Givology Staff's Blog

Returns on Investment in Education in the Developing World

By Brent Harlow
While education plays an important role in generating social cohesion, forming knowledgeable and engaged citizens, and promoting other individual and social goods, one of the most powerful ways of thinking about and measuring the impact of education that has emerged over the last half-century comes from the field of economics, and the understanding of education as “human capital.” This model has made it possible to measure returns to investment in education and quantify the relationship between education and the material prosperity of individuals and societies (which, many economists argue, may be the best proxy available for measuring overall well-being).
This has been done using both microeconomic models which analyze the relationship between household investment in education and private earnings, and macroeconomic models which study the connection between a country’s overall investment in education and economic growth. There is now very strong microeconomic, and increasingly good macroeconomic evidence that shows education to be positively correlated with economic productivity.
In poorer, developing countries, this means that education can be a crucial tool in helping individuals and communities begin to break out of intergenerational cycles of poverty. Especially in today’s rapidly changing, tech- and information-based global economy, it is more important than ever that less-advantaged children be trained in the skills and knowledge needed to perform the jobs that will help them begin to break out of long-standing “poverty traps.” In this context, research on human capital investments and the variability of returns on investment across gender, grade-level, socioeconomic and national boundary lines can be extremely useful in determining how and where to invest resources in education in the developing world.
In this article, my aim is to provide a very broad overview of what researchers have found— from the microeconomic and macroeconomic perspectives— as they look at the heterogeneous returns to investment in education in the developing world, and how these findings can help determine which investments are likely to generate the most impact.
[b]The microeconomic evidence on returns to investment in education [/b]
Empirical studies demonstrating the link between education and personal income gains date back to the 1960s and Theodore Schultz’s ground-breaking article “Capital Formation by Education” (1) and Gary Becker’s seminal book [i]Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education[/i] (2). These studies provided the initial articulations of the model that would be used to measure the relationship between individual education levels and private earnings— calculated as a “return on investment”— in numerous studies conducted over subsequent decades (3). These studies have consistently found that there is a positive correlation between an individual’s educational attainment and earnings.
While much of this work has focused on individuals in more affluent countries— due in part to the greater reliability and availability of data— there is a smaller contingent of researchers who have done very important work examining the relationship between individual investments in education and private earnings in poorer, less developed countries as well. Since the 1970s, George Psacharopoulos has compiled and analyzed data from upper- as well as middle- and lower-income countries, and found that individual income rises with years of schooling everywhere the link is studied— in developed as well as in poorer, developing countries (4).
Paul Glewwe has shown that this positive relationship between an individual’s education and income in developing countries is even more pronounced when educational achievement is measured by the actual skills a student has acquired in school, as opposed to the years of schooling he or she has completed (5). After conducting a review of the existing literature, he concludes that skills acquisition and specifically basic skills acquisition (literacy, numeracy, basic science skills, etc.) is the key variable driving the increased earnings (as opposed to other possible explanations for the correlation, including credentialism and screening). Glewwe further recognizes that studying the link between education and earnings in the developing world requires researchers to adapt the models initially used to study more affluent countries, so as to accommodate different variables that may be at play in those contexts. In poorer countries, for instance, where a greater percentage of an individual’s income is reaped outside of the formal wage economy, researchers will need to look at how education impacts productivity in various forms of self-employment and agricultural labor as well.
While Glewwe’s study was unable to draw any general conclusions about the link between cognitive skills and self-employment income, other studies have shown that basic skills acquisition and the completion of primary education can enhance the productivity of farmers and may even generate positive externalities in rural communities (6). However, these benefits seem to be dependent upon a certain level of technological development in the agricultural techniques used in the farmer’s community, illustrating a much more general point, i.e. that variability in returns to investment in education can in part be explained by the degree to which the surrounding economic, social, political, cultural and technological contexts provide the supports needed to reap the rewards of a given investment in education.
While much of this variability remains to be studied, Psacharopoulos and Patrinos have identified a number of interesting patterns in their study of returns to investment in education in eighty-three high- and low-income countries from 1958 to 1999 (7). The study finds that overall, the average return to investment for each additional year of schooling is 10%, but that there is significant variability by level of economic development, by region, and by level of schooling. They find that returns to investment are greatest in poorer, developing countries; in Latin America and the Caribbean (12% per year) and Sub-Saharan Africa (11.7% per year); and in primary school (18.9% per year). This last finding has been confirmed in a 2009 study conducted by Colclough, Kingdon, and Patrinos entitled, “The Patterns of Returns to Education and its Implications,” which concluded that investments in primary education have the highest rates of return (8, 9).
