2015 Conference- Session 8

Session 8 - Friday, March 20, 3:45-5:15 p.m.


A.8: Challenges and Opportunities for Economic Analysis of Risk Regulations (Marvin 309)

Chair: Jennifer Baxter (jbaxter@indecon.com), Industrial Economics, Incorporated

Panelists to Include:

1.     Elizabeth Ashley (Elizabeth_M_Ashley@omb.eop.gov), U.S. Office of Management and Budget

2.     Tony Cheesebrough (acheesebrough@gmail.com), National Protection and Programs Directorate, U.S. Department of Homeland Security

3.     Sandra Hoffmann (shoffmann@ers.usda.gov), Economic Research Service, U.S. Department of Agriculture

4.     Amber Jessup (Amber.Jessup@HHS.GOV), U.S. Department of Health and Human Services  

5.     Clark Nardinelli (clark.nardinelli@fda.hhs.gov), U.S. Food and Drug Administration

6.     Rosemarie Odom (rosemarie.a.odom@uscg.mil), U.S. Coast Guard

7.     Al McGartland (McGartland.al@epa.gov), U.S. Environmental Protection Agency

8.     Jack Wells (jackwells1@mac.com), U.S. Department of Transportation (retired)

Regulatory impact analysis (RIA) is an important tool for improving the quality of regulation and motivates the development of methods that are useful in other risk policy contexts. However, U.S. federal agencies face analytic challenges both common across the government and unique to their agencies. Developing and applying better tools both improves the quality of regulatory analysis and offers potential for improving agency decision making in other contexts. This panel brings together regulatory and other economists from across the government to share the particular analytic challenges and opportunities they observe in using these tools to promote evidence-based decision making.

B.8:Retrospective Review of Federal Regulations (Marvin 307)

Chair: Maeve Carey (MCAREY@crs.loc.gov), Congressional Research Service

Discussant: James Broughel (jbroughel@mercatus.gmu.edu), George Mason University


1.     Learning from Experience: An Assessment of the Retrospective Review of Agency Rules and the Evidence for Improving the Design and Implementation of Regulatory Policy, Joseph E. Aldy* (joseph_aldy@hks.harvard.edu), Harvard University

A well-functioning regulatory program makes the American people better off by promoting innovation; encouraging competition; protecting the air we breathe, water we drink, and food we eat; and improving the safety of our workplaces and the goods we buy. Determining if society is realizing the most out of its regulatory program requires rigorous analysis. Despite a long track record of prospective analysis for proposed regulations, the Federal government has a mixed record on retrospective review of existing regulations. Every administration dating back to the Carter Administration in 1978 has implemented some form of regulatory look-back. The ad hoc nature of the Presidential-mandated reviews, the apparent need for every administration to implement such a retrospective review, and the heterogeneity in approaches to retrospective review by agencies suggest that efforts to enhance and institutionalize retrospective review are merited. This paper evaluates the practice of the Obama Administration's retrospective review and places it in the context of the academic literature and past administrations' efforts at retrospective review. In particular, I analyze the processes and products of agency review plans, agencies' initial tranche of identified rules for review, and every economically significant rule finalized in fiscal years 2012 and 2013 in my examination of retrospective review under Executive Orders 13563, 13579, and 13610. This assessment identifies best practices among agencies, describes key lessons learned from the ongoing and past retrospective review efforts, and makes recommendations for ways to improve retrospective review.

2.     Looking Back at Retrospective Review: How Did Agencies Measure Up in 2014? Sofie E. Miller* (sofiemiller@gwu.edu), The George Washington University Regulatory Studies Center

Through a series of Executive Orders, President Obama has encouraged federal regulatory agencies to review existing regulations “that may be outmoded, ineffective, insufficient, or excessively burdensome, and to modify, streamline, expand, or repeal them in accordance with what has been learned.” Evaluating whether the intended outcomes of regulations are met ex post can be challenging, so multiple government guidelines instruct agencies to incorporate retrospective review plans into their proposals during the rulemaking process. To support this effort, the George Washington University Regulatory Studies Center examined significant regulations proposed in 2014 to assess whether they included plans for retrospective review, and provided recommendations for how best to do so. This paper examines how often agencies successfully incorporate plans for ex post analysis into their rules and provides agencies with five recommendations to facilitate transparency, public accountability, and measurement of their rules’ success.

