The NBER Reporter 2007 Number 2: Program and Working Group Meetings
In order to study the impact of offshoring on sectoral and economy-wide rates of unemployment, Mitra and Ranjan construct a two-sector general equilibrium model in which labor is mobile across the sectors and unemployment is caused by search frictions. They find that, contrary to general perception, wages increase and sectoral unemployment declines because of offshoring. This result can be understood to arise from the productivity enhancing (cost reducing) effect of offshoring. If the search cost is identical in the two sectors, or even if the search cost is higher in the sector that experiences offshoring, the economy-wide rate of unemployment decreases. The researchers also find multiple equilibrium outcomes in the extent of offshoring and therefore, in the unemployment rate. Furthermore, a firm can increase its domestic employment through offshoring. Also, such a firm's domestic employment can be higher than that of a firm choosing to remain fully domestic. When they modify the model to disallow intersectoral labor mobility, the negative terms-of-trade effect on the sector in which firms offshore some of their production activity becomes stronger. In such a case, it is possible for this effect to offset the positive productivity effect, and to result in a rise in unemployment in that sector. In the other sector, offshoring has a much stronger unemployment reducing effect in the absence of intersectoral labor mobility than in the presence of it. Finally, allowing for an endogenous number of varieties in the offshoring sector provides an additional indirect channel through which sectoral unemployment goes down.
Falling costs of coordination and communication have allowed firms in rich countries to fragment their production process and offshore an increasing share of the value chain to low-wage countries. Popular discussions about the aggregate impact of this phenomenon on rich countries have stressed either a (positive) productivity effect associated with increased gains from trade or a (negative) terms-of-trade effect linked with the vanishing effect of distance on wages. Rodriguez-Clare proposes a Ricardian model where both of these effects are present. He analyzes the effects of increased fragmentation and offshoring in the short run and in the long run (when technology levels are endogenous). The short-run analysis shows that when fragmentation is sufficiently high, further increases in fragmentation lead to a deterioration (improvement) in the real wage in the rich (poor) country. But the long-run analysis reveals that these effects may be reversed as countries adjust their research efforts in respon to increased offshoring. In particular, the rich country always gains from increased fragmentation in the long run, whereas poor countries see their static gains partially eroded by a decline in their research efforts.
Foreign-owned firms are often hypothesized to generate productivity "spillovers" to the host country, but both theoretical micro-foundations and empirical evidence for this are limited. Malchow-Moller, Markusen, and Schjerning develop a heterogeneous-firm model in which ex-ante identical workers learn from their employers in proportion to the firm's productivity. Foreign-owned firms have, on average, higher productivity in equilibriumbecause of entry costs, which means that low-productivity foreign firms cannot enter. Foreign firms have higher wage growth and, with some exceptions, pay higher average wages, but not when compared to similarly large domestic firms. The empirical implications of the model are tested on matched employer-employee data from Denmark. The authors find considerable evidence of higher wages and wage growth in foreign-owned firms; these effects are significantly reduced but generally not eliminated when controlling for firm size. The results are largely consistent with their model. Furthermore, productivity transfers appear to be important, as experience from foreign-owned and large firms is clearly beneficial for subsequent wages and self-employment earnings.
Product-level data on bilateral U.S. exports exhibit two strong patterns. First, most potential export flows are not present and the incidence of these "export zeros" is strongly correlated with distance and importing country size. Second, export unit values are positively related to distance. Baldwin and Harrigan show that every well-known multi-good general equilibrium trade model is inconsistent with at least some of these facts. They also offer direct statistical evidence of the importance of trade costs in explaining zeros, using the long-term decline in the cost of air shipment to identify a difference-in-differences estimator. To match these facts, they propose a new version of the heterogeneous-firms trade model pioneered by Melitz (2003). In their model, high quality firms are the most competitive, with heterogeneous quality increasing with firms' heterogeneous cost.
Lileeva and Trefler weigh into the debate about whether rising productivity is ever a consequence, rather than a cause, of exporting. They argue that improved market access should induce plants to invest in productivity-enhancing activities. Further, improved market access should not induce all plants to invest, only those that expect the largest returns to such investments. That is, there should be heterogeneous responses to improved market access. Using data on plant-specific tariff concessions afforded Canadian plants under the Canada-U.S. Free Trade Agreement as an instrument for exporting, the researchers compare the performance of new (post-Agreement) exporters with non-exporters. They find that relative to non-exporters, new exporters experienced higher rates of labor productivity growth and higher rates of investment in product development and advanced manufacturingtechnologies.
Why does FDI travel from rich to poor countries but non-FDI not do so? Could financial and property rights institutions play different roles in understanding capital flows? Unbundling both institutions and capital flows, Ju and Wei propose a simple model for studying the relationship between various domestic institutions financial system, corporate governance, and property rights protection -- and patterns of international capital flows. They describe conditions under which inefficient financial systems and poor corporate governance in a country may be bypassed by two-way capital flows in which domestic savings would leave the country in the form of outflows of financial capital but domestic investment would take place via the inflows of foreign direct investment. In this framework, financial, and property rights institutions can have different effects on capital flows.
Keller and Shiue study the emergence of the increasingly unified commodity market in Europe in the nineteenth century. During this period, there were major institutional changes in the form of currency agreements and the Zollverein customs liberalizations, as well as transport cost reductions in the form of building railways. In assessing the relative importance of these factors, the setting here has a number of clear advantages over existing studies. For one, almost all economies in this sample experienced changes over the course of the nineteenth century. Currency or trade arrangements did not exist between any of the states in the early 1800s, whereas by the closing years of the nineteenth century they existed between all German states. Similarly, railroads did not exist before the 1830s, whereas by the end of the century trains had arrived almost everywhere in the sample. The authors study changes in market integration in terms of the spatial dispersion of grain prices in 68 markets with about 10,000 observations, located in five different countries and fifteen different German states. They find that the emergence of integrated commodity markets in nineteenth century Europe is, in a major part, attributable to the transportation revolution in the form of the railways. Over a relatively short time horizon, the effect of customs liberalization is comparable in size, whereas in the long run, the impact of railways is larger. The researchers do not estimate a significant effect of currency agreements on market integration. Their results suggest that as significant as institutional factors were for the expansion of markets, technology factors may have been even more important.
