Working Paper: NBER ID: w9201
Authors: Jeffrey A. Frankel; Andrew K. Rose
Abstract: What is the effect of trade on a country's environment, for a given level of GDP? Some have observed an apparent positive correlation between openness to trade and measures of environmental quality. But this could be due to endogeneity of trade, rather than causality. This paper uses exogenous determinants of trade geographical variables from the gravity model as instruments to isolate the effect of openness. The finding is that trade may indeed have a beneficial effect on three measures of air pollution. Statistical significance is lacking for Particulate Matter, but is moderate for NO2, and high for SO2. Results for broader environmental measures are not as encouraging, but one can at least say that there is little evidence that trade has the detrimental effect on the environment that the race-to-the-bottom theory would lead one to expect. The larger effect appears to come via income itself: our results generally support the environmental Kuznets curve, which says that growth harms the environment at low levels of income and helps at high levels, and to support the proposition that openness to trade accelerates the growth process.
Keywords: Trade; Environment; Kuznets Curve; Pollution
JEL Codes: Q2; F1
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| openness to trade (F10) | real income per capita (E25) |
| output (C67) | pollution (Q53) |
| real income per capita (E25) | environmental quality (Q50) |
| income (E25) | environmental quality (Q50) |
| higher income levels (D31) | public demand for environmental quality (Q53) |
| public demand for environmental quality (Q53) | better environmental regulation (Q58) |
| trade (F19) | environmental outcomes (Q56) |