3.     A Framework for Evaluating Regulatory Outcomes, Kathryn Newcomer* (newcomer@gwu.edu), Susan Dudley, and Estelle Raimondo, George Washington University

Benefit-cost analysis is a key component of the regulatory impact analysis that agencies are required to conduct before introducing new regulations. However, the U.S. government has traditionally placed much less emphasis on ex-post evaluation of regulatory outcomes after they are in effect. This emphasis on ex-ante rather than ex-post regulatory impact analysis differs from the practice in other government policy tools, where evaluation of program outcomes is routine. Better ex-post measurement and evaluation of regulatory benefits and costs is essential for evidence-based decision making, and could be helpful not only for gauging regulatory effectiveness, but also for improving ex-ante predictions of regulatory impact. This paper will attempt to bring together the techniques of program evaluation and regulatory benefit-cost analysis, which are each well-established in their own right but have not generally been used to inform each other. We start with the premise that evaluating the impact of regulations is a process analogous to evaluating the impact of public programs, which depends on proper modeling and valid and reliable measurement of inputs, outputs and outcomes. We hope to develop a general framework for measuring regulatory impacts and test it with a case study.

C.8: State and Local Benefit-Cost Issues (Marvin 308)

Chair: Ronald Bird (rbird@uschamber.com), U.S. Chamber of Commerce

Discussant: Jerry Ellig (jellig@mercatus.gmu.edu), George Mason University


1.     Evidence-Based Policymaking: Integrating Cost-Benefit Analysis into a Broader Approach to Fund What Works in State and Local Government, Darcy White* (dwhite@pewtrusts.org) and Torey Silloway* (tsilloway@pewtrusts.org), The Pew Charitable Trusts

Governments make policy and budget choices each year have long-term impacts on their fiscal and social outcomes. Currently, governments often make these decisions based on inertia, anecdotes, and political and ideological factors, but they can achieve substantially better outcomes by using rigorous evidence to guide their choices. This approach, called evidence-based policymaking, can enable policymakers to strategically fund and operate their programs, and benefit-cost analysis plays a key role in this process. To date, most states and local governments have made limited use of benefit-cost analysis and evidence-based policymaking, and there has been no comprehensive roadmap to help guide them on this approach.  Since 2011, the Pew-MacArthur Results First Initiative (Results First) has worked in a growing number of states to help them customize and use a benefit-cost analysis model initially developed by the Washington State Institute for Public Policy. Results First has also developed an integrated framework of steps that governments can take to build and support a system of evidence-based policymaking. This framework, built on an extensive review of research and discussions with public officials, practitioners, and academic experts, includes five key components – program assessment, budget development, implementation, oversight, and evaluation. The framework will be published as a Pew report in the fall of 2014.  This presentation will present the framework and highlight the role of benefit-cost analysis in helping governments to think holistically about using research and benefit-cost analysis to inform their budget and policy choices. It will emphasize that while identifying and investing in cost-effective programs and practices are an essential step to achieving strong outcomes for citizens, it must be done in conjunction with implementation fidelity and outcome monitoring. It will also discuss how Results First states have used benefit-cost analysis to drive evidence-based policymaking in their states and the results they have seen to-date.

2.     Why Look Back? An Analysis of State Government Decisions to Analyze Existing Regulations, Stuart Shapiro* (stuartsh@rutgers.edu), Debra Borie-Holtz and Ian Markey, Rutgers University

Regulatory agencies have long been required to analyze the costs and benefits of their prospective regulations.  Yet once a regulation is in place, discussions of its effectiveness and its costs and benefits are, if they are examined at all, the province of academic analyses.  Rarely do agencies revisit earlier regulatory decisions.

In recent years “retrospective review” has started to become more common.  At the federal level, President Obama’s Executive Order 13563 required agencies to conduct some reviews of existing regulations.  Among the 50 states, exactly half have also implemented retrospective review requirements on regulatory agencies.  For proponents of retrospective review (also known as “look-backs”) this trend is an encouraging sign that governments are beginning to examine the benefits and costs of previous regulatory decisions, and hopefully revise policies that have not been successful.  However, good policy is only one possible motivation for this change.

