Search

textabstractWe draw on a discontinuity at a large university, wherein second-year students with a low first-year grade point average are allocated to a full year of forced, frequent, and regular attendance, to estimate the causal effect of additional structure on academic performance. We show that the policy increases student attendance but has no average effect on grades. The effects differ, however, depending on how course instructors handled unforced students, such that we observe significant grade decreases in courses where unforced students were given full discretion over their attendance. Our evidence suggests that grades decrease in these courses because the policy prevented forced students from picking their desired mix of study inputs.

We study the effects of a policy aimed at attracting more experienced and better qualified teachers in primary schools in disadvantaged neighborhoods in Uruguay. Teachers in these schools could earn higher salaries. Estimates from regression discontinuity models show that the policy increased experience by two to three years. The policy was especially successful in ‘hiring experience from other schools’, but also increased tenure. However, the effect on student outcomes appears to be small. The distinction between ‘hiring or keeping’ teachers seems important for explaining this result. Keeping teachers appears to be more beneficial for students than hiring experienced teachers. We also find that the effect of the policy is better for schools that replaced teachers with less than five years of experience.

Many studies have found a strong association between economic outcomes of nations and their performance on international cognitive tests. This association is often interpreted as evidence for the importance of cognitive skills for economic growth. However, noncognitive skills also affect performance on cognitive tests. Following Borghans and Schils (2012), we exploit exogenous variation in the ordering of questions asked by the Programme for International Student Assessment (PISA) to decompose student performance into two components: the starting performance and the decline in performance during the test. The latter component is interpreted as a measure of noncognitive skills. Students from different countries exhibit differences in performance at the start of the test and in their rates of deterioration in performance during the test. Both components have a positive and statistically significant association with economic growth, and the estimated effects are quite similar and robust. Our results show that noncognitive skills are also important for the relationship between test scores and economic growth.

This paper investigates the empirical validity of the setup of a large-scale government neighbourhood investment programme in the Netherlands. Selection of neighbourhoods into the programme was determined by a measure of neighbourhood quality. At first sight this is a textbook example for the application of a regression discontinuity design to estimate the causal effect of the programme on neighbourhood outcomes. Neighbourhoods close to the threshold should be similar before the programme starts. However, at the discontinuity threshold we observe a surprisingly large gap in the share of non-Western immigrants between neighbourhoods that were selected into the programme and neighbourhoods that were not. In addition, there is non-compliance with the assignment rule based on the quality index. The pattern of non-compliance is consistent with investing in neighbourhoods with a high share of non-Western immigrants. Finally, the way in which neighbourhoods were defined could be a likely explanation for the imbalance in the share of non-Western immigrants at the discontinuity threshold.

This paper investigates the effect of physical and sexual child maltreatment on several types of illegal and problematic behaviour. By using variation within pairs of twins, we are able to mitigate concerns about confounding factors from previous studies. Using ordinary least squares and twin fixed effects estimation approaches, we find that child maltreatment has a large effect on illegal and problematic behaviours such as drug abuse, conduct disorders, and crime. The estimated effects suggest an increase of illegal and problematic behaviour between 50 and 100%. Our findings are consistent with the so-called cycle of violence hypothesis.

This paper examines the causal link that runs from classroom quality to student achievement using data on twin pairs who entered the same school but were allocated to different classrooms in an exogenous way. In particular, we apply twin fixed-effects estimation to assess the effect of teacher quality on student test scores from second through eighth grade of primary education, arguing that a change in teacher quality is probably the most important classroom intervention within a twin context. In a series of estimations using measurable teacher characteristics, we find that (a) the test performance of all students improves with teacher experience; (b) teacher experience also matters for student performance after the initial years in the profession; (c) the teacher experience effect is most prominent in earlier grades; (d) the teacher experience effects are robust to the inclusion of other classroom quality measures, such as peer group composition and class size; and (e) an increase in teacher experience also matters for career stages with less labor market mobility, which suggests positive returns to on-the-job learning of teachers.

