Society For Risk Analysis Annual Meeting 2016

Session Schedule & Abstracts


* Disclaimer: All presentations represent the views of the authors, and not the organizations that support their research. Please apply the standard disclaimer that any opinions, findings, and conclusions or recommendations in abstracts, posters, and presentations at the meeting are those of the authors and do not necessarily reflect the views of any other organization or agency. Meeting attendees and authors should be aware that this disclaimer is intended to apply to all abstracts contained in this document. Authors who wish to emphasize this disclaimer should do so in their presentation or poster. In an effort to make the abstracts as concise as possible and easy for meeting participants to read, the abstracts have been formatted such that they exclude references to papers, affiliations, and/or funding sources. Authors who wish to provide attendees with this information should do so in their presentation or poster.

Common abbreviations

T4-D
Revolutions in Benefits Analysis

Room: Marina 6   3:30 pm–5:10 pm

Chair(s): Kevin Brand   kbrand@uottawa.ca

Sponsored by Economics and Benefits Analysis Specialty Group and Society for Benefit-Cost Analysis



T4-D.1  3:30 pm  Challenges to product labeling: consumer protection or opportunism? Cantor RA*, Cross PJ, Mackoul CV; Berkeley Research Group   rcantor@thinkbrg.com

Abstract: Investigations in the area of product labeling have expanded substantially in recent years. Regulators and consumer advocates are looking closely at labels designed to signal specific product attributes related to health, safety, environmental, and performance qualities. When the attribute information is challenged, it is becoming more common to encounter the related class-action litigation for damages to consumers from alleged mislabeling. Class-action litigation can be an expensive “bet-the-company” process for product producers. Given the growth of these matters and their high cost, it is worth investigating the economic foundations of the theory and methodology used to support mislabeling impact and damages claims. This paper examines how the economic theory of demand is employed to support a theory of impact and damages in mislabeling class actions. Other economic theories of supply and competition are often ignored, especially for their influence on product pricing. We also review the methodologies used by experts for plaintiffs to measure demand in mislabeling matters and contrast the reliability of stated preference versus revealed preference as a foundation for establishing class and estimating damages in tort litigation. We conclude with some thoughts regarding the social value of mislabeling class actions in the context of protecting consumers as opposed to merely enriching a legal community that might be exploiting advances in stated preference methodologies in a lucrative litigation area.

T4-D.3  4:10 pm  Using FDA adverse event data to estimate the avoided risk of allergic reactions from bakery products through recalls. Estrin AE*, Lasher AB, Nolan NM, Levine JC, Willig JA, Brewer VA, Chen Parker C, Markon AO, Nsubuga J, Wolpert, BJ; Grant EA; Federal government   andrew.estrin@fda.hhs.gov

Abstract: The Food and Drug Administration (FDA) Amendments Act of 2007 required FDA to establish the Reportable Food Registry (RFR), an electronic reporting system used by industry, to submit reports when there is reasonable probability that FDA-regulated food will cause serious adverse health consequences. According to RFR’s 2009-2013 annual reports, undeclared allergens account for 30%-44% of primary reports–most resulting in recalls. Bakery products account for the largest share of primary reports from undeclared allergens (20%-24%). We use adverse event reports citing allergic reactions from bakery products from FDA’s Center for Food Safety and Applied Nutrition Adverse Event Reporting System (CAERS) to estimate average risk per serving from bakery product allergens. After adjusting for self-reporting and under-reporting, we apply the per case health cost of an allergic reaction from Minor et al. (2014) ($2,563) to the total number of CAERS cases citing bakery product allergic events. Based on 2013 Nielsen scanner data, we estimate the number of annual retail bakery product servings sold (~38.5 billion); based on USDA food consumption data, we obtain the ratio of annual away-from-home to at-home grain consumption (~0.44843) and calculate total annual exposure at ~55.8 billion servings. We apply the estimated value of the risk per serving to the number of servings of recalled bakery products estimated from CFSAN recall information. This may overstate avoided risks from undeclared allergens to the extent that CAERS includes allergic reaction reports citing intake of products with labels that do and do not fully declare allergens on their labels. We incorporate uncertainty into the estimates of self-reporting and under-reporting biases, the health cost of an allergic reaction, and exposure to bakery products using @Risk simulations to compute a 95% confidence interval for the risk per serving and avoided costs of allergic reactions from bakery product recalls.

T4-D.4  4:30 pm  Innovative experiments to explore possible mis-estimation of the net benefits of environmental, public health, and safety regulations. Finkel AM*, Johnson BB; Univ. of Pennsylvania Law School and Univ. of Michigan School of Public Health   afinkel@law.upenn.edu

Abstract: Many aspects of cost-benefit analysis (CBA) are controversial, but ironically, the mechanics of how to quantify the effects of both risk reduction and cost on human welfare has received less attention. This new project presumes that altering conventional methods of benefit and cost valuation in two particular ways might change which regulations pass or fail the cost-benefit test, and might alter the maximally net beneficial regulatory option for a given regulation. First, stated-preference estimates of the “value of a statistical life” (VSL) deliberately preclude altruism, and confront subjects with tiny costs and minuscule probabilities of personal mortality in order to value public programs benefiting (and costing) the entire nation. Secondly, CBA tallies only the total number of lives saved by a regulation or its total cost, regardless of whether individual mortality risks are concentrated or whether the costs affect some businesses or consumers disproportionately (hence it is oblivious to the diminishing marginal utility of money). We are currently designing two large survey experiments will probe these simplified assumptions and offer principled, quantitative alternatives. The first survey will estimate the “VSL with shared purpose”; it will test for (and isolate) the effects of paternalistic versus non-paternalistic altruism, and will pose the tradeoffs as both a user-defined acceptable range of cost for a fixed number of lives saved and a user-defined minimum number of lives saved for a fixed regulatory cost. A second survey will test the assumption that individuals regard the welfare effects of risk or cost at any level as linear, using subjects’ ratings of how dire they view varying hypothetical individual probabilities of harm and varying personal costs. This experiment is designed to reveal whether there are de minimus levels of either risk or cost that can sensibly be rounded down to zero, and/or intolerably high levels that should not be counted as merely proportional to the effects at lower levels.

T4-D.5  4:49 pm  What if Revealed Preference Isn't So Revealing? Insights from Agent-Based Modeling and Complex Systems for the Practice of Benefit-Cost Analysis. Campbell HE*; Department of Politics and Policy, Claremont Graduate University   heather.campbell@cgu.edu

Abstract: One of the assumptions used in Benefit-Cost Analysis (and other Economics) is that revealed preferences provide a better measure of actors goals than do other methods such as surveys. In keeping with this, when there is a difference between preferences expressed in surveys versus observed social outcomes, the assumption is often made that the stated preferences are strategic, insincere, or otherwise untrue, and that the truth is shown through the outcomes rather than the statements. However, a key insight from the study of emergence in complex systems is that emergent outcomes may, in fact, be quite different from the preferred outcome of every actor. This can be tested through the use of Agent-Based Modeling (ABM), in which individual actors are assigned preferences and, due to the algorithmic nature of the process, they perform their stated preferences exactly, yet end up with undesired outcomes. Indeed, this insight fits with early work on segregation by Schelling, in which he was able to show that his agents ended up in neighborhoods that were more segregated than their preferences. But the implications of this for BCA, and the significant doubt thrown on the use of revealed preferences in complex systems such as cities, has not been brought into BCA practice. The proposed presentation will discuss evidence that revealed preference may be different from actual preference, and discuss implications for BCA practice.



[back to schedule]