Showing papers by "Institute for the Study of Labor published in 1995"

Identification and Estimation of Local Average Treatment Effects

Abstract: We investigate conditions sufficient for identification of average treatment effects using instrumental variables. First we show that the existence of valid instruments is not sufficient to identify any meaningful average treatment effect. We then establish that the combination of an instrument and a condition on the relation between the instrument and the participation status is sufficient for identification of a local average treatment effect for those who can be induced to change their participation status by changing the value of the instrument. Finally we derive the probability limit of the standard IV estimator under these conditions. It is seen to be a weighted average of local average treatment effects.

Sex Discrimination in Restaurant Hiring: An Audit Study

Abstract: This paper reports on a small-scale audit study that investigates sex discrimination in restaurant hiring. Comparably matched pairs of men and women applied for jobs as waiters and waitresses at 65 restaurants in Philadelphia. The 130 applications led to 54 interviews and 39 job offers. The results provide statistically significant evidence of sex discrimination against women in high-price restaurants. In high-price restaurants, job applications from women had an estimated probability of receiving a job offer that was lower by about .5, and an estimated probability of receiving an interview that was lower by about .4. These hiring patterns appear to have implications for sex differences in earnings, as informal survey evidence indicates that earnings are higher in high-price restaurants.

Estimating the Labor Market Impact of Voluntary Military Service Using Social Security Data on Military Applicants

Abstract: This study uses Social Security data on the earnings of military applicants to the all-volunteer forces to compare the earnings of Armed Forces veterans with the earnings of military applicants who did not enlist. Matching, regression, and Instrumental Variables (IV) estimates are presented. The matching and regression estimates control for most of the characteristics used by the military to select qualified applicants from the military applicant pool. The IV estimates exploit an error in the scoring of exams used by the military to screen applicants between 1976 and 1980. All the estimates suggest that soldiers who served in the early 1980s were paid considerably more than comparable civilians while in the military. Military service also appears to have led to a modest (less than 10 percent) increase in the civilian earnings of nonwhite veterans while actually reducing the civilian earnings of white veterans. Most of the positive effects of military service on civilian earnings appear to be attributable to improved employment prospects for veterans.

Examining the Role of Economic Opportunity and Amenities in Explaining Population Redistribution

Abstract: This paper develops a model of migration integrating equilibrium and disequilibrium components in which individuals and firms from rational expectations about future opportunities. Levels of migration are derived as functions of variations in factors influencing migrant labor demand ("economic opportunity") and migrant labor supply ("residential amenities"). The model is used to estimate the extent to which migration in the United States over the period 1950-1980 is determined by these two classes of exogenous factors.

Does Measured School Quality Really Matter? An Examination of the Earnings-Quality Relationship

Abstract: This paper examines the economic and empirical foundations of the aggregate evidence on the effect of schooling quality on earnings. A common framework is presented which nests all previous studies as special cases. We discuss two crucial identifying assumptions and test them. The first assumption is the absence of region of birth - region of resident interactions in the return to schooling. This rules out patterns of migration on the basis of realized earnings in the destination state. Both parametric and nonparametric versions of this hypothesis are tested. Using 1970, 1980 and 1990 Census data, it is decisively rejected. A second assumption is that log earnings equations are linear - or nearly linear in schooling. This assumption is false. We find that estimated earnings-quality relationships are sensitive to specification of the earnings function. When false linearity assumptions are relaxed, the only effect of measured schooling quality is on the returns for college graduates. The evidence for an aggregate earnings-quality relationship is weak once false empirical restrictions are relaxed.

Beyond the Incidence of Training: Evidence from a National Employers Survey

Abstract: This paper seeks to provide new insight into how school and post school training investments are linked to employer workplace practices and outcomes using a unique nationally representative survey of establishments in the U.S., the Educational Quality of the Workforce National Employers Survey (EQW-NES). We go beyond simply measuring the incidence of formal or informal training to examine the determinants of the types employers invest in, the relationship between formal school and employer provided training, who is receiving training, the links between investments in physical and human capital, and the impact that human capital investments have on the productivity of establishments. We find that the smallest employers are much less likely to provide formal training programs than employers from larger establishments. Regardless of size, those employers who have adapted some of the practices associated with what have been called `high performance work systems' are more likely to have formal training programs. Employers who have made large investments in physical capital or who have hired workers with higher average education are also more likely to invest in formal training and to train a higher proportion of their workers, especially in the manufacturing sector. There are significant and positive effects on establishment productivity associated with investments in human capital. Those employers who hire better educated workers have appreciably higher productivity. The impact of employer provided training differs according to the nature, timing and location of the employer investments.