We examine how the gender of a sibling affects labor market outcomes and family formation. Identification is complicated by parental preferences: if parents prefer certain sex compositions over others, children' s gender affects not only the outcomes of other children but also the existence of potential additional children. We employ two empirical strategies that both address this problem. First, we use a large sample of singletons to estimate whether first-borns are affected by the gender of their second-born sibling. Second, we look at a sample of dizygotic (i.e. non-identical) twins. We find that a same-sex sibling increases men's earnings and family formation outcomes (marriage and fertility), as compared to an opposite-sex sibling. The results for women are similar but the effects are smaller in magnitude and less robust. We argue that the income result for men could be driven by competition between brothers, as we find that men with brothers choose higher paying occupations. For women, we find suggestive evidence that the income premium may come partly from lower unemployment, which could be due to shared job search networks. The effects on family formation might stem from differential parental treatment for men, and from competition between sisters for women.

Many studies in economics use quarter of birth as an instrument for identifying the causal effect of schooling on outcomes such as earnings and health. The key assumption in these studies is that people born in different quarters of the year do not differ systematically in their unobserved abilities. This study uses genetic data from the US Health and Retirement Study to analyze the validity of the quarter of birth instrument. We find some evidence that genetic factors influencing education are not randomly distributed over the year. However, these factors only slightly change the effect of quarter of birth on schooling.

School dropout is an important social and economic problem. This paper investigates the effect of an intensive coaching program aimed at reducing school dropout rates among students aged 16–20. Students received support and guidance with their study activities, personal problems and internships in firms. The coaching program lasted one or two years. Students were randomly assigned to the coaching program. We find that one year of coaching reduced school dropout rates by more than 40% from 17 to 10 percentage points. The second year of coaching further reduced school dropout by 1 percentage point. The program is most effective for students with a high ex-ante probability of dropping out, such as students no longer obliged to be in formal education, male students, and students not living with both parents. Cost-benefit analysis suggests that one year of coaching is likely to yield a net social gain.

In many Dutch cities there are Geert Groote schools. This is no coincidence, because Geert Groote's (1340-1384) ideas form the foundation of modern education in the Netherlands and many other countries. Much less known is that his investments, and those of his followers, have put the Netherlands on the pathway towards the Golden Age. In this study, we describe the mechanism by which the influence of Groote so impressive. Furthermore, we present econometric estimates that indicate a long-lasting legacy. This research provides an explanation for high literacy, economic growth and societal developments in the Netherlands in the period before the Dutch Republic. We establish a link between the Brethren of the Common Life (BCL), a religious community founded by Geert Groote in the city of Deventer in the late fourteenth century, and the early development of the Netherlands. The BCL stimulated human capital accumulation by educating Dutch citizens without inducing animosity from the dominant Roman Catholic Church or other political rulers. Human capital had an impact on the structure of economic development in the period immediately after 1400. The educated workforce put pressure on the Habsburg monarchy leading to economic and religious resentment and eventually to the Revolt in 1572. The analyses show that the BCL contributed to the high rates of literacy in the Netherlands. In addition, there are positive effects of the BCL on book production and on city growth in the fifteenth and sixteenth century. Finally, we find that cities with BCL-roots were more likely to join the Dutch Revolt. These findings are supported by regressions that use distance to Deventer as an instrument for the presence of BCL. The results are robust to a number of alternative explanations. The statistical programme used for the regression analysis and the data used for this research can also be downloaded below (attachment, if necessary rename the file as a .zip).

We analyse the causal effect of education on starting and quitting smoking, using longitudinal data of Australian twins. We take into account the endogeneity of education, censoring of smoking durations and the timing of starting smoking versus that of completing education by using the flexible Mixed Proportional Hazard (MPH) specification. Unobserved effects in the specification are assumed to be twin specific and possibly correlated with completed education years. In addition, we use various unique control indicators reflecting the discounting behaviour of individuals that may affect both the smoking decision and the number of education years. In contrast to previous studies in our model specification, differences in the number of education years cannot explain differences in smoking behaviour at young ages. We find one additional year of education to reduce the duration of smoking with 9 months, but no significant effect of education on starting smoking. The effect of education on quitting smoking largely confines to male twins. This suggests that education policies that succeed in raising the level of education may improve public health through an increase of smoking cessation, but are not effective in preventing smoking at young ages.