During the 1990s, U.S. welfare policy underwent dramatic reforms aimed at promoting employment and reducing dependence on public assistance. While the immediate effects of welfare reform on adult employment and family income have been studied extensively, little is known about its long-run effect on the wellbeing of children in low-income families. Miller and Zhang evaluate the impact of welfare reform on the academic performance of children in low-income families using nationally representative data on math achievement that spans a ten-year period starting just prior to welfare reform. In a generalized difference-in-differences framework, they use children from more affluent families as a control group to simulate the unobservable counterfactual time trend. The key finding is an improvement in the relative math scores of fourth-grade students from low-income families, which is statistically significant for 2003 and 2005. No consistent differences are found by race, sex, or ability. The treatment effect is smaller for eighth-grade students, but the benefits from early exposure appear to persist and even increase with age. Preliminary analysis suggests that these beneficial effects are associated with relative improvements in low-income children's time use and interaction with parents.
The National Board for Professional Teaching Standards (NBPTS) uses videos and essays to assess teachers' effectiveness. In elementary schools in Los Angeles, they find, classrooms randomly assigned to more highly rated teachers outperform their comparison group by a wider margin than classrooms taught by poorly rated teachers. Analogous to Lalonde (1986), Kane and his co-authors compare experimental and non-experimental estimates of teacher impacts (both for the same sample of teachers in earlier years as well as for a non-experimental comparison group). The estimates yielded by the non-experimental techniques are similar. By validating against student performance, Kane and his co-authors also make a number of suggestions for improving the NBPTS scaling process.
Many of the "choice" schools to which some U.S. students now have access appear popular with parents. Yet recent studies have found that the causal effect of attending these schools on student achievement is at best small, at least on "basic" outcomes such as statewide test scores. One explanation for strong parental demand and weak achievement effects is that these schools have larger effects on "high-achievement" outcomes such as (for high schools) the probability of taking Advanced Placement courses and college admission prospects. Clark considers the impact of U.K. elite public schools. Students are assigned to these schools on the basis of a test taken in primary school, and he exploits this rule via an instrumental variables approach, comparing students on the border of elite and non-elite schools on a wide range of high school outcomes, including basic test scores, the probability of enrolling in advanced courses, and college entry. He shows that elite schools have small effects on test scor but much larger effects on advanced course-taking and college entry patterns. Investigating the mechanisms through which these effects operate, he finds that some can be attributed to the impact of elite schools on advanced course quality; some are likely attributable to the greater opportunities offered by elite schools.
The incentives and outcomes generated by public school choice depend to a large degree on parents' choice behavior. There is growing empirical evidence that low-income parents place lower weights on academics when choosing schools, but there is little evidence as to why. Hastings and her co-authors use a field experiment in the Charlotte-Mecklenburg Public School district (CMS) to examine the degree to which information costs affect parental choices and their revealed preferences for academic achievement. The authors provided simplified information sheets on school average test scores, or test scores coupled with estimated odds of admission, to students in randomly selected schools along with their CMS school choice forms. They find that receiving simplified information leads to a significant increase in the average test score of the school chosen. This increase is equivalent to a doubling in the implicit preference for academic performance in a random utility model of school choice. Receiving information on odds of admission further increases the effect of simplified test score information on preferences for test scores among low-income families, but dampens the effect among higher-income families. Using within-family changes in choice behavior, the authors provide evidence that the estimated impact of simplified information is more consistent with lowered information costs than with suggestion or saliency.
Efforts to achieve universal primary education remain an elusive goal in most developing countries. Resource constraints limit the extent to which demand-based subsidies can be used. Khwaja and his co-authors focus on a supply-side factor: the availability of low cost teachers and the resulting ability of the market to offer affordable education. Using data from rural Pakistan and official public school construction guidelines as an instrument, they find that private schools are three times more likely to exist in villages with girls' secondary schools. In contrast, there is little or no relationship between the presence of a private school and pre-existing girls' primary, or boys' primary and secondary schools. Moreover, villages that received girls' secondary schools not only show a more than doubling in educated women but also 20 percent lower teacher wages. In an environment with low female education levels and mobility, girls' secondary schools substantially increase the local supply of skille women. This lowers wages for women in the local labor market and allows the market to offer affordable education. These findings highlight the prominent role of women as teachers in facilitating educational access and resonates with similar historical evidence from developed economies. Higher education (also) matters because the students of today are the teachers of tomorrow.
Performance pay for teachers is frequently suggested as a way of improving educational outcomes in schools, but the empirical evidence to date on its effectiveness is limited and mixed. Muralidharan and Sundararaman present results from a randomized evaluation of a teacher incentive program implemented across a representative sample of government-run rural primary schools in the Indian state of Andhra Pradesh. The program provided bonus payments to teachers based on the average improvement of their students' test scores in independently administered learning assessments (with a mean bonus of 3 percent of annual pay). Students in incentive schools performed significantly better than those in control schools, by 0.19 and 0.12 standard deviations in math and language tests respectively. They scored significantly higher on "conceptual" as well as "mechanical" components of the tests suggesting that the gains in test scores represented an actual increase in learning outcomes. Incentive schools also performed better on subjects for which there were no incentives. The researchers find no significant difference in the effectiveness of group versus individual teacher incentives. Incentive schools performed significantly better than other randomly-chosen schools that received additional schooling inputs of a similar value.
Ho and her co-authors investigate how hospitals responded to changes in incentives introduced by the Health Care Reform Acts of the mid-1990s in New Jersey and New York. These legislative acts removed the centralized rate setting systems determining the prices paid to individual hospitals by private insurers (excluding HMOs in New York) and allowed these insurers to set rates through bilateral bargaining with hospitals. Preliminary results suggest that hospitals responded by forming bargaining coalitions (systems) and by reducing beds. There is no evidence, at the hospital level, to suggest that there was a systematic response in how patients were treated or in the adoption of new technology. Further research into the determinants of the modes of response continues.
