published as S. less than 10 percent. I then looked at quantity demanded) to compensate for Borenstein, “The Long-Run Efficiency of how much the losers might mitigate this the positive price/quantity correlation Real-Time Electricity Pricing,” Energy impact by responding to the price varia- and showed that this approach can lower Journal, 26(3), 2005, pp. 1–24. tion. Over the plausible range of short- bill volatility to be about the same as 5 S. Holland and E.T Mansur, “Is Real- run price elasticities, the effect is fairly under time-of-use pricing. Time Pricing Green? The Environmental

NBER Reporter • 2009 Number 1 11 Impacts of Electricity Demand Variance,” and Puller, eds., Electricity Deregulation: Journal, 28(2), 2007, pp. 131–49. NBER Working Paper No. 13508, Choices and Challenges, Chicago: 8 S. Borenstein, “Customer Risk from October 2007, and Review of Economics University of Chicago Press, 2005. Real-Time Retail Electricity Pricing: and Statistics, 90(3), 2008, pp. 550–61. 7 S. Borenstein, “Wealth Transfers from Bill Volatility and Hedgability,” NBER 6 I discussed many of the implementation Implementing Real-Time Retail Electricity Working Paper No. 12524, September issues in “Time-Varying Retail Electricity Pricing,” NBER Working Paper No. 2006, and Energy Journal, 28(2), 2007, Prices: Theory and Practice,” in Griffin 11594, September 2005, and Energy pp. 111–30.

Re-evaluating Learning

Esther Duflo*

Developing countries have rapidly have. Besides these challenges, teachers to Development Economics”, Abhijit increased access to primary school, but face lax incentives, so teacher motiva- Banerjee and I3 review the evolution the quality of education has remained tion is low: many teachers do not come of the use of randomization in devel- low. Many children are now in school, to school and even those who do come opment economics research. Much but they are hardly learning. In India, do not always teach. like earlier work in labor economics, for example, a 2007 nationwide survey What can be done to improve edu- health, and education, the experimen- by Pratham1, a large education non- cation quality in developing coun- tal research in development economics profit, found that 97 percent of the of- tries? My recent research suggests some started with concerns over the reliable age children are in primary school, but answers to this question. My approach identification of program effects in the only 51 percent of third graders could has centered on using randomized eval- face of complex and multiple chan- read a simple first-grade paragraph, uations to identify the causal effects of nels of causality. The central difficulty and only 33 percent could do simple promising education programs. that randomization seeks to address subtraction. If developing countries are In a randomized evaluation, from is selection bias. When program par- to attain meaningful universal primary the program’s inception the researcher ticipants are not randomly selected, education, they must improve the qual- works in close collaboration with their outcomes may differ systemati- ity of education. the practitioner. The program gets cally from those of non-participants. This is a formidable task: for assigned randomly to part of the sam- This makes it difficult to attribute any starters, rising enrollment, unaccom- ple — the treatment group — which is differences observed between partici- panied by additional budget outlays, compared to the rest, the comparison pants and non-participants to the pro- has increased pressure on available group. In recent years, there has been gram itself. For example, schools that resources. Classes in the lower grades an explosion in research using random- receive better inputs also may differ often are very large, and the children ized evaluations in development eco- systematically from the other schools arrive with wide-ranging levels of pre- nomics. Development economists have in other ways, for example in pedagogy paredness. These large and heteroge- pioneered the use of research partner- and teacher incentives. However, when neous classes can challenge pedagogy. ships with non-governmental organi- the program is randomly assigned, The curricula, set nationally and often zations (NGOs) or private companies. these initial differences even out and inherited in large part from the colo- These partnerships often allow greater selection bias disappears. Experiments nial period, are not adapted to local control over the research design and, allow researchers to vary one factor at challenges and needs. Too often, they increasingly often, input into the pro- a time by randomly assigning the pro- presuppose competencies that many gram design itself. Rachel Glennerster, gram to part of the sample, and there- of the first-generation learners do not Michael Kremer, and I2 describe the fore they yield internally valid esti- various ways of incorporating random mates of program effects. * Duflo is a Research Assistant in the assignment in the evaluation design, Thus, in the mid-1990s, develop- NBER’s Program on Children and a pro- and the practical challenges that go ment economists started doing experi- fessor of economics at MIT. Her profile with it. ments to answer basic questions about appears later in this issue. In “The Experimental Approach the education production function:

12 NBER Reporter • 2009 Number 1 Does better access to inputs (text- program in urban India. The nonprofit Yet disentangling the relative impor- books, teachers) affect school out- Pratham hired locals with some sec- tance of these two mechanisms is key for comes (attendance, test scores) — and ondary education, trained them for effective policy design, because nothing if so, by how much? The motivating two weeks, and deployed them to local constrains them a priori to be embodied theoretical framework was very sim- schools as teacher’s aides specializing in the same program. For example, many ple, but the results were surprising. in remedial instruction. The remedial more marginalized children could be For example: Glewwe, Kremer, and curriculum targeted students in grades taught basic competencies if the regular Moulin4 found that lowering the stu- three and four who did not have first- teachers were trained and instructed to dent-textbook ratio from 4 to 2 had no grade math and reading competencies. focus on them. Conversely, more moti- effect on average test scores. Banerjee, These students were pulled out of the vated teachers could teach the standard Jacob, and Kremer5 found that halv- regular classroom and worked with the curriculum to all the children, if moti- ing the student-teacher ratio also had teacher’s aide for half the four-hour vation were the salient factor. no effect on test scores. These nega- school day. Test scores in this group tive results prompted new reflection increased by 0.6 standard deviations, a Reorganizing the Classroom on the barriers to education in poor large effect. countries: If simply providing inputs The second project replicated Thus, a third project, conducted does not increase the quality of educa- this finding in a very different con- in rural Kenya, was set up to assess the tion in poor countries, then it must be text. Abhijit Banerjee, Rukmini Banerji, importance of the two factors; Pascaline necessary to change the organization Rachel Glennerster, and I8 evaluated Dupas, Michael Kremer, and I9 designed of teaching in schools, both the peda- Read India, another remedial educa- the experiment. When Kenya intro- gogy and the incentives faced by stu- tion program. Pratham gives rural vol- duced free primary education in 2003, dents and teachers. This led to a new unteers (educated youth from the vil- class sizes exploded in the lower grades. round of field experiments motivated lage) a week’s training in its reading At the beginning of the program, in by the general question: Can changing pedagogy and deploys them back to 2005, the average first-grade class in the the organization of teaching in schools their villages to run after-school read- area where we worked was 83 students, affect education outcomes? For the ing programs. We found that after a and in 28 percent of the classes it was most part, these more recent projects year, among students who could not more than 100. The program provided have varied more than one factor at a read at baseline, those who participated funds, starting in the second term, to time in different experimental groups, in Read India were 60 percentage points 140 schools, randomly selected out of making randomization a powerful tool more likely to be able to recognize let- 210 possibilities, to hire extra teachers for examining the role of incentives, ters than those in comparison villages. on one-year renewable contracts. (The spillovers, and other key questions in The findings already have affected pol- extra teachers were fully qualified but the economics of education. icy: based on this demonstrated effec- young and inexperienced, being recent I have contributed to this litera- tiveness, Pratham secured funding from teacher’s college graduates.) In 121 of ture with four projects. the Gates and Hewlett foundation to the 140 program schools, there was just extend the Read India to 100 districts, one first-grade class. These classes were Remedying Education covering millions of children. And so, split into two sections. In 60 randomly even when the instructor has no for- selected schools, students were quasi- One finding of Glewwe, Kremer, mal teacher’s training, remedial educa- randomly assigned to sections; in the and Moulin6 was that, while the average tion focusing on what children need to remaining 61, students were ranked by child did not benefit from textbooks, know to take advantage of the available prior achievement (first-term grades) students who were already proficient inputs can be highly effective. and the top and bottom halves were did benefit. A possible explanation for There are two main potential expla- assigned to different sections. In all 121 this, the authors conclude, could be that nations for these results. First, the reme- schools, the teachers were randomly the textbook (and the curriculum) was dial instruction, by focusing on what assigned to sections from a common too advanced for the majority of the students do not know rather than the pool of extra and regular teachers. students. Motivated by such evidence, inappropriate curriculum, allows them We compared test scores in 61 track- my research first examined programs to learn more effectively. Second, the ing schools and 60 non-tracking schools that seek to teach students what they teachers hired by Pratham were partic- after 18 months and found that students can learn, rather than what a centrally ularly motivated. Because the remedial in tracking schools scored 0.14 standard set curriculum says they should learn. instruction was always delivered by the deviations higher on average, regard- In the first of these projects Abhijit potentially more-motivated teacher, we less of their initial score. This suggests Banerjee, Shawn Cole, Leigh Linden, cannot distinguish the relative impor- that students benefit from being taught and I7 evaluated a remedial education tance of these two factors. in more homogenous peer groups. We