Parental Leave Policies in Europe and North America

Abstract: The authors describe and compare family leave policies in Europe and North America "First we provide a brief history of family leave legislation in Europe and North America and summarize arguments relating to the efficiency and incidence of mandated leave Second we have constructed a longitudinal data set detailing durations of job-protected leave in 17 countries during the 1960-89 period and use this information to examine recent trends in the regulationsThird we provide an exploratory investigation of the relationship between mandated leave policies and macroeconomic outcomes" (EXCERPT)

Culture and Language

Abstract: Common culture and common language facilitate trade between people. Minorities have incentives to become assimilated and to learn the majority language so that they have a larger pool of potential trading partners. The value of assimilation is larger to someone from a small minority than to one from a large minority group. When a society has a very large majority of individuals from one culture, individuals from minority groups will be assimilated more quickly. Assimilation is less likely when an immigrant's native culture and language is broadly represented in his new country. Also, when governments protect minority interests directly, incentives to be assimilated into the majority culture are reduced. Both factors may explain the recent rise in multiculturalism. Individuals do not properly internalize the social value of assimilation and ignore the benefits others receive when they learn the majority language and become assimilated. In a pluralistic society, a government policy that encourages diverse cultural immigration over concentrated immigration is likely to increase the welfare of the population. In the absence of strong offsetting effects, policies which encourage multi- culturalism reduce the amount of trade and have adverse welfare consequences. Conversely, policies that subsidize assimilation and the acquisition of majority language skills can be socially beneficial. The theory is tested and confirmed by examining U.S. Census data, which reveals that the likelihood that an immigrant will learn English is inversely related to the proportion of the local population that speaks his or her native language.

The Effect of New Jersey's Minimum Wage Increase on Fast-Food Employment: A Re-Evaluation Using Payroll Records

Abstract: We re-evaluate the evidence from Card and Krueger's (1994) New Jersey-Pennsylvania minimum wage experiment, using new data based on actual payroll records from 230 Burger King, KFC, Wendy's, and Roy Rogers restaurants in New Jersey and Pennsylvania. We compare results using these payroll data to those using CK's data, which were collected by telephone surveys. We have two findings to report. First, the data collected by CK appear to indicate greater employment variation over the eight-month period between their surveys than do the payroll data. For example, in the full sample the standard deviation of employment change in CK's data is three times as large as that in the payroll data. Second, estimates of the employment effect of the New Jersey minimum wage increase from the payroll data lead to the opposite conclusion from that reached by CK. For comparable sets of restaurants, differences-in-differences estimates using CK's data imply that the New Jersey minimum wage increase (of 18.8 percent) resulted in an employment increase of 17.6 percent relative to the Pennsylvania control group, an elasticity of 0.93. In contrast, estimates based on the payroll data suggest that the New Jersey minimum wage increase led to a 4.6 percent decrease in employment in New Jersey relative to the Pennsylvania control group. This decrease is statistically significant at the five-percent level and implies an elasticity of employment with respect to the minimum wage of -0.24.

The Effects of Minimum Wages on Teenage Employment and Enrollment: Evidence from Matched CPS Surveys

Abstract: The recent debate over minimum wages raises two questions. First, should policy makers no longer believe that minimum wages entail negative consequences for teenagers? Second, should economists discard the competitive labor market model? Our evidence for teenagers, using matched CPS surveys, suggests that the answer to both of these questions is no. We find that although increases in minimum wages have small net effects on overall teen employment rates, such increases raise the probability that more-skilled teenagers leave school and displace lower-skilled workers from their jobs. These findings are consistent with the predictions of a competitive labor market model that recognizes skill differences among workers. In addition, we find that the displaced lower-skilled workers are more likely to end up non-enrolled and non-employed. Thus, despite the small net disemployment effects for teenagers as a group, there are significant enrollment and employment shifts associated with minimum wage changes that should be of concern to policy makers.

