Discussion Paper No. 1378 examines the link between tax policy and economic growth, focusing in particular on the importance of the mix between direct and indirect taxes and, within direct taxes, between capital and labour income taxes. Enrique Mendoza, Research Fellow Gian Maria Milesi-Ferretti and Patrick Asea re-examine the theoretical and empirical relations between tax policy, investment and growth. By testing the predictions of standard two-sector endogenous growth models in which long-run growth is driven by human as well as physical capital accumulation, they provide evidence in support of Harberger's superneutrality conjecture, which contends that, although in theory the mix of direct and indirect taxes affects investment and growth, in practice tax policy is ineffective as an instrument to promote growth. A new methodology is used to construct aggregate effective tax rates on capital, labour and consumption for 18 OECD countries.

These tax measures are then exploited in a cross-country regression to study whether it is possible to detect significant effects of the tax structure on private investment and economic growth. The results show that the tax structure has statistically significant effects on the ratio of private investment to GDP, and helps explain a sizeable fraction of its variability. Notwithstanding its strong effect on investment, however, tax policy only has a small and in most regressions statistically insignificant impact on economic growth. These findings are robust to the introduction of other growth determinants such as initial income, the level of secondary school enrolment and fluctuations in the terms of trade.

On the Ineffectiveness of Tax Policy in Altering Long-Run Growth: Harberger's Superneutrality Conjecture
Enrique G Mendoza, Gian Maria Milesi-Ferretti and Patrick Asea