Peer Review for the Report,
"How Consumers Value Fuel Economy:
A Literature Review"
United States
Environmental Protection
Agency
-------
Peer Review for the Report,
"How Consumers Value Fuel Economy:
A Literature Review"
Assessment and Standards Division
Office of Transportation and Air Quality
U.S. Environmental Protection Agency
SER&
United States
Environmental Protection
Agency
EPA-420-S-10-001
April 2010
-------
March 3 0,2010
MEMORANDUM
SUBJECT: Peer Review for David Greene paper, "How Consumers Value Fuel Economy: A
Literature Review"
FROM: Gloria Helfand, Assessment and Standards Division
Office of Transportation and Air Quality, U.S. Environmental Protection Agency
In December 2009, EPA contracted with RTI International (RTI) to conduct a peer review of a
literature survey conducted by David Greene of Oak Ridge National laboratory. The draft study,
titled "How Consumers Value Fuel Economy: A Literature Review," looked at 22 papers that, in
some fashion, provided quantitative estimates of the role of fuel economy in consumer vehicle
purchase decisions.
The three peer reviewers selected by RTI were Drs. Carolyn Fischer of Resources for the Future,
Christopher Knittel of the University of California at Davis, and Walter McManus of the
University of Michigan. EPA would like to extend its appreciation to all three reviewers for their
efforts in evaluating this survey. The three reviewers brought useful and distinctive views in
response to the charge questions.
The first section of this document contains the final RTI report summarizing the peer review of
David Greene's literature survey, including the detailed comments of each peer reviewer and an
overview of the most significant comments compiled by RTI. The RTI report also contains the
peer reviewers' resumes, the charge letter, and cover letters from each reviewer explaining any
real or perceived conflicts of interest. The second major section contains our responses to the
peer reviewers' comments. In this section, we repeat the summarized comments provided by
RTI and, after each section of comments, provide our response. We have retained the
organization reflected in RTFs summary of the comments to aid the reader in moving from the
RTI report to our responses.
CONTENTS
I. Peer Review of EPA's Study of Vehicle Choice Models for Estimating Impacts of Fuel
Economy, conducted by RTI International
1. Background
2. Description of Review Process
3. Summary of Review Comments
4. References
Appendices
A. Resumes of Selected Reviewers
B. Charge Questions
C. Cover Letters
D. Reviews
II. EPA's Response to Peer Review Comments
-------
Memorandum
INTERNATIONAL
TO: Kent Helmer, Gloria Helfand, U.S. Environmental Protection Agency, Office of
Transportation and Air Quality (OTAQ)
FROM: Paramita Sinha, RTI
DATE: February 26, 2010
SUBJECT: Peer Review of EPA's Study of Vehicle Choice Models for Estimating Impacts of
Fuel Economy
1. Background
The U.S. Environmental Protection Agency's (EPA's) Office of Transportation and Air Quality
(OTAQ) is involved in exploring the regulation of CO2 and other greenhouse gas (GHG)
emission control measures in motor vehicles and equipment. An important component of the
costs of such regulations is the cost of improved technology, while the benefits include, among
others, the value of reduced fuel consumption. Understanding the role of fuel economy in
consumer vehicle purchases will contribute to EPA's assessment of the impacts of its rules on
vehicle sales.
EPA is assessing the use of consumer vehicle choice models to analyze the impacts of regulatory
programs that affect fuel economy. Coefficients from discrete choice models of consumer
vehicle choice are typically used to estimate the value of improving fuel economy. Other studies
use hedonic price methods. These vehicle choice models could help EPA anticipate market
responses to vehicle regulations and thus judge the effectiveness of regulations.
EPA is analyzing a set of existing consumer vehicle choice models to see whether there is a
robust estimate of the value of additional fuel economy to consumers. EPA is seeking a peer
review of its comparative study of consumer vehicle choice models and what each model may
predict for the value of improving vehicle fuel economy at the time of vehicle purchase. This
report contains documentation of the peer review process for the vehicle choice study.
Section 2 of this memorandum provides a description of the process for choosing reviewers,
administering the review process, and closing the peer review. Section 3 contains an overview of
the comments made by the panel of reviewers. The appendices to this memorandum provide the
resumes of the reviewers; charge letter to the peer reviewers, which describes their task and what was
RTI International is a trade name of Research Triangle Institute
-------
Kent Helmer, Gloria Helfand
February 26, 2010
Page 2
requested from them in terms of deliverables; and the cover letters and review comments from the
panel.
2. Description of Review Process
In December 2009, EPA's OTAQ contacted RTI International to facilitate the peer review of
EPA's study of vehicle choice models for estimating impacts of fuel economy regulations. This
study, authored by David L. Greene, is entitled "How Consumers Value Fuel Economy: A
Literature Review."
EPA provided a short list of subject matter experts from academia and industry (Appendix A of
the performance work statement) to RTI, and this served as a "starting point" from which we
assembled the list of peer reviewers. RTI selected three independent (as defined in Sections 1.2.6
and 1.2.7 of EPA's Peer Review Handbook) subject matter experts to conduct the requested
reviews. Subject matter experts familiar with economic valuation, discrete choice models and the
use of these models for valuation, and the use of these models for predicting automobile
purchases were selected.
To ensure that the review process was conducted in a timely manner, RTI contacted potential
reviewers within a week of submitting the work plan and determined that each reviewer would
be able to perform work during the period of performance. To make the review process as
credible as possible, RTI did not consult the Agency in the final determination of reviewers. RTI
obtained the resumes of the selected reviewers, and these are included in Appendix A.
RTI provided the panel reviewers with the final edited version of the analysis that each subject
matter expert was expected to review along with a set of charge questions (both provided by
EPA) in the first week of January.l The memo from RTI to the reviewers with the charge
questions is included in Appendix B of this report.
A teleconference between EPA, the reviewers, and RTI was organized to provide an opportunity
to the panel to discuss any questions or concerns they may have regarding the material provided
and expected deliverables. Completed reviews from the panel were sent to EPA by the requested
date. These reviews included the response to the charge questions and any additional comments
the reviewer may have had (e.g., margin notes on review materials). From each reviewer, RTI
1 Upon request from one of the reviewers, EPA provided RTI with two examples of previous peer reviews, and these were
subsequently forwarded to them. These are entitled "Peer Review for the RTI Report, Automobile Industry Retail Price
Equivalent and Indirect Cost Multipliers (Assessment and Standards Division, Office of Transportation and Air Quality, U.S.
Environmental Protection Agency, June 2009, EPA-420-R-09-004 ) and "Review: Vehicle Greenhouse Gas Emission
(VGHG) Emissions Cost and Compliance Model (Jonathan Rubin, January 2010).
-------
Kent Helmer, Gloria Helfand
February 26, 2010
Page3
obtained a cover letter stating the reviewer's name; the name and address of their organization if
applicable; a list of review documents/media received by the reviewer and which were actually
reviewed; and a statement of any real or perceived conflict(s) of interest. The cover letters and
reviews are included in Appendices C and D, respectively.
3. Summary of Review Comments
Carolyn Fischer (Resources for the Future), Christopher Knittel (Department of Economics,
University of California at Davis), and Walter McManus (University of Michigan,
Transportation Research Institute) reviewed "How Consumers Value Fuel Economy: A
Literature Review." This section provides a summary of the comments received from them.
3.1 OVERALL APPROACH AND METHODOLOGY OF THE STUDY
The reviewers are supportive of the overall approach and methodology.
Walter: "The overall approach and methodology are within the mainstream of critical review
practice in economics, and the execution in unexceptionable." He does comment on the fact that
"Greene (2009) adopts a conventional 'literature review' approach" and "both the conventional
approach to reviewing economic research and the conventional approach to economic research
itself do not give much help to decision makers who are understandably made uncomfortable by
the widely differing expert opinions about parameters that are crucial in predicting the impacts of
regulations and in choosing between alternatives. An overall critique of conventional practice is
beyond the scope of this review. However, I do have some suggestions for improvements in the
methodology for comparing studies and for dealing with the lack of consensus in expert
opinions. (The References section below lists some papers by Edward E. Learner, a professor at
UCLA, in which a critique of conventional practice is developed along with alternative practices
that would enhance the usefulness of economic research to policy makers.)"
Chris comments that the study does not discuss "why consumers undervalue gasoline, relative
society (if they do indeed undervalue fuel economy). This has important implications for which
policies are most effective at aligning discount rates." However, he recognizes that papers
dealing with such issues are lacking.
-------
Kent Helmer, Gloria Helfand
February 26, 2010
Page 4
3.2 APPROPRIATENESS OF THE DATASETS AND ANY OTHER
INPUTS
One reviewer commented that the choice of the papers was comprehensive. Another reviewer
recommended including some explanation of how the studies were selected for inclusion in
Greene (2009).
