When does investment deliver growth? Human capital and heterogeneous investment–growth associations in Sub-Saharan Africa

This study examines whether the association between investment and GDP per capita growth varies with human capital. Using 879 country-year observations across 37 Sub-Saharan African countries from 1998 to 2023, we estimate dynamic fixed-effects models with one-year-lagged Penn World Table (PWT) investment and human capital. Controls include lagged growth, lagged real GDP per capita, macroeconomic conditions, and country and year effects. We find that higher investment is associated with stronger subsequent growth at lower human capital levels, but this association declines as human capital rises. At the 5th percentile of the Human Capital Index, a one-standard-deviation (8.41 percentage points) increase in PWT investment share is associated with approximately 1.10 percentage points higher annual growth. At the 75th percentile, the association is close to zero. The investment–human capital interaction is negative (β = −0.150, SE = 0.052) and statistically significant under conventional country-clustered inference (p = 0.007) and the Webb wild-cluster bootstrap (p = 0.034). It remains negative across principal sensitivity analyses, with lower precision under country-specific trends and alternative measures. The findings demonstrate that the association between investment and growth varies with human capital levels, but do not establish that human capital itself causes this variation.