Minimum wage effects on school and work transitions of teenagers

Abstract: Minimum wages increase the probability that teenagers leave school to become employed or work more hours, and increase the probability that they leave school and become non-enrolled and non-employed. Minimum wages also increase the probability that lower-wage employed teenagers become non-enrolled and non-employed. This evidence suggests that (1) the competitive model of minimum wage effects is largely correct; and (2) that there are significant enrollment and employment effects associated with minimum wage changes that should be of concern to policy makers.

Split Sample Instrumental Variables

Abstract: Instrumental Variables (IV) estimates tend to be biased in the same direction as Ordinary Least Squares (OLS) in finite samples if the instruments are weak. To address this problem we propose a new IV estimator which we call Split Sample Instrumental Variables (SSIV). SSIV works as follows: we randomly split the sample in half, and use one half of the sample to estimate parameters of the first-stage equation. We then use these estimated first-stage parameters to construct fitted values and second-stage parameter estimates using data from the other half sample. SSIV is biased toward zero, rather than toward the plim of the OLS estimate. However, an unbiased estimate of the attenuation bias of SSIV can be calculated. We us this estimate of the attenutation bias to derive an estimator that is asymptotically unbiased as the number of instruments tends to infinity, holding the number of observations per instrument fixed. We label this new estimator Unbiased Split Sample Instrumental Variables (USSIV). We apply SSIV and USSIV to the data used by Angrist and Krueger (1991) to estimate the payoff to education.

Technological Change and the Skill Acquisition of Young Workers

Abstract: Using the National Longitudinal Survey of Youth (NLSY) and six proxies for industry rates of technological change, we study the impact of technological change on skill accumulation among young male workers in the manufacturing sector during the time period 1987 through 1992. Production workers in manufacturing industries with higher rates of technological change are more likely to receive formal company training, but not other types of training. An important finding is that, while more educated workers are more likely to receive formal company training, the training gap between the highly educated and the less educated narrows, on average, as the rate of technological change increases. The positive effect of technological change on hours of training is due largely to an increase in the incidence of training, not in the number of hours per training spell.

Conditioning on the Probability of Selection to Control Selection Bias

Abstract: Problems of sample selection arise in the analysis of both experimental and non-experimental data. In clinical trials to evaluate the impact of an intervention on health and mortality, treatment assignment is typically nonrandom in a sample of survivors even if the original assignment is random. Similarly, randomized training interventions like National Supported Work (NSW) are not necessarily randomly assigned in the sample of working men. A non- experimental version of this problem involves the use of instrumental variables (IV) to estimate behavioral relationships. A sample selection rule that is related to the instruments can induce correlation between the instruments and unobserved outcomes, possibly invalidating the use of conventional IV techniques in the selected sample. This paper shows that conditioning on the probability of selection given the instruments can provide a solution to the selection problem as long as the relationship between instruments and selection status satisfies a simple monotonicity condition. A latent index structure is not required for this result, which is motivated as an extension of earlier work on the propensity score. The conditioning approach to selection problems is illustrated using instrumental variables techniques to estimate the returns to schooling in a sample with positive earnings.

Labor Market Effects of Spinal Cord Injuries in the Dawn of the Computer Age

Abstract: What effect does a severe disability have on individuals' employment and earnings? Has the computer revolution lessened the adverse labor market consequences of severe disabilities? This paper investigates the labor market effects of severe, traumatic disabilities resulting from spinal cord injuries (SCIs). We compare the employment experiences of a sample of individuals with SCIs to those of former co-workers over the same period, and to two random samples of individuals in New Jersey. The analysis is based in large part on a 1994 telephone survey of New Jersey adults who had SCIs within the past ten years. Results indicate that the occurrence of an SCI causes a steep decline in employment, hours worked, and weekly earnings, but relatively little change in wage rates for those who work. The computer revolution has the potential to expand employment opportunities for people with disabilities. Our results indicate that having computer skills is associated with higher earnings, and a faster return to work and earnings recovery, for SCI individuals, after holding constant other variables such as education. There is no apparent earnings gap between SCI and non-SCI computer users, whereas among those who do not use computers at work the earnings of SCI employees lag behind those of non-SCI employees. Despite the benefits, individuals with SCIs are less likely to use computers than the general population.