Conflicting theories of the nonprofit firm have existed for several decades, yet empirical research has not resolved these debates, partly because the theories are not easily testable but also because empirical research generally considers organizations in isolation, rather than in markets. In their paper, Horwitz and Nichols examine three types of hospitals -- nonprofit, for-profit, and government -- and their spillover effects. They look at the effect of for-profit ownership share within markets on both the likelihood of offering profitable and unprofitable medical services, and operating margins. They find that nonprofit hospitals are more likely to offer profitable services and less likely to offer unprofitable services in markets with a higher concentration of for-profit hospitals, but there is no statistically significant effect of for-profit market share on operating margins. These results fit best with theories in which nonprofit hospitals maximize their own output.
Bartel and her co-authors use a unique panel dataset (Veterans Administration acute-care hospitals for the time period 2003 through 2006) to study the impact of human capital and relational capital in the nursing workforce on nursing-sensitive patient outcomes. Unlike most of the prior work on nursing and patient outcomes that uses the hospital as the unit of analysis and focuses on the impact of nurse staffing, their study uses the nursing unit as the unit of analysis and goes beyond estimating the impact of nurse staffing to consider the attributes of the nursing workforce. Since hospital nursing units are quite heterogeneous, using aggregated data is likely to mask the true relationship between nurse characteristics and patient outcomes. The approach in this paper thus provides a more accurate measure of this relationship. In this preliminary version, the authors study the impact of RN education and RN tenure in the current hospital on two measures of nursing-sensitive patient outcomes in intensive care units, the rate of infections and the failure to rescue rate. They find that RN education and RN tenure are negatively correlated with the rate of infections and the effect of RN tenure holds even when they control for nursing unit fixed effects. The latter finding suggests that policies that encourage RN retention should have beneficial effects on patient outcomes.
Skinner and Staiger examine productivity in the treatment of heart attacks across hospitals and over time. In their model, physicians search optimally over new technological innovations. The authors test the model using U.S. Medicare data on survival and factor inputs for 2.8 million heart attack patients during 1986-2004 combined with a 1994-5 survey on technology adoption. They find that the speed of adopting highly efficient and often low-cost innovations such as beta blockers, aspirin, and primary reperfusion explains a large fraction of variations in productivity, and swamps the impact of traditional factor inputs. The empirical patterns are also consistent with those found in macroeconomic cross-country studies: the persistence of rapid or slow adoption within institutions, a lack of convergence in output, substantial differences in long-run productivity, and the importance of a productivity "frontier." Large informational barriers to adoption are the best explanation for why some physicians fail to prescribe even aspirin for their patients, and could also explain puzzling empirical patterns in other sectors of the economy.
In his presentation, Rebitzer posed three questions: Why are quality improvement initiatives (such as pay for performance and cross-functional work teams) so hard to implement? How might new investments in information technology (both within and across organizations) influence care quality? How do coordination costs shape patterns of specialization, referral, and ownership? In discussing potential answers to these questions, Rebitzer applied ideas from organizational economics to highlight new avenues for economic research on hospitals.
Using multiple datasets and five different time periods, Babcock and Marks document and quantify changes in time use by full-time college students in the United States between 1961 and 2004. They find large and continuous declines in academic time investment over this period. In 1961 full-time college students appeared to allocate about 40 hours per week toward class and studying, whereas full-time students in 2004 appear to have invested about 23 to 26 hours per week. The declines were extremely broad-based and are not easily explained by changes in work choices or in the composition of students or schools. The authors explore the implications of this finding, focusing in particular on wage regressions and a recalculation of the college wage premium.
Stuen and his co-authors use panel data on 2300 science and engineering departments at 100 large American universities from 1973 to 1998 to estimate the impact of foreign and domestic graduate students in a department on the publications and patents produced by that department. Macroeconomic shocks and policy changes in source countries that differentially affect enrollments across fields and universities in the United States are used to instrument for the supply of students by region. The authors also aggregate micro data on every Ph.D. recipient at all major science and engineering departments in the United States between 1960 and 2005 to create student enrollment counts by source country. Further, they create citation-weighted publication counts for each academic department by automating thousands of Web of Science publication and citation searches. They also identify academic patents and allocate them to fields of inquiry. They show that both foreign and domestic graduate students are central inputs into knowledge creation, and that certain estimates of the foreign student contribution are biased downwards. The impact of an additional foreign doctoral student varies by source region and by the type of shock that sends the student to the United States.
In an increasingly globalized world, some firms with high-skilled jobs search world-wide for the best talent. Whereas identifying talent and the correct firm-employee fit are difficult and expensive under any circumstance, hiring foreign workers complicates the task of evaluating individual's skill sets (both cognitive and non-cognitive abilities) and their match with company and industry standards and cultural norms. To analyze the differences that U.S. selection committees face in identifying foreign versus domestic talent, Grove and Wu use applicants to a single top-five economics Ph.D. program in 1989 (N=344), although those individuals obtained economics Ph.D.s from 50 different U.S. doctoral departments and 11 foreign programs. Admission committee members judge students' potential economics talent (for example, their preparation, aptitude, drive, and creativity) based on the uniform evidence provided in the application folders: standardized test scores, the course selection and grades provid in transcripts, the quality of the undergraduate institution, fellow economists' evaluations of their potential economics talent, and other relevant information. Because non-U.S. citizens received 68 percent of economics doctoral degrees as of 2003, the signals used by economics Ph.D. admission committees (such as the globally-ranked quality of undergraduate schools or of the letter-of-reference writers) provide considerably less help in identifying economics talent for those educated outside U.S. borders. Grove and Wu reach two broad sets of conclusions. First, foreign undergraduates attended higher quality Ph.D. programs and were more likely to complete the doctorate than U.S. baccalaureates, but experienced no differences in initial job placement or research productivity 17 years later. Second, many of the results previously reported in the literature regarding determinants of some aspect of graduate school and career success vary according to whether the individuals received their undergraduate training in the United States or elsewhere. Four notable differences emerge: 1) GRE scores (quantitative scores matter more for those with U.S. baccalaureates but verbal scores for those from abroad); 2) gender (for graduate school outcomes and publishing success); and for initial job placement and early career research productivity, both 3) the quality of the Ph.D. program attended and 4) the dissertation advisor. For example, Ph.D. program quality affects the initial job placement of U.S. undergraduates but the publication success of foreign baccalaureates. By contrast, U.S. (but not foreign) undergraduates' research productivity increased for those with a globally-ranked researcher as a dissertation advisor.