3.3 DATA ANALYSIS CONDUCTED FOR THE STUDY
Two reviewers comment on the fact that the type of analysis done for the different papers varies
greatly.
Chris: "It strikes me that there are two ways to conduct a review such as this. One is to report the
results of papers with little, or no, discussion of the quality each paper. The other is to both
review and critique each paper. I found a lack of uniformity in this regard. For many of the
papers, the latter strategy is taken. The report discusses key weakness of these papers. Still many
papers are simply summarized, with little or no discussion of their strengths or weaknesses. This
gives the impression that the results in these papers are more accurate. This may actually be the
case, but if it is, this should be stated more explicitly.
For example, the first paper (Alcott and Wozny) is simply summarized with no editorializing.
One possibility is that there are no weaknesses in this paper. But, many of the issues raised later
for other papers also relate to this paper. For example, the report correctly states, when
discussing the Sawhill paper, that in order to get a consistent estimate of how consumers value
fuel economy, the author is required to correctly specify how consumers form expectations about
gas prices and drive their vehicles. If any of these are incorrectly specified, the estimate will be
biased. The same is true for the Alcott and Wozny paper. While I find Alcott and Wozny's
specification for how expectations are formed to likely be more accurate, ironically, they assume
consumers are hyper-rational in the sense that they form their gas price expectations based on
NYMEX futures prices. In some ways the myopia result is inconsistent with this key
assumption."
Walter: "It is difficult for the reader to verify that the same standards are being used to assess
each paper, especially since the amount of discussion by paper varies greatly. This leads the
reader to have less confidence in the reviewer's opinions of the individual studies (that this or
that one had plausible or implausible assumption, methodologies, or results)."
-------
Kent Helmer, Gloria Helfand
February 26, 2010
Page 5
3.4 APPROPRIATENESS OF THE CONCLUSIONS DRAWN
There was general consensus among the reviewers on the major conclusion of the document: that
the literature remains mixed as to whether consumers undervalue, overvalue, or approximately
value fuel economy.
One reviewer raised doubts that differences in assumptions, opinions, and methods have been
sufficiently examined to support even the weak proposition that "there do not appear to be clear
associations among methods or data sources and the resulting inferences."
One reviewer expressed doubts about what Greene means by "incorrect models" of consumer
decision making about fuel economy but suggested efficient markets hypothesis and economic
rationality to be a possible interpretation of this phrase.
3.5 RECOMMENDATIONS FOR ALTERNATE DATA AND/OR
ANALYSES
3.5.1 Additional Data
Two of the reviewers suggested several other studies that are relevant for this work. A list of
these studies is provided below.
Dreyfus, Mark K. and W. Kip Viscusi (1995). "Rates of Time Preference and Consumer
Valuations of Automobile Safety and Fuel Efficiency." Journal of Law and Economics
38: 79-98.
Hughes, Jonathan E., Christopher R. Knittel and Daniel Sperling (2008). "Evidence of a Shift in
the Short-Run Price Elasticity of Gasoline Demand." Energy Journal.
Kilian, Lutz and Eric Sims (2006). The Effects of Real Gasoline Prices on Automobile Demand:
A Structural Analysis Using Micro Data, Working Paper, University of Michigan (April).
Klier and Linn (2010 working paper) "The Price of Gasoline and New Vehicle Fuel Economy:
Evidence from Monthly Sales Data"
Klier, Thomas, and Joshua Linn (2008). "The Price of Gasoline and the Demand for Fuel
Efficiency: Evidence from Monthly New Vehicles Sales Data." Working Paper,
University of Illinois at Chicago (September).
Sallee and West (2009 working paper)
-------
Kent Helmer, Gloria Helfand
February 26, 2010
Page 6
Sallee, James, and Sarah West (2008). "Testing for Consumer Myopia: The Effect of Gasoline
Price Changes on the Demand for Fuel Economy in Used Vehicles." Working Paper,
Macalester College (December).
West, S. (2008). "The Effect of Gasoline Prices on the Demand for Sport Utility Vehicles."
Working Paper. Macalester College.
3.5.2 Additional Analysis
The reviewers suggested several additional analyses to strengthen the study.
3.5.ii a. The author should take advantage of the summary section to indicate whether
some approaches reviewed in the study are likely to be better than others.
3.5.ii b. The analysis to explain the difference in results could be improved.
Carolyn: "First, the discussion interweaves reasons that could explain why we can plausibly
believe any of a range of results (e.g., heterogeneous decision rules, uncertainty) with reasons for
why different studies get different results. It would be helpful to separate these discussions,
reminding readers of the former, and then focusing on the latter, which is the most important
contribution. This kind of analysis should help indicate best practices and directions for future
research. Can we suggest some experiments (e.g., new techniques on old data sets, or old
techniques on new data sets) that can help tease out what differences arise from alternative
econometric approaches? If most studies have focused on other aspects of vehicle demand, are
there better specifications for looking directly at the valuation of fuel economy?"
A summary of key differences in approaches and data and an analysis of their consequences
would be useful. These would include
levels of aggregation;
time horizons (how useful is information from the 1970s in estimating current
demand?);
incorporation of the supply side;
details of other attributes and variable definitions;
allowing for separate responses to MPG and fuel costs (this was raised as lacking in
the discussion of Gramlich, but it was not apparent which other studies had done
this);
assumptions of vehicle lifetime, VMT, discount rates, etc.; and
econometric approach.
-------
Kent Helmer, Gloria Helfand
February 26, 2010
Page 7
Although Table 10 provides some of these summary indicators (model type, data, and time),
additional analysis such as the following would be useful: "Do we observe any trends across
classes or time? Within a class of models, what drives the differences?"
Walter: "It would be useful to expand Table 10 (or to create a set of tables) to show other
differences that could be important sources of differences in estimates. The items that could be
included would show differences in:
1. Maintained assumptions reported (assumptions that are maintained across any alternative
specifications or sensitivities),
2. Reported sensitivity analyses (is one done? which assumptions are varied? how widely?),
3. Econometric challenges and approaches used to address them (Greene (2009) appears to
discuss these comprehensively in the text.), and
4. Out-of-sample predictive performance (are predictions made? about what? are the
predictions compared to actual outcomes? how accurate are the predictions?).
Adding these additional side-by-side comparisons would increase the reader's confidence that
the same standards are being used to form the reviewer's opinions about the relative usefulness
of each paper to inform decision making by EPA.
My opinion: The additional elements are responsible for more of the variation in the estimates of
consumer value for fuel economy than are the elements already included in Table 10."
He also suggested a measure of researcher/specification uncertainty and disagreement to
quantitatively assess whether we have made progress toward a consensus. He also provides
illustrative suggestions on ways to use and interpret this measure. "The idea is to compare the
range of parameter estimates to the parameter value predicted by mainstream theory."
The metric is called Multiple Extreme Estimates Test (MEET) and is defined as the ratio of the
value predicted by theory to the difference between the two extreme estimates.
^ lvalue Predicted by Theoryl V
MEET = ! - =
[Largest Estimate - Smallest Estimate] R
limMEET = 0 and \imMEET = oo
fl-><» fl->0
"Numerically, MEET measures the predicted value as a share of the range of expert opinion. The
range of expert opinion, R, is a crude and partial measure of the uncertainty or fragility of the
estimates due to differences in assumptions, specification, and opinions between researchers. The
metric MEET views this fragility compared to the theoretical value of the parameter, V."
-------
Kent Helmer, Gloria Helfand
February 26, 2010
PageS
3.5.ii.c. Though the issue of consumer heterogeneity has been mentioned in several
places, this may be important area for additional analysis, both to explain differences and
to suggest future research directions.
3.6 OVERALL CLARITY OF THE PRESENTATION
The reviewers agreed that the overall clarity of the document was high.
Reviewers had several suggestions about organizing the paper.
i. One reviewer suggested that since the articles are organized logically by class of data and
methods, it would be helpful to review those methods at the beginning of each section
"including the advantages and drawbacks, as well as key factors or assumptions that can
affect the results."
ii. It has also been suggested that it would be helpful to begin with the seminal papers in
each category, "so as to understand the evolution and improvements made subsequently."
iii. Two reviewers agreed that the publication status of papers should be recognized and one
reviewer suggested that the review should "focus first on published papers and then
second on unpublished papers."
iv. Chris Knittel: "Many of the papers reviewed did not focus on measuring the implicit
discount rate, or even on the issue of how gas prices affect vehicle demand. If a paper did
not focus on either of these two questions, it is difficult to gauge the robustness of their
results with respect to these questions, the quality of the variation in gas prices in the
data, etc. Perhaps a better method would be to first focus on those papers where gas
prices and vehicle choice are the central research question, and those papers where this is
more a tangential part of the analysis."