This paper reports about a randomized experiment in which training vouchers of €1000 were given to low-skilled workers. The vouchers increase training participation by almost 20 percentage points in two years, relative to a base rate of 0.45. This increased participation comes at a substantial deadweight loss of almost 60%. Consistent with predictions from human capital theory, we find that vouchers cause a shift towards more general forms of training. We do not find any significant impact of the program on monthly wages or on job mobility. The program does, however, have a significant impact on future training plans. Compared to always-takers, new trainees are more often male, more risk averse, work shorter hours and are less likely to have participated in training prior to treatment. Compared to never-takers, they are more often female, work longer hours and have a somewhat lower formal education level.

This paper examines whether gay and lesbian workers sort into tolerant occupations. With information on sexual orientation, prejudice and occupational choice taken from Australian Twin Registers, we find that gays and lesbians shy away from prejudiced occupations. We show that our segregation results are largely driven by those gay and lesbian workers with disclosed identities, and robust to the inclusion of unobserved factors that are inherited and observed factors that strongly correlate with productive skills and vocational preferences. Our segregation estimates are generally large and consistent with prejudice based theories of employer and employee discrimination against gay and lesbian workers.

The genetic influence on the association between contemporaneously measured intelligence and academic achievement in childhood was examined in nationally representative cohorts from England and The Netherlands using a whole population indirect twin design, including singleton data. We identified 1,056 same-sex (SS) and 495 opposite-sex (OS) twin pairs among 174,098 British 11 year-olds with test scores from 2004, and, 785 SS and 327 OS twin pairs among 120,995 Dutch schoolchildren, aged 8, 10 or 12 years, with assessments from 1994 to 2002. The estimate of intelligence heritability was large in both cohorts, consistent with previous studies (h (2) = 0.70 +/- A 0.14, England; h (2) = 0.43 +/- A 0.28-0.67 +/- A 0.31, The Netherlands), as was the heritability of academic achievement variables (h (2) = 0.51 +/- A 0.16-0.81 +/- A 0.16, England; h (2) = 0.36 +/- A 0.27-0.74 +/- A 0.27, The Netherlands). Additive genetic covariance explained the large majority of the phenotypic correlations between intelligence and academic achievement scores in England, when standardised to a bivariate heritability (Biv h (2) = 0.76 +/- A 0.15-0.88 +/- A 0.16), and less consistent but often large proportions of the phenotypic correlations in The Netherlands (Biv h (2) = 0.33 +/- A 0.52-1.00 +/- A 0.43). In the British cohort both nonverbal and verbal reasoning showed very high additive genetic covariance with achievement scores (Biv h (2) = 0.94-0.98; Biv h (2) = 0.77-1.00 respectively). In The Netherlands, covariance estimates were consistent across age groups. The heritability of intelligence-academic achievement associations in two population cohorts of elementary schoolchildren, using a twin pair extraction method, is at the high end of estimates reported by studies of largely preselected twin samples. Refereed/Peer-reviewed

This paper estimates the longer-term effects of childhood conduct disorder on human capital accumulation and violent and criminal behavior later in life using data of Australian twins. We measure conduct disorder with a rich set of indicators based on diagnostic criteria from psychiatry. Using ordinary least squares and twin fixed effects estimation approaches, we find that early-age (pre-18) conduct disorder problems significantly affect both human capital accumulation and violent and criminal behavior over the life course. In addition, we find that conduct disorder is more deleterious if these behaviors occur earlier in life.

In many countries there is a deep concern about shortages of Science and Engineering workers (S&E). This article focuses on the effectiveness of policies aimed at stimulating the supply of S&E workers in the Netherlands. Despite the ‘common wisdom’ of severe and increasing shortages, we do not find evidence for a tight labour market of S&E workers. Instead, the data suggest that S&E workers have become less scarce since 1996. Stimulating enrolment in S&E studies may not be an effective policy for increasing R&D activity in the Netherlands because the majority of Dutch S&E freshmen do not end up working in R&D. They drop out during their S&E study or choose other jobs. In addition, the internationalization of the market for S&E workers tends to counter the effects of supply-side policies because the growing supply of foreign S&E graduates puts downward pressure on wages. As a result, demand-side policies may be more effective because they are directly targeted at fostering R&D.

This paper investigates the effect of a major reform in the final years of the two highest levels of Dutch secondary education. The reform focused on increasing active and independent learning and aimed to improve the match between secondary and higher education. We use data from six graduation cohorts from Dutch higher education and exploit the fact that two recent graduation cohorts contain both students that were treated with the new curriculum and students that were treated with the old curriculum. The effects of the curriculum change have been estimated by using a difference-in-differences approach for matched samples of treated and untreated graduates. We find that the reform had no effect on student performance and labour market outcomes.