Long explores how variation in the availability of affirmative action practiced by nearby colleges affected the application and enrollment decisions of three cohorts of students in 1972, 1982, and 1992. He finds significant differences in the degree of preference given to minority applicants by various college characteristics, with less affirmative action used by private colleges and colleges in the Midwest (particularly in later cohorts), and more selective institutions giving larger preferences to minority applicants. Given the spatial distribution of college characteristics, Long finds substantial variation in the degree of affirmative action available to students at nearby colleges. However, this variation in nearby affirmative action had very small effects on the average quality of institutions at which minority and non-minority students apply and enroll. More nearby affirmative action increases the likelihood that minority students apply to a 4-year college, and increases the likelihood that highest ability non-minority college applicants submit more than one application.
Governments spend heavily on across-the-board tuition subsidies to make college accessible for high school graduates. Direct state appropriations to post-secondary schools currently exceed $65 billion per year. Despite the expensive nature of this intervention, remarkably little is known about its causal impact on college-going behavior. Uncertainty surrounding the efficacy of public subsidies in higher education is largely attributable to the virtual absence of plausible exogenous variation in college price. Using a large sample of Texas students, McFarlin exploits geographic variation in college price associated with community college taxing-district residency to estimate the impact of tuition policy on college attainment. His paper is a progress report on efforts to assess how well public subsidies shape college decisions. The early findings suggest that while public subsidies for community colleges play an important role in promoting access and also improving the likelihood of completing an associate's degree, they concurrently appear to "crowd out" enrollments at public four-year colleges and to decrease the chances of receiving a baccalaureate degree. This research also sheds light on the seemingly overlooked pervasiveness of college districts throughout the American higher education landscape.
Administrative skill is essential to organizational effectiveness. Yet, few studies have examined how human capital investments over a career affect selection into administration. McDowell, Singell, and Stater use panel data for economists to estimate the probability of choosing administration over a pure academic track. Their results show that, while research-specific human capital reduces the probability of becoming an administrator, general human capital increases it. There are also inferior administrative opportunities for women that have not improved over time and variation in the role of human capital according to institutional research mission. Thus, results suggest academic leaders are not merely born, but cultivated.
Gentzkow and Shapiro construct a new index of media slant that measures whether a news outlet's language is more similar to a congressional Republican or Democrat. They apply the measure to study the market forces that determine political content in the news. They estimate a model of newspaper demand that incorporates slant explicitly, estimate the slant that would be chosen if newspapers independently maximized their own profits, and compare these ideal points with firms' actual choices. Their analysis confirms an economically significant demand for news slanted toward one's own political ideology. Firms respond strongly to consumer preferences, which account for roughly 20 percent of the variation in measured slant in their sample. By contrast, the identity of a newspaper's owner explains far less of the variation in slant, and they find little evidence that media conglomerates homogenize news to minimize fixed costs in the production of content.
Fiori and his co-authors provide a systematic empirical investigation of the effect of product market liberalization on employment when there are interactions between policies and institutions in product and labor markets. Using panel data for OECD countries over the period 1980-2002, they show that product market deregulation is more effective at the margin when labor market regulation is high. Moreover, there is evidence in their sample that product market deregulation promotes labor market deregulation. They show that these results are mostly consistent with the basic predictions of a standard bargaining mode such as Blanchard and Giavazzi (2003) extended to allow for a richer specification of the fall back position of the union and for taxation.
When democracy is new, it is often fragile and not fully consolidated. Brender and Drazen investigate how the danger of a collapse of democracy may affect fiscal policy in comparison to countries where democracy is older and often more established. They argue that the attitude of the citizenry towards democracy is important in preventing democratic collapse and expenditures therefore may be used to convince them that "democracy works". This contrasts with much of the literature that concentrates on policy directed towards anti-democratic elites. The authors consider the inference problem that citizens solve in looking at economic outcomes and forming their beliefs about the efficacy of democracy. They argue that the implications of the model are broadly consistent with the empirical patterns generally observed, including the existence of political budget cycles at the aggregate level in new democracies that are not observed in old democracies.
Bandiera, Prat, and Valletti propose a distinction between active waste (a situation where the presence of waste benefits the public decisionmaker, as in the case of bribery) and passive waste (pure inefficiency, possibly because of excessive red tape). They analyze purchases of several standardized goods by over 200 Italian public bodies and exploit a policy experiment that introduced a national procurement agency. A revealed preference argument implies that the decision to buy from the new procurement agency rather than from traditional suppliers can be used to distinguish between active and passive waste. Their results indicate that: 1) different public bodies pay widely different prices for observationally equivalent goods, with centralized bodies paying on average at least 25 percent more than semi-autonomous bodies; 2) price differences are mostly because of passive rather than active waste on average passive waste accounts for 79 to 92 percent of estimated waste; and 3) there is no trade-off between passive and active waste.
Does economic inequality affect redistributive policy? Ramcharan turns to U.S. county and state data on land inequality over the period 1890-1930 to help address this fundamental question in political economy. To facilitate causal inference, he uses indicators of weather risk -- rainfall, growing degree days, and topographical variability as instruments for land inequality. The evidence strongly supports those theories that posit a causal connection between the distribution of wealth, political institutions, and redistributive policies The instrumental variables estimates indicate that greater inequality is associated with less political competition, and less redistribution.
