One way of gauging how global warming will affect an economy is to look at the economic performance of countries that are warmer. This paper looks at the income-temperature relationship for a cross-section of 156 countries in 1999. As is well known, hotter countries are poorer on average. The widespread belief is that this relationship is mostly historical; that is, due to a past effect of climate. Acemoglu, Johnson, and Robinson have recently made great gains in identifying a specific historical path. They posit that mortality rates of early colonizing settlers had a profound effect on the institutions that were set up in those colonies. These institutional differences persist to this day, they argue, and have strong effects on current incomes. Because colonial mortality and average temperature are highly correlated, the mortality-income relationship also manifests itself as an income-temperature relationship. There is, however, sufficient evidence to warrant continued examination of the income-temperature relationship. First, we find a strong income-temperature relationship within OECD countries, a result that does not appear to be predicted by the colonial mortality model and that various authors seem to disavow. Second, we find that the income-temperature relationship is essentially the same within the OECD and non-OECD countries, a striking yet unremarked and as-yet unexplained result. Third, we find an exceptionally strong income-temperature relationship within the fifteen countries of the former Soviet Union, where colonial institutions would seem to have been wiped out. Our best measure of the effect of temperature on income, after accounting for the influence of colonial mortality, is that a one percent increase in temperature leads to a -0.9 percent decrease in per capita income. Thus, a temperature increase of 3 degrees Fahrenheit would result in a 4.6 percent decrease in world GNP.
原文链接
分享
引用