NBER Reporter • 2009 Number 1 13 argue that greater homogeneity allowed ulum. Glewee, Ilias, and Kremer10 find, absence fell from 40 percent to 20 per- teachers to tailor their teaching to what for example, that when teachers in Kenya cent. What’s more, there is no evidence the students did not know. We found, were offered such rewards, test scores that when they were in school the teach- for example, that students assigned to rose in the short term. Because the test- ers were less likely to teach or that they the bottom section seemed to gain most score gains did not persist, the authors taught differently. With teaching time in the easier competencies and least in suggest that the teachers may have been increased, test scores increased by 0.17 the hardest competencies. “teaching to the test.” standard deviations. This suggests that We also found, however, that com- Another possibility is to reward direct, attendance-based incentives— pared to those assigned to regular teach- teacher effort directly — if it can be applied systematically—can improve ers, students assigned to the extra observed. In developing countries, there learning. teacher have significantly (0.18 stan- is a significant margin of improvement in dard deviation) higher test scores, both one relatively easy-to-observe dimension Re-empowering the Parents? in tracking and non-tracking schools. of teacher effort, namely, the amount There were other differences between of time the teacher spends in front of It may be more difficult, though, to these two groups — for example, stu- the classroom. The Kenya tracking study apply such incentives on teachers already dents assigned to the extra teacher were also found that teachers who face strong in government service. They are politi- more likely to always be taught by the incentives do come to school regularly: cally empowered and they are accus- same teacher, whereas the regular teach- the teachers hired on short contracts tomed to lax enforcement of incentive ers often adopted a rotation system by were more likely to be in school during structures. On paper, the teachers answer which different teachers teach different random checks than the regular teach- to the government which answers to the subjects. Even so, the test-score differ- ers. It seems relatively easy to monitor parents. Many international organiza- ence does suggest that motivation is teacher presence, so would penalizing tions, such the World Bank, have argued important. The young and inexperi- chronic absence (or rewarding presence) that one way to strengthen teacher incen- enced but highly motivated teacher improve teacher presence and learning? tives is to empower the parents and to get seems to be more effective than several A priori, it is not evident that direct them involved in the schools. Parents, experienced but unmotivated teachers attendance-based teacher incentives the argument goes, can monitor teachers put together. would improve learning. Teachers could better and they are more motivated to Thus, the findings suggest that both always come to school but not teach: in improve school quality than faraway gov- pedagogy and incentives matter — abil- the Kenya tracking study, only 54 per- ernment officials; increasing their aware- ity to adapt what is taught in the class- cent of the regular teachers (compared ness of poor school quality, through room to what the students can learn to 84 percent of the extra teachers) in information, and empowering them to benefits everyone, but teacher moti- school on a given day were teaching in do something about it, by increasing vation makes a difference as well. The the classroom, the rest being in the teach- their control of school resources, should findings also confirm that just increas- er’s room. And, in a five-country study, lead to improvements in school quality. ing inputs, without any other changes, Chaudhury et al11 found that 19 percent A finding from the second project is not effective: students who were of teachers were absent and only half of suggests caution. Alongside our eval- assigned to the regular teacher in non- those present actually were teaching at uation of Read India, my co-authors tracking schools did not perform signif- the time of the unannounced visit. Abhijit Banerjee, Rachel Glennerster, icantly better than students in compari- Thus, to address this empirical ques- Rukmini Banerji, and I13 also examined son schools. tion, in a fourth project, Rema Hanna the impact of providing parents with and I12 evaluated the impact of direct, information on learning levels and on Restructuring Teacher attendance-based incentives on teacher the resources available to them to change Incentives presence, and student learning. The their school. Despite days spent in vil- NGO Seva Mandir runs single-teacher lages conducting meetings, to get parents So, teacher motivation matters, schools in remote rural Rajasthan, India. to effectively engage with the school sys- but how can teachers be incentivized? The teachers were given durable cameras tem and teachers to change their behav- One possibility is to reward teachers with date and time functionality and ior, the information and mobilization for improved test scores. But, as stud- asked to photograph themselves with the campaign had no effect. If confirmed ies in the United States suggest, this can children at the beginning and at end of in further research, this finding would lead to teachers focusing on the proxi- each school day. Attendance was deter- suggest that, in the short run, govern- mal (rewarded) outcome, rather than mined based on the number of valid ments should retain the responsibility the ultimate (policy target) outcome. In photographs and the teacher’s pay was of getting the schools to work for poor particular, teachers can focus on acing based on attendance. Not surprisingly, people. the test, rather than learning the curric- the teacher presence increased. Chronic

14 NBER Reporter • 2009 Number 1 Re-evaluating Learning— 2 E. Duflo, R. Glennerster, and M. Participatory Programs: Evidence from a Summing Up Kremer, “Using Randomization in a randomized evaluation in education Development Economics: A toolkit,” in India,” NBER Working Paper No. Together, a series of randomized eval- NBER Technical Working Paper No. 14311, September 2008. uations of education programs in devel- 333, December 2006. 9 E. Duflo, P. Dupas, and M. Kremer, oping countries have taught us some- 3 A. Banerjee and E. Duflo, “The “Peer Effects and the Impact of Tracking: thing about how education in developing Experimental Approach to Development Evidence from a Randomized Evaluation countries can be improved: focus teach- Economics,” NBER Working Paper No. in Kenya,” NBER Working Paper No. ing on skills students need to progress 14467, November 2008. 14475, November 2008. further; find ways to motivate teachers. 4 P. Glewwe, M. Kremer, and S. 10 P. Glewwe, N. Ilais, and M. Kremer, Neither of these is necessarily an easy, Moulin, “Many Children Left Behind? “Teacher Incentives,” NBER Working ready-to-implement prescription. Much Textbooks and Test Scores in Kenya,” Paper No. 9671, May 2003; a substantial more work is needed to develop pro- NBER Working Paper No. 13300, revision is P. Glewwe, N. Ilais, and M. grams that can achieve these two objec- August 2007; forthcoming in American Kremer, “Teacher Incentives,” Harvard tives on a large enough scale, especially Economic Journal: Applied Economics. University mimeo, June 2008. given the political economy of education 5 A. Banerjee, S. Jacob, and M. Kremer, 11 N. Chaudhury, J. Hammer, M. in developing countries. While neither with J. Lanjouw and P. Lanjouw, Kremer, K. Muralidharan, and F. H. suggests plug-and-play prescriptions, “Moving to Universal Education! Costs Rogers, “Teacher Absence in India: A they do give us ample direction about and Tradeoffs,” MIT mimeo, 2005. Snapshot,” Journal of the European where to search. 6 P. Glewwe, M. Kremer, and S. Economic Association, 3:2–3 (April- What’s more, these experiments have Moulin, “Many Children Left Behind? May 2005), pp. 658–67. also taught us something about how to Textbooks and Test Scores in Kenya,” 12 E. Duflo and R. Hanna, “Monitoring search, how we can learn about learning. NBER Working Paper No. 13300, Works: Getting Teachers to Come to Each experiment answers some questions August 2007; forthcoming in American School,” NBER Working Paper No. and asks new ones; the next study builds Economic Journal: Applied Economics. 11880, December 2005. on the previous one, progressively sug- 7 A. Banerjee, S. Cole, E. Duflo, and 13 A. Banerjee, R. Banerji, E. Duflo, gesting a model of education which is L. Linden, “Remedying Education: R. Glennerster, and S. Khemani, ready to be enriched over time. Evidence from Two Randomized “Pitfalls of Participatory Programs: Experiments in India,” NBER Working Evidence from a randomized evalu- Paper No. 11904, December 2005, and ation in education in India,” NBER 1 Pratham Resource Center, Annual Quarterly Journal of Economics, 122 Working Paper No. 14311, September State of Education Report, Mumbai, (3) (August 2007), pp. 1235–64. 2008. India: Pratham Resource Center, January 8 A. Banerjee, R. Banerji, E. Duflo, R. 2008. Glennerster, and S. Khemani, “Pitfalls of

NBER Reporter • 2009 Number 1 15 Teen and Non-Marital Childbearing

Melissa S. Kearney*

There is a widespread consensus reports her pregnancy to be unintended. teens by over 4 percent. Scaling these among the American public that rates Furthermore, there is ethnographic estimates by the estimated proportion of teen pregnancy and unintended preg- and anecdotal evidence that a substan- of women in a state made newly eligi- nancies to young, unmarried women are tial number of teen pregnancies may ble by expanded coverage, we find that too high. Approximately 30 percent of be deliberate. Policies on contraception births to newly-eligible non-teens fell by teenage girls in the United States become will be effective only to the extent that almost 9 percent. Moreover, our analysis pregnant, and 20 percent give birth by teenagers or other young women are of individual-level data from the NSFG age 20. In addition, half of all pregnan- committed to avoiding pregnancy and implies that the reduction in fertility cies in the United States are reported by to the extent that they serve women who associated with income-based waivers is the mother as unintended. More than were not already using contraception. attributable to greater contraceptive use; one-third of these (1.1 million pregnan- Research that I conducted with Phil we find no evidence of an effect on sex- cies in 2001) are to unmarried women in Levine suggests that expanded access ual activity. Based on the cost per recipi- their twenties. Rates of teen pregnancy to publicly provided family planning ent of family planning services, we find and unplanned pregnancy are higher services results in a moderate reduc- that each birth avoided would cost on among young unmarried women, lower tion in overall births and in births to the order of $6,800. income women, women with lower lev- teens.1 Between December 1993 and Another policy approach to address- els of education, and minority women. March 2007, 25 states received waivers ing teen and non-marital childbearing is The concern about rates of teen from the federal government to extend to alter the financial costs and incen- pregnancy and non-marital pregnancy is the coverage of family planning ser- tives for childbearing. This approach was driven in large measure by the observed vices to women who do not otherwise taken during the reform of welfare, first inferior outcomes for children born into qualify for Medicaid health insurance as part of state waiver demonstrations disadvantaged situations — in particular, coverage. We conduct difference-in-dif- and then with the 1996 national wel- those born to young, unmarried moth- ference analyses to identify the causal fare reform legislation. In previous work ers. In this research summary, I describe impact of these waiver policies using I found that family cap policies imple- some of my work on the policy determi- a wide array of data sources, includ- mented as part of welfare reform in the nants of teen and non-marital childbear- ing Vital Statistics birth data, abortion early- and mid-1990s were not effective ing and the socioeconomic differences data from the Guttmacher Institute, and at reducing birth rates among targeted in rates of teen childbearing and paren- microdata on sexual activity and con- women.2 The family cap was the pri- tal time investment in children. traceptive use from the 1988, 1995, mary welfare reform policy targeted at and 2002 National Surveys of Family reducing non-marital childbearing — it Policy determinants of Teen Growth (NSFG). Using data from the ended the practice of providing families and Non-marital Pregnancy Centers for Medicaid and Medicare on welfare with additional cash benefits Services (CMS), we also confirm that when a new child was born into the fam- There are several commonly sug- these waivers increased the number of ily. The motivation behind the policy gested policy approaches to reducing women receiving Medicaid-funded fam- was to eliminate the financial incentive the level of teen and unintended preg- ily planning services. for conceiving an additional child while nancies. One is increased access to con- We find that extending Medicaid on welfare and thereby to reduce births traception. As an initial matter, the term family planning services to women at among the target population. “unintended” as captured in surveys is higher levels of income dramatically I use the variation across states in fraught with measurement and inter- increased the number of women receiv- the timing of family cap implementation pretation issues, and it is not always ing those services. By itself, this does to identify whether this policy leads to clear what is meant when a woman not necessarily indicate an effect on a reduction in births. The primary eco- behavior, as it could merely reflect a nomic question is whether the avail- * Kearney is a Faculty Research Fellow in crowding out of privately provided ser- ability of fewer resources at the mar- NBER’s Program on Children and an assis- vices. However, we demonstrate that the gin decreases a woman’s propensity to tant professor at the University of Maryland. waiver policies reduced overall births bear additional children. The potential Her profile appears later in this issue. to non-teens by about 2 percent and to direct effect of the policy is to reduce