We examine the recent wave of retrospective review in the 50 states in an attempt to determine the motivation for states to pursue this policy reform.  We have collected data on the use of retrospective reviews in each state as well as political and economic variables that may have influenced the decision by governors and legislatures to undertake a look-back at older regulations.  Using survival analysis, we model the decision to undergo retrospective review of regulations as a function of these independent variables.  The results of this analysis are then used to draw conclusions about the likely fate of retrospective review as political and economic conditions change.

3.     Ex-Ante Cost-Benefit Analyses of Community-based DRR Interventions in the Caribbean, Meenakshi Jerath* (mjera001@fiu.edu) and Juan Pablo Sarmiento, Florida International University

The evidence for the effectiveness and impact of disaster risk reduction (DRR) interventions concerns humanitarian organizations, donor agencies and communities alike. The cost benefit analysis (CBA) of DRR projects can demonstrate the attractiveness of these interventions by enumerating the benefits of lower costs to donor agencies and reduced damages to beneficiaries. This paper presents an approach to analyze community-based DRR interventions through the findings of CBA within the larger context of project management. The study aims to assist leaders and practitioners in the humanitarian community in analyzing a DRR project within its institutional and community setting with a focus on capacity building of personnel for evidence based decision-making. We conducted an ex ante CBA of several DRR interventions to improve climate change resilience in Caribbean communities by estimating the costs incurred by the society and the benefits accrued to the community in general. The current accepted rate of the central bank of each country was selected as the discount rate for analysis. The effect of varying discount rates (3–10%) on the benefit cost ratio (BCR) was tested. All the DRR interventions analyzed, including safer shelters, water and health micro-projects, obtained a BCR above 1, justifying the implementation of all the interventions. For interventions other than safer shelters, the BCR ranged from 2.6 to 215. The study revealed several areas for policy consideration within humanitarian organizations: the need to comply with standards of a rigorous CBA, capacity building in the areas of CBA and project management, need for timely integration of economic analysis of DRR projects within the project cycle, importance of data collection and record keeping, building informational sources, and use of more simplified related forms of CBA such as cost effectiveness and least cost analysis.

D.8: Assessing Benefits in Consumer Protection Regulation (Marvin 302)

Chair: Lisa A. Robinson (Lisa.A.Robinson@comcast.net), Harvard Center for Risk Analysis

Panelists to Include:

1.     Howell Jackson (hjackson@law.harvard.edu), Harvard Law School

2.     Paul Rothstein (paul.rothstein@cfpb.gov), Consumer Financial Protection Bureau

3.     Michael Livermore (Livermore@exchange.law.nyu.edu), University of Virginia

4.     Art Fraas (fraas@rff.org), Resources for the Future

5.     Dick Morgenstern (morgenst@rff.org), Resources for the Future

In the United States, consumer protection regulation has not traditionally been the focus of extensive benefit-cost analysis.  But recent decisions of the Federal Court of Appeals for the District of Columbia, new statutory mandates under the Dodd-Frank Act of 2010, and a flurry of law review articles have brought the topic to the fore.  Drawing on a new survey of seventy-two recent consumer protection rulemakings, this panel will review current approaches to benefit analysis in consumer finance and analogous areas of consumer protection.  The presentation will offer an overview of how federal agencies are currently analyzing and quantifying the benefits of consumer protection regulation, including statistics about the incidence and intensity of benefit analysis across the survey sample and selected subsamples.  The presentation will also include a more in-depth discussion of eighteen “exemplars” of benefit analysis across nine different benefit types.  Together these exemplars could be said to represent the current state-of-the-art benefit analysis in consumer protection regulation.  Finally, we hope to propose a handful of specific future research projects that could improve the quality of benefit analysis in consumer protection regulation and explore institutional arrangement to improve the quality of benefit analysis in federal agencies.