This paper investigates the effect of the timing of tracking on completion of higher education by exploiting unique variation from the Dutch education system. At the age of 12 Dutch students can enrol in tracked schools or in comprehensive schools. The comprehensive schools postpone enrolment into tracked classes by one or two years. OLS- and IV-estimates, using regional variation in the supply of schools as instruments, show that early tracking has a detrimental effect on completion of higher education for students at the margin of the Dutch high and low tracks. The negative effects of early tracking are larger for students with relatively high ability or students with a higher socioeconomic background. In addition, we find no negative effects of comprehensive classes on higher ability students. These results suggest that increasing participation in comprehensive classes would increase graduation from higher education.

This paper investigates the effects of a teacher strike on student achievement. From May 1990 until November 1990 teachers in the French community of Belgium stroked to obtain a salary increase. We exploit the political division of Belgium in a French community and a Flemish community, with similar educational institutions, for estimating the long-term effects of the strikes. Based on a difference-in-differences approach, using data from two different surveys, we find some evidence that the strikes reduced educational attainment and increased class repetition. We also find that the strikes led to a significant reallocation of students to a lower level of higher education. Overall, the results suggest that teacher strikes can lead to substantial costs for those not involved in the conflict.

We use a novel quasi-experimental strategy to estimate the effect of expanding early schooling enrollment possibilities on early achievement. It exploits two features of the school system in The Netherlands. The first is rolling admissions; children are allowed to start school immediately after their 4th birthday instead of at the beginning of the school year. The second is that children having their birthday before, during and after the summer holiday are placed in the same class. These features generate sufficient exogenous variation in children’s enrollment opportunities to identify its effects on test scores. Making available one additional month of time in school increases language scores of disadvantaged pupils by 6 percent of a standard deviation and their math scores by 5 percent of a standard deviation. For non-disadvantaged pupils we find no effect.

This paper estimates the causal effect of teenage childbearing on educational attainment using two cohorts of Australian twins and their relatives. Our main finding is that the negative effect of teenage childbearing on educational attainment appears to be small. We find no difference in educational attainment between teen mothers and their identical twin sisters. Data on the relatives of the twins enable us to compare a teen mother with both her twin sister and her other sibling sisters. When twin sisters are used as a control group instead of sibling sisters, the estimated difference in educational attainment is much smaller.

This article investigates the effect of Dutch primary school inspections on test scores. These inspections are assessments of the educational quality of schools. The authors use two approaches to estimate the effect of the inspections. The first approach, based on a standard fixed effects model, indicates that inspections lead to slightly better school performance: Test scores increase by 2% to 3% of a standard deviation. However, the fixed effects model might be sensitive to selection bias. The second approach exploits a sample of randomly selected schools originally drawn for the purpose of compiling the annual report of the state of Dutch education. Based on this approach, the authors find that inspections do no harm but seem to have little or no effect on student performance.

This paper investigates whether studying abroad increases the propensity to live abroad later on. We use an instrumental variable approach based on cut-offs in the ranking of Dutch higher education graduates who applied for a scholarship programme for outstanding students. Applicants ranked above the cut-off received a scholarship to study abroad. Applicants ranked below the cut-off were denied a scholarship. Assignment of a scholarship increases the probability to study abroad and the number of months spent studying abroad. Studying abroad and the number of months spent studying abroad increase the probability of currently living abroad.

This study investigates the impact of two local calamities-an explosion at a firework factory in the city of Enschede on 13 May and a fire at a discotheque on 1 January 2001 during a New Years Eve party in the town of Volendam-on the educational achievement of children at the end of their primary education. Based on a quasi-experimental design with both control groups and pre-tests, we found that in the three years following the two tragedies, the test scores of girls in those areas closest to the events were on average 0.2 standard deviations lower. This corresponds to a downward shift in the distribution of girls' test scores. Boys' test scores, meanwhile, were not significantly affected by the disasters, and nor were the scores of pupils from nearby areas. In the three years following the calamities, girls' test scores in one of the areas (Volendam) have slowly recovered, although they remain well below their pre-event level.