16 NBER Reporter • 2009 Number 1 higher-order births: a decrease in mar- involved. My review of the most com- regarding, for example, contraceptive ginal resources raises the price of an pelling evidence concludes that absti- use, educational attainment, and labor additional child and may thereby deter nence education programs tend to be market training. Alternatively, schools a woman from having additional births. ineffective at reducing rates of sexual and labor market conditions in their Insofar as the policy sends a message that activity. However, compared to non- communities may be sufficiently weak welfare is less generous than previously, abstinence-focused sexual education that staying in school and avoiding early it may also lead a woman to delay child- courses, these programs do not lead to childbearing might not be seen as offer- bearing until she is financially secure and lower rates of contraceptive use among ing any real benefit. In addition, some thereby reduce first births as well. sexually active teens, as some critics have ethnographic evidence suggests that The vital statistics birth data for claimed. Certain contraceptive-focused those who grow up in disadvantaged the years 1989 to 1998 offer no evi- sex education programs may be effective situations are simply more inclined to dence that family cap policies lead to at reducing risky sexual behavior among have children because of what might be a reduction in births among women participants. And finally, a couple of appropriately described as differences in ages 15 to 34. After I control for state high-profile multi-component interven- preferences. effects, month effects, and state-specific tions show promise of being effective, Levine and I confirm an empirical linear time trends, I find that a decline but would be very difficult and expen- relationship between individual rates in births of more than 1 percent can be sive to replicate in other settings. of socioeconomic disadvantage and rejected at the 95 percent confidence In sum, the past two decades have rates of early childbearing but find that level. (The upper bound of the confi- seen numerous and varied efforts — by socioeconomic disadvantage plays only dence interval is an increase in births community groups, schools, non-prof- a small role in the aggregate.4 Using of 1.1 percent.) This finding — of no its, and all levels of government — to the Panel Study of Income Dynamics effect on births — is maintained across bring down rates of teen pregnancy and (PSID) we observe that growing up in multiple specification checks. The set childbearing in this country. Research in socioeconomic disadvantage is associ- of confidence intervals around six alter- this area, at this stage, is far from being ated with substantially higher rates of native estimates has a lower bound of a able to offer a conclusive answer to the teen childbearing. The main empirical 1 percent decline and an upper bound question of what drove the rise and sub- contribution of this paper is a cohort- of a 2 percent increase. I also find that sequent decline in teen pregnancy. Nor based analysis: we want to know whether the data reject large declines in higher- can we conclusively answer why the level cohorts of women with higher rates order births among demographic groups of teen childbearing and unintended of socioeconomic disadvantage at birth with relatively high welfare participa- (as reported by the mother) childbear- subsequently have higher rates of early tion rates. ing are so much higher in the United childbearing. We use Vital Statistics A third approach to addressing high States than in other developed coun- microdata from 1968 through 2003 rates of teen pregnancy involves targeted tries. A long-term goal of my research to conduct an analysis of the relation- interventions, or programs run at the is to address this issue of international ship between rates of socioeconomic school or community level. In a recently comparisons. disadvantage of a birth cohort and the completed draft for an NBER confer- cohort’s early childbearing experiences. ence volume, I review the evidence on Socioeconomic Differences As a proxy for disadvantage at birth, the effectiveness of teenage pregnancy we use four alternate factors, all based prevention programs.3 Teen pregnancy The economics literature on teen on the mother’s characteristics: having prevention programs can be usefully cat- and non-marital childbearing has tended been born to a mother with a low level egorized into three types: 1) sex educa- to focus on policy and environmental of education; to an unmarried mother; tion programs with an abstinence focus; determinants in a rational choice frame- to a mother under age 18; or to a mother 2) sex education with a contraception work. But parallel literatures in other under age 20. focus; and 3) multi-component youth social as well as clinical sciences empha- Our cohort-based analysis implies development programs that include sex size the role of socioeconomic disadvan- an even tighter correlation between education as one of many features. Some tage and related factors during child- rates of background disadvantage and programs are based in schools and are hood. Socioeconomic disadvantage can early childbearing than is observed in compulsory, others are school-based but lead to early childbearing through a the PSID data at the individual level. voluntary, and others are run through number of different mechanisms. The But, when our analysis econometrically community centers and groups. There poor may lack the resources available to controls for fixed state and year of birth is substantial variation across programs know about the different opportunities effects, the relationship between rates of in terms of the types of populations available to them or to take advantage disadvantage and early childbearing is served, including racial and ethnic dif- of those opportunities. This could hin- found to be quite modest. For example, ferences as well as ages of the teenagers der their ability to make optimal choices a 10 percent increase in the proportion

NBER Reporter • 2009 Number 1 17 of women who were themselves born use and education or income in time “Subsidized Contraception, Fertility, and to a teen mother (a measure of disad- spent in child care is the opposite of the Sexual Behavior,” earlier version released vantage) is associated with a less than gradient observed for typical leisure and as NBER Working Paper No. 13045, 1 percent increase in the proportion home production activities. April 2007, forthcoming in Review of who give birth before age 18 (a mea- Using a Beckerian framework of Economics and Statistics. sure of early childbearing). This sug- time allocation, we conclude that child 2 M.S. Kearney, “Is There an Effect of gests that broader, societal forces are far care is best modeled as being distinct Incremental Welfare Benefits on Fertility more important in determining rates of from either typical home production or Behavior? A Look at the Family Cap,” early childbearing than rates of socio- leisure activities. In addition, our results NBER Working Paper No. 9093, August economic disadvantage. Our results lead suggest that time spent with one’s chil- 2002, and Journal of Human Resources, us to conclude that the impact of a fairly dren is more highly valued by indi- 2004, 39(2): pp. 295–325. large shock to socioeconomic disadvan- viduals with a higher opportunity-cost 3 M.S. Kearney, “Teen Pregnancy tage would have only a modest impact of their time, as measured by earnings Prevention as an Anti-Poverty on rates of early childbearing. potential. What might account for this? Intervention: A Review of the Evidence,” Another area of socioeconomic It could arise if caring for children is a forthcoming in Targeting Investments differences in childrearing behaviors “luxury good,” if more educated parents in Youth: Fighting Poverty when involves parental time-use patterns.5 have a lower elasticity of substitution Resources are Limited, Phillip Levine Jonathan Guryan, Erik Hurst, and I use between own and market-based child and David Zimmerman eds., University data from the recent American Time Use care (or just a higher relative preference of Chicago Press. Surveys as well as from Multinational for time spent with their children), or 4 M. S. Kearney and P. Levine, Time Use Surveys to examine parental if the returns to investing in the chil- “Socioeconomic Disadvantage and Early time allocated to the care of their chil- dren of more educated parents are rela- Childbearing,” NBER Working Paper dren. We draw three major empirical tively higher. The fact that the children No. 13436, September 2007, forthcom- conclusions about parental child care of higher-educated parents receive more ing in An Economics Perspective on time. First, higher earnings or earnings time with the active attention of their the Problems of Disadvantaged Youth, potential are associated with more time parents, or conversely, that the chil- Jonathan Gruber ed., University of spent with children, even though higher dren of lower-education parents receive Chicago Press. earning parents also work more hours in less, may have important implications 5 J. Guryan, E. Hurst, and M.S. the labor market. Second, this relation- for the intergenerational transmission of Kearney, “Parental Education and ship appears to hold within the United human capital. Parental Time with Children,” NBER States, across other countries, and within Working Paper No. 13993, May 2008, other countries examined. And third, and Journal of Economic Perspectives this positive gradient of parental time 1 M. S. Kearney and P. Levine, 22(3), Summer 2008, pp. 23–46.

