Lower and upper bounds of returns to schooling: An exercise in IV estimation with different instruments

Several recent studies based on `exogenous' sources of variation in educational outcomes show IV estimates of returns to schooling that are substantially higher than the corresponding OLS estimates. Card (1995a, Earnings, schooling, and ability revisited. Research in Labor Economics 14, 23–48) suggests that these results are explained by the existence of heterogenity in individual returns and by the fact that these studies are based on instruments that influence only the educational decision of individuals with high marginal returns due to either liquidity constraints or to high ability. This conclusion is consistent with the local average treatment effect (LATE) interpretation of IV (Imbens and Angrist (1994, Identification and estimation of local average treatment effects. Econometrica 62, 467–475) according to which IV identifies only the average returns of those who comply with the assignment-to-treatment mechanism implied by the instrument. We show evidence for Germany suggesting that returns to schooling are heterogeneous, instruments matter and the LATE interpretation of IV makes sense. With an appropriate choice of instruments we also show how IV can be used to approximate the range of variations of returns to schooling in Germany.     

Income inequality and education revisited: persistence,endogeneity and heterogeneity

This article presents new results on the relationship between income inequality and education expansion – that is, increasing average years of schooling and reducing inequality of schooling. When dynamic panel estimation techniques are used to address issues of persistence and endogeneity, we find a large, positive, statistically significant and stable relationship between inequality of schooling and income inequality, especially in emerging and developing economies and among older-age cohorts. The relationship between income inequality and average years of schooling is positive, consistent with constant or increasing returns to additional years of schooling. While this positive relationship is small and not always statistically significant, we find a statistically significant negative relationship with years of schooling of younger cohorts. Statistical tests indicate that our dynamic estimators are consistent and that our identifying instruments are valid. Policy simulations suggest that education expansion will continue to be inequality reducing. This role will diminish as countries develop, but it could be enhanced through a stronger focus on reducing inequality in the quality of education.     

Returns to schooling for urban and migrant workers in China: a detailed investigation

We use a new data set, the 2009 Rural Urban Migration in China (RUMiC) to estimate returns to schooling in China using instrumental variable (IV) estimation. After identifying a set of instruments, we conduct comprehensive validity and relevance testing of different combinations of instruments as well as robustness analysis of our estimates for rural-to-urban migrants and urban residents in China. We find that our point estimates are in the 6–9% range for urban workers compared to 7–8% for migrant workers. Returns for men (at 8–9%) are slightly higher than for women (at 6–7%). Thus, private returns to education in urban China in 2009 were not as high as other transition and developing countries, but substantial and have increased over time. Comparing OLS and IV estimates, we also find that the attenuation bias due to measurement error is generally large and more important in the migrant sample compared to the urban sample.     

Entry into the Schooling Market: How is the Behaviour of Private Suppliers Influenced by Public Sector Decisions?

This research examines the location choice of private schools entering the California schooling market in 1979–80. We find that entrants are more likely to locate in public school districts with lower levels of per–pupil expenditure and higher fractions of public school students who reside in low–income households. In addition, we provide evidence of differences in the responsiveness of different types of private schools to the underlying conditions. Also, in comparing our results to those of previous research, we find that the determinants of the location choices of entrants appear to be the same as the determinants of the location pattern of incumbent private schools.     

The effects of generalized school choice on achievement and stratification: Evidence from Chile's voucher program

In 1981, Chile introduced nationwide school choice by providing vouchers to any student wishing to attend private school. As a result, more than 1000 private schools entered the market, and the private enrollment rate increased by 20 percentage points, with greater impacts in larger, more urban, and wealthier communities. We use this differential impact to measure the effects of unrestricted choice on educational outcomes. Using panel data for about 150 municipalities, we find no evidence that choice improved average educational outcomes as measured by test scores, repetition rates, and years of schooling. However, we find evidence that the voucher program led to increased sorting, as the “best” public school students left for the private sector.     

Returns to schooling in Kazakhstan

This paper examines rates of return to schooling in Kazakhstan using OLS (Ordinary Least Squares) and instrumental variable (IV) methodologies. We use spouse's education and smoking as instruments. We find that spouse's education is a valid instrument and that conventional OLS estimates that assume the exogenous nature of schooling, and hence do not control for endogeneity bias, may underestimate the true rates of return. The results indicate that the returns to schooling in Kazakhstan have increased with transition. This may reflect the relative scarcities of highly educated people in Kazakhstan with human capital that employers require and, following the market reforms, reward accordingly.     