Acemoglu and Robinson construct a model to study the implications of changes in political institutions for economic institutions and economic outcomes. The model society consists of elites and citizens who have different preferences over economic institutions. The main idea is that equilibrium economic institutions are a result of the exercise of de jure and de facto political power. A change in political institutions for example a move from nondemocracy to democracy alters the distribution of de jure political power, but the elite can intensify their investments in de facto political power, for example by lobbying or using paramilitary forces, to partially or even fully offset their loss of de jure power. Thus equilibrium changes in political institutions may have little or no effect on the (stochastic) equilibrium distribution of economic institutions. The interplay between de jure and de facto political power also leads to a number of new comparative static results. The authors then show that when changing political institutions is more difficult than altering economic institutions, the model implies a pattern of captured democracy, whereby a democratic regime may survive, but chooses economic institutions favoring the elite. The model provides conditions under which economic or policy outcomes will be invariant to changes in political institutions and economic institutions themselves will persist over time.
Comerton-Forde and her co-authors use an 11-year panel of daily specialist revenues on individual NYSE stocks to explore the relationship between market-maker revenues and liquidity. If market makers suffer substantial trading losses, lenders may respond by increasing funding costs or reducing credit lines, and market makers should respond by reducing liquidity provision. The data indicate that when specialists in aggregate lose money on their inventories, market-wide effective spreads widen in the days or weeks that follow, even after controlling for stock returns, volatility, and volume. This suggests an important role for market-maker financial performance in explaining liquidity time-variation. Revenues at the specialist firm level explain liquidity changes in that firm's assigned stocks. Revenues at the individual stock level do not explain changes in individual stock liquidity, consistent with a financial constraints model with broadly diversified intermediaries. Aggregate specialist reven are increasing in conditional return volatility, as is revenue volatility. Specialist margins (specialist revenue per dollar of trading volume) are essentially constant across stocks, implying limited scope for cross-subsidization.
Foster and his co-authors assess the role and viability of an order-crossing or market-clearing mechanism that is automatically triggered only when a minimum number of shares can be crossed. Such a mechanism is naturally more attractive to traders who do not require much immediacy for their trades, as liquidity is cheaper in this market than in a continuous-auction market. The volume condition that they propose is crucial to the effectiveness with which this market complements the continuous-auction market in two important ways. First, when appropriately set, the volume condition endogenously adjusts the probability that market-clearing is triggered and so keeps impatient traders and highly informed traders away. Second, because market-clearing with a large volume condition reduces the effects of adverse selection in this market, patient traders are more willing to place orders in it. As the researchers show, these effects often combine into a Pareto-dominating equilibrium when the continuous-aucti market and the crossing mechanism with the right volume condition are both open.
A fundamental role of financial markets is to gather information on firms' investment opportunities, and thus to help guide investment decisions. Dow and his co-authors study the incentives for information production when prices perform this allocational role. If firms cancel planned investments following poor stock market response, the value of their shares will become insensitive to information on investment opportunities, so that speculators will be deterred from producing information ex ante. Based on this insight, the researchers derive several new results on the determinants of information production and the resulting firm value and investment policy. They show that information production on investment opportunities is distinctly different than that on assets in place, and argue that some overinvestment increases firm value.
A general goal for stock exchanges is to increase participation by firms and investors. Recent research has highlighted the role of ambiguity in affecting participation. Easley and O'Hara show the role that microstructure can play in reducing the ambiguity confronting traders. They develop a model with objective expected utility maximizing traders and naive traders, and they show how these naïve traders can choose to participate or not participate in markets. The researchers then show how specific features of the microstructure can reduce the perceived ambiguity, and induce participation by both firms and issuers. Their analysis demonstrates how designing markets to reduce ambiguity can benefit investors through greater liquidity, exchanges through greater volume, and issuing firms through a lower cost of capital.
Bessembinder and his co-authors study limit order traders' joint decisions regarding order price, order size, and order exposure in markets where the option to hide a portion of order size exists. Using order-level data from Euronext-Paris, the authors document that hidden orders are used extensively by market participants, representing approximately 44 percent of order volume. After controlling for known determinants of order price aggressiveness, order exposure, and order size, and allowing for simultaneity in the decisions, they show that the traders electing to post aggressively priced orders tend to expose their orders, while the traders placing orders away from the best quotes tend to hide their orders. Further, traders choose to hide a larger portion of their orders when they have also selected larger orders. All else equal, hidden orders are associated with smaller opportunity costs and lower implementation shortfall costs. However, the offsetting costs are that hidden orders are associated with lower probability of full execution and longer times to execution. Overall, the evidence indicates that hidden orders are used primarily by uninformed traders to lower the option value of orders that are likely to be left standing in the book.
Osler and her co-authors make three contributions to our understanding of the price discovery process in currency markets with this paper. First, they show that this process cannot be the familiar one based on adverse selection and customer spreads, because such spreads are inversely related to a trade's likely information content. Second, they suggest three potential sources for the pattern of customer spreads, two of which rely on the information structure of the market. Third, they present an alternative price discovery process for currencies, centered on inventory management strategies in the interdealer market, and provide preliminary evidence for that process. They also suggest more broadly that the price discovery process will vary with market structure, and that their proposed mechanism may apply to liquid two-tier markets in general.
The impact of circumstances early in life on later life outcomes in the United States can now be explored using data linking individuals from Social Security records back to the manuscript schedules of federal population censuses shortly after their birth. This allows Ferrie and his co-authors to assess the effect of individual, household, and community level influences on longevity. They find that these all had substantial affects on age at death, as did characteristics measured at enlistment into the U.S. Army in World War II.
The inability to smooth consumption in developing countries is thought to make health vulnerable to sudden economic downturns. However, studies suggesting this relationship often examine events that influence health independently of economic conditions. Miller and Urdinola investigate how world coffee price shocks influence infant and child mortality in Colombia's coffee-growing regions. As in wealthy country studies, here they find evidence of procyclical mortality and countercyclical health investments that appear linked to changes in the opportunity cost of time. These results suggest that in rural Colombia, any adverse health consequences of reduced transitory income during bad economic times are dominated by increases in time-intensive health investments, and the relative price of health may be a more powerful determinant of mortality than income.
