Working Paper

Preferences for Equality in Environmental Outcomes

Sep 2, 2016 | Maureen L. Cropper, Alan J. Krupnick, William Raich

Summary

This survey demonstrates that people are willing to accept programs that results in a higher total environmental health risks (such as risks of cancer and lung disease), as long as the risks are equally distributed across the population.

Key Findings

  • Survey respondents are more averse to inequality in health risks than to inequality in income.
  • Nearly a third of respondents prefer a program that distributes health risks equally to a program with an unequal distribution of risks that makes everyone better off.
  • While respondents’ aversion to income inequality can be predicted based on their sociodemographic and political profile, it is much more difficult to explain why individuals are averse to inequality in health risks.
  • This survey demonstrates that eliciting inequality aversion to health risks is possible and, given the high rates of inequality aversion, important to consider when evaluating policy.

Abstract

Benefit-cost analyses traditionally evaluate the economic efficiency of health regulations—ignoring the equity of the regulations. To help address the importance of equity, we survey respondents to identify their preferences toward equality in health risks stemming from environmental causes (specifically risks of cancer and lung disease). Survey responses are used to parameterize an Atkinson index over environmental health risks. We compare these results to similar questions in the income context and find that respondents are significantly more averse to inequality in health risks than in income. The mean respondent is willing to accept a 22 percent increase in average health risk if risks are equally distributed in the population, but willing to accept a decrease of only 5 percent in average income if incomes are equally distributed in the population. We find that 30 percent of respondents answer health risk questions lexicographically—always preferring an equal distribution of risks to an unequal distribution, even if the latter makes everyone better off.