[b]Positive externalities and social returns to investment in education[/b]
The microeconomic approach discussed above, while necessary, does not provide an adequate way of measuring returns to investment in education, which seem to exceed private household earnings and spill over into the larger community where they provide other, more difficult to quantify social benefits. Some economists have even suggested that education may “constitute an ‘infinite’ engine of economic growth” in that it generates “positive externalities through which individual decisions interact with each other and result in outcomes greater than the sum of individual ones” and produces “self-amplifying virtuous cycle[s]” of knowledge and productivity (10).
Although the microeconomic approach discussed above is unable to capture these positive externalities, it is possible that by taking a macroeconomic perspective and looking at the relationship between a society’s overall investment in education and its total economic output, these positive externalities would be included in calculations of rate of return. Since the 1990s, economists have been trying to test out this idea as they conduct empirical studies examining the relationship between flows or stocks of “human capital” and GDP in different countries and at different times. In the next section, I will discuss some of the main findings as well as some problems that remain to be addressed in that literature.
[b]The macroeconomic evidence on returns to investment in education [/b]
In 1991, Robert Barro published a paper entitled “Economic Growth in a Cross-section of Countries,” which presented data on school enrollment rates and economic growth in a cross-section of countries, and concluded that “poor countries tend to catch up with rich countries if the poor countries have high human capital per person (in relation to their level of per capital GDP), but not otherwise” (11). Barro’s study was received with enthusiasm among researchers looking for empirical evidence confirming the positive relationship between education and economic growth within a macroeconomic framework.
Subsequent studies, however, showed mixed results, and found that while initial levels of human capital were positively correlated with growth, increasing investment in education did not universally lead to greater growth (12). One possible explanation for this fact is that the social, political, cultural and technological contexts in which people are educated and work in countries across the world are simply too heterogeneous to allow for general conclusions to be drawn about how increasing investment in education will affect economic growth.
It has been argued that this variability points to the fact that different factors affect growth in developing and developed countries, and that it is therefore necessary to analyze the relationship between human capital and growth in developing countries separately. This is the approach taken by UNESCO and the Organisation for Economic Co-Operation and Development (OECD) in their analysis of data from the World Education Indicators Programme (WEI), which was launched in 1997 to generate a database of quality, standardized, and internationally-comparable statistics on educational performance among participating countries (13). The report finds that when this is done, and more sophisticated measuring techniques are used, the result is “consistently strong and positive association between improvements in the stock of human capital and economic growth among WEI countries, an association that is even greater than that observed among OECD countries.” The report goes on to state that “the results suggest that for every single year for which the average level of schooling of the adult population in WEI countries is raised, there is a corresponding increase of 3.7 per cent in the long-term economic growth rate” (8). The report also finds—consistent with the microeconomic evidence discussed above—that the association is strongest in developing countries and in Latin America.
It has also been argued that “measurement errors in the education variable, especially in developing countries” were to blame for the discrepancy between the microeconomic and the macroeconomic findings. For instance, if instead of looking at enrollment rates, one looks at actual knowledge or skills acquired for the education variable, then the macroeconomic findings become much stronger. With better data on learning outcomes in developing countries becoming increasingly available—e.g. through programs like the WEI Programme—this provides a promising direction for further macroeconomic research on how investment in education affects economic growth (14).
[b]Some conclusions from the research[/b]
The microeconomic research shows unequivocally that household investments in education pays off in increased private earnings, with the greatest returns on investment being in developing countries, in Latin America, and in primary education. Research suggests that primary education can also result in increased productivity for the self-employed and farmers provided a certain level of technological advancement in farming practices has been achieved in the context in which they work. The association between individual education and income is, furthermore, even stronger when cognitive skills—as opposed to years of schooling—are taken as the education variable, contributing further evidence to the consensus view that it is student learning, as opposed to credentialism or screening, that is driving the positive correlation between an individual’s education and earnings.
There are very good reasons to believe that households investments in education benefit not only the students in that household, but also the larger community, generating positive externalities that have historically been difficult to quantify. These externalities can theoretically be caught by macroeconomic models that measure the relationship between a given country’s stocks of human capital and economic growth. However—because of measurement difficulties, incomplete and poor data, and the heterogeneous conditions affecting growth in developing versus developed countries—economists have not yet provided empirical evidence that there is a universal positive association between investment in education and increases in GDP. Studies attempting to make headway in these areas are currently underway, however, and it is likely that a clearer macroeconomic picture will emerge with better models and data. Until then, however, all researchers agree that investments in human capital are at least a necessary, if not a sufficient, condition for growth.