Previous studies based Oil Population cohorts born at least 35 years ago, have reported appreciable childhood cognitive deficits for twins. We compared longitudinal IQ scores from approximately 188,000 singletons and some 6000 twins who went to primary school in the Netherlands from 1994 to 2003. In addition, we used a family-based design in which IQ scores of adult twins (N = 196) were compared with those of their adult singleton siblings (N = 589). After correcting for such Confounding factors as the year of testing, gender, age at the time of the test, and parents' education and ethnicity, twins aged 6 scored 16% of a standard deviation lower than non-twins in language and 17% of a standard deviation lower in arithmetic. I-or twins aged 8 the difference with non-twins in language and arithmetic reduced to 5% and 2% of a standard deviation and for twins aged 10 and 12 the differences were not statistically significant. For IQ scores, twins scored 0.09 points lower than non-twins at age 8 and 0.83 points lower at 10. However, twins scored higher at age 12 by 0.14 points. The only significant difference found was at age 10. Using the family-based adult sample, no differences in IQ scores were found between twins and their singleton siblings. These results suggest that in a very recent generation of school children in the Netherlands, there was a small but significant cognitive deficit for twins aged 6 and 8. However, the difference disappeared by the time the children were 12, and was also insignificant in the adult population. Previous studies, based oil cohorts born more than 35 years ago in Britain, reported much larger cognitive deficits in twins. Whatever the reason of the cognitive deficit at age 6, which Could include prenatal growth, shorter gestation and parental care, twins caught up and the cognitive cost of being a twin in the Netherlands seems to be minor and temporary. (C) 2007 Elsevier Inc. All rights reserved.

This paper evaluates the effects of two subsidies targeted at schools with large proportions of disadvantaged pupils. The first scheme gives primary schools with at least 70% disadvantaged minority pupils extra funding for personnel. The second scheme gives primary schools with at least 70% pupils from any disadvantaged group extra funding for computers and software. The cutoffs provide a regression discontinuity design that we exploit in a local difference-in-differences framework. For both subsidies we find negative point estimates, which are for some outcomes significantly different from 0. Extra funding for computers and software seems especially detrimental for girls' achievement. The negative effects of extra funding for computers and software are consistent with results from other recent studies casting doubt on the efficacy of computers in schools. Copyright by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.

We examine how the gender of a sibling affects earnings, education and family formation. Identification is complicated by parental preferences: if parents prefer certain sex compositions over others, children’s gender affects not only the outcomes of other children but also the very existence of potential additional children. We address this problem by looking at dizygotic twins. In these cases, the two children are born at the same time, so parents cannot make decisions about one twin based on the gender of the other twin. We find that the gender of the sibling influences both men and women, but in a different way. Men with brothers earn more and are more likely to get married and have children than men with sisters. Women with sisters obtain lower education and give birth earlier than women with brothers. Our analysis shows that the family size channel cannot explain the findings. Instead, the most likely explanation is that siblings affect each other via various social mechanisms.

This study investigates the long-term effects of peace-time military conscription on educational attainment and earnings by exploiting a policy change that exempted a complete birth cohort from military service. We find that compulsory military service decreases the proportion of Dutch university graduates by 1.5 percentage points from a baseline of 12.3 per cent. In addition, being a conscript reduces the probability of obtaining a university degree by almost four percentage points. The effect of military service on earnings is also negative and long-lasting. Approximately 18 years after military service, we still find a negative effect of 3 to 4 per cent. The effect of conscription on educational attainment does not fully explain the wage reduction. Jel classification H56; J31; J24

This paper investigates the impact of student support on performance and time allocation of students in Dutch higher education. In 1996 the maximum duration of grants was reduced by 1 year, and thereby limited to the nominal duration of the study program. This reform could have had substantial financial consequences for students. We evaluate the effects of the reform using a difference-in-differences approach. Our main findings are that after the reform, students early in their study (i) switched less to other programs, (ii) obtained higher grades, while (iii) they did not spend more time studying or working. In addition, for students not older than 20 years when they started their study we find larger effects on all performance variables (switching, percentage of completed courses, graduation in the first year and grade point averages). These findings are consistent with recent evidence on heterogeneous treatment effects for higher ability students.