Reconciling the Evidence of Employment Effects of Minimum Wages: A Review of Our Research Findings

Abstract: In this paper, we review the evidence that we have accumulated on the employment effects of minimum wages. We point out specific areas of agreement and disagreement between our research and that of others, and where possible, offer our reconciliation of the conflicting results. Our conclusion is that the revisionist view that minimum wages do not reduce employment is not compelling, and that much of the conflicting evidence can be reconciled with the competitive view of minimum wages and low-wage labor markets.

Relative Income Concerns and the Rise in Married Women's Employment

Abstract: We ask whether women's decisions to be in the labor force may be affected by the decisions of other women in ways not captured by standard models. We develop a model that augments the simple neoclassical framework by introducing relative income concerns into women's (or families') utility functions. In this model, the entry of some women into paid employment can spur the entry of other women, independently of wage and income effects. This mechanism may help to explain why, over some periods, women's employment appeared to rise faster than could be accounted for by the simple neoclassical model. We test the model by asking whether women's decisions to seek paid employment depend on the employment decisions of other women with whom relative income comparisons might be important. In particular, we look at the effects of sisters' employment on women's own employment. We find strong evidence that women's employment decisions are positively related to their sisters' employment decisions. We also take account of the possibility that this positive relationship arises from heterogeneity across families in unobserved variables affecting the employment decision. We conduct numerous empirical analyses to reduce or eliminate this heterogeneity bias. We also look at the relationship between husbands' relative income and wives' employment decisions. In our view, the evidence is largely supportive of the relative income hypothesis.

The role of labor unions in fostering economic development

Abstract: In this essay on how labor unions work, the author proposes a framework for the law on collective bargaining in developing countries. The structure of developing economies cannot sustain as high a level of unionism as in industrial economies. Typically less (often much less) than a quarter of the workers in a developing country are covered by collective bargaining agreements-and those covered (the labor elite) are likely to be employed by the state and by large private sector employers. In this setting, the author says, states sometimes adopt either a patronage regime (nourishing unionism and collective bargaining) or an obstructionist regime (undermining and subverting it). Patronage regimes are found in Bangladesh, India, and certain African and West Indian countries (many of them former British colonies), countries in which close ties exist between political parties (sometimes including the governing party), and labor unions. Some of the features of obstructionist regimes are found in certain countries in Southeast Asia and North Africa. In both patronage and obstructionist regimes, unions are highly politicized. Because that state routinely figures in defining the union's effective environment, the union's relationship with state and political leaders becomes more important than its dealings with the employers of the workers they represent. Not only may agreements bear no relation to a firm's economic circumstances, but they increase dissonance between workers and union leaders. What is needed instead is a system of collective bargaining that directs unions'efforts to the ultimate lasting source of their members welfare: the firm they work for. Ultimately, improving workers'standard of living required growth in productivity, argues the author. Raising a worker's earning by redistributing income from profits, dividends, and interest cannot sustain a persistent rise in earnings. And mandating or encouraging high wage policies (as in Latin America and the Caribbean) discourages the economic growth that is the ultimate durable source of improvements in workers living standards. Unions can help raise productivity in the workplace by participating with management in the search for better ways of organizing production. It is important for workers not to feel alienated from the system and to believe they have a stake in it. They value the fact that they or their agents help to shape the working environment. In determining the"rules of the game"in which labor unions operate, societies have wrestled with the problem of finding the proper balance between upholding the principle of free association, on the one hand, and on granting entitlement that result in resource inefficiencies at best and ultimately in challenges to the authority of the democratic state at worst. Governments'main responsibility in labor relations is to set up the regulations that underpin labor market interactions, including the legal framework should neither encourage nor discourage unionism, says the author, but should keep the activities of unions in the domain where they can be productive: the enterprise.

