اثر انتخاب درآمد بر ساخت جانبداری در تصمیمات سیاسی با استفاده از تجزیه و تحلیل هزینه - فایده

Cost–benefit analysis (CBA) is often used in policy decisions to determine the economic value of various choices. Although the effects of income disparity are well accepted, one of the reasons for the success of this type of analysis is that many policy-makers and analysts consider the results of a well-crafted analysis to be largely free of partisan or political bias. Here I propose that tradeoffs between the benefits of various possible jobs or careers introduce a significant bias into values imputed based upon willingness to pay (WTP) measures. In particular, I demonstrate why we might expect the values of various segments of society to be differentially appreciated by CBA. Empirical evidence is presented, indicating that there is a significant negative correlation between the importance an individual places on income and the willingness of that individual to forgo consumption in favor of environmental improvement, as well as a correlation between the importance of income and actual income. These results suggest that CBA as a tool of welfare economics is likely to be biased against environmental protection because concerned individuals are likely to choose careers which do not maximize consumption, thus decreasing their ability to pay for both real and hypothetical environmental improvements. Moreover, the size of this bias appears to be large.

The study of public economics has three main functions: (1) to understand economic systems, (2) to develop efficient methods for carrying out public policies that have already been decided upon, and (3) to suggest which policies to implement. The first and second functions have been called positive economics, while the third is often termed welfare economics. In this paper, I will suggest that the use of economic measures for welfare applications can allow for the de facto preference of one set of values over other sets, thereby inherently biasing the political decisions to which it is applied. This is likely to result in a hidden political advantage for those adhering to the preferred value set. The use of cost–benefit analysis (CBA) in public policy has been growing over the last three decades. Its early use in the review of environmental and health policy began with attempts by U.S. Presidents Nixon, Ford, and Carter to increase executive influence over the regulatory process, under titles such as “Quality of Life Reviews” and “Inflation Alerts”. Then, in 1981, President Ronald Reagan issued Executive Order (EO) 12291 that required agencies to submit Regulatory Impact Analyses (RIAs) on all major proposed regulations for review by the Office of Management and the Budget. This order required that “the potential benefits outweigh the costs” and “of all the alternative approaches to the given regulatory objective, the proposed action will maximize net benefits to society” (for a discussion, seeMorgenstern, 1997 and Pildes and Sunstein, 1995). By 1983, the U.S. Environmental Protection Agency (EPA) had issued its Guidelines for Performing Regulatory Impact Analysis, which it recently updated in order to reflect the flurry of new activity which this area has undergone ( EPA, 2000). Examples are the Unfunded Mandates Reform Act of 1995, Executive Order 12898 (Federal Actions to Address Environmental Justice in Minority Populations and Low-Income Populations), and, most importantly, the 1993 Executive Order 12866 (Regulatory Planning and Review which superceded EO 12291). This last directive, issued by President Clinton, adds some flexibility for agencies to consider distributional and equity effects, but retains from EO 12291 the requirements for CBA and the examination of alternative approaches. In the mid-1990s, attempts by the Republican-controlled Congress to strengthen CBA requirements failed ( Morgenstern, 1997), but John Graham, George W. Bush's Administrator of the Office of Information and Regulatory Affairs, has been more active than his predecessors in his enforcement and interpretation of EO 12866 ( EPA, 2000, Graham, 2002 and Craig, 2003). CBA is today involved in numerous decisions regarding “whether, when, and how to regulate” ( Portney, 1990). This increased reliance on CBA in recent years stems largely from a desire to increase reliance on objective, quantitative methods for policy comparison and optimization. Given that not all public projects can be fully funded, CBA allows one to compare various options and reject the allocation of scarce funds to those projects which do not provide sufficient margin of return or which do not pass some other criterion used in quantitative or semiquantitative decision-making. The classic criticism that CBA ignores things of value which are not traded in markets, or ignores nonuse values like existence value, has been mitigated to some extent by the increasing use of nonmarket mechanisms (hedonic pricing, contingent valuation, etc.) to impute willingness to pay (WTP) for these goods, although standard regulatory cost–benefit analyses continue to apply these techniques conservatively if at all (see, for example, Johansson et al., 1995, Bateman and Willis, 1999 and Farber et al., 2002). The ability of CBA to fully overcome this criticism remains a point of discussion in the literature.

