The “Lee thesis” claims that there is a tradeoff between a country’s respect for human rights and its economic growth. This argument has been used for justifying the lack of protection of human rights in developing countries. Without acknowledging the legitimacy of the argument, this paper investigates the claim. It does this by analyzing the effects of a subset of human rights (physical integrity rights) on economic growth. The paper presents a brief review of previous theoretical and empirical contributions to understanding how human rights practices affect various economic variables, including foreign direct investment, domestic investment, foreign aid, and trade.
The main focus of the paper is on the empirical level, by means of econometric analysis on cross-sectional time-series data. With country-years as the unit of analysis, the paper presents estimations on the effect of physical integrity rights on economic growth through diverse econometric techniques. Using OLS, a cross-country estimation is performed on values averaged over time, and this estimates significant positive effects of physical integrity rights. However, deeper analysis using the panel data and dummy variables for each value of the physical integrity index (range 0-8) allows a nonlinear relationship to be observed.
The results are striking. There appears to be a strong nonlinear and non-monotonic relationship between the physical integrity index (range 0-8) and GDP growth. The results from OLS with panel corrected standard errors, random effects, fixed effects and nonparametric matching all indicate a similar structure. While a score of 0 on the physical integrity index is associated with the lowest growth rates, a score of 2 is associated with the highest growth, in all estimations. Scores above 2 are generally associated with substantially lower growth rates, though a relatively large increase in growth is observed as the score goes from 7 to 8.
The non-monotonic result is consistent with Robert Barro’s inverse U-shaped relationship between democracy and economic growth. His argument is that an initial reduction of repression facilitates higher growth through reduced fear and higher security, allowing a country’s inhabitants to be more productive. A subsequent reduction of repression leads to less growth through demands for redistribution or political uncertainty. However, in contrast to Barro, here a final increase in growth is observed as the repression reaches the lowest level.
Further analysis indicates that this relationship is robust to various specifications. Regressions on disaggregated indices indicate that there are heterogeneous effects of the variables indicating torture, disappearances, extrajudicial killings and political imprisonment, which together make up the physical integrity index. These are estimated to have various effects on growth, possibly leading to the nonlinear and non-monotonic relationship between the aggregated physical integrity index and growth.
While human rights practices affect growth, growth also affects human rights practices. An attempt to control for reverse causality using two stage least squares was performed, and this reports a negative, though insignificant, effect of physical integrity rights on growth when using a single linear estimator. Nonlinear estimation using the technique did not give meaningful results. However, there are indications of endogeneity bias, and the empirical results using other techniques must be interpreted with caution.
The Lee thesis cannot be rejected on the findings in this paper. However, the Lee thesis can be rejected by rejecting the premises of the argument.