Here is the paper I submitted for the AERE 2nd Annual Summer Conference:
Paper ID: 369
Paper Title: WILLINGNESS-TO-PAY FOR OYSTER CONSUMPTION MORTALITY RISK REDUCTIONS
Select Session Type: Contributed Session
Organized Session: No
Abstract: Oyster consumers are at risk from the Vibrio vulnificus bacteria. Vibrio is found in coastal waters and is most widespread in the warm waters of the Gulf of Mexico during the summer months. Vibrio can be transmitted to humans through the consumption of raw shellfish harvested from waters containing the organism. The ingestion of the Vibrio bacteria typically poses little risk of illness when consumed by a healthy adult with a normally functioning immune system. However, a small percentage of the oyster consumer population is immunocompromised and at a greater risk for contracting infections. The risk of life threatening illness from consuming oysters arises primarily if the oysters are consumed raw or undercooked. At an average annual rate, approximately 30 consumers become seriously ill from consuming raw Gulf oysters, of which 15 die.
Over the last decade, the FDA has educated at-risk consumers with fact sheets and brochures detailing the risks associated with raw oyster consumption. Despite this, the frequency of illness has remained constant. With the ineffective education strategies, the FDA mandated that states implement controls to reduce the incidence of illness. These plans focused on encouraging the use of post-harvest processing (PHP) technologies for reducing bacteria levels. There are four FDA-approved oyster processing methods that may be used to reduce bacteria in oysters to non-detectable levels: irradiation, individual quick-freezing, cool pasteurization, and hydrostatic pressure.
In October 2009 the FDA announced a new policy, requiring only the use of PHP Gulf oysters for the raw half-shell market during the months of April through October, with an effective date of May 2011. Significant industry concern surrounded the potential negative impact on the oyster industry. Based on these concerns, the FDA has since issued a letter postponing implementation until additional research into the consequences of such a ban could be completed.
In this paper we use data from an internet-based survey and estimate the benefits of the PHP policy with the contingent valuation method. Previous oyster safety policy analysis has relied on willingness-to-pay estimates from the labor market literature for valuing mortality risk. A number of studies find that the value of mortality risk reduction varies by context. These studies raise the question of whether existing oyster safety policy analyses have accurately compared benefits and costs. In addition to providing a context specific estimate of willingness-to-pay for oyster health risk reductions, we consider two unresolved issues in the contingent valuation health risk literature. First, a number of studies in the mortality risk reduction literature find that willingness-to-pay is not sensitive to the scope of the risk change. One recognized problem is that survey respondents have difficulty understanding small changes in probabilities regardless of how they are presented.
Another strategy is to change the format of the valuation question. In this paper we ask respondents about their willingness-to-pay for lives saved directly, without requiring an understanding of small probabilities. Second, a number of studies have addressed the theoretical aspects of altruism in benefit-cost analysis and argue that willingness-to-pay that includes altruism towards others should not necessarily be included in benefit-cost analysis. Previous research has attempted to isolate this effect by comparing public and private programs. In contrast, we are able to identify those at risk respondents who would personally benefit from the policy. The residual willingness-to-pay allows a decomposition of willingness to pay for personal mortality risk reductions and altruistic willingness-to-pay.
The sample consists of oyster consumers from Florida, Alabama, Georgia, Mississippi, Louisiana, Texas, and California and the survey was administered in 2010. The response rate was 53 percent and our analysis is conducted with a sample of 1849. We asked a number of revealed and stated preference questions related to PHP oysters. The average number of oyster meals that consumers state they would eat falls significantly with the PHP policy.
The stated preference section of the survey included three contingent valuation questions. Each willingness-to-pay question is in the referendum format. The first question considered a policy that would reduce a randomly assigned number of lives by mandating PHP oysters. The payment vehicle is an increase in the price of oyster meals. Those respondents who voted against or were undecided were presented with a second referendum question that reduced the number of deaths to zero. Respondents who voted against the second referendum or were undecided were asked a third referendum question that changed the policy to only be applied to those oysters consumed raw from April 1 through October 31.
Considering the first referendum, 34% of respondents would vote for the policy, 29% are undecided and 37% would vote against. The votes significantly vary across the three outcomes with the price increase. The Turnbull lower bound nonparametric willingness-to-pay is $2.13 per meal. Considering the large number of uncertain respondents we analyze the referendum votes with the ordered logit and multinomial logit models. Considering the two-thirds of the sample who would not decrease oyster consumption with the PHP policy, the referendum votes vary in expected ways with perceived health with PHP oysters. Considering the issue of scope, referendum votes vary significantly with the numbers of lives saved and the effect is enhanced if the respondent is at-risk. Considering those consumers who are not at risk (86% of the sample), altruistic willingness-to-pay per meal to save one life is $2.44. Considering those consumers who are at risk, willingness-to-pay per meal to save one life is $2.71. For both groups of consumers a theoretical decomposition of willingness to pay suggests that $1.08 is the value of a reduction in morbidity risk per meal. Given these estimates and a number of market-based assumptions the aggregate willingness to pay to save one life is almost $5 million. These results will be used to estimate the aggregate benefits of the PHP policy. Considering the market costs (i.e., increased production costs) and nonmarket costs (i.e., lost consumer surplus from reductions in oyster consumption), we will conduct a benefit cost analysis of the PHP policy.
Keyword(s): Oyster consumption, mortality risk, contingent valuation method
Field(s): Environmental valuation + CBA: Contingent valuation, Resources and Ecosystem Studies: Fisheries
Your Role: Presenter
Morgan, Ash, Appalachian State University (Co-Author)
Huth, Bill, University of West Florida (Co-Author)
Martin, Greg, University of Northern Kentucky (Co-Author)
Sjolander, Rich, University of West Florida (Co-Author)