I have shown here that the values resulting from CBA are likely to not be unbiased, but instead to be more strongly influenced by individuals who value consumption potential over other, nonmonetized benefits in their career choice. I have also shown that these individuals tend to be less willing to pay for environmental benefits than others who value consumption less. This analysis indicates that the use of an income-dependent monetary metric in policy-making is value-laden in a society where career choice includes premarket benefits and where consumption beyond a given level may be seen as nonproductive by a significant sector of society. If individuals pursue income with different levels of vigor, then it seems unreasonable to expect income-dependent measures to reflect meaningful societal values, even if we are willing to accept the consequences of large income disparities in determining consumption levels. In essence, the use of CBA can be said to partially disenfranchise citizens who voluntarily eschew consumption-maximizing career paths. Along these lines, many observers have suggested the need for a large-scale, voluntary transition toward a less-consumptive, more sustainable society. It is important to keep in mind that, during the transition period, we would expect the impact of the disparity discussed here to increase as a larger fraction of the population accepted compensation in nonconsumptive form. If such a transition took place because a fraction of the population was concerned about ecological sustainability (or for any other reason associated with a particular set of values), then, during the transition, we would expect that public policy based upon economic measures would become increasingly weighted towards those whose career choices remained driven mostly or largely by income. This is an important point and applies to the use of any economic measures in the decision-making process, not only CBA; as values change, so should the fundamental assumptions of economists when they attempt to explain behavior and recommend policies based upon their models. More work needs to be done to determine the current significance of this result. Given that the income difference between those who consider income to be of primary importance and those who give it the least importance is about 10% of median income, this might make a reasonable rough estimate for the strength of the effect upon the results of CBA. The value of CBA to policy-makers would certainly not be nullified by a 10% bias, although such a strong, consistent bias would need to be factored into a policy-maker's intuitive welfare function. At the same time, this survey did not include individuals with incomes above $163,000 (in 1986 dollars), and so it is possible that in leaving out the tails of the distribution significant effects related to motivation and market power have not been captured. Because of this, the likelihood of temporal changes in the strength of the effect, and for other reasons, the main emphasis of this paper is the direction of the bias, rather than its quantification. Given that relevant policy choices are often not continuous, but binary (a project is undertaken or it is not, a regulation is promulgated or it is not), a consistent bias, even if fairly weak, will often have the effect of driving nonoptimal policy choices if the decision variables are near the indifference point. Therefore, this paper does not claim to argue that CBA is so biased as to be unusable for welfare economics, only that the bias demonstrated here is significant and warrants discussion. In addition, this analysis begs the interesting question of how one might correct for the observed bias in future work with CBA. One possible improvement would be to increase the use of the willingness to accept compensation (WTA) criterion, rather than WTP. Although WTA can still be sensitive to income, it is not limited by a strict income constraint, and so a person is free to consider their entire income, rather than simply their monetary income, in deciding how valuable would be an increase in their monetary wealth. This would then obviate the current problem of whether a person takes their compensation premarket or not. In fact, many, and perhaps most, environmental CBAs deal with questions where something is being lost and the question is how much to curb or hinder economic activities that will lead to the loss. On theoretical grounds, these CBAs should be based on WTA instead of WTP anyway. However, WTP is often used instead (see discussions in Bateman and Willis, 1999; and for recent examples, Turner et al., 2003 and Turpie, 2003). This is in part because the unboundedness of WTA causes suspicion among many economists, especially when using nonmarket valuation methods. This paper simply indicates that the choice of one or the other may be more significant than is currently thought due to the specific bias demonstrated here, and that it is not acceptable to substitute WTP for WTA as is often done in conventional CBA. There may also be other methods that would allow analysts to consider an individual's total income (in the form of money, time, self-satisfaction, etc.), rather than simply monetary income, when determining prices for nonmarket goods. Essentially, the important distinction is that because consumer goods are purchased with individual consumption in mind, market prices are usually sufficient to determine their market value; however, when determining private values for public goods in CV studies, a broader definition for “income” in WTP needs to be considered.