On a related note, another reviewer commented on studies that do not explicitly estimate the
value of fuel economy: "Since few econometric studies have explicitly estimated the value of
fuel economy, a significant contribution of the paper is to translate a wide variety of results into
more consistent indicators. Yet, they could still be more consistent. In Table 10, can all of the
results be converted into the same WTP metric, preferably using the same assumptions about
vehicle lifespan, VMT, discounting, etc.?"
Another comment on the presentation of the paper was that "the author's baseline assumptions
need to be stated clearly early in the paper, so careful attention can be given throughout to
assumptions that deviate from that baseline. Indeed, the statement made in the Allcott and
Wozny analysis ('Calculating the discounted present value of fuel costs requires a number of
assumptions...') should be made earlier in a general context, so that the variety and range of
assumptions can be understood and explored."
-------
Kent Helmer, Gloria Helfand
February 26, 2010
Page 9
3.7 OTHER CLARIFICATION COMMENTS
One reviewer provided the following comments to clarify or rephrase certain statements.
1. The authors should "use the phrase 'at society's discounted expected value...' (top of
page 5) instead of 'does the market value fuel economy improvements at the discounted
expected value of future fuel savings over the lifetime of the vehicle, or less, or more?'"
2. The statement (on page 5) that "On the other hand, if consumers are myopic and consider
only the first three years of fuel savings, for example, fuel economy standards can
increase welfare even based solely on private costs and benefits" should be made clearer.
3. One of the reviewers comment on the statement on page 5 that "it is surprising that there
is no basic research on how consumers consider fuel economy." "This is a strange use of
the term basic research, which I take to mean research on pure science. Does the author
equate basic science with interviews? I have never heard basic science used in this
manner."
Two of the reviewers also provided several detailed comments on specific paper reviews, Section
2 of the document, and editorial comments and corrections to typographical errors.
4. References
Rubin, J. January 2010. "Review: Vehicle Greenhouse Gas Emission (VGHG) Emissions Cost
and Compliance Model."
U.S. Environmental Protection Agency, Office of Transportation and Air Quality, Assessment
and Standards Division. June 2009. "Peer Review for the RTI Report, Automobile
Industry Retail Price Equivalent and Indirect Cost Multipliers. EPA-420-R-09-004.
-------
Appendix A:
Resumes of Selected Reviewers
-------
CAROLYN FISCHER
Resources for the Future Phone: (202) 328-5012
1616 P St., NW Fax: (202) 939-3460
Washington, D.C. 20036 E-mail: fischer@rff.org
EDUCATION
University of Michigan, Ann Arbor, MI. 1991-97.
Ph.D. in Economics with fields in public finance and natural resource economics.
M.A. in Economics, 1993.
University of Pennsylvania, Philadelphia, PA. 1986-90.
B.A. in International Relations with honors (major Valedictorian) and Economics,
minor in French, magna cum laude.
PROFESSIONAL EXPERIENCE AND AFFILIATIONS
Resources For the Future, Washington, DC. 1997-present.
Senior Fellow. Indefinite appointment (tenure), 2004.
Fellow, CESifo Research Network, Energy and Climate Economics Area, 2009-present.
Board of Directors, Association of Environmental and Resource Economists, 2008-2010.
Editorial Board, Resource and Energy Economics, 2006-2009.
Leadership Committee for the Energy Collaborative Analysis Initiative, NREL, 2007-present.
Fellow, Center for Advanced Study at the Norwegian Academy of Science and Letters, with the
group "Environmental Economics: Policy Instruments, Technology Development, and
International Cooperation," 2005.
Johns Hopkins University, 1999. Instructor, Natural Resource and Environmental Economics.
Council of Economic Advisors, Washington, DC. 1994-95. Staff economist.
The WEFA Group, Philadelphia, PA. 1990-91. Economic analyst in World Service division.
SELECTED PUBLICATIONS
Current Work
Fischer, C. and T.P. Lyon (2009) "Competing Environmental Labels." RFF Working Paper.
Fischer, C. and C. Coleman (2009) "Imperfect Competition, Consumer Behavior, and the
Provision of Fuel Efficiency in Light-Duty Vehicles"
Fischer, C. and A. K. Fox (2009) "Comparing Policies to Combat Emissions Leakage: Border
Tax Adjustments versus Rebates" RFF DP 09-02. (under review at JEEM)
Fischer, C. and A. K. Fox (2009) "Combining Rebates with Carbon Taxes: Optimal Strategies
for Coping with Emissions Leakage and Tax Interactions" (under review at JPubE)
Fischer, C. and R. Morgenstern (2008) "Metrics for Evaluating Policy Commitments in a
Fragmented World: The Challenges of Equity and Integrity" Discussion Paper 08-17,
Harvard Project on International Climate Agreements.
Fischer, C. and T. Sterner (2008) "Climate Policy, Prudence, and the Role of Technological
Innovation." (under review at JEEM)
-------
Journal Articles
Fischer, C. (forthcoming) "Does trade help or hinder the conservation of natural resources?"
Review of Environmental Economics and Policy.
Fischer, C. (forthcoming) "When Do Renewable Portfolio Standards Lower Electricity Prices?"
The Energy Journal 30 (4).
Fischer, C., R. Morgenstern, and E. Herrnstadt (forthcoming) "Understanding Bias in EIA
Forecasts of Energy Demand." The Energy Journal.
Fischer, C. and R.G. Newell (2008) "Environmental and Technology Policies for Climate
Mitigation" Journal of Environmental Economics and Management. 55 (2): 142-162.
Fischer, C. (2008) "Emissions Pricing, Spillovers, and Public Investment in Environmentally
Friendly Technologies," Energy Economics. 30 (2): 487-502.
Fischer, C. and A.K. Fox (2007) "Output-Based Allocation of Emissions Permits for Mitigating
Tax and Trade Interactions," Land Economics. 83: 575-599.
Fischer, C., W. Harrington, and I.W.H. Parry (2007) "Do Market Failures Justify Tightening
Corporate Average Fuel Economy (CAFE) Standards?" The Energy Journal 2% (4): 1-30.
Bernard, A.L., C. Fischer, and A.K. Fox (2007) "Is There a Rationale for Output-Based Rebating
of Environmental Levies?" Resource & Energy Economics 29 (2): 83-101.
Fischer, C. (2005) "On the Importance of the Supply Side in Demand-Side Management,"
Energy Economics, 27 (1): 165-180.
Margolis, M., J.F. Shogren and Carolyn Fischer (2005) "How trade politics affect invasive
species control," Ecological Economics, 52 (3): 305-313.
Fischer, C. and R. Laxaminarayan (2005) "Sequential Development and Exploitation of an
Exhaustible Resource: Do Monopoly Rights Promote Conservation?" Journal of
Environmental Economics and Management. 49 (3): 500-515.
Fischer, C. and R. Laxaminarayan (2004) "Monopoly Extraction of an Exhaustible Resource
with Two Markets," Canadian Journal of Economics, 37 (1): 178-188.
Fischer, C. (2004) "The Complex Interactions of Markets for Endangered Species Products,"
Journal of Environmental Economics and Management, 48 (2): 926-953.
Fischer, C., I.W.H. Parry and W.A. Pizer (2003) "Instrument Choice for Environmental
Protection when Technological Innovation is Endogenous," Journal of Environmental
Economics and Management 45(3): 523-545.
Fischer, C. (2001) "Read This Paper Later: Procrastination with Time-Consistent Preferences."
Journal of Economic Behavior and Organization, (46)3 pp. 249-269.
Recent Policy Briefs:
Fischer, C. (2009) "Are Consumers or Fuel Economy Policies Efficient?" in Sperling, Daniel and
James S. Cannon, Eds., Reducing Climate Impacts in the Transportation Sector. Springer.
Fischer, C. and R.G. Newell (2008) "What's the Best Way to Promote Green Power? Don't
Forget the Emissions Price." Resources, Summer (169): 10-13.
Fischer, C. and C. Egenhofer. (2008). "The Critical Role of Technology for International Climate
Change Policy," in Beyond Bali: Strategic Issues for the international climate change
negotiations, Brussels, Belgium: CEPS; and Washington, DC: Brookings Institution.