18 NBER Reporter • 2009 Number 1 Organizations in the Aggregate Economy

Esteban Rossi-Hansberg*

Economists generally agree that an economies that suggest that the long- need to show that aggregation yields important feature of any modern mac- term growth rate of organizations is the necessary linearity in production. roeconomic theory is an explicit aggre- fairly stable over time. This suggests the The key, however, is that we can do so gation of the microeconomic behavior need for theories that exhibit constant while still being consistent with the of all agents in the economy. In the last returns to scale in the factors that can rich data on establishment sizes and century, the profession has gone from be accumulated over time — a feature dynamics. Wright and I argue that the the formulation of some general aggre- that most endogenous growth theories size distribution and growth patterns gate relationships governing the evolu- share. How can intermediate levels of of establishments can be aggregated in tion of the economy to detailed theories organization affect the required lin- a way that implies balanced growth but that explicitly incorporate the observed earity in aggregate production? Mark that also explains important cross-sec- heterogeneity in many characteristics of Wright and I argue that the organiza- toral differences in the observed estab- agents and firms. Adding these micro- tion of agents in cities is closely related lishment growth rates.2 The key, we economic details has resulted in new to aggregate technologies with con- argue, is to recognize that fluctuations insights on policy, as well as better and stant returns to scale.1 Agents orga- in prices of industry-specific factors more detailed descriptions of modern nize production and their lives in cit- lead to changes in the use and accumu- economies. Although it is obviously ies because they obtain benefits from lation patterns of these factors, which important to recognize that an aggre- agglomeration: there are increasing in turn affect the employment size and gate economy is formed by individual returns at the local level. We claim that growth of establishments. agents making explicit decisions, the to reconcile the increasing returns at These theories all incorporate a standard aggregate models still abstract the local level with constant returns at rich pattern of economic organization from a precise description of how these the aggregate level, one must under- in industries, cities, and production agents interact in small (or not so small) stand the role of cities. We thus pro- plants, but they model individuals, and groups to produce, live, and consume. pose a theory by which the number and their human capital, as an industry- Most economic activity occurs in inter- sizes of cities react to industry produc- specific but uniform factor of produc- mediate levels of aggregation: organi- tivity shocks in a way that exhausts the tion. For the theory, hiring ten work- zations. Firms or plants, but also cit- increasing returns at the local level and ers with some knowledge is the same as ies, families, international production yields constant returns in the aggre- hiring one expert and nine people with chains, political parties, and religious gate. According to our findings, cities no knowledge, as long as total units of organizations, among many others, are are the reason to obtain aggregate bal- human capital are kept fixed. This dis- examples of such organizations. Most of anced growth, but our mechanism also tinction is important, because model- my recent research has concentrated on yields a size distribution of cities that ing the production process as simply a incorporating these organizations into very closely resembles the one observed function of total inputs implies that the general equilibrium theories in order in the data. This is a stark example in organization of agents in specific teams to understand their implications for which considering intermediate organi- is irrelevant. In contrast, we may believe aggregate outcomes. zations (in this case, spatial agglomera- that sometimes having knowledge con- A starting point of this agenda is tion in cities) is fundamental to under- centrated in one individual is more an understanding of how organizations standing aggregate outcomes. efficient than having it evenly distrib- affect economic growth. There is a set A similar argument can be made uted, because we can organize the tasks of fairly consistent facts for developed for considering the formation and char- of production in ways that will exploit acteristics of firms. In the presence of knowledge more intensively. The any form of fixed cost of production, it important distinction, I believe, is that *Rossi-Hansberg is a Research Associate in the NBER’s Program on Economic is hard to rationalize constant returns knowledge is embedded in individuals Fluctuations and Growth and Inter­ to scale at the plant level. Given this, and these individuals have limited time. national Trade and Investment and a how can we reconcile the organization Organizations maximize their access to Professor of Economics and International of economic activity in establishments the knowledge embedded in individuals Affairs at Princeton University. His pro- of particular sizes with the behavior by not making them do standard tasks file appears later in this issue. of the aggregate economy? Again we that other, less able agents also can per-

NBER Reporter • 2009 Number 1 19 form. Heterogeneous agents form pro- stand that the tasks required for pro- the role of organizations in trade.8 As duction teams and the characteristics of duction are heterogeneous in terms of we argue in our survey, most of the this process, and the ability and knowl- their difficulty or the ability or knowl- available research on the international edge of their co-workers, determines edge needed to solve them. Therefore, organization of production in aggre- their productivity and compensation. the location of their performance is gate theories studies the case of trade Of course, once incorporated into an determined by the distribution of skills in tasks between a developed and an aggregate framework, these consider- in the different countries. But many underdeveloped country (north-south ations have key implications for the dis- tasks required for production are het- offshoring). Trade in tasks is, how- tribution of wages in the economy, the erogeneous in other dimensions not ever, by no means restricted to north- size and characteristics of production related to ability or knowledge. A par- south relationships. It is probably more teams, and aggregate productivity. In ticularly important dimension for trade important between developed econo- a series of papers, Luis Garicano and I is heterogeneity in offshoring costs, mies. Of course, the motive for trade is have studied: the implications of hier- namely, the cost of performing a task less transparent in this case. One needs archical organization for wage inequal- away from a firm’s headquarters. Some to incorporate some form of increas- ity, the cross-sectional implications of tasks can require very basic skills but ing returns. One avenue is to use inter- changes in information and communi- need to be performed close to head- nal increasing returns as in the “New cation technology (ICT),3 and, more quarters (like janitorial or transporta- Trade Theory.” The difficulty with this recently, the effect of ICT on innova- tion services), while others require sub- approach is that it implies full special- tion and growth.4 stantial knowledge but can easily be ization at the task level — an unap- Incorporating the explicit decision performed far away from headquarters pealing implication. Another avenue, of whom, and not only how many work- (like tax accounting, or many business which Grossman and I have favored ers, to hire implies that the distribu- services). Heterogeneity in offshoring in our work, is to use external increas- tion of abilities in the population has costs implies that, as tasks performed by ing returns.9 Of course, the potential important consequences for total out- workers of all talents can be offshored, problem with using externalities in a put and efficiency. Naturally, it also any worker could win or lose from off- competitive model is the existence of has important consequences for trade shoring. Gene Grossman and I show multiple equilibriums attributable to and the international organization of that if the motive for trade is cost dif- standard coordination problems. We production. An economy with many ferences, then under some conditions show that having a continuum of tasks, talented workers can organize more reductions in the costs of offshoring all of which are required for produc- efficiently by using some less talented will lead to Pareto gains in the source tion, and letting firms outsource tasks individuals to perform the most com- country.7 All workers may gain from and compete in prices, are enough to mon and straightforward tasks. This offshoring, because firms that inten- eliminate the potential multiplicity of implies that there are gains from trad- sively use low-skill tasks obtain a rela- equilibriums. The reason is that an out- ing production tasks internationally. tive cost advantage, even though tasks sourcing firm can break an equilibrium Pol Antràs, Garicano, and I show that performed by, say, low-skilled work- in which the performance of the task is in such a framework the organization ers are being offshored. This leads to a concentrated in a sub-optimal location of international teams leads to more higher demand for low-skill labor and a by moving on its own, competing in wage inequality in the south, gains from higher low-skill wage. Of course, there prices, and obtaining the whole world trade, and merchandise trade deficits in may be other, more standard effects on market for the task. Importantly, since the north.5 We also study the role of factor prices if there are labor mobil- tasks are small relative to total produc- intermediate managerial skills in gen- ity frictions or if task trade leads to tion, the firm does not gain monop- erating trade in tasks across countries.6 changes in relative prices. oly power. We use this framework to We find that having intermediate skilled It is clear from this research that study the specialization pattern of tasks agents is essential to being a good target modeling the production problem and across developed economies.10 Our for offshoring if communication tech- the resulting organization in more main result is that if two countries trade nology in the target country is not detail has allowed us to better under- tasks, and are identical except for their particularly good. The logic is simple: stand the aggregate implications of glo- size, the smaller one will specialize in intermediate managers are necessary balization. The emergence of global the tasks with the low offshoring costs to save on international communica- production chains is an organizational and will have lower wages. tion costs, but their opportunity cost phenomenon that can have important To conclude, my work emphasizes is to setup a firm on their own (which effects on factor prices, income lev- the need to incorporate a rich set of increases with the quality of communi- els, trade patterns, and growth. Antràs organizations in our theories in order cation technology). and I recently reviewed the progress to understand aggregate phenomena. The papers discussed above take the made by this literature in exploring As is evident from this discussion, many

20 NBER Reporter • 2009 Number 1 aspects of intermediate organizations in American Economic Review 97, no. Helpman, E., D. Marin, and T. Verdier, have not yet been explored. Key among 5 (December 2007), pp. 1639–66. The Organization of Firms in a Global them are the dynamic implications of 3 L. Garicano and E. Rossi-Hansberg, Economy, Harvard University Press, organization for factor investment and “Organization and Inequality in a 2008. knowledge acquisition. Organizations Knowledge Economy,” NBER Working 7 G. Grossman and E. Rossi-Hansberg, are a key determinant of the ability to Paper No. 11458, July 2005, published “Trading Tasks: A Simple Theory of innovate and exploit efficiently avail- in Quarterly Journal of Economics Offshoring,” NBER Working Paper able technologies. How can we study 121, no. 4 (November 2006), pp. 1383– No. 12721, December 2006, forthcom- growth without paying full attention 1435. ing in American Economic Review to the way in which these organizations 4 L. Garicano and E. Rossi-Hansberg, (December 2008). develop over time? “Organizing Growth,” NBER Working 8 P. Antràs and E. Rossi-Hansberg, Paper No. 13705, December 2007. “Trade and Organizations,” NBER 5 P. Antràs, L. Garicano and E. Rossi- Working Paper No. 14262, August 1 E. Rossi-Hansberg and M. Wright, Hansberg, “Offshoring in a Knowledge 2008. “Urban Structure and Growth,” NBER Economy,” NBER Working Paper No. 9 G. Grossman and E. Rossi-Hansberg, Working Paper No. 11262, April 2005, 11094, January 2005, published in “External Economies and International published in Review of Economic Quarterly Journal of Economics 121, Trade Redux,” NBER Working Paper Studies 74, no. 2 (April 2007), pp. no. 1 (February 2006), pp. 31–77. No. 14425, October 2008. 597–624. 6 P. Antràs, L. Garicano and E. Rossi- 10 G. Grossman and E. Rossi-Hansberg, 2 E. Rossi-Hansberg and M. Wright, Hansberg, “Organizing Offshoring: “Task Trade between Similar Countries,” “Establishment Size Dynamics in the Middle Managers and Communication mimeo, October 2008, http://www. Aggregate Economy,” NBER Working Costs,” NBER Working Paper No. princeton.edu/~erossi/TTSC.pdf. Paper No. 11261, April 2005, published 12196, May 2006, published in

✴

NBER Reporter • 2009 Number 1 21 NBER Profile: Severin Borenstein

Severin Borenstein is a Research centered on the airline, oil, gasoline, and Associate in the NBER’s Programs on electricity industries. His current research Industrial Organization and Environmental involves retail electricity pricing and con- and Energy Economics. He is also the E.T. sumption, the markets for alternative energy, Grether Professor of Business and Public and climate change policy. He is a past edi- Policy at Haas School of Business, University tor of Journal of Industrial Economics, and of California, Berkeley, and Director of the currently serves on the editorial boards of University of California Energy Institute. the Journal of Economic Literature and the Borenstein received an A.B. in eco- American Economic Journal: Economic Policy nomics from Berkeley in 1978 and a Ph.D. and is an associate editor of the Review of in economics from MIT in 1983. Before Economics and Statistics. joining the Berkeley faculty in 1996, he Borenstein lives in Orinda, California taught at the University of Michigan and with his wife, Margaret Stewart, and infa- University of California, Davis. mous dog, Wilbur. He enjoys hiking with Borenstein’s research has focused on both of them. policy-driven issues in the fields of indus- trial organization and regulation, mainly