Jackknife Instrumental Variables Estimation

Abstract: Two-stage-least-squares (2SLS) estimates are biased towards OLS estimates. This bias grows with the degree of over-identification and can generate highly misleading results. In this paper we propose two simple alternatives to 2SLS and limited-information-maximum-likelihood (LIML) estimators for models with more instruments than endogenous regressors. These estimators can be interpreted as instrumental variables procedures using an instrument that is independent of disturbances even in finite samples. Independence is achieved by using a `leave-one-out' jackknife-type fitted value in place of the usual first-stage equation. The new estimators are first-order equivalent to 2SLS but with finite-sample properties superior to those of 2SLS and similar to LIML when there are many instruments. Moreover, the jackknife estimators appear to be less sensitive than LIML to deviations from the linear reduced form used in classical simultaneous equations models.

The Schooling Quality-Earnings Relationship: Using Economic Theory to Interpret Functional Forms Consistent with the Evidence

Abstract: This paper investigates the economic and empirical foundations of the evidence relating earnings to schooling quality We replicate the Card-Krueger model for Census years 1970, 1980 and 1990 and find that it consistently produces a strong relationship between schooling quality and the rate of return to schooling We test key identifying assumptions used by Card and Krueger and others Several assumptions are rejected When they are relaxed, the evidence for a strong effect of schooling quality on earning is greatly weakened A crucial identifying assumption is the absence of selective migration on the basis of earnings Nonparametric tests strongly reject this hypothesis The conventional assumption of linearity of the earnings- schooling relationship widely used in the literature is also rejected The only surviving evidence of any schooling quality effect is in the return to college education We also test and reject conventional efficiency unit models of the pricing of labor services The empirically concordant model of earnings is a model of heterogeneous human capital in which regional shocks affect the prices of less- skilled labor

Differential Mortality and Wealth Accumulation

Abstract: The issue of asset accumulation and decumulation is central to the life cycle theory of consumer behavior and to many policy questions. One of the main implications of the life cycle model is that assets are decumulated in the last part of life. Most empirical studies in this area use cross-sectional data of estimate mean or median wealth-age profiles. The use of cross-sections to estimate the age profile of assets is full of pitfalls. For example, if wealth and mortality are related, in that poorer individuals die younger, one overestimates the last part of the wealth-age profile when using cross-sectional data because means (or other measures of location) are taken over a population which becomes 'richer' as it ages. This paper examines the effect of differential mortality on cross-sectional estimates of wealth-age profiles. Our approach is to quantify the dependence of mortality rates on wealth and use these estimates to 'correct' wealth-age profiles for sample selection due to differential mortality. We estimate mortality rates as a function of wealth and age for a sample of married couples drawn from the Survey of Income and Program Participation (SIPP). Our results show that accounting for differential mortality produces wealth profiles with significantly more dissaving among the elderly.

Is High School Employment Consumption or Investment

Abstract: Using data from the National Longitudinal Survey of Youth, this study examines whether employment by high school students improves or worsens economic attainment 6 to 9 years after the scheduled date of high school graduation. There is no indication that light to moderate job commitments ever have a detrimental impact and hours worked during the senior grade are positively correlated with future earnings, fringe benefits, and occupational status. These results are robust across a variety of specifications and suggest that employment increases net investments in human capital and facilitates the school- to-work transition, particularly towards the end of high school and for students not continuing on to college.