In recent years, a wave of new studies on the effects of educational interventions on student performance has emerged. The realization that inputs in the education process are endogenous is important for the validity of traditional findings. Because of ignoring endogeneity bias, all traditional estimates might be wrong. Recent studies exploit exogenous variation in interventions in education produced by controlled or natural experiments. Results generated by this methodological innovation differ substantially from the traditional findings. This article reviews this new literature, illustrates new methods for identifying causal effects of interventions in education and compares the findings with the traditional literature.

This paper examines the causal link that runs from classroom quality to student achievement using data on twin pairs who entered the same school but were allocated to different classrooms in an exogenous way. In particular, we apply twin fixed-effects estimation to assess the effect of teacher quality on student test scores from second through eighth grade, arguing that a change in teacher quality is probably the most important classroom intervention within a twin context. In a series of estimations using measurable teacher characteristics, we find that: (a) the test performance of all students improve with teacher experience; (b) teacher experience also matters for student performance after the initial years in the profession; (c) the teacher experience effect is most prominent in earlier grades; (d) the teacher experience effects are robust to the inclusion of other classroom quality measures, such as peer group composition and class size; and (e) an increase in teacher experience also matters for career stages with less labor-market mobility which suggests positive returns to on the job training of teachers.

This paper analyzes the determinants of choosing a technical study at university level and of staying in it. We find that – in The Netherlands – there is little correlation between the probability of a student choosing a technical study and the probability of completing it. This implies that a substantial number of technically talented people choose non-technical studies. Especially female students and students from high income families are unlikely to attent a technical study but these students are relatively successful in such studies. A large part of these technically talented students are attracted to medical studies and law schools, where they are no more likely to persist than other medical and law students. This finding is predicted by the tournament model in which rewards are based on relative performance instead of absolute performance.

This study provides the first estimates of the causal effect of time in school on cognitive skills for many countries around the world, for multiple age groups and for multiple subjects. These estimates enable a comparison of the performance of education systems based on gain scores instead of level scores. We use data from international cognitive tests and exploit variation induced by school entry rules within a regression discontinuity framework. The effect of time in school on cognitive skills strongly differs between countries. Remarkably, we find no association between the level of test scores and the estimated gains in cognitive skills. As such, a country’s ranking in international cognitive tests might misguide its educational policy. Across countries we find that a year of school time increases performance in cognitive tests with 0.2 to 0.3 standard deviations for 9-year-olds and with 0.1 to 0.2 standard deviations for 13-year-olds. Estimation of gains in co gnitive skills also yields new opportunities for investigating the determinants of international differences in educational achievements.

This paper evaluates the effects of a special program designed to increase school enrolment and employment among multi-problem youths. Treated youths are guided by personal coaches and receive a comprehensive treatment of educational, work, and health services. We investigate the impact of the program by implementing a specific assignment rule such that treatment status depends in a deterministic way on an individual’s application date. We find evidence that assignment to the program increases criminal activity compared to standard intervention, especially among the subpopulation of youths who were suspected of a crime at the time of entry. Peer effects caused by grouping at-risk youths together may explain the adverse impact on criminal behaviour. We find statistically insignificant effects on school enrolment and employment. Keywords: disadvantaged youths, school dropout, criminal behaviour, program evaluation

This paper investigates the effect of a new type of financial incentive in education targeted at regional authorities. Previous studies have focused on financial incentives for students, teachers or schools. We identify the effect by exploiting the gradual introduction of a new policy aimed at reducing school dropout in the Netherlands. The introduction of the policy in 14 out of 39 regions and the use of a specific selection rule for the participating regions allow us to estimate local difference-in-differences models. Using administrative data for all Dutch students in the year before and the year after the introduction of the new policy we find no effect of the financial incentive scheme on school dropout. In addition, we find suggestive evidence for manipulation of outcomes in response to the program.

