Human capital in OECD countries: A new measure and its policy drivers

Jarmila Botev, Balázs Égert, Zuzana Smidova, David Turner 06 January 2021

a

A

Many conventional measures of a country’s human capital are based on the mean years of schooling of its population. Yet, macroeconomic cross-country growth regressions provide mixed evidence on the relationship between such measures of human capital and economic outcomes. For instance, a meta-analysis of 60 studies published over the period of 1989-2011 suggests that around 20% of the reported coefficient estimates on human capital have the wrong (negative) sign (Benos and Zotou 2014). When focusing on a set of about a dozen papers by Robert J. Barro, based on comparable specifications, techniques and datasets, almost half of the coefficient estimates on education are negative and/or statistically not significant at the 10% significance level. And Bayesian model averaging (BMA) also provides mixed evidence for the robustness of human capital in growth regressions. 

Recent OECD studies looking at OECD countries confirm the difficulty of finding a robust positive effect of human capital on income per capita or productivity levels. The estimated effect is sensitive to the measure of human capital and to the estimation method (Guillemette et al. 2017). Including a large number of control variables in the regression analysis tends to reduce or eliminate the statistically significant positive effect. This may be due to a correlation of human capital with other institutions, in particular those representing good governance, leading to an indirect effect through these variables, and weakening the estimated effect of human capital (Fournier and Johansson 2016). Using common time fixed effects appears to further weaken the estimated effect of human capital due to a similar time trend across OECD countries (Égert 2017). 

Taking a closer look at measures of human capital, many empirical studies using mean years of schooling also assumed decreasing marginal returns to education. This meant that primary education had the biggest marginal returns, followed by secondary education, with tertiary education having the lowest returns (Hall and Jones 1999, Caselli 2004, Feenstra et al. 2015). Moreover, returns were considered to be constant across countries and over time. But the most recent and authoritative data on returns to education paint a different picture. They suggest that average returns to primary, secondary and tertiary education are U-shaped relative to the time spent in education, not linear as previously assumed (Psacharopoulos and Patrinos 200, Montenegro and Patrinos 2014). The returns also vary substantially across countries. Differences within OECD countries can be as large as 10 percentage points and returns in the BRICS and the rest of the world are substantially higher than in the OECD. The data also indicate that average returns have increased over time in both OECD countries and the BRICS (see Figure 1). 

Figure 1 Rate of return on education

A) Rate of return over time

      

B) Rate of return across countries

 

Source: Botev et al. (2019) based on Psacharopoulos and Patrinos (2004) and Montenegro and Patrinos (2014)

A new measure of human capital which is positively correlated with productivity

Correcting for these shortcomings, in Botev et al. (2019) we build a new measure of human capital based on the observed U-shaped or increasing returns to additional years of schooling and allowing for variation across countries and over time. The data on returns to education are combined with a 2018 update of the mean years of schooling constructed by Goujon et al. (2016). Our measure is based on rates of return, which differ across five groups of countries (advanced OECD, converging OECD, Easter European OECD, emerging market economic, rest of the world) and across three time periods (1979-1989, 1990-2000, 2001-2012). Our macroeconomic measure of human capital shows a positive and statistically significant correlation with multifactor productivity in time-series cross-country panel data regressions in which country and time fixed effects are included, with results robust to the time period, estimation methods, and the set of controls included.

Policy drivers of the new measure of human capital

Empirical analysis identifies a number of policy drivers of our macroeconomic measure of human capital. The importance of pre-primary education, delaying tracking, teaching resources as well as school autonomy is in line with earlier findings of the microeconomic literature (Égert et al. 2019, Smidova 2019). 

Greater enrolment in pre-primary education has a positive influence on human capital. As suggested by the microeconomic literature, the effect is stronger for countries with an above-average share of disadvantaged children. Another finding is that teaching resources matter. Given the absence of appropriate measures for teaching quality, Égert et al. (2019) use the student-teacher ratio as a crude measure of teaching quality. Streaming children at a later age into different education tracks, such as vocational and grammar schools, based on their ability or achievement also has a positive impact. School autonomy as measured by the PISA index on school autonomy is good for human capital. In line with the microeconomic literature, this positive effect is greater in countries with external central exams that captures external accountability imposed on schools. Moreover, countries in which universities have more autonomy in how they can allocate their resources, have higher human capital and the ease of access to individual financing of university education helps to raise a country’s human capital. It should be noted, however, that these results are less robust. 