-------
CHRISTOPHER ROLAND KNITTEL
University of California, Davis
Department of Economics
One Shields Ave
Davis, CA 95616
Office: 530.302.1032
Fax: 530.752.9382
crknittel@ucdavis. edu
http://www. econ. ucdavis. edu/faculty/knittel
CURRENT APPOINTMENTS:
2006-present, Associate Professor of Economics, University of California, Davis
2008-present, Chancellor's Fellow, University of California, Davis
2007-present, Research Associate, National Bureau of Economic Research. Groups:
Environmental Economics and Energy, Industrial Organization, and Productivity
2003-present, Visiting Research Fellow, University of California Energy Institute
2005-present, Faculty Affiliate, Institute of Transportation Studies, UC Davis
2006-present, Strategy and Policy Thread Leader for STEPS
2006-present, Associate Editor, The Journal of Industrial Economics
2007-present, Associate Editor, American Economic Journal - Economic Policy
2007-present, Associate Editor, The Journal of Energy Markets
2008-present, Member, Economic and Allocation Advisory Committee for AB32's cap-and-
trade program, State of California
PREVIOUS APPOINTMENTS:
2002-2006, Assistant Professor of Economics, University of California, Davis
2004-2007, Faculty Research Fellow, National Bureau of Economic Research. Groups:
Environmental Economics and Energy, Industrial Organization, and Productivity
1999-2002, Assistant Professor of Finance and Economics, School of Management, Boston
University
1996-1999, Research Assistant, University of California Energy Institute
1994-1996, Teaching Assistant, University of California, Davis
EDUCATION:
Ph.D., University of California, Berkeley, 1999 (Economics)
M.A., University of California, Davis, 1996 (Economics)
B.A., California State University, Stanislaus, summa cum laude, 1994 (Economics and Political
Science)
PUBLICATIONS:
Knittel, Christopher R. and Jason J. Lepore. "Tacit Collusion in the Presence of Cyclical
Demand and Endogenous Capacity Levels." Forthcoming in The International Journal of
Industrial Organization.
Version: 11/1/09
-------
CHRISTOPHER R. KNITTEL
Stewart, Scott, John J. Neumann, Christopher R. Knittel, and Jeffrey Heisler. "Absence of
Value: An Analysis of Investment Allocation Decisions by Institutional Plan Sponsors,"
Financial Analyst Journal, 65(6), November/December 2009.
Knittel, Christopher R. and Victor Stango. "How Does Incompatibility Affect Prices?:
Evidence from ATMs," The Journal of Industrial Economics, LVII(3), September 2009, pp.
557-582.
Holland, Stephen P., Jonathan E. Hughes and Christopher R. Knittel. "Greenhouse Gas
Reductions under Low Carbon Fuel Standards?," The American Economic Journal -
Economic Policy, 1(1), February 2009, pp. 106-146.
Borenstein, Severin, James Bushnell, Christopher R. Knittel and Catherine Wolfram. "Trading
Inefficiencies in California's Electricity Markets," The Journal of Industrial Economics,
LVI(2), June 2008, pp. 347-378.
Feenstra, Robert and Christopher R. Knittel. "Re-Assessing the Quality Adjustment to
Computer Prices: Do U.S. Procedures Overstate the Gains?," forthcoming Price Index
Concepts and Measurement, NBER and the Chicago Press.
Knittel, Christopher R. and Konstantinos Metaxoglou. "Diagnosing Unilateral Market Power
in Electricity Reserves Market," The Journal of Energy Markets, 1(1), Spring 2008.
Knittel, Christopher R. and Victor Stango. "Incompatibility, Product Attributes and Consumer
Welfare: Evidence from ATMs," The BE Journal of Economic Analysis & Policy, Advances,
8(1), January 2008. Available at: http://www.bepress.com/bejeap/vol8/issl/artl.
Hughes, Jonathan E., Christopher R. Knittel and Daniel Sperling. "Evidence of a Shift in the
Short-Run Price Elasticity of Gasoline." The Energy Journal, 29(1), January 2008.
Heisler, Jeffrey, Christopher R. Knittel, John J. Neumann and Scott Stewart. "Why Do
Institutional Plan Sponsors Hire and Fire their Investment Managers?" Best Paper Award for
the 31st NBEA Conference. The Journal of Business and Economics Studies, 13(1), Spring
2007, pp. 88-116.
Kim, Dae-Wook and Christopher R. Knittel "Biases in Static Oligopoly Models? Evidence
from the California Electricity Market," The Journal of Industrial Economics, LIV(4),
December 2006, pp. 451-470.
Knittel, Christopher R. "The Adoption of State Electricity Regulation: The Role of Interest
Groups," The Journal of Industrial Economics, LIV(2), June 2006.
Knittel, Christopher R. and Michael R. Roberts. "Financial Models of Deregulated Electricity
Prices: An Application to the California Market," Energy Economics, 27(5), September 2005,
pp. 791-817.
Knittel, Christopher R. "Regulatory Restructuring and Incumbent Price Dynamics: The Case
of Local Telephone Markets," Review of Economics and Statistics, 86(2), May 2004, pp. 614-
625.
Knittel, Christopher R. and Victor Stango. "Price Ceilings as Focal Points for Tacit Collusion:
Evidence from the Credit Card Market," The American Economic Review, 93(5), December
2003, pp. 1703-1729.
Knittel, Christopher R. "Market Structure and the Pricing of Electricity and Natural Gas," The
Journal of Industrial Economics, LI(2), June 2003, pp. 167-191.
Knittel, Christopher R. "Alternative Regulatory Methods and Firm Efficiency: Stochastic
Frontier Evidence the US Electricity Industry," Review of Economics and Statistics, 84(3),
August 2002, pp. 530-540.
-------
CHRISTOPHER R. KNITTEL
Borenstein, Severin, James Bushnell, and Christopher R. Knittel. "Market Power in Electricity
Markets: Beyond Concentration Measures," The Energy Journal, 20(4), October 1999, pp.
65-88.
Knittel, Christopher R. "Long Distance Rates: Search Costs, Switching Costs, and Market
Power," Review of Industrial Organization, 12(4), August 1997, pp. 519-536.
WORKING PAPERS:
"Automobiles on Steroids: Product Attribute Trade-offs and Technological Progress in the
Automobile Sector" Revisions requested from The American Economic Review.
Knittel, Christopher R. and Victor Stango. "Strategic Incompatibility in ATMs." Revisions
requested from The Journal of Banking and Finance.
"The Implied Cost of Carbon Dioxide under the Cash for Clunkers Program" Revisions
requested from The BE Journal of Economic Analysis & Policy,.
Busse, Meghan, Christopher R. Knittel and Florian Zettelmeyer. "Pain at the Pump: How
Gasoline Prices Affect Automobile Purchasing." In submission.
Knittel, Christopher R. and Konstantinos Metaxoglou. "Estimation of Random Coefficient
Demand Models: Challenges, Difficulties and Warnings." In submission.
Fowlie, Meredith, Christopher R. Knittel and Catherine Wolfram. "Sacred Cars: Optimal
Regulation of Stationary and Non-stationary Pollution Sources." In submission.
Knittel, Christopher R. and Victor Stango. "The Productivity Benefits of IT Outsourcing."
