Showing papers by "Joshua D. Angrist published in 2001"

Instrumental Variables and the Search for Identification: From Supply and Demand to Natural Experiments

Abstract: Instrumental variables was first used in the 1920s to estimate supply and demand elasticities and later to correct for measurement error in single equation models. Recently, instrumental variables have been widely used to reduce bias from omitted variables in estimates of causal relationships. Intuitively, instrumental variables methods use only a portion of the variability in key variables to estimate the relationships of interest; if the instruments are valid, that portion is unrelated to the omitted variables. We discuss the mechanics of instrumental variables and the qualities that make for a good instrument, devoting particular attention to instruments derived from "natural experiments."

Estimation of Limited Dependent Variable Models With Dummy Endogenous Regressors: Simple Strategies for Empirical Practice

Abstract: Applied economists have long struggled with the question of how to accommodate binary endogenous regressors in models with binary and nonnegative outcomes. Iargue here that much of the dife culty with limited dependent variables comes from a focus on structural parameters, such as index coefe cients, instead of causal effects. Once the object of estimation is taken to be the causal effect of treatment, several simple strategies are available. These include conventional two-stage least squares, multiplicative models for conditional means, linear approximation of nonlinear causal models, models for distribution effects, and quantile regression with an endogenous binary regressor. The estimation strategies discussed in the article are illustrated by using multiple births to estimate the effect of childbearing on employment status and hours of work.

How Do Sex Ratios Affect Marriage and Labor Markets? Evidence from America's Second Generation

Abstract: Sex ratios, i.e., relative numbers of men and women, can affect marriage prospects, labor force participation, and other social and economic variables. But the observed association between sex ratios and social and economic conditions may be confounded by omitted variables and reverse causality. This paper uses variation in immigrant flows as a natural experiment to study the effect of sex ratios on the children and grandchildren of immigrants. The flow of immigrants affected the second generation marriage market because second generation marriages were mostly endogamous, i.e., to members of the same ethnic group. The empirical results suggest that high sex ratios had a large positive effect on the likelihood of female marriage, and a large negative effect on female labor force participation. Perhaps surprisingly, the marriage rates of second generation men appear to be a slightly increasing function of immigrant sex ratios. Higher sex ratios also appear to have raised male earnings and the incomes of parents with young children. The empirical results are broadly consistent with theories where higher sex ratios increase female bargaining power in the marriage market.

Instrumental Variables And The Search For Identification: From Supply And Demand To Natural Experiments

Abstract: The method of instrumental variables was first used in the 1920s to estimate supply and demand elasticities, and later used to correct for measurement error in single-equation models. Recently, instrumental variables have been widely used to reduce bias from omitted variables in estimates of causal relationships such as the effect of schooling on earnings. Intuitively, instrumental variables methods use only a portion of the variability in key variables to estimate the relationships of interest; if the instruments are valid, that portion is unrelated to the omitted variables. We discuss the mechanics of instrumental variables, and the qualities that make for a good instrument, devoting particular attention to instruments that are derived from "natural experiments." A key feature of the natural experiments approach is the transparency and refutability of identifying assumptions. We also discuss the use of instrumental variables in randomized experiments.

Consequences of employment protection? : the case of the Americans with Disabilities Act

Abstract: The Americans with Disabilities Act (ADA) requires employers to accommodate disabled workers and outlaws discrimination against the disabled in hiring, firing, and pay. Although the ADA was meant to increase the employment of the disabled, the net theoretical effects are ambiguous. For men of all working ages and women under 40, Current Population Survey data show a sharp drop in the employment of disabled workers after the ADA went into effect. Although the number of disabled individuals receiving disability transfers increased at the same time, the decline in employment of the disabled does not appear to be explained by increasing transfers alone, leaving the ADA as a likely cause. Consistent with this view, the effects of the ADA appear larger in medium‐size firms, possibly because small firms were exempt from the ADA. The effects are also larger in states with more ADA‐related discrimination charges.

New Evidence on Classroom Computers and Pupil Learning

Abstract: The question of how technology affects learning has been at the center of recent debates over educational inputs. In 1994, the Israeli State Lottery sponsored the installation of computers in many elementary and middle schools. This program provides an opportunity to estimate the impact of computerization on both the instructional use of computers and pupil achievement. Results from a survey of Israeli school-teachers show that the influx of new computers increased teachers’ use of computer-aided instruction (CAI) in the 4th grade, with a smaller effect on CAI in 8th grade. Although many of the estimates are imprecise, on balance, CAI does not appear to have had educational benefits that translated into higher test scores. OLS estimates show no evidence of a relationship between CAI and test scores, except for a negative effect on 8th grade Math scores in models with town effects. IV estimates for 4th graders show lower Math scores in the group that was awarded computers, with smaller (insignificant) negative effects on language scores.