NBER Profile: Esther Duflo

Esther Duflo is a Research Associate pleted a Ph.D. in Economics at MIT in in the NBER’s Programs on Children, 1999. Upon completing her Ph.D. she was Education, and Aging and is the Abdul appointed assistant professor of econom- Latif Jameel Professor of Poverty Allevia­ ics at MIT, and has been there ever since, tion and Development Economics in aside from being on leave to Princeton MIT’s Economics Department. She is University in 2001–2. also a founder and director of the Jameel Duflo has received numerous aca- Poverty Action Lab (J-PAL), a research demic honors and prizes including the network specializing in randomized eval- American Economic Association’s Elaine uations of social programs, which won Bennett Prize for Research (2003); the the 2009 BBVA Foundation “Frontier of “Best French Young Economist Prize” (Le Knowledge” award in the development Monde/Cercle des economistes, 2005); the cooperation category. Médaille de Bronze (Centre National de la Duflo serves on the board of the Bureau Recherche Scientifique, 2005); and the for Research and Economic Analysis of Prix Luc Durand-Reville (Académie des Development (BREAD), and is Director of Sciences Morales et Politiques, 2008). In the Center for Economic Policy Research’s 2008–9 she was the inaugural holder of the development economics program. Her international chair “Knowledge Against research focuses on microeconomic issues Poverty” at the College de France. in developing countries, including house- Duflo has been a co-editor of the hold behavior, education, access to finance, Journal of Development Economics and the health, and policy evaluation. Review of Economics and Statistics, and she Duflo completed her undergraduate currently serves as the founding editor of studies at L’Ecole Normale Supérieure in the AEJ Applied Economics. Paris in 1994, received a master’s degree In her spare time, Esther enjoys rock from DELTA in Paris in 1995, and com- climbing and mountaineering.

22 NBER Reporter • 2009 Number 1 NBER Profile: Melissa Schettini Kearney

Melissa Schettini Kearney is a Faculty Kearney’s fields of specialization are Research Fellow in the NBER’s Programs on public economics and labor. Her research Public Economics and Children and an assis- interests include the effectiveness and design tant professor in the Economics Department of anti-poverty programs; the effect of gov- at the University of Maryland. She also ernment policies on fertility and related serves on the Research Advisory Council of outcomes; issues of income distribution and the National Campaign to Prevent Teen and inequality; and the nature of risk preferenc- Unplanned Pregnancy and on the Policy es and perceptions and, in particular, how Council of the Association of Public Policy these relate to individual decisions with and Management (APPAM). regard to gambling and savings behavior. Kearney received her A.B. in econom- She lives in Washington, D.C. with her ics from Princeton University in 1996 and husband Dan (a lawyer). They have great her Ph.D. from MIT in 2002. Prior to join- fun playing with their children, William ing the Maryland faculty, she was a resident (age 3.5) and Sophia (almost 2), in all of fellow at the Brookings Institution. From their favorite places, including Shenandoah 2002 to 2004, she was an assistant pro- National Park in VA; Park City, Utah; Long fessor of economics at Wellesley College. Beach Island, NJ; Fenway Park; and closer Before attending graduate school, she to home, the National Gallery of Art, Rock worked as a research assistant/program- Creek Park, and their little backyard strewn mer at Mathematica Policy Research in with outdoor toys. Washington, D.C.

NBER Profile: Esteban Rossi-Hansberg

Esteban Rossi-Hansberg is a Research ics, international trade, and urban econom- Associate in the NBER’s Programs on ics. Rossi-Hansberg’s research focuses on the Economic Fluctuations and Growth and internal structure of cities, the distribution of International Trade and Investment and a economic activity in space, economic growth, Professor of Economics and International offshoring and the international organization Affairs at Princeton University. He received of production, the role of information tech- his B.A. and M.A. in economics from ITAM nology on wages and organization, and firm (Mexico) in 1996 and 1997 respectively, and dynamics and the size distribution of firms. his Ph.D. in economics from the University He is an editor of the B.E. Journal in of Chicago in 2002. Macroeconomics and an associate editor of Rossi-Hansberg began his academic the Quarterly Journal of Economics and the career at Stanford University and joined the Journal of Monetary Economics. He lives in faculty at Princeton University in 2005. His Princeton with his wife María José and their areas of specialization are macroeconom- one-year old son, Julián.

NBER Reporter • 2009 Number 1 23 Conferences

Micro and Macroeconomic Effects of Financial Globalization

An NBER/Universities Research Conference on “Micro and Macroeconomic Effects of Financial Globalization” took place in Cambridge on December 5 and 6. The organizers were Ross Levine, NBER and Brown University, and Carlos Vegh, NBER and University of Maryland. These papers were discussed:

• Matteo Iacoviello, Boston College, and Raoul Minetti, Michigan State University, “Foreign Lenders in Emerging Economies”

• Thorsten Beck, Tilburg University, and Maria Soledad Martinez Peria, The World Bank, “Foreign Bank Participation and Outreach: Evidence from Mexico”

• Todd Gormley, Washington University, St. Louis, “Costly Information, Foreign Entry, and Credit Access”

• Geert Bekaert, NBER and Columbia University; Campbell R. Harvey, NBER and Duke University; Christian T. Lundblad, University of North Carolina, Chapel Hill; and Stephan Siegel, University of Washington, “What Segments Equity Markets?”

• Sebnem Kalemli-Ozcan, NBER and University of Houston; Bent Sorensen, University of Houston; and Vadym Volosovych, Florida Atlantic University, “Deep Financial Integration and Volatility”

• Scott Davis, Vanderbilt University, “The Effects of Globalization on International Business Cycle Co-Movement: Is All Trade and Finance Created Equal?”

• Uluc Aysun, University of Connecticut, and Adam Honig, Amherst College, “Bankruptcy Costs, Liability Dollarization, and Vulnerability to Sudden Stops”

• Filippo Brutti, Universitat Pompeu Fabra, “Legal Enforcement, Public Supply of Liquidity, and Sovereign Risk”

• Laura Alfaro, NBER and Harvard Business School, and Andrew Charlton, London School of Economics, “Intra- Industry Foreign Direct Investment” (NBER Working Paper No. 13447)

Summaries of these papers may be found at: http://www.nber.org/confer/2008/URCf08/summary.html

Sticky Prices and Inflation Dynamics

The 22nd Annual NBER-TCER-CEPR Conference on “Sticky Prices and Inflation Dynamics” took place in Tokyo, Japan on December 17–18, 2008. The conference organizers were: Jordi Gali, the Barcelona Graduate School of Economics and NBER; Takeo Hoshi, University of California, San Diego and NBER; Anil K Kashyap, University of Chicago and NBER; and Tsutomu Watanabe, Hitotsubashi University. These papers were discussed:

• Chihiro Shimizu, Reitaku University; Kiyohiko Nishimura, former Professor of Economics, University of Tokyo; and Tsutomu Watanabe, “Residential Rents and Price Rigidity: Micro Structure and Macro Consequences”

24 NBER Reporter • 2009 Number 1 • Kenn Ariga, Kyoto University, and Ryo Kambayashi, Hitotsubashi University, “Employment and Wage Adjustments at Firms under Distress in Japan: An Analysis Based upon a Survey”

• Fabio Canova, ICREA-UPF, and Tobias Menz, University of Bern, “Does Money have a Role in Shaping Domestic Business Cycles? An International Investigation”

• Morten Ravn, European University Institute and CEPR; Stephanie Schmitt-Grohe and Martin Uribe, Columbia University and NBER; and Lenno Uuskula, European University Institute, “Deep Habits and the Dynamic Effects of Monetary Policy Shocks”

• Naohito Abe and Daiji Kawaguchi, Hitotsubashi University, “Incumbent’s Price Response to New Entry: The Case of Japanese Supermarkets”

• Oleksiy Kryvtsov, Bank of Canada, and Virgiliu Midrigan, New York University and NBER, “Inventories and Real Rigidities in New Keynesian Business Cycle Models”

• Kevin Clinton, Consultant; Marianne Johnson, Bank of Canada; and Ondra Kamenik and Douglas Laxton, International Monetary Fund, ”Deflation Risks Under Alternative Monetary Policy Rules”

These papers will be published in the Journal of the Japanese and International Economies. Summaries of these papers may be found at: http://www.nber.org/confer/2008/trio08/summary.html

Tenth Annual Conference in India

On January 10–12, 2009, the NBER, India’s National Council for Applied Economic Research (NCAER), and the Indian Council for Research on International Economic Relations (ICRIER) sponsored a meeting that united NBER researchers with about thirty economists from Indian universities, research institutions, and government departments. The theme of this year’s meet- ing, the tenth such gathering, was “Sustaining Growth in Troubled Times.” NBER President James M. Poterba of MIT organized the conference jointly with Suman Bery of NCAER.

The NBER participants in this meeting were: NBER Board Chairman John Clarkeson; John Y. Campbell, Mihir A. Desai, and Martin S. Feldstein, NBER and Harvard University; Don Fullerton, NBER and University of Illinois; Roger H. Gordon and Michelle J. White, NBER and University of California, San Diego; Abhijit Banerjee, Simon Johnson, and James M. Poterba, NBER and MIT; Anne O. Krueger, NBER and Johns Hopkins University; and Emily Oster, Raghuram Rajan, and Jesse M. Shapiro, NBER and University of Chicago.

The participants focused their discussion on a number of issues that are related to the current global economic situation. Montek Ahluwalia, Deputy Chairman of the Indian Planning Committee, provided an overview of current economic developments and policy challenges in India. Clarkeson and Feldstein, along with Tarun Das of the Confederation of Indian Industry, participated in a panel discussion on the U.S. economy and the near-term economic outlook. Other core topics that were discussed during the two-day meeting included fiscal policy and monetary management, trade and tariff policy, and the role of social programs in provid- ing social protection for low-income families in both India and the United States.

NBER Reporter • 2009 Number 1 25 Economic Growth

NBER Faculty Research Fellows Matthias Doepke, University of California, Los Angeles, and Benjamin Jones, Northwestern University, organized a “Conference on Economic Growth,” which took place at the Federal Reserve Bank of San Francisco on February 12. These papers were discussed:

Michele Boldrin, Washington University, and David K. Levine, NBER and Washington University, “Quality Ladders, Competition, and Endogenous Growth”

Diego Comin, NBER and Harvard Business School, and Bart Hobijn, Federal Reserve Bank of San Francisco, “An Exploration of Technology Diffusion” (NBER Working Paper No. 12314)

Alexander Monge, Pennsylvania State University, “Foreign Firms, Domestic Entrepreneurial Skills, and Development”

Alice Schoonbroodt, University of Southampton, and Michele Tertilt, NBER and Stanford University, “Who Owns Children and Does It Matter?”