The 1960s ushered in a new era in U.S. demographic history characterized by sharp reductions in family size. Using cross-state variation in the restrictiveness of anti-obscenity statutes, Bailey quantifies the importance of oral contraception in reducing married women's fertility. Her analysis has two central conclusions: First, bans on the sale of contraception limited the use of the birth control pill before the 1965 Griswold decision. Second, bans on the sale of contraception slowed the decline in birth rates in the early 1960s. Whereas differences in birth rates between states with and without bans on sales are indistinguishable from zero in the pre- and post-period, they increased by approximately 6 births per 1000 women from 1959 to 1965, a figure comparable to back-of-the-envelope calculations that make use of differences in failure rates between the pill and other available methods. These estimates suggest that at least 20 percent of the fertility decline from 1960 to 1965 can be attributed the introduction of the birth control pill.
During the nineteenth century, the U.S. birthrate fell by half. While previous economic literature has emphasized demand-side explanations for this decline that rising land prices and literacy caused a decrease in demand for children historians and others have emphasized changes in the supply of technologies to control fertility, including abortion and birth control. Ananat and Lahey exploit the introduction during the nineteenth century of state laws governing American women's access to abortion to measure the effect of changes in the supply of fertility technologies on the number of children born. They estimate an increase in the birthrate of 3 to 8 percent when abortion is restricted, which lies within the ranges of estimates found for the effect of fertility control supply restrictions on birthrates today. Restriction of birth control access led to an increase of 3 to 4 percent, numbers also consistent with modern evidence. By demonstrating the importance of legal abortion and birth control reducing nineteenth-century birthrates, they account for a previously unexplained portion of the demographic transition. Moreover, they show that there has long been a demand, often unmet, for fertility control.
The effect of population health on aggregate income remains an open question, despite the substantial literature showing positive and long-lasting effects of health on productivity at the individual and cohort levels. Bleakley looks at several eradication campaigns against parasitic disease in the Americas. Previous work indicated that cohorts exposed to these campaigns as children had higher productivity and human capital. As these more productive cohorts entered the labor force, average incomes rose as well, above and beyond that attributable to the changing cohort compositon of workers. This suggests, on net, the presence of positive spillovers from health capital. Bleakley also estimates the impact of average childhood exposure to the campaign on aggregate (state-level) income, where the effect is at least as large as the sum of estimated direct and spillover effects.
Household economies of scale arise when households with multiple members share public goods, making larger households better off at lower per capita expenditures. Research into household scale economies has yet to consider how household economies of scale change over time. Logan uses American household expenditure surveys, covering 1888 to 1935, to produce the first comparable historical estimates of household scale economies in consumption. Scale economies in clothing, entertainment, and housing declined from 1888 to 1935, consistent with market expansion and increasing substitutes for these expenditures over time. Households in the past had fewer scale economies in food than today, however, exactly the opposite of what theory would predict and deepening a well known puzzle in the literature. Overall, Logan finds that scale economies changed significantly from 1888 to 1935 for all expenditure categories. He then consider the implications of changing scale economies for estimates of real income. Previous estimates of CPI bias based on Engel curves do not account for changing scale economies in the household, and this can lead to omitted variable bias. His estimates of the annual rate of CPI bias from 1888 to 1935 are reduced by at least 25 percent once changing scale economies are accounted for, consistent with household economies of scale having a large, material effect on estimates of real income.
Almond and Mazumder use maternal fasting during the Islamic holy month of Ramadan as a natural experiment for evaluating the long-term effects of prenatal nutrition. Because the timing of Ramadan varies by year, they can disentangle the effect of fasting from seasonal effects (for example weather, or virus exposure). They use the 2002 Uganda Census, which collects information on a range of adult outcomes (such as health, education, fertility) for a large sample of Muslims and non-Muslims. They find that the occurrence of Ramadan nine months before birth increases the likelihood among Muslims of having a physical disability in adulthood by close to 20 percent. Effects are found for blindness, deafness, disability involving the lower extremities, and mental disability. No corresponding effect is found among non-Muslims. Ramadan falling during the second month of pregnancy is positively associated with years of schooling and literacy for men. The authors speculate that this may be attributable to selective attrition during a period of high fetal loss, which previous work has found to be more important among male fetuses. They find no evidence that negative selection in conceptions during Ramadan account for our results.
Wilson notes that, from the outset, the Union Army pension was nominally color-blind. But color-blind policy does not always prevent discrimination. In the early years of the pension, blacks were subject to "institutional discrimination" that resulted in dramatically lower application rates for blacks. Even after controlling for medical military history, blacks who often faced a much harder burden in proving their identify and their military service, as well as being less able to navigate the intricacies and costs associated with pension application fared much worse than whites. Soon after the war, they also began to experience "discretional discrimination." Popular support for increasing pension assistance resulted in increasing discretion exercised by the Pension Board. This discretion, however, was applied to a much greater extent to whites than blacks. Documented evidence from the War Department provided only a partial protection for black applicants, as whites with similar medical historie pulled away from blacks. In the decade prior to the 1890 liberalization of the pension system, both blacks and whites applied for pensions in increasing numbers, but the approval rate for black pensions plummeted as the rate for whites held stable. The 1890 law which negated part of the discretionary role of the Pension Board narrowed the enrollment gap considerably, but specific disabilities continued to be approved for whites at a higher rate than for blacks. This was most notably true for disabilities that were harder to diagnostically verify. By the turn of the century, the enrollment gap had become predominantly a result of discretional discrimination rather than differential application rates. Finally, although blacks and whites received similar pension awards when they entered the system, increases for white pensioners outpaced those given to blacks.
Lee explores how broadly-defined technological changes (including organizational and managerial transformations as well as innovations in production methods) in the U.S. manufacturing industries affected the probabilities of long-term unemployment and of retirement of older male workers in the early-twentieth-century United States. For this purpose, industry-level statistics reported in the 1899 and 1909 manufacturing census were linked to the IPUMS of the 1910 census, and to a longitudinal sample of Union Army veterans. The results suggest that the rapid technological changes had both favorable and adverse impacts on the employment of older workers. On one hand, technological progress improved the employment prospect of older workers by enhancing labor productivity and by formalizing the workplace management. On the other hand, emergence of large corporations and technological shifts toward more capital- and technology-intensive productions made it increasingly difficult for older workers to remai in the labor market, perhaps by increasing the requirements for physical strength, mental agility, and ability to acquire new skills. It is likely that the overall impact of technological changes on the employment of older workers during the industrial era was negative.
