مدارس کاتولیک و یا کیفیت مدرسه؟ بررسی اثرات مدارس کاتولیک بر نتایج بازار کار

This paper studies the short and medium run impact of highly skilled immigrants from the Former Soviet Union to Israel on natives' wages and employment. If immigrants are relatively good substitutes for native workers, the impact of immigration will be largest immediately upon the immigrants' arrival, and may become smaller as the labor market adjusts to the supply shock. Conversely, if immigrants upon arrival are poor substitutes for natives, the initial effect of immigration is small, and increases over time as immigrants acquire local labor market skills and compete with native workers. We empirically examine these alternative hypotheses using data from Israel between 1989 and 1999. This paper studies the effects of attending a Catholic high school on students’ labor market outcomes. Using data from the Wisconsin Longitudinal Study, I find that Catholic schooling is significantly associated with higher wages over the careers even after taking into account possible selection into Catholic schools with instruments. Using matched school quality data for public and Catholic schools, I further find that Catholic and public schools are different in various aspects of school quality measures and that these differences explain most of Catholic school effects. Among the school quality variables, teacher quality and the number of math courses taken are estimated to matter the most for students’ later earnings in the long run.

With increased concerns and debates on the quality of public schools, researchers from various fields have paid attention to the performance of Catholic schools. See, for example, Coleman, Hoffer, and Kilgore (1982). According to national statistics reported in 2000, the average high school graduation rate in the U.S. was 86.5 percent, with 64 percent of those high school graduates pursuing a college education, whereas 99 percent of Catholic high school students obtained a high school degree and 97 percent of them moved on to higher education.1 These figures suggest that Catholic school students performed substantially better than their public school counterparts. However, since Catholic schools are private schools, there exists the issue of selection bias in the evaluation of Catholic schooling. The effect of attending a Catholic high school has been also a controversial issue in the economics literature. Among others, Evans and Schwab, 1995, Sander and Krautmann, 1995 and Neal, 1997, and Altonji, Elder, and Taber (2005b) examine Catholic school effects by taking account of selection into Catholic schools and find that attending a Catholic high school substantially raises the probability of high school graduation or college attendance. On the other hand, some studies find no effect from Catholic schooling. Figlio and Stone (1999) show that only minority students in Catholic schools perform better than their counterparts in public and other private schools in math and science tests. Other than these educational outcomes, Mocan, Scafidi, and Tekin (2002) examine the effect of Catholic schooling on risky behaviors, such as drug use and teenage crime, and find no evidence that Catholic schools affect these behavioral outcomes, whereas Figlio and Ludwig (2000) find Catholic schooling to be associated with a reduction in teenage sexual activity, drug use and arrests. This study investigates the effects of attending a Catholic high school on students’ subsequent wages. Despite great interest in the economic value added by attending a Catholic school, little is known as to whether Catholic schooling is associated with success in the labor market other than educational achievement. Tyler (1994) and Neal (1997) find that Catholic schooling is associated with better performance at the early stage of careers in the labor market. I contribute to the literature by looking at long-run effects of Catholic schooling on wages, carefully addressing the issue of possible selection bias. In addition, I account for the quality of Catholic schools as well as that of public schools for the first time in the literature. I make use of school quality data collected from Catholic schools to examine specific factors within Catholic schools that help explain the positive wage premium that is estimated. The data used for this study come from three different sources. First, I use data from the Wisconsin Longitudinal Study (WLS). The WLS is a unique dataset of randomly selected Wisconsin high school graduates in 1957. It collects extensive information from students and their families, and contains detailed information on students’ labor market experience. Moreover, it allows me to match students with the high school they had attended. Second, for Catholic school students, I collected detailed school and teacher quality data from schools and religious orders or communities for most of the Catholic schools that were attended by the WLS respondents. The basic school data were extracted from the Official Catholic Directories and the Annual School Reports from the Milwaukee archdiocese. Lastly, for students who attended public schools, I use the school data reported by the Wisconsin department of public instruction, which were compiled by Olson and Ackerman (2004) for the years when respondents of the WLS attended high schools. The selection issues with Catholic schooling present a serious challenge to estimation of Catholic schooling effects. This has been understood in the previous literature and in critiques of previous research. See, for example, Goldberger and Cain (1982) and Altonji, Elder, and Taber (2005a). I bring a new and rich dataset to the literature which allows me to combine an instrumental variable approach with detailed information on school quality for the estimation of Catholic school effects. As instruments for Catholic school attendance, I use Catholic religion and the proportion of Catholic population in the county of residence in 1890. The historical density of Catholic population was used in a recent Catholic school study by Cohen-Zada and Elder (2009). Catholic religion has been also used in the previous literature, including Evans and Schwab, 1995 and Neal, 1997 and Grogger and Neal (2000). However, use of Catholic religion as an instrument has been criticized by Altonji et al. (2005a) for the potential correlation with unobserved effects that are associated with student outcomes. To demonstrate the validity of my instruments, Catholic religion in particular, I provide a host of evidence. For example, I find that the mean wage difference by Catholic religion is almost zero for both years that I examine unlike the marked difference in educational outcomes by Catholic religion in Altonji et al. (2005a). Furthermore, I explore the reduced-form relationship between Catholic religion and wages for a sample of students for whom Catholic high school was not a choice, following Altonji et al. (2005a) and Cohen-Zada and Elder (2009). I find that among students from the counties with no Catholic high school, the reduced-form relationship is not significantly different from zero. The overidentifying restriction test results also suggest that my instruments are jointly valid for the wage analysis. My main findings are as follows. In a baseline model where only individual and family background variables are controlled for, it appears that Catholic school students earn significantly more 17 and 35 years after high school graduation even after taking account of selection bias. Furthermore, I find that there is a substantial difference in school quality between public and Catholic schools. Catholic schools tend to have more highly educated teachers and offer a more extensive curriculum in math and foreign language courses. Although these school quality variables may capture unobserved attributes of a student in the wage regressions, most of the Catholic school effects are explained by these school quality measures. Specifically, teacher quality and the number of math courses taken in high school remain significant in extended wage regressions. The results suggest that school quality plays a significant role for the success of Catholic school students in the labor market.

