Working Paper: NBER ID: w17788
Authors: Yinwong Cheung; Michael P. Dooley; Vladyslav Sushko
Abstract: This paper revisits the association between investment and growth. The empirical findings highlight substantial heterogeneity for the effect of investment on growth and suggest a possible negative association. Results based on a battery of cross-sectional and time-series regressions show that the link between investment and growth has weakened over time and that investment in high-income countries is more likely to have a negative effect on growth. The adverse effect for high-income countries appears to have increased over time. An implication is that uphill capital flows could be associated with negative or zero returns. The result is robust to the presence of control variables that are commonly included in growth studies.
Keywords: investment; growth; high-income countries; low-income countries; diminishing returns
JEL Codes: F43; O40; O57
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Investment (G31) | Growth (O00) |
| High-income countries (O57) | Negative growth effects from investment (E22) |
| Investment share in GDP (E20) | Growth (O00) |
| High-income countries + Investment share (F21) | Negative impact on growth (F69) |
| Investment (G31) | Negative growth effects in high-income countries (F69) |
| Investment and growth link (E22) | Weakened over time (N13) |