Jorgenson and Nomura compare total factor productivity (TFP) levels in the United States and Japan for the period 1960-2004 and allocate the observed gap to individual industries. They carefully distinguish the various concepts of purchasing power parity (PPP) and measure them within the framework of a U.S.-Japan bilateral input-output table. They also measure industry-level PPPs for capital, labor, energy, and materials inputs and output for 42 industries common to both countries, based on detailed estimates for 164 commodities, 33 assets, including land and inventories, and 1596 labor categories. The U.S .-Japan productivity gap shrank during three decades of rapid Japanese economic growth, 1960-90. The Japanese manufacturing sector achieved parity with its U.S. counterpart by the end of the period. With the collapse of the Japanese economic bubble at the end of the 1980s, the U.S.-Japan productivity gap reversed course and expanded to 79.5 percent by 2004. This can be attributed to ra productivity growth in the IT-producing industries in the United States during the late 1990s and the sharp acceleration of productivity growth in the IT-using industries in the United States during 2000-4. Wholesale and Retail Trade emerged as the largest contributor to this gap, accounting for 25.1 percent of the lower TFP of the Japanese economy.
The 1998 passage of the Land Revaluation Law in Japan provided regulatory forbearance to Japanese banks in the form of a regulatory capital infusion. Allen and Chakraborty test whether this divergence from international bank capital requirements had an effect on Japanese bank lending behavior. Because this natural experiment created an exogenous supply shock, the researchers can use it to disentangle demand and supply effects so as to determine the impact on Japanese bank lending in the United States and Japan. They find that the infusion of regulatory capital had no aggregate impact on Japanese bank lending in Japan, but it did change the allocation of loans. Well-capitalized Japanese banks shifted their lending from low margin, less capital intensive mortgage lending toward higher yielding, more capital intensive commercial loans. Moreover, they find evidence consistent with a shifting of Japanese bank lending activity away from U.S. lending (which is predominately real estate based) to domestic lending to fund manufacturing. Thus, they discover that divergences from international capital standards have significant allocative effects on lending, as well as on bank profitability.
van Leuvensteijn and co-authors are the first to apply a new measure of competition, the Boone indicator, to the banking industry. This approach measures the competition of bank market segments, such as the loan market, versus many well-known measures of competition that only can consider the entire banking market. One caveat of the Boone-indicator may be that it assumes that banks generally pass on at least part of their efficiency gains to their clients. Like most other model-based measures, this approach ignores differences in bank product quality and design, as well as the attractiveness of innovations. The researchers measure competition on the lending markets in the five major EU countries and, for comparison, the United Kingdom, the United States, and Japan. Bearing the aforementioned caveats in mind, the findings indicate that from 1994-2004 the United States had the most competitive loan market, whereas overall loan markets in Germany and Spain were among the most competitive in the EU. Th Netherlands occupied a more intermediate position, whereas in Italy competition declined significantly over time. The French, Japanese, and UK loan markets were generally less competitive. Turning to competition among specific types of banks, the commercial banks tend to be more competitive, particularly in Germany and the United States, than savings and cooperative banks.
Otani and his co-authors empirically analyze the interaction between the distortions in the real side of the economy (real distortion) and those in the financial side of the economy (financial distortion) in Japan after the 1990s. They focus on protracted economic stagnation after the bursting of the asset price bubble, and the subsequent recovery in economic activity and restoration of financial system stability. They use the intersectoral differences in relative factor prices as an indicator for the real distortion, and use the gap between the actual and benchmark loan portfolios, based on the mean-variance approach to maximizing risk adjusted returns from loan portfolios, as an indicator for financial distortion. They show that both distortions sharply deteriorated in the late 1990s and improved in the 2000s. In addition, they conduct panel data analyses using data at the individual bank level as well as at the industry level to examine the interaction between the two distortions: the interac was negative in the late 1990s, but reversed to positive in the 2000s.
Estimating the effect of official foreign exchange market intervention is complicated by the fact that intervention at any point entails a "self-selection" choice made by the authorities and that no counterfactual is observed. To address these issues, Fatum and Hutchison estimate the "counterfactual" exchange rate movement in the absence of intervention by introducing the method of propensity score matching to estimate the "average treatment effect" (ATE) of intervention. To derive the propensity scores, they introduce a new intervention reaction function that includes the difference between market expectations and official announcements of macroeconomic developments that can influence the decision to intervene. They estimate the ATE for daily official intervention in Japan over the January 1999 to March 2004 period. This sample encompasses a remarkable variation in intervention frequencies as well as unprecedented frequent intervention towards the latter part of the period. They find that the effects of intervention vary dramatically and inversely with the frequency of intervention: intervention is effective over the 1999 to 2002 period and ineffective (or possibly counterproductive) during 2003 and 2004. These results hold up to a variety of robustness tests. Only sporadic and relative infrequent intervention appears to be effective.
It is common to define benefit eligibility for small business policies by restrictions on the firm size. Onji investigates the incentives for a large firm to "masquerade" as many small firms by separately incorporating business segments, focusing on the case of the Japanese value-added tax. He finds that the masquerading behavior was pervasive and took place quickly after the introduction of tax incentives. Tax avoidance accounted for 3.4 percent of the overall revenue drain attributable to the preferential tax treatment of small businesses, but the efficiency consequence would not have been severe. This study suggests that the strategy may be commonplace in other settings.
Does popular participation in political processes affect economic policy outcomes? To answer this long-standing but recently re-visited question, Horiuchi and Saito estimate the causal effects of political participation on distributive policy outcomes, using a large municipality-level dataset from Japan. The theoretical rationale is that risk-averse incumbents listen to "voices" (policy needs) of participants more than to those of non-participants. To test various hypotheses derived from this theory, the existing studies typically had used voter turnout, but these researchers argue that this overlooks an important dimension of political participation. In addition to voting in elections, some citizens communicate with incumbents more effectively, through more direct modes of political participation. Methodologically, a valid measure should be voter-turnout weighted by how much those who turn out to vote can pitch their voice to policymakers so that their policy needs can be heard more effectively than those of the "silent majority." With effective instrumental variables, the authors show that their estimates are far larger than the conventional estimates under the assumption of homogeneity among voters.