E.8: The Effectiveness of Policies Involving Health Warning Labels and Signage: Cigarettes, e-Cigarettes, and Alcohol (Marvin 310)

Chair: Trudy Ann Cameron (cameron@uoregon.edu), University of Oregon

Discussant: Joseph Cordes (cordes@gwu.edu), The George Washington University


1.     Will New Warning Labels Encourage or Discourage the Use of Electronic Cigarettes? Evidence from Experimental Markets, Don Kenkel* (dsk10@cornell.edu), Cornell University

Electronic cigarettes are battery-powered devices that deliver various concentrations of nicotine in an aerosol mist. The use of e-cigarettes is rising rapidly and is expected to reach $3 billion in US sales in 2015. Emerging research suggests that e-cigarettes are much less harmful than regular cigarettes and hold promise as smoking cessation devices, although they are not approved as medical devices. E-cigarettes are also not currently required to carry warning labels. However, the 2009 Family Smoking Prevention and Tobacco Control Act gave the FDA regulatory authority over tobacco products. The FDA recently announced that it intends to use that authority to require e-cigarettes to carry a warning label that they contain nicotine which is derived from tobacco and is an addictive chemical. Also recently, some e-cigarette manufacturers have voluntarily adopted somewhat stronger warning labels. Under another regulatory scenario, e-cigarette manufacturers could apply to market their products as a modified risk tobacco product that could carry a more favorable label. We conduct discrete choice experiments to estimate the impact of alternative warning labels on adult smokers’ use of e-cigarettes. The data are from on-line surveys that present smokers with choices between their current tobacco product, an e-cigarette, and a smoking cessation product. The e-cigarette warning label is randomly assigned to vary across different choice scenarios.  The econometric discrete choice model yields estimates of the impact of the different warning labels on the probability smokers choose e-cigarettes. In addition, the model yields an estimate of smokers’ willingness to pay for the different warning labels. 

2.     What are the Potential Benefits of Graphic Warning Labels? Richard M. Peck* (rmpeck@uic.edu) and John A. Tauras, University of Illinois at Chicago

Huang et al. (2013) estimate that in the United States, graphic warning labels (GWL) would result in 5.3 to 8.6 million less smokers in 2013, reducing smoking prevalence by 12.6 percent to 20.4 percent.   Using the framework of Murphy and Topel (2006), we find that the annual gross benefits of GWL arising from lower levels of premature death would be about $32 billion to $52 billion dollars annually. This is higher than the FDA estimates of 225 to 675 million dollars annually – their approach is similar but there are important methodological differences (page 36723 of Federal Register, V. 76, No 120, 2011). Their estimates of the efficacy of GWL are also much lower.  If we use recently proposed adjustments for loss of consumer surplus, then the net benefits of GWL are $10 to $16 billion annually, for premature death alone. An adjustment is also made to take into account the fact that on average current smokers have lower than average income so that the average value of a statistical life that is used to calibrate the parameters of the model is lower (4.2 million dollars).  Initial prevalence rates vary by state, gender and age and the population numbers, from the Census Bureau, are for 2011.  The model is linear, so the consequences of adjusting wages, or the calibrating value of statistical life are straightforward to compute.

3.     The Effects of Posted Point-of-Sale Warnings on Alcohol Consumption During Pregnancy and on Birth Outcomes, Gulcan Cil* (gcil@uoregon.edu), University of Oregon

In 23 states and in the District of Columbia, all alcohol retailers are required by law to post signs that warn against the risks of drinking during pregnancy. This study examines the effects of these point-of-sale alcohol warning signs (AWS) on alcohol consumption during pregnancy and on birth outcomes using the variation in the timing of AWS legislation across states. The contributions of this study are two-fold. First, this research is an investigation of the effectiveness of AWS as a policy tool in reducing prenatal alcohol use. Moreover, using a quasi-experimental setting that results from a plausibly exogenous policy change that affects all pregnant women, this study aims to establish a causal link between prenatal alcohol exposure and birth outcomes. Using the National Vital Statistics data and a specification that accounts for state and year fixed effects, state-specific time trends, and individual- and state-level covariates, I find a statistically significant reduction in the odds of drinking during pregnancy in response to AWS laws. This finding is supported by the results obtained using the Behavioral Risk Factor Surveillance System data in a model that compares the change in alcohol consumption before and after AWS laws among pregnant women with that of non-pregnant women. In the reduced form regressions for birth outcomes, I find that AWS laws are associated with decreases in the odds of very low birth weight and very pre-term births.