Unemployment insurance and duration of unemployment: evidence from Slovenia's transition

Abstract: Between 1990 and 1992 in Slovenia, recipients of unemployment insurance (UI) benefits tended to remain (formally) unemployed until their benefits expired, before taking a job. Institutional set-up suggests, and labor surveys show, that many of the recipients were actually working while collecting UI benefits. In the spirit, if not in the letter of the law, the UI system was abused. The author shows that the escape rate of the recipients of unemployment compensation to employment increased dramatically just before the potential exhaustion of unemployment benefits - and decreased equally dramatically after benefits were exhausted. When grouped by the potential duration of benefits, unemployment length varies significantly. The unemployed with longer potential benefits stay unemployed longer. Because these groups differ in their characteristics (for example, in age), this does not prove the"waiting behavior"of the recipients. However, exits to employment dramatically increase just before exhaustion - and that does prove waiting behavior. The pattern of an increased escape rate just before benefits are exhausted and its dramatic fall thereafter is more rigorously demonstrated using hazard model estimation. Possibilities for informal employment are abundant in Slovenia, and the environment of transition economies generally seems conducive to misuse of the UI system. Legislative loopholes and failure to enforce the labor code allowed the unemployed to work and to collect benefits. The monitoring of job searches was also lax. The author's calculations suggest that reducing the duration of benefits would reduce the incidence of unemployment, its duration, the amount spent on UI benefits, and the inefficiencies generated by raising taxes to finance unemployment insurance. At the same time, reducing the duration of benefits would not impair job matches or crowd out jobs for nonrecipients. True, despite increased efficiency generally, the workers with the least job mobility might suffer hardships for the least mobile group and greater efficiency generally would have to be resolved in the political sphere. Redesigning the system for better targeting would be less controversial. One way to reduce UI spending without seriously curtailing incentives to work would be to reduce the benefits in proportion to earnings from irregular work. Another possibility is stricter monitoring of the job searches of the unemployed. To reduce spending and make"double dipping"less attractive, old-age insurance could be removed from the package of benefits the UI system offers. Also, counselors who help the unemployed find jobs (and who may thus develop a close relationship with them) should perhaps not be expected to be able to make impartial decisions about disqualifications for benefits; someone else should do that. In addition to better targeting, a"benefit transfer program"- a voluntary program that converts UI benefits Also, counselors who help the unemployed find jobs (through vouchers) into hiring subsidies - seems particularly attractive for Slovenia and other transition economies. In a way, such a program would legalize the"double-dipping"that has been taking place in Slovenia and possibly elsewhere. It would legalize practices that have undermined the system's credibility. But it might improve fiscal savings while sustaining the incentive to find jobs.

Beauty, Productivity and Discrimination: Lawyers' Looks and Lucre

Abstract: We propose several models in which an ascriptive characteristic generates earnings differentials and is sorted across sectors. The general approach shows how to distinguish the ultimate sources of labor-market returns to such characteristics; the specific example uses longitudinal data on a large sample of attorneys who graduated from one law school. Beauty is measured by ratings of their matriculation photographs. 1) Better-looking attorneys who graduated in the 1970s earned more after 5 years of practice than their worse- looking classmates, other things equal, an effect that grew even larger by the fifteenth year of practice. There is no impact of beauty on earnings among 1980s graduates. 2) Attorneys in the private sector are better-looking than those in the public sector, with the differences rising as workers sort across sector based on their beauty. 3) Male attorneys' probability of attaining an early partnership rises with beauty. The results support a theory of dynamic sorting and the role of customer behavior. We cannot determine whether this is because clients discriminate or because better-looking lawyers are able to obtain greater pecuniary gains for their clients.

Average Causal Response with Variable Treatment Intensity

Abstract: In evaluation research, an average causal effect is usually defined as the expected difference between the outcomes of the treated, and what these outcomes would have been in the absence of treatment. This definition of causal effects makes sense for binary treatments only. In this paper, we extend the definition of average causal effects to the case of variable treatments such as drug dosage, hours of exam preparation, cigarette smoking, and years of schooling. We show that given mild regularity assumptions, instrumental variables independence assumptions identify a weighted average of per-unit causal effects along the length of an appropriately defined causal response function. Conventional instrumental variables and Two-Stage Least Squares procedures can be interpreted as estimating the average causal response to a variable treatment.