These policy effects are estimated using the methodology proposed by Lorenzoni et al. (2018). Spending on education per capita has a direct effect on human capital and educational policies impact  human capital by leveraging the effect of spending on education. This approach overcomes the (very) limited time series availability of educational policies by assuming that they are relatively time invariant. In other words, educational policies amplify or attenuate the impact of education spending on human capital.

Figure 2 The long-run impact of education reforms on per capita income when policies move from the median to the 90th percentile of the OECD sample (% change in GDP per capita)

Note: Policy effects are conditional on a one standard deviation increase in public spending on education. A one standard deviation change in spending (stripped of country and time fixed effects) represents about 5% of the average spending level in OECD countries (this corresponds to about 1% of public spending on education per full-time student equivalent relative to GDP per capita).
Source: Égert et al (2019).

Some of these educational policies – namely, increasing attendance in pre-primary education, greater university autonomy and lower barriers to funding for students in tertiary education – represent ‘good value for money’ because they offer a double dividend of boosting human capital as well as reducing spending pressures. Increasing school autonomy at primary and secondary level enhances educational outcomes but does not reduce spending pressures, while higher student-to-teacher ratios, higher age of first tracking and a reduction in the extent of tracking also boost human capital but at a higher cost.

References

Benos, N and S Zotou (2014), “Education and Economic Growth: A Meta-Regression Analysis”, World Development 64 (C): 669-689. 

Botev, J, B Égert, Z Smidova and D Turner (2019), "A new macroeconomic measure of human capital with strong empirical links to productivity", OECD Economics Department Working Paper No 1575. 

Caselli, F (2004), “Accounting for Cross-Country Income Differences”, NBER Working Paper No 10828. 

Égert, B (2017), “Regulation, Institutions and Productivity: New Macroeconomic Evidence from OECD Countries”, OECD Economics Department Working Papers 1393.

Égert, B, J Botev and D Turner (2019), "Policy drivers of human capital in the OECD’s quantification of structural reforms", OECD Economics Department Working Paper No 1576. 

Feenstra, R, R Inklaar and M Timmer (2015), “The Next Generation of the Penn World Table”, American Economic Review 105(10): 3150-3182. 

Goujon, A, S K C, M Speringer, B Barakat, M Potancokova, J Eder, E Striessnig, R Bauer and W Lutz (2016), “A Harmonized Dataset on Global Educational Attainment between 1970 and 2060 - An Analytical Window into Recent Trends and Future Prospects in Human Capital Development”, Journal of Demographic Economics 82(8): 315-363. 

Guillemette, Y, A Kopoin, D Turner and A De Mauro (2017), “A Revised Approach to Productivity Convergence in Long-Term Scenarios”, OECD Economics Department Working Papers No 1385. 

Hall, R and C Jones (1999), “Why Do Some Countries Produce So Much More Per Worker Than Others?”, The Quarterly Journal of Economics 114(1): 83-116. 

Lorenzoni, L, F Murtin, L-S Springare, A Auraaen and F Daniel (2018), “Which policies increase value for money in health care?”, OECD Health Working Paper No 104, OECD Publishing, Paris.

Montenegro, C and H Patrinos (2014), “Comparable Estimates of Returns to Schooling Around the World”, World Bank Policy research Working Paper No 7020. 

Psacharopoulos, G and H Patrinos (2004), “Returns to Investment in Education: A Further Update”, Education Economics 12(2): 111-134.

Smidova, Z (2019), "Educational outcomes: A literature review of policy drivers from a macroeconomic perspective”, OECD Economics Department Working Paper No 1577.

a

A

Economist, OECD

Senior Economist, OECD

Economist, OECD

Head, Macroeconomic Analysis Division, OECD

CEPR Policy Research