Francesco Caselli, NBER and London School of Economics, and Guy Michaels, London School of Economics, “Resource Abundance, Development, and Living Standards: Evidence from Oil Discoveries in Brazil”

Hyeok Jeong, Vanderbilt University, and Yong Kim, University of Southern California, “Complementarity and Transition to Modern Economic Growth”

Summaries of these papers may be found at http://www.nber.org/confer/2009/EGCs09/summary.html

✴

26 NBER Reporter • 2009 Number 1 NBER News

NBER Researchers in Public Service

Research Associate Christina D. member at Harvard University, is the of Harvard University and former NBER Romer, the former co-director of NBER’s new director of the National Economic Board member Roger Ferguson of TIAA- Program on Monetary Economics, a for- Council. CREF are also members of this new mer member of the Business Cycle Dating Research Associates Cecilia E. Rouse Board. Committee, and a faculty member at the of Princeton University and Austan Research Associate Jeremy C. Stein University of California, Berkeley, has Goolsbee of the University of Chicago of Harvard University is a Senior Advisor been confirmed as Chair of the President’s have been nominated as members of the to the Secretary of the Treasury. And, Council of Economic Advisers. Council of Economic Advisers. Goolsbee Research Associate Jeffrey B. Liebman, Research Associate Lawrence H. will also serve as the chief economist and also of Harvard, is serving as Executive Summers, a member of several NBER staff director for the President’s Economic Associate Director of the Office of Programs including Public Economics Recovery Advisory Board. NBER Management and Budget. and Monetary Economics and a faculty President-Emeritus Martin S. Feldstein

2008–9 Awards and Honors A number of NBER researchers and was appointed to the editorial board Society for Global Health. received honors, awards, and other forms of the Journal of Health Economics. Nicholas Bloom and Alex Mas won of professional recognition during 2008 Anirban Basu received the Alan the Labor and Employment Relations and early 2009. A list of these honors, Williams Health Economics Fellow­ Association’s 2007 John T. Dunlop excluding those that were bestowed by ship for 2008 from the University of York Scholar Award for “outstanding aca- the researcher’s home university and list- (U.K.) He was also the 2008 Program demic contributions to research by recent ing researchers in alphabetical order, is Chair for the Health Policy Statistics entrants to the field.” presented below. Section of the American Statistical Jeffrey R. Brown was nomi- Yacine Ait-Sahalia was elected a Association. nated by President Bush for the posi- Fellow of the American Statistical Assoc­ Paul Beaudry was awarded the 2008 tion of Public Trustee of the Social iation and the Journal of Econometrics John Rae prize for the most significant Security and Medicare Trust Funds; he and delivered the Hermann Otto contribution to economic research in was awarded the Early Career Scholarly Hirshfeld Lectures, Humboldt University, Canada over the last five years. Achievement Award from the American Berlin, in 2008; she also received the David Blanchflower presented the Risk and Insurance Association; and he John Simon Guggenheim Memorial Esmée Fairbairn Memorial Lecture at and Amy Finkelstein received the 2008 Fellowship for 2008–9. Lancaster University on October 30, TIAA-CREF Paul A. Samuelson Award Elizabeth Oltmans Ananat became 2007. for their paper, “The Interaction of Public a Research Associate of the National Rebecca Blank received the and Private Insurance: Medicaid and the Poverty Center at the University of Distinguished Alumni Award at the Long-Term Care Insurance Market.” Michigan. University of Minnesota. She also will Richard V. Burkhauser is President- Torben Andersen was elected a deliver the Sulzberger Lecture at Duke Elect of the Association for Public Policy Fellow of the Econometric Society dur- University’s Sanford Institute of Public Analysis and Management. His term ing 2008. Policy. will begin in 2010. He also received Ashish Arora served on the Francine D. Blau has been named the 2008 Felix Buchal Award for schol- “Measuring Innovation in the 21st one of the five inaugural Academic arship in interdisciplinary longitudi- Century” Economy Advisory Committee Fellows of the Labor and Employment nal research using the German Socio- to the U.S. Secretary of Commerce. Relations Association (LERA). Economic Panel. Finally, in 2008 he was Katherine Baicker was elected to the David Bloom was appointed an the R.I. Downing Visiting Fellow at the Board of Directors of Academy Health Ambassador in the Paul G. Rogers University of Melbourne.

NBER Reporter • 2009 Number 1 27 John Y. Campbell received an hon- Economics at Columbia University. Association for Business Economics. orary doctorate from the University Steven J. Davis has been appointed Amy Finkelstein received the 2008 of Maastricht, in the Netherlands, on the Editor of the American Economic American Economic Association’s Elaine January 8, 2009. Journal: Macroeconomics. Bennett Research Prize “awarded every Dennis W. Carlton, and his co- Angus Deaton received an honorary other year to recognize, support, and authors Joshua Gans and Michael Doctor of Letters from the University of encourage outstanding contributions by Waldman, received the Robert F. St. Andrews (Scotland) in June 2008. young women in the economics profes- Lanzilotti Prize, awarded by the Peter DeMarzo won the Michael sion.” She and Jeffrey R. Brown also International Industrial Organization Brennan Award Second-Place Prize for his received the 2008 TIAA-CREF Paul Society, for the Best Paper in Antitrust paper “Agency and Optimal Investment A. Samuelson Award for their paper: Economics. Their paper is “Why Tie Dynamics” (with Michael Fishman), pub- “The Interaction of Public and Private a Product Consumers Do Not Use?” lished in the Review of Financial Studies. Insurance: Medicaid and the Long-Term (NBER Working Paper No. 13339). He was also elected Vice President of the Care Insurance Market” (AER, June 2008 Amitabh Chandra won the Eugene Western Finance Association. and NBER Working Paper No. 10989). Garfield award with Douglas O. Staiger Mihir A. Desai, C. Fritz Foley, and Price Fishback was appointed for the Impact of Medical Research. He James R. Hines, Jr. co-authored the co-editor of the Journal of Economic was named a co-editor of the Journal of paper “Dividend policy inside the multi- History. Human Resources and is on the editorial national firm” (NBER Working Paper No. Richard Freeman has been named board of the American Economic Review 8698), which was awarded the Pearson one of the five inaugural Academic (Applied). Prize by the Financial Management Fellows of the Labor and Employment Raj Chetty received an Alfred P. Association. Another paper by Desai, Relations Association (LERA). Sloan research fellowship and the Alexander Dyck, and Luigi Zingales, Ken Froot won the Witt Award American Young Economist Award. titled “Theft and Taxes” (NBER Working of the American Risk and Insurance Richard H. Clarida, Jordi Gali, Paper No. 10978), was a second place win- Association for his article, “Risk and Mark Gertler won the First Prize ner of the Jensen Prize for the Best Papers Management, Capital Budgeting, and Award for Best Paper presented at Published in the Journal of Financial Capital Structure Policy for Insurers and the NBER’s International Seminar on Economics in the Areas of Corporate Reinsurers” (NBER Working Paper No. Macroeconomics during its first 25 years Finance and Organizations, 2007. 10184). (on the occasion of ISOM’s thirtieth Erwin Diewert became a fellow of Nicola Fuchs-Schündeln was anniversary in June 2008). The paper, the American Economic Association. awarded the “2008 DAAD Prize for written in 1998, was “Monetary policy Darrell Duffie was elected President Distinguished Scholarship in German rules in practice - Some international of the American Finance Association. and European Studies” by the American evidence” (NBER Working Paper No. He also gave the Princeton Lectures in Institute for Contemporary German 6254). Finance, and was the Nash Lecturer at Studies. Courtney Coile became a mem- Carnegie-Mellon University. Mikhail Golosov received an NSF ber of the National Academy of Social Esther Duflo received the “Prix Luc CAREER grant, the NSF’s major award Insurance. Durand-Reville” from the Academie des in support of early-career scientists, and a Philip J. Cook and Jacob Vigdor Sciences Morale et Politiques, France in Sloan Fellowship, in 2008. (with co-authors Robert MacCoun and 2008. Robert J. Gordon delivered the A. Clara Muschkin) received the Raymond William Easterly won the Hayek W. H. Phillips invited lecture at the Vernon Memorial Prize for the best arti- Award for Best Book of the Last Two Australasian meetings of the Econometric cle published in the Journal of Policy Years from the Manhattan Institute and Society in July 2008. He also presented Analysis and Management in 2008. delivered the Dobriansky Lecture at invited lectures at the 14th Dubrovnik Mario J. Crucini was appointed to a Georgetown University. Economic Conference in Dubrovnik, three-year term as Associate Editor of the Ronald G. Ehrenberg was awarded Croatia, in June; at the International Journal of Monetary Economics, effective an honorary Doctor of Science degree Workshop on Chinese Productivity, July 1, 2008. He became a Senior Fellow by the State University of New York Hangzhou, China, in September; and at the Globalization and Monetary Policy and was named a fellow of the American at the Seventh Macroeconomic Policy Institute, Federal Reserve Bank of Dallas, Educational Research Association. Research Workshop of the Hungarian in June 2008. He was also the Weatherall Isaac Ehrlich was appointed by Central Bank and CEPR, Budapest, in Visiting Fellow, Queen’s University, Governor David Patterson to serve on October. Canada, from July 1–August 30, 2008. his new Council of Economic Advisers. Pierre-Olivier Gourinchas won the Janet Currie is the inaugural Martin Feldstein received the 2008 prize for the best young French holder of the Sami Mnaymneh Chair in 2008 Butler Award from the New York economist (under age 40) awarded by