Inference approaches for instrumental variable quantile regression

We consider asymptotic and finite sample confidence bounds in instrumental variables quantile regressions of wages on schooling with relatively weak instruments. We find practically important differences between the asymptotic and finite sample interval estimates.     

Exploring the spatial determinants of children’s activities: evidence from India

This article investigates the choice of children’s activities in India and provides recommendations for areas where policy intervention to promote schooling and combat child labor would be most successful. First, we recognize that child schooling and labor are not the sole activities that children can engage in and include idleness as one of the choices. Second, we use a hierarchical model with spatially correlated random effects to analyze the determinants of the choice of children’s activities. Lastly, we recommend that pro-schooling intervention be implemented in districts with favorable attitudes toward schooling and unfavorable attitudes toward idleness, while anti-child-labor interventions be implemented in districts where attitudes toward child labor are less favorable. We thus identify two groups of Indian districts to target appropriate government interventions.     

Early Tracking and the Misfortune of Being Young

Recent research suggests that the relative age of a student within a grade has a causal effect on educational achievement, and that this effect fades with the duration of schooling. In this study, we estimate the causal relative‐age effect on track choice in Austria, a country where students are first tracked in grade 5 (at the age of 10 years), and again in grade 9. We find a strong positive relative‐age effect on track choice in grades 5–8. The age effect persists beyond grade 8 for students from less‐favorable socioeconomic backgrounds and students in urban areas.     

What factors drive individual misperceptions of the returns to schooling in Tanzania? Some lessons for education policy

Evidence on educational returns and the factors that determine the demand for schooling in developing countries is extremely scarce. Building on previous studies that show individuals underestimating the returns to schooling, we use two surveys from Tanzania to estimate both the actual and perceived schooling returns and subsequently examine what factors drive individual misperceptions regarding actual returns. Using OLS and instrumental variable methods, we find that each additional year of schooling in Tanzania increases earnings, on average, by 9 to 11%. We find that on average individuals underestimate returns to schooling by 74 to 79% and three factors are associated with these misperceptions: income, asset, poverty and educational attainment. Shedding light on what factors relate to individual beliefs about educational returns can inform policy on how to structure effective interventions in order to correct individual misperceptions.     

The Decline of Schooling Productivity in OECD Countries

Based on Baumol's cost-disease model, we develop two alternative measures of the change in the productivity of schooling. Both productivity measures are based on changes in the relative price of schooling. We find that in most OECD countries the price of schooling has increased faster in 1970–94 than would be compatible with constant schooling productivity. In addition, we show that the average performance of pupils has remained constant at best in most OECD countries. Our results imply a larger decline in the productivity of schooling in many OECD countries than in the United States.     

An empirical analysis of assortative mating in India and the U.S.

This paper uses a conditional logit model to analyze empirically how individuals sort themselves through marriage into households in India and the U.S. The results support positive assortative mating of spouses with respect to age and schooling. We find no evidence in favor of Becker's theory of labor market specialization in couples. Moreover, while similarity in age is the strongest predictor of marital choice in India, education of a prospective spouse plays a more important role in the U.S. Finally, we find that while dowry increases the likelihood of women marrying men with characteristics dissimilar to their own, availability of a mate has a positive effect on the degree of stratification in India.     

The impact of rural electrification on income and education: Evidence from Bhutan

We investigate the impact of a rural electrification program on household income and children's schooling in rural Bhutan. Using propensity score matching, we find that electrification had a statistically significant impact on nonfarm income and education. Nonfarm income increased by 61 percent and children gained 0.72 additional years of schooling and 9 minutes of study time per day. We do not observe significant effects on farm income. Results are consistent and robust to different matching algorithms. Our findings indicate that investments in reducing energy deficit may help improve human welfare in Bhutan.     

