102
CHAPTER 2
GLOBAL PRODUCTIVITY
TABLE 2.1 Recent developments in productivity drivers Proximate sources
Driver
Are EMDEs approaching advanced economies?
Innovation
Partially. Divergence in innovation measures excluding China.
Capital
Partially. Higher investment ratio in EMDEs; divergence in financial development indicators.
Education
Partially. Convergence in secondary education and divergence in tertiary education.
Health
Partially. Convergence in infant mortality and divergence in old age life expectancy.
Demographya
No. Working-age population share in EMDEs has started to decline.
Supporting Institutions environment
Partially. Significant improvements in rule-based fiscal and monetary policy, but limited improvement in institutional measures.
Macroeconomic stability
Yes. Milder inflation and fewer financial crises compared to 1980-90s.
Gender inequality
Partially. Gender gaps are narrowing except in some regions.
Income inequality
No clear uniform trend in the last 25 years.
a
Market Trade/Complexity development
Stalled. Divergence of GVC participation. No convergence in economic complexity, with a few exceptions.
FDIa
Stalled. No increase in FDI.
Urban
No. Continued large gap in urbanization rate.
Sample period is the longest available: typically, 1960-2018, but significantly shorter for some drivers. Source: World Bank. Note: EMDEs = emerging market and developing economies; FDI = foreign direct investment; GVC = global value chain. a. These drivers are not necessarily lower in EMDEs than in advanced economies. The answers in the row are about absolute improvements, rather than improvements relative to advanced economies.
per capita—and in EMDEs improvements in other drivers, such as institutions and economic complexity, have stalled (table 2.1).
Analyzing the effects of drivers Thus far, the analysis has considered individual drivers in isolation. This section considers them together: it examines the partial correlations between productivity growth and various drivers, and how they have changed over time. Methods. To study the role of drivers in productivity growth, cross-country regressions are used. These regressions are useful for uncovering associations between initial conditions and later growth. The sample comprises 60 countries, including 38 EMDEs, observed from 1960 to 2018. The time span is longer than in many previous studies and should ensure that the results are not confounded by short-run or cyclical effects. The use of initial values of the drivers, rather than averages or changes during the sample period, helps to address potential concerns over reverse causality. Nevertheless,