28 NBER Reporter • 2009 Number 1 Cercle des Economistes and Le Monde. College of Business, Drexel University. the inaugural Distinguished Finance Michael Grossman was awarded the Anil K Kashyap, Raghuram Rajan, Educator Award by the Financial 2008 Victor Fuchs Award. Also, Nankai and Jeremy C. Stein co-authored Education Association in September University has announced the establish- NBER Working Paper No. 6962, which 2008. ment of the Nankai-Grossman Center was selected by the New York Times Olivia S. Mitchell was awarded of Health Economics and Medical Magazine in December 2008 as one of the 2008 Roger F. Murray Prize from Insurance. Professor Grossman will the “ideas of the year.” the Institute for Quantitative Research serve as the Center’s Honorary Director John Kennan was elected a Fellow in Finance. She also received the and Chairperson of its Academic of the Society of Labor Economists. 2008 Carolyn Shaw Bell Award of the Committee. Michael Kremer was named a fel- Committee on the Status of Women in Philip Haile was named a Fellow of low of the Econometric Society. the Economics Profession. the Econometric Society. Alan B. Krueger was awarded the Randall Morck received the J. Michael Haines was recently elected Susan C. Eaton Award by the Labor and Gordin Kaplan Award for Excellence in Vice President of the Economic History Employment Relations Association. The Research from the University of Alberta Association. award “is given annually to a member of in 2008. Robert E. Hall was selected to run the Association who has achieved distinc- Enrico Moretti won the Carlo unopposed for President of the American tion as both a scholar and practitioner in Alberto Medal, presented to an Italian Economic Association, was elected, and the labor field.” He also was selected economist under the age of 40 for out- will serve for the year 2010. to give the Distinguished Lecture in standing research contributions to the James D. Hamilton presented the Economics and Government by the field of economics. Esther Peterson Lecture at the Joint Society of Government Economists at Dale Mortensen was named a Meeting of the American Agricultural the January 2009 ASSA meeting. Distinguished Fellow of the American Economics Association and the American Naomi Lamoreaux is president-elect Economic Association. Council on Consumer Interests. In addi- of the Economic History Association; Aldo Musacchio won the Arthur tion, he was keynote speaker at the 5th she will become president in September H. Cole Prize in 2008, awarded by the Colloquium on Modern Tools for Busi­ 2009. Economic History Association for the ness Cycle Analysis in Luxembourg and Eric Michael Leeper spent two best paper published in the Journal of at the Sixteenth Annual Symposium of months as a Professorial Fellow in Economic History between September the Society for Nonlinear Dynamics and Monetary Economics and Finance at 2007 and September 2008. Econometrics. Victoria University Wellington and the Joseph P. Newhouse was named Eric A. Hanushek became a Fellow Reserve Bank of New Zealand. to the (U.S.) Comptroller General’s of the American Educational Research Jonathan D. Levin was elected a Advisory Committee. Association; he also became chair of Fellow of the Econometric Society. Gerard Padro i Miquel received the the Board of Directors for the National Nuno Limão received the Kiel ESRC (Economic and Social Research Board for Education Sciences. Institute’s “Excellence Award in Global Council) first grant in July 2008, the Geoffrey Heal received an Honorary EconomicAffairs”, which aims “to build a organization’s main award in support of Doctorate from the University of Paris, community of the brightest young early-career researchers. The ESRC is the Dauphine, in November 2008. researchers in the area of global eco- main research council for social sciences James J. Heckman became a nomic affairs.” in the U.K. Resident Member of the American Jin-Tan Liu received the Academic Lubos Pastor received the Philosophical Society, a Fellow of the Award from the Ministry of Education, NASDAQ Award for the best paper on International Statistical Institute, and a Taiwan in 2008. capital formation at the 2008 Western Fellow of the American Association for Robert A. Margo is a Visiting Finance Association conference, for the Advancement of Science, in 2008. Scholar at the Russell Sage Foundation, “Entrepreneurial learning, the IPO deci- Takatoshi Ito becomes the fourth New York City, from September 2008 sion, and the post-IPO drop in firm prof- holder of the Tun Ismail Ali Chair of until June 2009. itability” (NBER Working Paper No. Faculty of Economics at the University Alexandre Mas and Nicholas 12792), joint with Lucian Taylor and of Malaya. Bloom received the John T. Dunlop Pietro Veronesi. He also received the Michael C. Jensen received the 2007 award from the Labor and Employment 2008 Q Group Research Award, for “Are Herbert Simon Award for outstanding Relations Association for outstanding stocks really less volatile in the long run?” contributions to business research by research by someone who has received joint with Robert F. Stambaugh. Rajk Laszlo College, Budapest, Hungary, a terminal degree within the last ten Monika Piazzesi became a Fellow of and the Dean’s Leadership Award in years. the Econometric Society. Corporate Governance from LeBow Robert C. Merton was given Rob Porter was selected as the

NBER Reporter • 2009 Number 1 29 Distinguished Fellow of the Industrial Kathryn Shaw delivered the Economics Dynamics in January 2009. Organization Society for 2009. Adam Smith Lecture, European Labor Joel Waldfogel won the “Best James Poterba is the President of Economics Association, in Amsterdam Article of the Year” Prize for 2006 from the National Tax Association and Vice- in September 2008; she also was the Journal of Industrial Economics. The President of the American Economic elected a Fellow of the Society of Labor article, co-authored with Lu Chen, is Association. Economists. “Does Information Undermine Brand? Edward C. Prescott was elected a Pablo T. Spiller is the President- Information Intermediary Use and member of the National Academy of Elect of the International Society for Preference for Branded Web Retailers” Science. New Institutional Economics. (NBER Working Paper No. 9942). Manju Puri won the Fama/DFA Robert W. Staiger was elected a Michael Whinston’s article with Prize for the best paper in the Journal Fellow of The Econometric Society in Ilya Segal, “Antitrust in Innovative of Financial Economics in the Areas of 2008. Industries” (NBER Working Paper No. Capital Markets and Asset Pricing (the Richard Sutch received the 2008 11525 and American Economic Review, second prize award) for “Optimism Thomas Jefferson Award from the Society 97, December 2007, pp. 1703–30), won and Economic Choice” (with David for History in the Federal Government. the 2008 Compass Lexicon Prize for best Robinson), NBER Working Paper No. John Tyler was one of three national article on antitrust economics. 11361. recipients of the Senior Urban Research Jeffrey G. Williamson presented Yi Qian received a research grant Fellowship from the Council of Great the: Joseph Fisher Public Lecture, from the Ewing Marion Kauffman City Schools for research he will be University of Adelaide (Australia); Trevor Foundation in the summer of 2008. conducting in the Cincinnati School Swan Distinguished Lecture, Australian “Does Ethnic Link Pay?” written with District on how teacher use of student National University; John Hicks Lec­ Yasheng Huang (MIT) and Li Jin (HBS), performance data, as provided by a web- ture, Oxford University; Fundacion won the best paper award at the TCFA based application, affects their classroom Areces Public Lecture, Madrid; and the (The Chinese Financial Association) instruction. University Inaugural Lecture, University conference in November 2008. Pietro Veronesi’s paper, Pompeu Fabra (Spain). Daniel L. Rubinfeld was awarded an “Entrepreneurial Learning, the IPO Richard Zeckhauser’s article with honorary doctorate from the University Decision, and the Post-IPO Drop in Firm Richard Frank, “Custom-Made Versus of Basel, Switzerland, in November Profitability” (NBER Working Paper No. Ready-to-Wear Treatments: Behavioral 2008. 12792), with Lubos Pastor (University of Propensities in Physicians’ Choices” Antoinette Schoar was awarded Chicago and NBER) and Lucian Taylor (NBER Working Paper No. 13445), won the Ewing Marion Kauffman Prize (University of Pennsylvania) won the the National Institute of Health Care Medal for Distinguished Research in NASDAQ Award for the best paper on Management’s best paper of the year Entrepreneurship, 2009. capital formation at the 2008 Western award for 2008. Zeckhauser also was the José A. Scheinkman was elected Finance Association conference. Lee Kwan Yew Distinguished Visitor to a Member of the National Academy of Gianluca Violante will become Singapore in 2008. Sciences. Managing Editor of The Review of

Program Director News After 19 years as Director of the that university’s Economics Department Program, Research Associate Richard NBER’s Program on Health Care, and an NBER affiliate since 1991, will Freeman of Harvard University will also Research Associate Alan M. Garber of become Director of the NBER’s Economic step down as program director on June Stanford University will step down on Well-Being of Children Program on 30, 2009. His successor will be Research June 30, 2009. His successor will be July 1, 2009. An active member of that Associate David Card of the University Jonathan Gruber of MIT, who has served Program for many years, Currie also served of California at Berkeley, who has been as Director of the NBER’s Program on as Program Director in 1997–8 while a member of the Labor Studies Program the Economic Well-Being of Children Gruber was a Deputy Assistant Secretary since 1982. Card has held faculty posi- since 1996. at the U.S. Treasury Department. tions at the University of Chicago and Research Associate Janet Currie of After more than thirty years as Princeton University as well as Berkeley. Columbia University, the current chair of Director of the NBER’s Labor Studies

30 NBER Reporter • 2009 Number 1 Victor Zarnowitz Dead at Age 89 Victor Zarnowitz, a long-time since 1999. He also was Professor Emeritus Life and Times, Zarnowitz received his NBER Research Associate and a mem- of Economics and Finance, Graduate Ph.D. from the University of Heidelberg ber of the NBER’s Business Cycle Dating School of Business, The University of in 1951. He joined the NBER in New Committee, passed away on February 21 Chicago. York City in 1952, and his work con- at the age of 89. At the time of his After a harrowing and remarkable tributed greatly to the study of economic death, Zarnowitz was Senior Fellow and youth, chronicled in his 2008 autobi- fluctuations. Economic Counselor to The Conference ography Fleeing the Nazis, Surviving the Board in New York, where he had worked Gulag, and Arriving in the Free World: My

Program and Working Group Meetings

Measuring Economic Activity in Markets for Ideas, Innovation, and Other Intangibles

The NBER’s Program on Technological Change and Productivity Measurement met in Cambridge on December 5. Program Director Ernst R. Berndt of NBER and MIT and Wesley M. Cohen of NBER and Duke University organized the meeting. The fol- lowing topics were discussed:

• James Bessen, Boston University, “More Machines or Better Machines”

• Rachel Soloveichik, Bureau of Economic Analysis, “Theatrical Movies as a Capital Asset”

