Page 15 - Education and Inclusive Growth --Jong-Wha Lee Korea University

Education and Inclusive Growthc161

change can raise the relative demand for skilled workers and skill premium,

thus promoting human capital accumulation. By ignoring those interactions, the
growth accounting result can over- or under-estimate the true contribution of
human capital to output growth. In addition, growth accounting does not explain

where the growth of inputs and total factor productivity comes from. The result
does not necessarily imply a causal effect of human capital in explaining the
growth rates of per worker output across economies over time. For example,

some exogenous factors can increase both educational attainment and output.
As an effort to identify the independent impact of human capital on
output growth, researchers have adopted a regression-based approach. After

controlling other important explanatory variables, the framework allows us to
investigate the independent role of educational attainment, measured by overall
years of schooling, in economic growth (Barro, 1991; Barro and Lee, 1994 and

2015). The regression often adopts cross-country panel data in order to explain
differences in growth performance across country,
The basic empirical framework is expressed as
(6) Dy = β + β log(y ) + β h + β X + ε .
i,t
it
i,t
0
3
2 i,t
1
i,t
where Dy is country i’s per worker GDP growth rate in period t, y is
it
i,t
country i’s per worker GDP at the beginning of period t, h is average years
i,t
of schooling at the beginning of period t, and x indicates a wide variety of
i,t
external environmental variables that influence long-term GDP growth rates.
As suggested by previous empirical researches, investment rate, institutions,
fertility, international trade, inflation and other policy factors are major
determinants of long-run per worker output (Barro and Xala-i-Martin, 2004;
Barro and Lee, 2015).
As can be seen in Equation (6), the estimate of the coefficient β measures
2
the impact of initial educational stock on subsequent per-worker output growth