Knittel, Christopher R., Douglas Miller, and Nick Sanders. "Caution, Drivers! Children
Present. Traffic, Pollution and Infant Health"
Huckfeldt, Peter, and Christopher R. Knittel. "Patents, Pharmaceutical Use and Pharmaceutical
Prices"
AWARDS, HONORS, AND GRANTS:
Chancellor's Fellowship, UC Davis (one of five faculty members), 2008
Barry D. McNutt Award for Excellence in Automotive Policy Analysis (with Jonathan
Hughes and Dan Sperling), 2008
National Science Foundation Grant (with Victor Stango), 2008-2010, $240,000
Chevron Bio-Fuel Research Grant, 2007-2008, $127,000
Chevron Bio-Fuel Research Grant, 2007-2008, $77,000
Chevron Bio-Fuel Research Grant (Co-Pi), 2007-2009, $370,000
Woods Institute for the Environment Leadership Scholar Training, 2007
Distinguished Paper, 2006 Academy of Finance
University of California Energy Institute Research Grant, 2005-2006, $50,000
Best Paper Award for the 31 st NBEA Conference
ASUCD Excellence in Teaching Award, 2004
University of California Energy Institute Research Grant, 2003
Faculty Research Grant, UC Davis, 2002, 2003, 20004, 2005, 2006
Institute of Governmental Affairs Junior Faculty Grant, 2002, 2003, 2004, 2005
Junior Faculty Research Grant, Boston University, 2001
Graduate Fellowship, University of California, Berkeley, 1997-1999
Graduate Fellowship, University of California, Davis, 1994-1996
Institute of Transportation Fellow, University of California, Davis, 1995-1996
Student Commencement Speaker, California State University, Stanislaus, 1994
-------
CHRISTOPHER R. KNITTEL
REFEREE SERVICES:
Agricultural Economics, American Economic Review, Bulletin of Economic Research, Census
Bureau, Econometrica, Economic Inquiry, The Economic Journal, Economics Letters, Energy
Economics, The Energy Journal, Energy Studies Review, European Economic Review,
International Journal of Industrial Organization, International Journal of Power and Energy
Systems, Journal of Banking and Finance, The Journal of Business, Journal of Business and
Economic Statistics, Journal of Economic Behavior and Organization, Journal of Economic
Education, Journal of Economics and Management Strategy, Journal of Futures Markets, Journal
of Industrial Economics, Journal Institutional and Theoretical Economics, Journal of Law and
Economics, Politics and Economics, Quarterly Journal of Economics, Rand Journal of
Economics, Resource and Energy Economics, Review of Economic Studies, Review of Economics
and Statistics, Review of Industrial Organization, Review of Network Economics, Southern
Economic Journal, Socio-Economic Planning Sciences, Utilities Policy, University of California
Energy Institute Grant Program, NSF Grant Program
RECENT INVITED PRESENTATIONS:
Energy Institute @ Haas, Policy Conference, Sacramento, October 2009
Institute of Transportation Studies, UC Davis, October 2009
TREE Seminar, Raleigh/Durham/Chapel Hill, North Carolina, October 2009
UC Berkeley, ARE, February 2009
NBER Winter IO Meeting, February 2009
Iowa State, Economics, January 2009
NBER Summer EEE Meeting, July 2009
UC Berkeley lO/Innovation Seminar, November 2008
UC Berkeley lO/Innovation Seminar, October 2008
Department of Justice, March 2008
University Retirement Community, February 2008
New American Foundation, February 2008
Tainjin Chinese Delegation at UC Davis
UCEI Policy Conference, December 2007
University of California Energy Institute, July 2007
Institute of Transportation Studies, UC Davis, October 2007
NBER EEE, Summer Meeting, July 2007
NBER Winter IO Meetings, January 2008
UC Berkeley, Department of Economics, November 2007
University of Alberta and Calgary University Industrial Organization Conference, October
2007
Washington University, Olin School of Business, November 2007
REGULATORY FILINGS:
Arons, S.M., A.R. Brandt, M.A. Delucchi, A. Eggert, A.E. Farrell, B.K. Haya, J. Hughes,
B.M. Jenkins, A.D. Jones, D.M. Kammen, S.R. Kaffka, C.R. Knittel, D.M. Lemoine, E.W.
Martin, M.W. Melaina, J.M. Ogden, R.J. Plevin, D. Sperling, B.T. Turner, R.B. Williams, C.
Yang, 2007. "A Low-Carbon Fuel Standard for California, Part 1: Technical Analysis."
Available Online: http://www.lcfs.ucdavis.edu.
Brandt, A.R., A.E. Farrell, B.K. Haya, J. Hughes, B.M. Jenkins, A.D. Jones, D.M. Kammen,
C.R. Knittel, M.W. Melaina, M. O'Hare, R.J. Plevin, D. Sperling, 2007. "A Low-Carbon Fuel
-------
CHRISTOPHER R. KNITTEL
Standard for California, Part 2: Policy Analysis." Available Online:
http://www.lcfs.ucdavis.edu.
Peer Review Comments on AB 1493, California Environmental Protection Agency Air
Resource Board, September 2004.
"Comments on the Use of Computer Models for Merger Analysis in the Electricity Industry," (Joint
with Severin Borenstein and James Bushnell), Federal Energy Regulatory Commission. Docket No.
PL98-6-000, June 1998.
"A Cournot-Nash Equilibrium Analysis of the New Jersey Electricity Market," December
1997. (Joint with Severin Borenstein and James Bushnell). Filed with the New Jersey Public
Utility Commission as testimony on the potential for market power in a deregulated
Pennsylvania-Jersey-Maryland Power Pool.
CONSULTING:
Customers First! Coalition, Energy Information Agency, Korean Electric Power Company,
California Air Resource Board, City of West Sacramento
PH.D. COMMITTEES (FIRST JOB):
UC Davis:
Anson Soderbery (on-going)
Nick Sanders (chair, on-going)
Chai-Wen Chen (chair, on-going)
Jonathan Hughes (chair, University of Colorado, Boulder)
Adib Bagh (University of Kentucky, Math and Economics)
Seungjoon Lee (Korean Insurance Research Institute)
Jason Lepore (chair, Cal Poly)
Wei-Min Hu (Peking University)
Byeongil Ahn (Gyeongsang University)
Konstantinos Metaxoglou (chair, Bates and White LLC)
Lan Li (University of Melbourne)
Neil Norman (Cornerstone Research)
Dae-Wook Kim (chair, Korean Institute for Industrial Economics and Trade)
Boston University:
Gustavo Genoni (2002, Finance, IAE, School of Business, Universidad Austral)
John Neumann (2003, Finance, St. John's University)
TEACHING:
UC Davis
o Graduate Empirical Industrial Organization (6 times)
Ratings: Mean 4.9 (out of 5)
o Transportation Economics (4 times)
Ratings: Mean 4.7
o Intermediate Microeconomics (1 time),
Ratings: Mean 4.8
o Undergraduate Industrial Organization (9 times)
Ratings: Mean 4.8
Boston University
o Modeling Business Decision Making,
Spring 2000, Spring 2001 and Spring 2002
-------
CHRISTOPHER R. KNITTEL
Ratings: 4.53 (out of 5), 4.77, 4.70
o Modeling Business Decision Making (honors),
Spring 2001 and Spring 2002
Ratings: 4.88, 4.70
UNIVERSITY SERVICE:
UC Davis:
2007-2008, Co-writer (with Jean Vandergehst) of a proposal for a Graduate Program in "Energy
Science and Technology" and "Energy Policy and Management"
2006-Present, Member, Energy Institute Steering Committee
2008, Founding Faculty Member, UC Davis Energy Institute
2005-2006, Hiring Committee and Interviewing Committee
2004-2005, Hiring Committee and Interviewing Committee
2002-2003, Hiring Committee and Interviewing Committee
2002-2007, Graduate Advisor
Oral committees: Dae-Wook Kim, Konstantinos Metaxoglou, Neil Norman (chair), Seungjoon
Lee, Wei-Min Hu, Lan Li (ARE), Sunhwa Lee, Byeongil Ahn (ARE), Michele Amaral, David
Ong, Adib Bagh, Jason Lepore, Bei Li, Chenguang Li (ARE), Tina Saitone (ARE), Carlo Russo
(ARE), Sandhya Patlolla (ARE), Jon Hughes (TTP), Peter Huckfeldt, Kyungwon Rho, Nick
Sanders, Chia-Wen Chen, Joeri de Witt (ARE), In-Sung Lee (TTP), Anson Soderbery, Nils
Johnson (TTP), David McCollum (TTP)
Boston University:
2000-2001, Finance Hiring Committee and Interviewing Committee
1999-2000, Finance Hiring Committee
RECENT MEDIA CITATIONS:
Print: Alameda Times-Star, Arizona Daily Star, Argus, ATMmarketplace.com, Austin-American
Statesman, Boston Globe, Buffalo News, California Aggie, Contra Costa Times, PE.com,
bankrate.com, marketwatch.com, Grain's Business Report (New York), Credit Card Magazine,
Kiosk Marketplace News, LA Observed, LA Times, International Herald Tribune, Northwestern
Herald, Oakland Tribune, Oregonian, Philadelphia Inquirer, Providence Journal, New York
Times, Sacramento Bee, St. Petersburg Times, Salon.com, San Diego Union Tribune, Salt Lake
Tribune, San Diego Union Tribune, SF Chronicle, San Mateo County Times, Santa Rosa Press
Democrat, Sarasota Herald-Tribune, Scripps News (DC), Tuscaloosa News Sun Herald, Quad
City News (Iowa), Winston-Salem Journal, Worcester Telegram
Radio: KQED's "Forum", KXJZ, KFBK, KUOP, KCBS, KNX, WHYY with Marty Moss-
Coane, WPR with Kathleen Dunn, Bloomberg Radio, Lambasted by Rush Limbaugh
Television: KCRA-3, CBS-13 Sacramento, NBC Nightly News, ABC World News, CBS
Evening News, ABC Good Morning America
REFERENCES:
Severin Borenstein
E.T. Grether Professor of Business Administration and Public Policy
Haas School of Business
University of California at Berkeley
borenste@haas.berkelev.edu
-------
CHRISTOPHER R. KNITTEL
Michael Greenstone
3M Professor of Environmental Economics
Department of Economics
Massachusetts Institute of Technology
mgreenst@mit.edu
Frank Wolak
Professor of Economics
Stanford University
wolak@zia. Stanford, edu
Catherine Wolfram
Associate Professor
Haas School of Business
University of California at Berkeley
wolfram@haas.berkelev.edu
-------
Resume
Dr. Walter McManus
Research Scientist and Director
Economics, Energy, and Environment Research Group
University of Michigan Transportation Research Institute
2901 Baxter Rd.
Ann Arbor, MI
(248) 821-0493
watsmcm@umich. edu
Biography
Dr. Walter McManus is a Research Scientist at the University of Michigan Transportation Research
Institute (UMTRI) and Director of the Economics, Energy, and Environment Research Group. Dr.