• Jaison R. Abel, Federal Reserve Bank of New York, and Todd M. Gabe, University of Maine, “Human Capital and Economic Activity in Urban America”

• Dietmar Harhoff, University of Munich, “Patent Families, Equivalents, and Patent Value”

• Carol A. Robbins, Bureau of Economic Analysis, “Measuring Payments for the Supply and Use of Intellectual Property”

• Panel Discussion on “Useful and Needed Research on Measuring Economic Activity in Markets for Ideas, Innovation, and Other Intangibles” Chair: Wesley M. Cohen Panel: Avi Goldfarb, University of Toronto; Jonathan Haskell, Imperial College Business School; Charles R. Hulten, University of Maryland and NBER; and Margaret Blair, Vanderbilt University

• Carol Corrado, The Conference Board, “Report of the NSF/Conference Board Workshop on Innovation Data”

• Julie Lane, National Science Foundation, “Improvements and Future Challenges for the Research Infrastructure: Administrative Transaction Data”

Summaries of these papers may be found at: http://www.nber.org/confer/2008/prf08/summary.html

NBER Reporter • 2009 Number 1 31 International Trade and Investment

The NBER’s Program on International Trade and Investment met at the University of California, San Diego on December 5 and 6. Program Director Robert C. Feenstra of the University of California, Davis chose these papers to discuss:

• Costas Arkolakis, Yale University and NBER, “Market Penetration Costs and Trade Dynamics”

• Costas Arkolakis, and Marc-Andreas Muendler, University of California, San Diego and NBER, “The Extensive Margin of Exporting Goods: A Firm-Level Analysis”

• Pravin Krishna, Johns Hopkins University and NBER, and Mine Z. Senses, Johns Hopkins University, “International Trade and Labor Income Risk in the United States”

• Jeffrey Grogger, University of Chicago and NBER, and Gordon H. Hanson, University of California, San Diego and NBER, “Income Maximization and the Selection and Sorting of International Migrants”

• Dhammika Dharmapala, University of Connecticut; C. Fritz Foley, Harvard University and NBER, and Kristin J. Forbes, MIT and NBER, “The Unintended Consequences of the Homeland Investment Act: Implications for Financial Constraints, Governance, and International Tax Policy”

• Ram C. Acharya, Industry Canada, and Wolfgang Keller, University of Colorado and NBER, “Estimating the Productivity Selection and Technology Spillover Effects of Imports” NBER( Working Paper No. 14079)

• Pinelopi K. Goldberg, Princeton University and NBER; Amit Khandelwal, Columbia University; Nina Pavcnik, Dartmouth College and NBER, and Petia Topalova, IMF, “Imported Intermediate Inputs and Domestic Product Growth: Evidence from India” (NBER Working Paper No. 14416)

• Stefania Garetto, Princeton University, “Input Sourcing and Multinational Production”

Summaries of these papers may be found at: http://www.nber.org/confer/2008/itif08/summary.html

Law and Economics

NBER’s Law and Economics Program met in Cambridge on February 6. Program Director Christine Jolls, NBER and Yale Law School, organized the meeting. These papers were discussed:

• Rosalind Dixon, University of Chicago Law School, and Richard Holden, MIT and NBER, “Amending the Constitution: Article V and the Effect of Voting Rule Inflation”

• Vikrant Vig, London Business School, “Access to Collateral and Corporate Debt Structure: Evidence from a Natural Experiment”

• Alberto Galasso, University of Toronto, and Mark Schankerman, London School of Economics, “Patent Thickets and the Market for Innovation: Evidence from Settlement of Patent Disputes”

• Haresh Sapra, University of Chicago; Ajay Subramanian, Georgia State University; and Krishnamurthy Subramanian, Emory University, “Corporate Governance and Innovation: Theory and Evidence”

• Edward L. Glaeser, Harvard University and NBER, and Gergely Ujhelyi, University of Houston, “Regulating

32 NBER Reporter • 2009 Number 1 Misinformation” (NBER Working Paper No. 12784)

• Marco Ottaviani and Abraham L. Wickelgren, Northwestern University, “Approval Regulation with Learning”

• Ernesto Dal Bo and Marko Tervio, University of California, Berkeley and NBER, “Self-Esteem, Moral Capital, and Wrongdoing” (NBER Working Paper No. 14508)

• Daniel L. Chen and Jasmin Sethi, Harvard University, “The Effects of Sexual Harassment Law on Gender Inequality”

Summaries of these papers may be found at: http://www.nber.org/confer/2009/LEs09/summary.html

Economic Fluctuations and Growth

NBER’s Program on Economic Fluctuations and Growth met at the Federal Reserve Bank of San Francisco on February 13. NBER Research Associates Peter J. Klenow of Stanford University and Valerie A. Ramey, University of California, San Diego, orga- nized the meeting. These papers were discussed:

• Francisco J. Buera, NBER and University of California, Los Angeles; Joseph P. Kaboski, Ohio State University; and Youngseok Shin, Washington University in St. Louis, “Finance and Development: A Tale of Two Sectors”

• Diego Comin, NBER and Harvard University; Mark Gertler, NBER and New York University; and Ana Maria Santacreu, New York University, “Technology Innovation and Diffusion as Sources of Output and Asset Price Fluctuations”

• Glenn D. Rudebusch and Eric T. Swanson, Federal Reserve Bank of San Francisco, “The Bond Premium in a DSGE Model with Long-Run Real and Nominal Risks”

• Eric M. Leeper, NBER and Indiana University; Todd B. Walker, Indiana University, and Shu-Chun Susan Yang, Congressional Budget Office, “Fiscal Foresight and Information Flows” (NBER Working Paper No. 14630)

• Alessandra Fogli, Federal Reserve Bank of Minneapolis, and Laura Veldkamp, NBER and New York University, “Nature or Nurture? Learning and the Geography of Female Labor Force Participation” (NBER Working Paper No. 14097)

• James Feyrer, Dartmouth College, “Trade and Income — Exploiting Time Series in Geography”

Summaries of these papers may be found at: http://www.nber.org/confer/2009/EFGw09/summary.html

Industrial Organization

The NBER’s Program on Industrial Organization, directed by Nancy L. Rose of MIT, met at the NBER’s California offices on February 20-21. Matthew Gentzkow, NBER and University of Chicago, and Matthew White, NBER and University of Pennsylvania, organized the meeting. These papers were discussed:

• Justine S. Hastings, Yale University and NBER, “Wholesale Price Discrimination and Regulation: Implications for Retail Gasoline Prices”

NBER Reporter • 2009 Number 1 33 • Nuno Cassola, European Central Bank; Ali Hortaçsu, University of Chicago and NBER; and Jakub Kastl, Stanford University, “The 2007 Subprime Market Crisis in the Euro Area through the Lens of ECB Repo Auctions” • Liran Einav, Stanford University and NBER; Amy Finkelstein, MIT and NBER; and Mark R. Cullen, Yale University, “Estimating Welfare in Insurance Markets Using Variation in Prices” (NBER Working Paper No. 14414)

• Jean-Pierre Dubé and Jeremy T. Fox, University of Chicago and NBER, and Che-Lin Su, University of Chicago, “Improving the Numerical Performance of BLP Static and Dynamic Discrete Choice Random Coefficients Demand Estimation”

• Andrew Sweeting, Duke University and NBER, “Equilibrium Price Dynamics in Perishable Goods Markets: The Case of Secondary Markets for Major League Baseball Tickets” (NBER Working Paper No. 14505)

• Patrick Bajari, University of Minnesota and NBER, and Gregory Lewis, Harvard University, “Procurement Contracting with Time Incentives: Theory and Evidence”

• Meghan R. Busse, Northwestern University and NBER; Christopher R. Knittel, University of California, Davis and NBER; and Florian Zettelmeyer, Northwestern University and NBER, “Pain at the Pump: How Gasoline Prices Affect Automobile Purchasing in New and Used Markets”

Summaries of these papers may be found at: http://www.nber.org/confer/2009/IOs09/summary.html

✴

34 NBER Reporter • 2009 Number 1 Bureau Books

Developments in the Economics of Aging Developments in the Economics of the economic issues that will confront ers of 401(k)s. It also addresses the com- Aging, edited by David A. Wise, will be policymakers as they seek to design the plicated relationship between health and available from the University of Chicago appropriate policies to protect the eco- economic status, including why health Press this spring for $99.00. nomic and physical health of these older behaviors vary across populations and The number of Americans eligible Americans. how socioeconomic measures correlate to receive Social Security benefits will This volume looks at such topics as with health outcomes. increase from 45 million to nearly 80 mil- what factors influence work and retire- Wise directs the NBER’s Program lion in the next twenty years. Retirement ment decisions at older ages; the changes on the Economics of Aging and is John systems therefore must adapt to meet in life satisfaction associated with retire- F. Stambaugh Professor of Political the demands of the largest aging pop- ment; and the shift in responsibility for Economy at Harvard’s Kennedy School ulation in our nation’s history. In this managing retirement assets, from pro- of Government. NBER Conference Volume, Wise and a fessional money managers of traditional distinguished group of analysts examine pension plans to individual account hold-

Financial Sector Development in the Pacific Rim Financial Sector Development in the finance, providing serious competition reform. The comparative analysis offered Pacific Rim, Volume 18 in the NBER- to the more traditional financial centers here can help to answer broad questions East Asia Seminar on Economics (EASE) of London and New York. This book about economic development and the series, will be available this April from the provides a rich collection of theoretical future of Asia. University of Chicago Press. The editors, and empirical analyses of the growing Ito and Rose are NBER Research who serve as organizers of the EASE con- capital markets in the region. It brings Associates in the Program on International ferences, are Takatoshi Ito and Andrew together authors from various East Asian Finance and Macroeconomics. Ito is also K. Rose. and Pacific nations, and examines the a professor of economics at the University The reform in Asian financial sec- institutional factors influencing financial of Tokyo. Rose is a professor of eco- tors — especially in banking and stock innovation; the consequences of financial nomics at The Haas School of Business, markets — has been remarkable since development; widespread consolidation University of California, Berkeley. the currency crisis of 1997–8. East Asia occurring through mergers and acquisi- The price of the clothbound volume is now a major player in international tions; and the implementation of policy will be $99.00.

NBER Reporter • 2009 Number 1 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