Private Tutoring: Evidence from India

Drawing on the nationally representative “Participation and Expenditure in Education” surveys, we document the incidence and cost of private tutoring at different stages of schooling over the last two decades in India. As private tutoring involves two decisions: (1) whether to take private tuition or not, and (2) how much to spend on private tutoring conditional on positive decision in (1), we analyze the determinants of the two decisions separately using a Hurdle model. We find that private tutoring is not a new phenomenon in India: a significant proportion of students at each stage of schooling took private tutoring even in 1986/87, and there has been no dramatic increase in those proportions. Students from urban areas, private schools, and from better economic backgrounds are more likely to take private tutoring. Furthermore, the demand for private tutoring is inelastic at each stage of schooling, which implies that private tutoring is a necessary good in the household consumption basket. We also find evidence of pro‐male bias in both decisions regarding private tutoring.     

Does education affect risk aversion? Evidence from the British education reform

Individual risk attitudes are frequently used to predict decisions regarding education. However, using risk attitudes as a control variable for decisions about education has been criticized because of the potential for reverse causality. Causality between risk aversion and education is unclear, and disentangling the different directions it may run is difficult. In this study, we make the first attempt to investigate the causal effects of education on risk aversion by examining the British education reform of 1972, which increased the duration of compulsory schooling from age 15 to age 16. Using regression discontinuity design, we find that this additional year of schooling increases the level of risk aversion, which is contrary to previous findings in the literature, and we also find that this result is particularly strong for individuals with less education. This positive causal effect of education on risk aversion might alleviate concerns regarding the endogeneity/reverse causality issue when using risk aversion as an explanatory variable for decisions about education; the sign would remain credible because the coefficients are underestimated.     

The intergenerational transmission of education: Evidence from Taiwanese adoptions

This paper examines the causal effect of parental schooling on children’s schooling using a large sample of adoptees from Taiwan. Using birth-parents’ education to help control for selective placement of children with adoptive parents, we find that adoptees raised with more highly educated parents have higher educational attainment, measured by years of schooling and probability of university graduation. We also find evidence that adoptive father’s schooling is more important for sons’ and adoptive mother’s schooling is more important for daughters’ educational attainment. These results support the notion that family environment (nurture) is important in determining children’s educational outcomes, independent of genetic endowment.     

Long-term impact of investments in early schooling — Empirical evidence from rural Ethiopia

We examine the cumulative impact of early schooling investments on later schooling outcomes using enrollment status and relative grade attainment as short-run and long-run measures of schooling. Using a child-level longitudinal data set from rural Ethiopia, we estimate a dynamic conditional schooling demand function where the coefficient estimate on the lagged dependent variable captures the impact of all previous period schooling inputs and resources. We find that a child who is enrolled in the prior period is 33 percentage points more likely to be enrolled currently. These lagged effects are stronger for girls and for children from higher income households.     

Are We Spending Too Many Years in School? Causal Evidence of the Impact of Shortening Secondary School Duration

During the last decade, most of the German states have abolished the final year of higher secondary schooling while keeping academic content almost unaltered. We evaluate the effects of the reform in Saxony‐Anhalt for the double cohort of graduates in 2007. In 2003, the 13th year of schooling was eliminated for students in grade 9, while tenth grade students were unaffected. This provides a natural experiment for analyzing the impact on schooling achievements and educational choice. We find negative effects on grades in mathematics, but no effects in German literature. Moreover, a significant share of females were found to delay university enrollment.     

Fertility and development: the roles of schooling and family production

This paper presents a quantitative theory of development that highlights three mechanisms that relate schooling, fertility, and growth. First, we point out that in the early stages of development, fertility and schooling may rise together as the schooling of younger children increases their relative contribution to family income when they turn working age. Second, the model contains a supply-side theory of schooling that generates a rise in schooling independent of technological change. Third, we introduce a direct negative effect of industrialization on fertility that does not operate through human capital and the quantity-quality tradeoff. An initial quantitative assessment of the theoretical mechanisms is conducted by calibrating and applying the model to United States history from 1800 to 2000. We find that the demise in family production is an important factor reducing fertility in the 19th century and schooling of older children is dominant factor reducing fertility in the 20th century. The same model is applied to England from 1740 to 1940, where we offer two complimentary explanations for the rise in fertility from 1740 to 1820. The first is based on the rapid expansion in the cottage industry and the second on the increased relative productivity of children. We also find that the subsequent fall in fertility from 1820 to 1940 cannot be explained without introducing child labor/compulsory schooling laws.