Recent empirical evidence has documented some forms of local pollution (airborne lead, sulfur) that have declined significantly in industrialized countries despite robust economic growth. Poor countries appear relatively unpolluted, middle-income countries more polluted, and rich countries clean again. Because this evidence comes from reduced-form, country-level regressions of pollution on per-capita GDP, the cause of this inverse-U pattern is uncertain and the policy implications are ambiguous. Some have inferred that poor countries are not efficiently regulating externalities. Others have concluded that environmental progress eventually becomes an automatic consequence of economic growth. If nothing else, these disparate claims highlight the need for a causal explanation of the relationship between economic growth and environmental quality.
In very recent years, several theoretical explanations have circulated. Because these theories were designed to generate inverse-U-shaped pollution income patterns, they cannot be distinguished empirically on the basis of those patterns. However, the theories have different implications for people's marginal "willingness to pay" for environmental cleanup, and therefore may have testable empirical predictions using international surveys of environmental priorities.
We begin by noting that the inverse-U-shaped pollution-income relationship may be based entirely on individual preferences. This "individual preferences" explanation demonstrates that the observed pattern of pollution and income is not sufficient evidence that low-income countries set either Pareto-efficient or inefficient policies. Without any institutional barriers to accommodating individual preferences, this explanation has no implications for policy. Moreover, the stark prediction of this explanation is that people's marginal willingness to pay (MWTP) for environmental quality will be the same at different income levels, ceteris paribus.
We then identify three theoretical explanations that depend on institutional differences between countries with varying levels of per-capita GDP. These include constraints on technologies, constraints on government institutions, and returns to scale in pollution abatement. Some of the models predict that MWTP increases with income, some predict increasing and then decreasing MWTP, while still others have no implications for the pattern of MWTP and income. Furthermore, because the models rely on different mechanisms to generate the inverse-U-shaped pollution income path, they have different policy implications.
In what follows we examine patterns of willingness to pay across households with different incomes and countries with different levels of GDP, using household-level data from the third wave of the World Values Survey (WVS), a poll of about 70,000 respondents in 48 countries. Among other questions, the survey asked "would you be willing to pay 20 percent higher prices to protect the environment?" and "which is more important: environmental quality or economic growth?" While these do not conform to standard contingent valuation protocols, we are not seeking specific dollar measures of willingness to pay. Instead, we only wish to see whether the patterns of marginal willingness to pay conform to those predicted by the models. For this purpose the WVS provides a reasonable first look at these previously unstudied empirical implications of the growth and environment theories.