From the beginning of 2003 to the spring of 2004, Japan's monetary authorities conducted large-scale yen-selling/dollar-buying foreign exchange operations in what Taylor (2006) has labeled the "Great Intervention." Watanabe and Yabu empirically examine the relationship between this "Great Intervention" and the quantitative easing policy the Bank of Japan (BOJ) was pursuing at that time. Using daily data on the amount of foreign exchange interventions and current account balances at the BOJ, they conclude first that, while about 60 percent of the yen funds supplied to the market by yen-selling interventions were immediately offset by the BOJ's monetary operations, the remaining 40 percent were not offset and remained in the market for some time. This is in contrast to the preceding period, when almost 100 percent were offset. Second, comparing foreign exchange interventions and other government payments, the extent to which the funds were offset by the BOJ was much smaller in the case of foreign exchange interventions, and the funds also remained in the market longer. This suggests that the BOJ differentiated between and responded differently to foreign exchange interventions and other government payments. Third, the majority of financing bills issued to cover intervention funds were purchased by the BOJ from the market immediately after they were issued. For that reason, no substantial decrease in current account balances linked with the issuance of FBs could be observed. These three findings indicate that it is highly likely that the BOJ, in order to implement its policy target of maintaining current account balances at a high level, intentionally did not sterilize yen-selling/dollar-buying interventions.
Rogerson and Wallenius build a life-cycle model of labor supply that incorporates changes along both the intensive and extensive margin and use it to assess the consequences of changes in tax and transfer policies on equilibrium hours of work. They find that changes in taxes have large aggregate effects on hours of work. Moreover, they find that there is no inconsistency between this result and the empirical finding of small labor elasticities for prime age workers. In their model, micro and macro elasticities are effectively unrelated. Their model is also consistent with other cross-country patterns.
Large and persistent global financial imbalances need not be the harbinger of a world financial crash. Instead, Mendoza and his co-authors show that these imbalances can be the outcome of financial integration when countries differ in financial markets deepness. In particular, countries with more advanced financial markets accumulate foreign liabilities in a gradual, long-lasting process. Differences in financial deepness also affect the composition of foreign portfolios: countries with negative net foreign asset positions maintain positive net holdings of non-diversifiable equity and FDI. Abstracting from the potential impact of globalization on financial development, liberalization leads to sizable welfare gains for the more financially-developed countries and losses for the others. Three empirical observations motivate this analysis: 1) financial deepness varies widely even among industrial countries, with the United States ranking at the top; 2) the secular decline in the U.S. net foreign asset position started in the early 1980s, together with a gradual process of international capital markets liberalization; 3) net exports and current account balances are negatively correlated with indicators of financial development.
Researchers have used unanticipated changes in monetary policy to identify preference and technology parameters of macroeconomic models. Dupor and his co-authors use changes in technology to identify the same set of parameters. Estimates based on technology shocks differ substantially from those based on monetary policy shocks. In the post World War II United States, a positive technology shock reduces inflation and increases hours worked, significantly and rapidly in both cases. Relative to policy shock identification, technology shock identification implies: 1) long duration durability in preferences instead of short duration habit, and 2) built-in inflation inertia disappears and price flexibility increases. In response to technological improvement, consumption durability increases hours worked because households temporarily increase labor supply to accumulate durables towards new, higher steady-state levels. Limited nominal rigidities allow inflation to fall because firms are able to immediately cut prices when households' labor supply increases.
What is the effect of increasing life expectancy on economic growth? To answer this question, Acemoglu and Johnson exploit the international epidemiological transition, the wave of international health innovations and improvements that began in the 1940s. They obtain estimates of mortality by disease before the 1940s from the League of Nations and national public health sources. Using these data, they construct an instrument for changes in life expectancy, referred to as predicted mortality, which is based on the pre-intervention distribution of mortality from various diseases around the world and dates of global interventions. They document that predicted mortality has a large and robust effect on changes in life expectancy starting in 1940, but no effect on changes in life expectancy before the interventions. The instrumented changes in life expectancy have a large effect on population; a 1 percent increase in life expectancy leads to an increase in population of about 1.5 percent. Life expectancy has a much smaller effect on total GDP both initially and over a 40-year horizon, though. Consequently, there is no evidence that the large exogenous increase in life expectancy led to a significant increase in per capita economic growth. These results confirm that global efforts to combat poor health conditions in less developed countries can be highly effective, but also shed doubt on claims that unfavorable health conditions are the root cause of the poverty of some nations.
Jaimovich and Siu investigate the consequences of demographic change for business cycle analysis. They find that changes in the age composition of the labor force account for a significant fraction of the variation in business cycle volatility observed in the United.States and other G7 economies. During the postwar period, these countries experienced dramatic demographic change, although details regarding extent and timing differ from place to place. Using panel-data methods, they exploit this variation to show that the age composition of the workforce has a large and statistically significant effect on cyclical volatility. They conclude by relating these findings to the recent decline in U.S. business cycle volatility. Using both simple accounting exercises and a quantitative general equilibrium model, they find that demographic change accounts for a significant part of this moderation.
Menzio builds a model marketplace populated by a finite number of sellers each producing its own variety of the goodand a continuum of buyerseach searching for a variety he likes. Using the model, he studies the response of a seller's price to privately observed fluctuations in its idiosyncratic production cost. He finds that the qualitative properties of this response critically depend on the persistence of the production cost. In particular, if the cost is i.i.d., the seller's price does not respond at all. If the cost is somewhat persistent, the seller's price responds slowly and incompletely. If the cost is very persistent, the seller's price adjusts instantaneously and efficiently to all fluctuations in productivity. Menzio argues that these findings can explain why the monthly frequency of a price change is so much lower for processed than for raw goods.