McManus has been a member of the UMTRI faculty since March 2005. Immediately prior to that, he was
Executive Director of Forecasting and Analytics at the global marketing information company, J.D. Power
and Associates. His business experience also includes nine years with General Motors in forecasting,
marketing analysis and strategy, and new-product development. (He also spent a year as a production
supervisor in a GM manufacturing plant.)
Dr. McManus graduated from Louisiana State University (BA 1977) and earned a doctorate in economics
from the University of California, Los Angeles (PhD 1983). Dr. McManus pursues a research program that
is focused on issues arising from the interaction of transportation, society, and the environment. The
research program generates knowledge through excellent, creative research on the social, economic, and
environmental dynamics that are producing change in the automotive transportation energy sphere; with the
goal of developing and delivering useful applications of findings to assist policy makers, industry, and
stakeholders in anticipating future societal needs.
Research Interests
Transportation economics, energy, and environment; the automotive industry; adoption and diffusion of
clean transportation technologies
Tools
Economic analysis (market demand and supply models, strategic behavior of firms, economic history),
econometrics, forecasting and simulation, finance, public speaking
Education
PhD, Economics, University of California, Los Angeles, 1983
BA, Economics, Louisiana State University, 1977
Awards
2008 UMTRI Research Excellence Award for the article in Business Economics 2007
NABE Abramson Award for the best article published in Business Economics 2007
GM Chairman's Honors for innovations enhancing performance in new-product development 1991 & 98
Sidney Stern Fellow; University of California, Los Angeles; 1979 - 82
Affiliations
Automotive Industry Expert Panel, U.S. Government Accountability Office, 2009 - present
Ceres Stakeholder Committee on Sustainability, Ford Motor Company, 2009 - present
Fellow, Michigan Memorial Phoenix Energy Research Institute, 2007 - present
Executive Committee, Michigan Center for Advancing Safe Transportation throughout the Lifespan, 2007
- present
Transportation Energy Committee, Transportation Research Board
Transportation Working Group, Energy Futures Coalition, 2003 - 04
American Economic Association
National Association for Business Economics
Society of Automotive Engineers
-------
Walter McManus
Professional History
Research Scientist and Director, Economics, Energy, and Environment Research Group, University of
Michigan Transportation Research Institute, Mar 2005 - Present
Visiting Scholar and Research Engineer, Transportation Sustainability Research Center, Institute of
Transportation Studies, University of California, Berkeley, Mar 2009 - Oct 2009
Executive Director of Forecasting and Analytics, J.D. Power and Associates, Oct 1999 - Jan 2005
Director of Marketing, Textron Automotive Company, Dec 1998 - Sept 1999
Leader, Industry Analysis Group, General Motors Corporation, July 1996 - Nov 1998
Manager, North American Market Analysis, General Motors Corporation, Jan 1994 - June 1996
Economist, Delco Remy Division, General Motors Corporation, Anderson, IN, Aug 1991 - Dec 1993
(Memo: included development assignment as Manufacturing Supervisor, Jan 1993 - Dec 1993)
Economist, General Motors Corporation, Detroit, MI, June 1989 - Aug 1991
Associate Professor of Economics and Fellow, Center for the Study of Business and Government, Baruch
College, New York, NY, July 1988- May 1989
Assistant Professor of Economics, University of Florida, Gainesville, FL, July 1983 - June 1988
Testimony and Briefings
U.S. EPA and NHTSA Public Hearing, Proposed Rulemaking to Establish Light-Duty Vehicle Greenhouse
Gas Emission Standards and Corporate Average Fuel Economy Standards, October 21, 2009
Investor Briefing, Citigroup Investment Research, CAFE Panel Conference Call & Briefing, April 2009.
Testimony, Committee on Assessment of Technologies for Improving Light-Duty Vehicle Fuel Economy,
National Research Council, March 16, 2009
Testimony, US EPA, Hearing on California Greenhouse Gas Waiver, March 3, 2009
Testimony, Environmental Regulation Commission Hearing, Greenhouse Gas Emissions Reduction-Florida
Clean Car Emission Rule, Florida Department of Environmental Protection, October 29, 2008
US Congressional Briefing, Environmental and Energy Study Institute & Investor Network on Climate
Risk, December 4, 2007
Investor Briefing, Citigroup Investment Research, CAFE and the U.S. Auto Industry: A Growing Auto
Investor Issue, 2012-2020, October 31, 2007
Public Briefing, National Commission on Energy Policy and the International Council on Clean
Transportation, Fuel Economy: Technology Trends and Policy Options, Washington, DC. October 1, 2007.
Congressional Testimony, U.S. Senate Finance Subcommittee on Energy, "Advanced Technology
Vehicles: The Road Ahead", May 1, 2007
Congressional Testimony, U.S. Senate Energy and Natural Resources Committee, "the Consumer Market
for Fuel Economy", January 30, 2007
Publications
-------
Walter McManus
Rogozhin, A., Gallaher, M, Helfand, G., and McManus, W. (2010 forthcoming), Using indirect cost
multipliers to estimate the total cost of adding new technology in the automobile industry, International
Journal of Production Economics.
McManus, W., Senter, R., Curtin, R., and Garver, S. (2009), The demographic threat to Detroit's
automakers, Targeting, Measurement and Analysis for Marketing 17:81-92
Senter, R. and McManus, W. (2009), General Motors in an age of corporate restructuring, in the second
automobile revolution: the automobile firms' trajectories at the beginning of the 21st century (Chapter 9),
Edited by M. Freyssenet, New York: Palgrave Macmillan
McManus, W. (2007), The link between gasoline prices and vehicle sales: economic theory trumps
conventional Detroit wisdom. Business Economics 1.42:54-60
McManus, W. and Griffor, E. (2006), Toward a science of driving: Safety in rules-based versus adaptive
self-regulating traffic systems, SAE Convergence, 2006-21-0064
McManus, W. (1985), Estimates of the deterrent effect of capital punishment: the importance of the
researcher's prior beliefs, Journal of Political Economy 93:417-25
McManus, W. (1985), Labor market assimilation of immigrants: the importance of language skills,
Contemporary Economic Policy 3:77-89
McManus, W. (1985), Labor market costs of language disparity: an interpretation of Hispanic earnings
differences, American Economic Review 75:818-27
Theil, H., Rosalsky, M., and McManus, W. (1985), Lp-norm estimation of non-linear systems
Economics Letters 17(1-2):123-125
McManus, W. and Rosalsky, M. (1985), Are all asymptotic standard errors awful? Economics Letters
17(3):243-245
McManus, W., Gould, W., and Welch, F. (1983), Earnings of Hispanic men: the role of English language
proficiency, Journal of Labor Economics 1:101-30
Technical Reports and Working Papers
McManus, W. and Senter, R., Market Models for Predicting PHEV Adoption and Diffusion, UMTRI-2009-
37, August 2009.
McManus, W. and Kleinbaum, R., Fixing Detroit, How Far, How Fast, How Fuel-Efficient, UMTRI-2009-
26, June 2009
Senter, R. and McManus, W., Reshaping the Big Three, GERPISA, June 2009
Rogozhin, A., Gallaher, M., Helfand, G., and McManus, W., Automobile Industry Retail Price Equivalent
and Indirect Cost Multipliers, EPA-420-R-09-003, Feb 2009
McManus, W., The Impact of Attribute-Based Corporate Average Fuel Economy (CAFE) Standards:
Preliminary Findings, Automotive Analysis Division, University of Michigan Transportation Research
Institute (UMTRI), July 2007
McManus, W., Economic Analysis of Feebates to Reduce Greenhouse Gas Emissions from Light Vehicles
for California, Automotive Analysis Division, University of Michigan Transportation Research Institute
(UMTRI), May 2007
-------
Walter McManus
McManus, W., Can Proactive Fuel Economy Strategies Help Automakers Mitigate Fuel-Price Risks?
