Private and Social Returns to Education: Causal Effects, Displacement, and Experimental Evidence from Ghana

This lecture delves into the complex relationship between private and social returns to education, highlighting the critical challenge of establishing causal effects due to pervasive selection issues like ability bias and omitted variables. Private returns encompass individual benefits such as increased earnings, improved health, and the ability to raise more educated future generations. In contrast, social returns refer to the broader value education provides to society, including enhanced technological adoption and stronger institutional support. A central argument is that these returns often diverge, particularly in developing economies where education can function as a "ticket" to scarce, rent-seeking government jobs, leading to significant displacement effects for other job seekers.

The discussion meticulously distinguishes between these two types of returns and explores the nuances of their measurement. The lecture emphasizes that simple correlations between education and outcomes are insufficient, as wealthier families often have both better educational access and job opportunities. It contrasts positive spillovers, where educated individuals benefit society, with negative externalities like the displacement effect, a concern echoed by Richard Freeman's "The Over-Educated American" regarding the potential collapse of returns to education. The Ghana study, a key focus, specifically examines the returns for a unique cohort: individuals who academically qualified for secondary education but faced financial barriers.

Methodologically, the lecture explores strategies to overcome selection bias, primarily through randomized assignment (e.g., scholarships in Ghana) and natural experiments (e.g., school construction). The Ghana scholarship experiment, a long-running study, reveals that while education demonstrably improves cognitive scores and political knowledge, its impact on labor market earnings is surprisingly modest, particularly for men. Employment gains are often concentrated in secure, public sector jobs, which, while offering protection during crises like COVID-19, are limited in number. This suggests that expanding educational access without addressing underlying labor market structures or parental expectations can lead to an overestimation of returns and potential disillusionment among graduates.

The findings from Ghana carry significant implications for education policy in developing nations, where the "ticket" phenomenon for government employment is widespread. The lecture underscores the necessity of considering general equilibrium effects when expanding education and the potential for displacement. The study's extended timeline highlights the importance of long-term follow-up to capture delayed impacts and the resilience of certain job sectors. Furthermore, the research sheds light on gender dynamics, showing a larger treatment effect for girls in accessing secondary education, which may indicate that parents initially underestimate the value of investing in their daughters' schooling.

"what are the private return to education well we discussed them a little bit they are on the one hand impact on the labor market and on the other hand uh the um the impact on get making it easier to raise a future generation that itself has a lot of Education which we can take as a shortcut for human capital"

"there could be selection issues or various kinds ability bias omitted variable of the kind that people who have enough money to feed their children also have enough money to make sure that they get the right apprenticeship"

"the pilot returns are for any individual how much money they make with or without education and in fact from the point of view of the parents you could add a layer of private estimated return which is how much they think the child is going to make with or without education and then there are the social returns which is the value for Society of a particular individual getting more education"

"in the Ghana paper it's pretty obvious because education is used as a ticket for Trend seeking jobs and then there are only so many of those it's not that you can create more"

"this sort of cues for those government job is uh I've come to believe a very disturbed has a very destructive impact on the whole labor market"

"this is what Richard Freeman now many many decades ago was worried about when he wrote this book The over-educated American is that a lot of people would go to school and then they would you know work at this point going to college and they would acquire skills and then the return to education would collapse"

"one of the pretty striking results in the context of this Improvement in human capital which are real is the fact that there is really not much happening on the liberal Market"

"the hope that they had of secondary education being a ticket was dashed squashed by what happened over time"

"the extremely long timeline I don't know if it's very cool or if it's we it means we cannot get around to write a paper because we continue finding them uh but that's useful because we are certainly learning things over the years"

"we are certainly learning things over the years it's necessary because people uh you know some amount of time was definitely necessary because you need to let people finish secondary school start in the labor market"