The Survival of New Plants: Start-Up Conditions and Post-Entry Evolution

Abstract: Studies the longevity of startups and the determinants of survival for new manufacturing plants, using econometric duration models to analyze the timing of plant exits. A database of 17,000 Portuguese manufacturing plants opened in the 1980s is used to determine effect of plant size on the likelihood of exit (through 1990). The regression results indicate that current (rather than initial) plant size correlates with the chances of survival. Since larger organizations tend to be more efficient, they are less likely to exit. In addition, plants that have started smaller but experienced faster post-entry growth have a greater probability of survival. Patterns of survival differ by type of entrant -- e.g., whether a plant is established in the same industry as its parent. Results also show that, in industries characterized by high entry rates where new firms face more competition from other startups, post-entry survival is more difficult. However, in fast growing industries where market entry can be achieved without causing harm to competitors, plants tend to survive longer. (SFL)

Who Works When? Evidence from the U.S. and Germany

Abstract: This study uses data for the U.S. from the May 1991 CPS and for Germany from the 1990 wave of the Socioeconomic Panel (GSOEP) to analyze when people work during the day and week. The evidence shows: 1) Work in the evenings or at night is quite common in both countries, with around 7 percent of workers on the job even at 3AM; 2) Such work is performed mostly by people who are not shift workers; 3) Work at these times is inferior, in that it is performed disproportionately by people with little human capital; 4) Minority workers in the U.S. and the foreign-born in Germany are especially likely to work at these undesirable times; 5) Evening and night work is least likely in large metropolitan areas; 6) Spouses tend to work at the same time of the day; but 7) Young children break down the joint timing of spouses' work, with the burden of evening and night work falling disproportionately on working mothers. The findings demonstrate the gains to basing the analysis of work and leisure on data describing instantaneous time use.

Alcohol Policies and Highway Vehicle Fatalities

Abstract: This study investigates the impact of beer taxes and a variety of alcohol-control policies on motor vehicle fatality rates, using fixed- effect models with data for the 48 contiguous states over the 1982 through 1988 time period. The econometric findings highlight the fragility of the parameter estimates to reasonable changes in model specifications. Special attention is paid to omitted variables biases resulting from failing to adequately control for grassroots efforts to reduce drunk driving, the enactment of other laws which simultaneously operate to reduce highway fatalities, and the economic conditions existing at the time of the legislation. In the preferred specifications, most of the regulations have little or no impact on traffic mortality. By contrast, higher beer taxes are associated with reductions in crash deaths and this result is relatively robust across specifications. These findings suggest the limited ability of further regulatory action to reduce drunk-driving but point to a potentially significant role for higher alcohol taxes.

Unemployment Insurance and Household Welfare: Microeconomic Evidence 1980-93

Abstract: This study examines the relative economic well-being of households that receive unemployment insurance (UI) benefits, as measured by consumption flows that are derived from information on households' spending in the Consumer Expenditure Surveys from 1980- 1993. For each quarter during this period we obtain the per-capita and equivalence-scale adjusted economic welfare of the two types of households. Adjusting for differences in the households' characteristics, we find: 1) The average UI recipient household during this period had a level of economic well-being that was on average between 3 and 8 percent below that of otherwise identical households (depending on the welfare measure used); 2) During a substantial part of this time the economic well-being of households that received UI benefits was at least that of other households; and 3) There is no cyclical variation in the relative well-being of UI recipient households compared to others. The findings imply that during the 1980s and early 1990s states' UI programs did a satisfactory job of maintaining the well-being of UI recipients. Emergency programs enacted during recessions raised potential duration sufficiently to prevent the economic position of the average UI recipient from deteriorating.

Non-Parametric Demand Analysis with an Application to the Demand for Fish

Abstract: Instrumental variables (IV) estimation of a demand equation using time series data is shown to produce a weighted average derivative of heterogeneous potential demand functions. This result adapts recent work on the causal interpretation of two-stage least squares estimates to the simultaneous equations context and generalizes earlier research on average derivative estimation to models with endogenous regressors. The paper also shows how to compute the weights underlying IV estimates of average derivatives in a simultaneous equations model. These ideas are illustrated using data from the Fulton Fish market in New York City to estimate an average elasticity of wholesale demand for fresh fish. The weighting function underlying IV estimates of the demand equation is graphed and interpreted. The empirical example illustrates the essentially local and context-specific nature of instrumental variables estimates of structural parameters in simultaneous equations models.