Automotive Analysis Division, University of Michigan Transportation Research Institute (UMTRI),
September 2006
McManus, W., Baum, A., Hwang, R., Luria, D., and Baura, G., In The Tank - How Oil Prices Threaten
Automakers' Profits and Jobs, Office for the Study of Automotive Transportation, July 2005
McManus, W. and Berman, B., The 2005 OSAT - HybridCars.com Survey of Owners and Shoppers,
Office for the Study of Automotive Transportation (OSAT), University of Michigan Transportation
Research Institute (UMTRI), 2005.
McManus, W., The Effects of Higher Gasoline Prices on U.S. Light Vehicle Sales, Prices, and Variable
Profit by Segment and Manufacturer Group, 2001 and 2004. Office for the Study of Automotive
Transportation (OSAT), University of Michigan Transportation Research Institute (UMTRI), June 2005.
Greene, D., Duleep, K., and McManus, W., Future Potential of Hybrid and Diesel Powertrains in the US
Light-Duty Vehicle Market, Report to Department of Energy, July 2004.
McManus, W., Consumer Acceptance of Alternative Powertrains, OE Industry Review. Troy, MI: Original
Equipment Suppliers Association, 2004.
McManus, W., "Diesel vs. Hybrid-Electric Powertrains: Assessing Dependability," Power Report (July
2004)
McManus, W., "Interest in Diesel GrowsQuietly," Power Report (June 2004)
McManus, W., Consumer Acceptance of Alternative Powertrains Study. Westlake Village, CA: J.D. Power
and Associates, 2004.
Malesh, T. and McManus, W., Clean Diesel Market Acceptance Study. Westlake Village, CA: J.D. Power
and Associates, 2003.
McManus, W., Analysis of Tax Credits to Stimulate Consumer Demand for Advanced-Technology Fuel-
Efficient Vehicles: Final Report to Energy Future Coalition Transportation Working Group. Westlake
Village, CA: J.D. Power and Associates, 2003.
McManus, W., Generation Y Automotive Market Assessment. Westlake Village, CA: J.D. Power and
Associates, 2002.
McManus, W., Interaction Between New and Used Vehicle Sales in the U.S. Market. Westlake Village,
CA: J.D. Power and Associates, 2002.
McManus, W., Telematics Forecast. Westlake Village, CA: J.D. Power and Associates, 2002.
McManus, W., Satellite Radio Forecast. Westlake Village, CA: J.D. Power and Associates, 2002.
McManus, W. and Bussmann, W., Isuzu in the U.S., Westlake Village, CA: J.D. Power and Associates,
2001
McManus, W., Adaptive Cruise Control Forecast, Westlake Village, CA: J.D. Power and Associates, 2001
-------
Appendix B:
Charge Questions
-------
RTI
3040 Cornwallis Road « PO Box 12194 * Research Triangle Park, NC 27709-2194 « USA
Telephone 919.541.6000 » Fax 91 9.541.5985 « www.rti.org
TO:
Carolyn Fischer
Walter McManus
Christopher Knittel
FROM:
CC:
Paramita Sinha
Michael Gallaher
DATE: December 31, 2009
SUBJECT: Charge Questions for Peer Review of EPA's Study of Vehicle Choice Models
The US EPA's Office of Transportation and Air Quality is currently analyzing a set of existing
consumer vehicle choice models to see whether there is a robust estimate of the value of
additional fuel economy to consumers. EPA is seeking a peer review of its comparative study of
consumer vehicle choice models and what each model may predict for the value of improving
vehicle fuel economy at the time of vehicle purchase.
EPA has provided direction and charge questions for this review and these are included below. A
teleconference call will also be arranged so that EPA can provide technical and/or background
information on the analysis. Any future questions you may have can be directed back through
RTI for resolution with EPA.
The review will involve a written report that includes the response to the charge questions and
any additional comments you may have, e.g., margin notes on review materials. Comments
should be provided in an enclosure to a cover letter that clearly states the reviewer's name, the
name and address of their organization if applicable, which model review documents/media were
received by the reviewer and which were actually reviewed and a statement of any real or
perceived conflict(s) of interest.
ELEMENTS TO BE ADDRESSED IN THE CHARGE TO THE
REVIEWERS OF EPA'S ANALYSIS OF THE VALUE OF ADDITIONAL
FUEL ECONOMY TO CONSUMERS
EPA's report on using vehicle choice models to estimate the impacts of fuel economy regulations
will inform its analyses of the effects of regulatory programs that affect vehicle fuel economy.
This report details an analysis of the value of additional fuel economy to consumers, estimated
RTI International is a trade name of Research Triangle Institute
-------
Carolyn Fischer
Walter McManus
Christopher Knittel
Page 2
December 31,2009
from consumer vehicle choice models. No independent data analysis will be required for this
review.
Specifically, EPA is seeking the reviewers' expert opinions on the data, concepts, and
methodologies upon which the analysis relies, whether or not the analysis was conducted
correctly, and whether the analysis draws appropriate conclusions. Toward this end, we ask that
each subject matter expert review and comment on the following items:
1. overall approach and methodology of the study;
2. appropriateness of the datasets and any other inputs;
3. data analysis conducted for the study;
4. appropriateness of the conclusions drawn;
5. recommendations for alternate data and/or analyses; and
6. overall clarity of the presentation.
In making their comments, the reviewers should distinguish between recommendations for
clearly defined improvements that can be readily made based on data or literature reasonably
available to EPA, and improvements that are more exploratory or dependent on information not
readily available to EPA. Any comment should be sufficiently clear and detailed to allow a
thorough understanding by EPA or other parties familiar with the model. EPA requests that the
reviewers not release the peer review materials or their comments to anyone else until the
Agency makes its report and supporting documentation public.
If a reviewer has questions about what is required in order to complete this review or need
additional background material, please direct the reviewer to contact (RTI project manager/Pi). If
a reviewer has any questions about the EPA peer review process itself, please have the reviewer
contact Ms. Ruth Schenk in EPA's Quality Office, National Vehicle and Fuel Emissions
Laboratory by phone (734-214-4017) or through e-mail (schenk.ruth@epa.gov.
-------
Appendix C:
Cover Letters
-------
To:
Gloria Helfand
US EPA, Assessment and Standards Division (OTAQ)
2000 Traverwood Drive
Ann Arbor, Michigan 48105
From:
Dr. Carolyn Fischer
3215 Pauline Dr.
Chevy Chase, MD 20815
January 27, 2010
Cover Letter to Accompany "Review of Greene 'How Consumers Value Fuel Economy'"
Greetings:
The documents that I received from EPA (or RTI International) were a memo containing the
charge questions and the draft report by Greene (2009). I also furnished versions of some
unpublished papers relevant to the report, including Alcott and Wozny (2009); Busse, Knittel,
and Zettelmeyer (2009); and Sallee, West, and Fan (2009).
I reviewed all of the documents that I received in developing my expert opinion as contained in
the "Review of Greene 'How Consumers Value Fuel Economy'," submitted on January 25.
I declare that there are no real or perceived conflicts of interest concerning my involvement in
this review for the EPA.
Best regards,
Carolyn Fischer
C-l
-------
To:
Gloria Helfand
US EPA, Assessment and Standards Division (OTAQ)
2000 Traverwood Drive
Ann Arbor, Michigan 48105
From:
Christopher R. Knittel
Department of Economics
University of California, Davis
Davis, CA 95616
January 27, 2010
Cover Letter to Accompany Review How Consumers Value Fuel Economy: A Literature
Review, by David Greene.
Greetings:
The documents that I received from EPA (or RTI International) were a memo containing the
charge questions, the draft report by Greene (2009), and two unpublished papers that are covered
in the draft report. The unpublished papers are Alcott and Wozny (2009) and Sallee, West, and
Fan (2009).
I reviewed all of the documents that I received in developing my expert opinion as contained in
the enclosed Review of How Consumers Value Fuel Economy: A Literature Review, by David
Greene.
I declare that there are no real or perceived conflicts of interest concerning my involvement in
this review for the EPA.
Sincerely,
Christopher R. Knittel
C-2
-------
To:
Gloria Helfand
US EPA, Assessment and Standards Division (OTAQ)
2000 Traverwood Drive
Ann Arbor, Michigan 48105
From:
Walter McManus
University of Michigan Transportation Research Institute
2901 Baxter Rd.
Ann Arbor, MI 48109-2150
January 25, 2010
Cover Letter to Accompany Review of Greene, D.L. (2009) How Consumers Value Fuel
Economy: A Literature Review
Greetings:
The documents that I received from EPA (or RTI International) were a memo containing the
charge questions, the draft report by Greene (2009), and three unpublished papers that are
covered in the draft report. The unpublished papers are Alcott and Wozny (2009); Busse, Knittel,
and Zettelmeyer (2009); and Sallee, West, and Fan (2009).