[b]Notes: [/b]
(1)Published in the [i]Journal of Political Economy[/i] 6(6): 571-583. 1960.
(2)New York: National Bureau of Economic Research: Distributed by Colombia University Press. 1964.
(3)A technical elaboration of the models commonly used by contemporary researchers can be found at http://ieg.worldbankgroup.org/sites/default/files/Data/reports/education_public_investments_wp.pdf, pp. 4.
(4)http://documents.worldbank.org/curated/en/512891468739485757/123523322_20041117181555/additional/multi0page.pdf
(5)Glewwe, Paul. “Schools and Skills in Developing Countries: Education Policies and Socioeconomic Outcomes” in [i]Journal of Economic Literature[/i]. Vol. XL (June 2002) pp 436-482.
(6)http://ieg.worldbankgroup.org/sites/default/files/Data/reports/education_public_investments_wp.pdf, pp. 7.
(7)http://documents.worldbank.org/curated/en/512891468739485757/123523322_20041117181555/additional/multi0page.pdf
(8)https://www.ssoar.info/ssoar/bitstream/handle/document/6918/ssoar-2009-colclough_et_al-the_pattern_of_returns_to.pdf?sequence=1, pp. 4.
(9)These conclusions, derived from data taken from over eighty countries over a span of forty years, in what is the most comprehensive study to date, have nonetheless not gone unchallenged. See, for example, Gunther Fink and Evan Peet’s 2014 (available at https://cdn1.sph.harvard.edu/wp-content/uploads/sites/1288/2015/06/PGDA_WP_120_Fink.pdf), which uses World Bank Household Surveys to gather data on education and income for a later period (1985 to 2012), and finds highest rates of return for Latin America and the Caribbean and Africa (as do Psacharopoulos and Patrinos), highest rates of return for tertiary and lowest for primary school investments (contrary to Psacharopoulos and Patrinos), and no higher rates of return for developing countries (also contrary to Psacharopoulos and Patrinos).
(10)http://uis.unesco.org/sites/default/files/documents/financing-education-investments-and-returns-en_4.pdf, pp. 20.
(11)[i]The Quarterly Journal of Economics[/i] [Vol. 106, No. 2 (May, 1991) pp. 407-443. 437.
(12)http://siteresources.worldbank.org/INTMENA/Resources/EDU_02-Chap02-Education.pdf, pp. 42.
(13)http://uis.unesco.org/sites/default/files/documents/financing-education-investments-and-returns-en_4.pdf
(14)http://ieg.worldbankgroup.org/sites/default/files/Data/reports/education_public_investments_wp.pdf, pp. 10-11.

31 Comments

Must be logged in to comment.