Recently, there has been considerable debate about a reform of the Dutch system of student support, in which grants will be (partly) replaced by loans. The discussion focuses on the effects on student enrollment decisions. Surprisingly, no study has yet analysed the effect of receiving a grant on parental contribution. Parents may decrease their contribution when their child receives a grant, in which case subsidies meant for the students unintentionally end up with the parents. Understanding the corresponding parental behaviour will contribute to a more in-depth discussion on the financial aid system. This paper focuses on the effect of the supplementary grant on the parental contribution in the Netherlands. The supplementary grant is meant to support students from disadvantaged families. Parents from students with the supplementary grant have less disposable income, which probably implies a lower contribution. Our identification strategy separates this income effect from the effect due to the payments of the supplementary grant. The results suggest substantial substitution. Each additional euro spent on supplementary grant reduces the parental contribution with approximately 20-60 cents. A broad range of sensitivity analyses support our main estimation results. Nevertheless, some caution in interpreting the results is needed because of data limitations.

This paper examines the impact of early tracking on enrollment in and completion of higher education. We compare pupils that are directly tracked in lower general secondary education (‘mavo’) to pupils that postpone their choice of education level by entering secondary education in a combined first-grade class. Potential self-selection problems are addressed in two ways. First of all, using micro data allows us to control for a large set of individual background characteristics including tests of cognitive ability. Second, we exploit differences in regional supply of particular school types. The estimates show that early tracking has a detrimental effect on enrollment in and completion of higher education for pupils who leave primary education with a mavo advice. In addition, we find no evidence that pupils who leave primary education with a higher general secondary education (‘havo’) advice would be negatively affected by being in a comprehensive class together with the mavo advice pupils. Enrollment in and completion of higher education can be increased by stimulating participation in combined first-grade classes that keep pupils with a mavo or havo advice together for an additional one or two years.

Early school-leaving is considered to be one of the major problems in Dutch education. In order to reduce the number of dropouts in the school year 2006-2007 the Dutch government has offered a financial incentive scheme to 14 out of 39 regions. This scheme provides a reward of 2000 euro per school dropout less in 2006-07. The target of the scheme was a reduction of the total number of school dropouts by at least 10 percent in one year. This paper evaluates the effectiveness of this school dropout policy by comparing the change in school dropout in these 14 regions with the change in the remaining 25 regions before and after the introduction of the policy. We observe a modest decline in the probability of dropping out in the 14 covenants regions. However, the decline in the non-covenant regions was equally large. We therefore find no significant effect on the probability of dropping out in the post-covenant year. In both regions, the number of dropouts has fallen by 3 percent in the year after the covenants. This nationwide decline can be largely assigned to changes in the student populations among the pre- and post-covenant year. The covenants also gave a reward to regions for a successful reintegration of dropouts in order to reduce school dropout in that way. However, estimates for the effect on the re-enrolment of previous dropouts are statistically insignificant as well. We conclude that 2006 covenant policy has not been effective in reducing early school-leaving.

This paper investigates the question whether crime reduces investment in human capital or whether education reduces criminal activity by using fixed effect estimation on data of Australian twins. The study takes genetic and socio-economic factors shared by the twins into account. We find that early arrests (before the age of 18) have a strong effect on human capital accumulation. In addition, we find that education decreases crime. However, controlling for early arrests and early behaviour problems reduces the estimated effect of human capital on crime to less than on third of the previously estimated association. From this, we conclude that the strong association between human capital and crime is mainly driven by the effect of early criminal behaviour on educational attainment. The strong detrimental effects of early criminal behaviour become also transparent if we consider the estimated effects of early arrests on three measures of crime. We also find large effects of early criminal behaviour on participation in crime later on. This suggests that programs that succeed in preventing early criminal behaviour might yield high social and private returns.

Skill gaps are widely seen as a problem that lowers aggregate productivity growth. A question for the European Commission is whether and how governments should take action with education and training policies to reduce skill gaps and make Europe the best performing region in the world. European citizens can best decide for themselves on the type of education. Distribution of information on occupation prospects is effective to influence their choice of education. Moreover, it is important that the education system is sufficiently flexible to absorb unexpected shocks in skill needs of employees. Policies stimulating education targeted at government-assigned sectors are risky policies. Intensification of general education at the cost of specific education, and intensification of training of employees find little support.

The prevalence of overweight and obesity is growing rapidly in many countries. Education policies might be important for reducing this increase. This paper analyses the causal effect of education on the probability of being overweight by using longitudinal data of Australian identical twins. The data include self-reported and clinical measures of body size. Our cross-sectional estimates confirm the well-known negative association between education and the probability of being overweight. For men we find that education also reduces the probability of being overweight within pairs of identical twins. The estimated effect of education on overweight status increases with age. Remarkably, for women we find no negative effect of education on body size when fixed family effects are taken into account. Identical twin sisters who differ in educational attainment do not systematically differ in body size. Peer effects within pairs of identical twin sisters might play a role.