Debate on the benefits of Catholic schooling in the literature has been concentrated on educational outcomes that have significant impacts on the labor market success. In this study I directly examine how Catholic schooling is linked to performance in the labor market. I find that Catholic schooling is significantly associated with higher wages at both early and later stages of careers. By investigating school quality resources for Catholic and public schools, I also find that the typical school resources that are commonly addressed in public school studies are the driving force behind the relative success of Catholic schools. It is worth noting that this study reinforces the importance of teacher quality reported in the previous literature and identifies other crucial components of school quality and their long-run impacts in the labor market. Previous studies on school resources including Rivkin, Hanushek, and Kain (2005) have agreed on teacher quality as a key determinant of student outcomes, but there has been little evidence that links the estimated teacher impacts with the observed teacher characteristics such as education levels.18 Compared to the literature, for example Aaronson, Barrow, and Sander (2007) which examine teacher quality impacts on high school student achievements, the effect of teacher education levels found in this study appears to be very large. One of the reasons may be that this study focuses on the impact of having a higher fraction of teachers who obtained a master or higher degree in the 1950s or earlier when the average education level was quite low. In the 1950s, less than 10 percent of people had a bachelor degree even among young adults. Another possible reason is that the effect of being taught by teachers from religious orders may be larger than that corresponding to other types of teachers. Teachers from religious orders were relatively homogeneous and tended to have spent more years in their education. Thus teachers’ education levels may contain other unobserved qualities that result in higher impacts on students. The smaller effect of teacher education from public school students supports this case. Greeley (1982) also finds that teachers from religious orders tended to provide higher quality of instruction. Therefore the measured teacher quality impacts may not necessarily be comparable to the ones in the literature that examine teacher quality from more recent data. My data set is based on a sample of students who attended high school in the peak era of Catholic schools. Enrollment in Catholic schools expanded until the early 1960s. At the same time, they were able to provide well-educated teachers with relatively small costs thanks to having religious staff. The proportion of religious staff and the student enrollment in Catholic schools have been decreasing since then, although the quality of teachers were maintained to some extent. For example, according to the Survey of Catholic Schools in the State of Wisconsin in 1970, less than 40 percent of the teaching staff were religious teachers but 30 percent of all teaching and other professional staff in Catholic schools had a master or higher degree. In this spirit, the estimates of the Catholic school effects from this study are larger than the one from more recent data that have been previously used in the literature. Yet considering that some of the key components of Catholic school effects are still prominent features of contemporary Catholic schools, one might expect the positive effects of Catholic schooling on wages documented in this study to persist until today. This is an issue well worth further study.