How Large are Human-Capital Externalities? Evidence from Compulsory-Schooling Laws

Abstract: Many economists and policymakers believe that education creates positive externalities. Indeed, average schooling in U.S. states is highly correlated with state wage levels, even after controlling for the direct effect of schooling on individual wages. We use variation in child labor laws and compulsory attendance laws over time and across states to investigate whether this relationship is causal. Our results show external returns to education around 1% and not significantly different from zero.

Does Teacher Training Affect Pupil Learning? Evidence from Matched Comparisons in Jerusalem Public Schools

Abstract: Most research on the relationship between teacher characteristics and pupil achievement focuses on salaries, experience, and education. The effect of in‐service training has received less attention. We estimate the effect of in‐service teacher training on achievement in Jerusalem elementary schools using a matched‐comparison design. Differences‐in‐differences, regression, and matching estimates suggest training in secular schools led to an improvement in test scores. The estimates for religious schools are not clear cut, perhaps because training in religious schools started later and was implemented on a smaller scale. Estimates for secular schools suggest teacher training provided a cost‐effective means of increasing test scores.

Vouchers for Private Schooling in Colombia: Evidence from a Randomized Natural Experiment

Abstract: Colombia's PACES program provided over 125,000 pupils from poor neighborhoods with vouchers that covered approximately half the cost of private secondary school. Since many vouchers were allocated by lottery, we use differences in outcomes between lottery winners and losers to assess program effects. Three years into the program, lottery winners were 15 percentage points more likely to have attended private school, had completed .1 more years of schooling, and were about 10 percentage points more likely to have finished 8 th grade, primarily because they were less likely to repeat grades. The program did not significantly affect dropout rates. Lottery winners scored .2 standard deviations higher on standardized tests. There is some evidence that winners worked less than losers and were less likely to marry or cohabit as teenagers. On average, lottery winners increased their educational expenditure by about 70% of the value of the voucher. Since winners also worked less, they devoted more total resources to education. Compared to an equivalent expansion of the public education system, the voucher program increased annual government educational expenditure by about $24 per winner. But the costs to the government and to participants were probably much less than the increase in winners' earnings due to greater educational attainment.

Instrumental Variables and the Search for Identification: From Supply and Demand to Natural Experiments

Abstract: Instrumental variables was first used in the 1920s to estimate supply and demand elasticities and later to correct for measurement error in single equation models. Recently, instrumental variables have been widely used to reduce bias from omitted variables in estimates of causal relationships. Intuitively, instrumental variables methods use only a portion of the variability in key variables to estimate the relationships of interest; if the instruments are valid, that portion is unrelated to the omitted variables. We discuss the mechanics of instrumental variables and the qualities that make for a good instrument, devoting particular attention to instruments derived from "natural experiments."

Protective or Counter-Productive? European Labor Market Institutions and the Effect of Immigrants on EU Natives

Abstract: We estimate the effect of immigrant flows on native employment in Western Europe, and then ask whether the employment consequences of immigration vary with institutions that affect labor market flexibility. Reduced flexibility may protect natives from immigrant competition in the near term, but our theoretical framework suggests that reduced flexibility is likely to increase the negative impact of immigration on equilibrium employment. In models without interactions, OLS estimates for a panel of European countries in the 1980s and 1990s show small, mostly negative immigration effects. To reduce bias from the possible endogeneity of immigration flows, we use the fact that many immigrants arriving after 1991 were refugees from the Balkan wars. An IV strategy based on variation in the number of immigrants from former Yugoslavia generates larger though mostly insignificant negative estimates. We then estimate models allowing interactions between the employment response to immigration and institutional characteristics including business entry costs. These results, limited to the sample of native men, generally suggest that reduced flexibility increases the negative impact of immigration. Many of the estimated interaction terms are significant, and imply a significant negative effect on employment in countries with restrictive institutions.

Protective or Counter-Productive? European Labor Market Institutions and the Effect of Immigrants on EU Natives

Abstract: We estimate the effect of immigrant flows on native employment in Western Europe, and then ask whether the employment consequences of immigration vary with institutions that affect labor market flexibility Reduced flexibility may protect natives from immigrant competition in the near term, but our theoretical framework suggests that reduced flexibility is likely to increase the negative impact of immigration on equilibrium employment In models without interactions, OLS estimates for a panel of European countries in the 1980s and 1990s show small, mostly negative immigration effects To reduce bias from the possible endogeneity of immigration flows, we use the fact that many immigrants arriving after 1991 were refugees from the Balkan wars An IV strategy based on variation in the number of immigrants from former Yugoslavia generates larger though mostly insignificant negative estimates We then estimate models allowing interactions between the employment response to immigration and institutional characteristics including business entry costs These results, limited to the sample of native men, generally suggest that reduced flexibility increases the negative impact of immigration Many of the estimated interaction terms are significant, and imply a significant negative effect on employment in countries with restrictive institutions