A shortage of teachers in primary and secondary schools in the Netherlands is expected in the near future as a large part of the current workforce will retire. Recently, the Dutch government has decided to increase wages of teachers. This paper qualitatively assesses the impact of three options for increasing teaching wages on teacher supply in the medium term: increasing wages at the start of the career, increasing wages at mid career or increasing wages at the end of the career.

This article analyzes the effect of classroom separation of twins on their cognitive abilities, measured at different ages in Dutch primary education. We use a large longitudinal school-based sample of twins and their classmates. The analysis tries to reduce the bias by unobserved factors due to the nonrandom assignment of twins by taking into account differences in school environment, previous test scores and variation in class assignment between years. We find that classroom separation matters for language in Grade 2. Nonseparated twins score higher on language, and the difference is larger for same-sex pairs. This finding is robust for various methods that take unobserved effects into account. In addition, there is some evidence for higher scores in arithmetic in Grade 2. For the higher grades we find no effect of classroom separation on cognitive ability. In the analysis of the effect of a separation of at least 3 years we find that separation increases language performance between Grade 6 and 8 for opposite-sex pairs.

This report surveys the recent literature on human capital and productivity. Recent studies suggest that the right-hand side of the skill distribution is important for productivity, especially in countries that already have a high level of productivity. An empirical analysis of the Dutch skill distribution reveals that the Netherlands is not positioned among the best-performing countries at the right-hand side of the distribution. On average, the Dutch skill level is high, but this level is mainly based on the relatively high skill level at the left-hand side of the skill distribution. The Dutch position declines when moving to the right-hand side. At the very highest skill level, the Netherlands is not among the best of the world. This is true for both secondary education and higher education. The Dutch share of graduates from higher education is also not among the highest in the OECD. The findings on the skill distribution are robust for several skill surveys, age groups and over time. This robustness may be the result of the structure of the Dutch educational system. The findings indicate that there is scope for improvement of skills at the right-hand side of the distribution. Therefore, policies that raise the Dutch performance at high- and top skill levels in higher education or in earlier stages of education may improve Dutch productivity. Further research is needed to assess these policies.

Inspectors from the Dutch Inspectorate of Education inspect primary schools, write inspection reports on each inspected school, and make recommendations as to how each school can improve. We test whether these inspections result in better school performance. Using a fixed-effects model, we find evidence that school inspections do lead to measurably better school performance. Our assessment of school performance is based on the Cito test scores of pupils in their final year of primary school. Therefore school improvement means increased Cito test scores. The results indicate that the Cito test scores improve by 2% to 3% of a standard deviation of the test score in the two years following an inspection. The arithmetic component shows the largest improvement. Our estimates are the result of an analysis of two types of school inspections performed between 1999 and 2002, where one type was more intensive than the other. In one fixed-effects model, we assume that the effect of the two types of school inspections was the same. We cannot, however, be sure that the estimates from this model are free from the problem of endogeneity bias. Therefore, we also obtain estimates for a less restrictive fixed-effects model. In this less restrictive model, we make use of the fact that a subset of the more intensive school inspections occurs at a representative selection of primary schools. Based on this smaller, essentially randomly drawn sample of schools, we can be confident that these estimates of the effect of school inspections are free from endogeneity bias. Due to the limited number of inspections at randomly selected schools, these estimates are not significantly different from zero. These estimates are, however, consistent with the effects found based on all inspections. The less restrictive model also allows for the effect of the more intensive inspections to differ from that for the less intensive ones. We find evidence that the more intensive inspections are responsible for larger increases in the Cito test scores than the less intensive ones.

This paper analyses the causal effect of teenage childbearing on smoking, drinking and body size using a sample of Australian twins and their relatives. Fixed effects estimates on samples of siblings, all twin pairs and identical twin pairs show that teenage mothers smoke more during their lives. Teen mothers tend to have a higher probability of being overweight, especially if they are older than 40 years. Their spouses are more likely to smoke and drink more. The quality of the spouse seems to be an important mechanism through which teenage childbearing affects subsequent maternal health.