The limits of the environmental Kuznets curve?
Posted by Daniel Hall on September 21, 2007
National Geographic News has a series of photos of the 10 most polluted places on Earth. Like they say, a picture is worth a thousand words, and it’s hard not to look and think, “There, but for the
grace of God environmental Kuznets curve, go we.” These places are almost all in poor countries, and the environmental Kuznets curve seems to promise that things will get better once people get richer, as happened in the United States. But I wonder if there are not limits.
The environmental Kuznets curve (EKC) hypothesis posits an empirical U-shaped relationship between pollution and income (i.e. per capita GDP).* Although the research literature has tried to stay away from theoretical models and focus on the empirics, the popular explanation is that countries start out poor but with a relatively pristine environment, go through a phase of economic growth and pollution, and then once they get rich enough develop a taste for environmental quality and so clean up, thus generating the U-shaped relationship.
This relationship is found to hold empirically for local pollutants — indoor air pollution (smoke) or fecal coliform (crap) in water — and is generally observed for local-regional pollutants like NOx or SOx, but the evidence becomes more mixed as pollutants become more spatially dispersed. (See the tables at the end of this paper for a review of the literature findings.) Once you get to CO2, a global pollutant, the empirical evidence is very mixed indeed; most studies suggest that if there is a U-shaped relationship, the peak is beyond reasonable levels of income (well beyond current U.S. per capita income of ~$42,000 — which is obviously far higher than the global figure). As Free Exchange noted last week, there are logical arguments for why the EKC relationship might not hold for global pollutants, implying that rising incomes may do much less to help with climate change than some hope.
But there is also a possibility that the EKC relationship may not describe a sustainable path to reducing all local pollutants either. As this succinct definition explains, economists — to the degree that they theorize about the reasons for the EKC relationship — think of it as being composed of 3 effects: scale, composition, and technique. Scale pushes you up the curve: you get richer, produce more, and have more pollution. Technique pulls you down: you increase production efficiency, install pollution controls, and generally develop cleaner ways of doing the same things. The composition effect is uncertain: you change the mix of economic activity as income changes. In practice this seems to have played itself out as wealthier nations having shifted from manufacturing into service- and knowledge-based economies, while poorer countries have been industrializing rapidly — both processes aided by increased international trade and investment. Thus composition is a part of driving some countries up the curve and of helping other countries move back down. Suri and Chapman (1998) [gated] found that this effect was an important component of the EKC relationship.
But if composition effects are a driving some of the empirically observed EKC relationship, then we are playing a game where someone is going to get left
holding the short end of the stick hosting the manufacturing plants. Vapi, India may be able to look forward to a day where it is wealthier and cleaner. But until we start literally offshoring — or offworlding — our mining and manufacturing activities, or we completely lose our appetite for manufactured goods, it seems likely that someone, somewhere is going to have to live with the consequences.
*In fact originally Simon Kuznets described this U-shaped relationship between income inequality and per capita GDP; it was Grossman and Krueger who extended this idea to pollution.