I reviewed all of the documents that I received in developing my expert opinion as contained in
the enclosed Review of Greene, D.L. (2009) How Consumers Value Fuel Economy: A Literature
Review.
I declare that there are no real or perceived conflicts of interest concerning my involvement in
this review for the EPA.
Best regards,
Walter McManus
C-2
-------
Appendix D:
Reviews
-------
Review of Greene
"How Consumers Value Fuel Economy"
by Carolyn Fischer
David Greene is contributing an in-depth review of the recent literature on the demand for
vehicle fuel economy that will be extremely valuable to researchers and policymakers alike.
Although ultimately he is unable answer the key questionhow do consumers value fuel
economy?he does provide considerable insight into how researchers have attempted to answer
this question and why we see so much variation in the results.
This review is structured to provide feedback and suggestions aimed at improving this work and
ultimately enhancing its impact.
Presentation
Overall, the writing is excellent. Some modest changes could help the reader.
The reviewed articles are organized logically by class of data and methods. At the beginning of
each section, it would be helpful to review those methods in general fashion, including the
advantages and drawbacks, as well as key factors or assumptions that can affect the results. Also,
it might be easier to begin with the seminal papers in each category, so as to understand the
evolution and improvements made subsequently.
The status of the papers should also be recognized, since many of the important new studies are
still unpublished working papers and subject to change.
Since few econometric studies have explicitly estimated the value of fuel economy, a significant
contribution of the paper is to translate a wide variety of results into more consistent indicators.
Yet, they could still be more consistent. In Table 10, can all of the results be converted into the
same WTP metric, preferably using the same assumptions about vehicle lifespan, VMT,
discounting, etc.?
Also, the author's baseline assumptions need to be stated clearly early in the paper, so careful
attention can be given throughout to assumptions that deviate from that baseline. Indeed, the
statement made in the Allcott and Wozny analysis ("Calculating the discounted present value of
fuel costs requires a number of assumptions...") should be made earlier in a general context, so
that the variety and range of assumptions can be understood and explored.
Additional Analysis
The paper reviews many approaches, and the author should take advantage of the summary
section to indicate whether some approaches are likely to be better than others.
That section does point out some factors that may explain such different results, but this analysis
could be improved. First, the discussion interweaves reasons that could explain why we can
plausibly believe any of a range of results (e.g., heterogeneous decision rules, uncertainty) with
D-l
-------
reasons for why different studies get different results. It would be helpful to separate these
discussions, reminding readers of the former, and then focusing on the latter, which is the most
important contribution. This kind of analysis should help indicate best practices and directions
for future research. Can we suggest some experiments (e.g., new techniques on old data sets, or
old techniques on new data sets) that can help tease out what differences arise from alternative
econometric approaches? If most studies have focused on other aspects of vehicle demand, are
there better specifications for looking directly at the valuation of fuel economy?
It would be useful to summarize key differences in approaches and data and analyze their
consequences. Some that were gleaned along the way:
levels of aggregation,
time horizons (how useful is info from the 1970s in estimating current demand?),
incorporation of the supply side,
details of other attributes and variable definitions,
allowing for separate responses to MPG and fuel costs (this was raised as lacking in the
discussion of Gramlich, but it wasn't apparent what other studies had done this),
assumptions of vehicle lifetime, VMT, discount rates, etc.
econometric approach
Some of these summary indicators (model type, data, and time) are given in Table 10, but some
additional analysis would be useful. Do we observe any trends across classes or time? Within a
class of models, what drives the differences?
The paper raises the issue of consumer heterogeneity in several places, but the importance of this
fact should be pushed further. Fischer (2009) points out that, facing consumers with
heterogeneous preferences over fuel economy, manufacturers with market power have distorted
incentives for providing fuel economy. More generally, failure to capture such consumer
heterogeneity can lead to significant errors in predicting the distribution of effort in complying
with regulation, as well as the calculation and distribution of the benefits.2 Portions of this
review highlight certain studies that captured differences in tastes for fuel economy (BLP,
Goldberg, Brownstone et al.), while most studies ignored such heterogeneity. This seems like an
important area for additional analysis, both to explain differences and to suggest future research
directions.
Additional Empirical Literature
I am aware of a few other studies that likely deserve attention as well.
Sallee and West (2009 working paper) "The Effect of Gasoline Prices on the Demand for Fuel
Economy in Used Vehicles: Empirical Evidence and Policy Implications"
Abstract: Fossil fuel consumption and greenhouse gas emissions from the personal transportation
sector pose serious challenges to today's policy-makers. If consumers acknowledge the full value
of fuel economy, a tax on gasoline could be as efficient as a tax on greenhouse gas emissions,
since the amount of carbon released by a gallon of gasoline is independent of the manner in
which it is combusted. If, however, consumers do not sufficiently value fuel economy when
2 Fischer, Carolyn (2009) "Imperfect Competition, Consumer Behavior, and the Provision of. Fuel Efficiency in Vehicles."
Working Paper, Washington, DC: Resources for the Future.
D-2
-------
making the decision about what car to buy, an additional complementary policy, such as a fuel
economy standard or a feebate, can be welfare improving. We use microdata on used vehicle
prices and a unique identification strategy based on micro-level variation in vehicle odometers to
test whether used car prices change by the amount predicted by a fully rational asset pricing
model. Our baseline results indicate that used car prices adjust by about 80% of the amount
predicted by theory, and sensitivity analysis suggests that for a reasonable discount rate, they
adjust by the full amount. These results contrast with recent literature, which finds adjustment
closer to 25%, and suggest a limited role for complementary policy tools.
Klier and Linn (2010 working paper) "The Price of Gasoline and New Vehicle Fuel Economy:
Evidence from Monthly Sales Data"
This paper uses a unique data set of monthly new vehicle sales by detailed model from 1978-
2007, and implements a new identification strategy to estimate the effect of the price of gasoline
on individual vehicle model sales. We control for unobserved vehicle and consumer
characteristics by using within model-year changes in the price of gasoline and sales. We find a
significant sales response, suggesting that the gasoline price increase from 2002-2007 explains
nearly half of the decline in market share of U.S. manufacturers. On the other hand, an increase
in the gasoline tax would only modestly raise average fuel economy.
(Their estimate implies that a one dollar increase in the price of gasoline raises average fuel
economy by 0.5-1 MPG.)
Dreyfus, Mark K. and W. Kip Viscusi (1995). "Rates of Time Preference and Consumer
Valuations of Automobile Safety and Fuel Efficiency." Journal of Law and Economics 38: 79-
98.
This seems like an earlier, influential study, of the same era as BLP, so worth some comparison,
although fuel economy demand was a secondary target of interest.
Description from Howarth (2004): "Dreyfus and Viscusi (1995) undertook a hedonic price
analysis of the U.S. automobile market to assess consumers' willingness to pay for improved
safety and energy efficiency. On the assumption that consumers use a common discount rate in
evaluating both safety and fuel economy, the study calculates an implicit discount rate that
ranges from 11 to 17 percent in alternative specifications. More tellingly, Dreyfus and Viscusi
conclude that only 35 percent of the present-value cost savings provided by improved energy
efficiency is capitalized in the purchase price of vehicles." (derived from coefficients from
operating cost and that interacted with weight)
Hughes, Jonathan E., Christopher R. Knittel and Daniel Sperling (2008). "Evidence of a
Shift in the Short-Run Price Elasticity of Gasoline Demand." Energy Journal.
West, S. (2008). "The Effect of Gasoline Prices on the Demand for Sport Utility Vehicles."
Working Paper. Macalester College.
(mentioned in Helfand and Wolverton)
D-3
-------
Other Points
2.0 in the discussion of Turrentine and Kurani, recent research on "The MPG Illusion" may
be relevant to mention, supporting the difficulty of consumers in calculating fuel cost
savings.
Larrick, R. P., & Soil, J. B. (2008). The MPG illusion. Science, 320, 1593-1594.
http://faculty.fuqua.duke.edu/~larrick/bio/Reshighlights.htm
2.1 The author talks about the global vehicle market; it seems relevant to give the statistics
for the U.S. market.
2.2 The review of the literature on the demand for energy efficiency should include
Gillinham et al., who "review economic concepts underlying consumer decision making
in energy efficiency and conservation and examine related empirical literature. In
particular, we provide an economic perspective on the range of market barriers, market
failures, and behavioral failures that have been cited in the energy efficiency context."
Gillingham, K.T., R.G. Newell and K.L. Palmer (2009). "Energy Efficiency Economics
and Policy." RFF Discussion Paper 09-13, April 2009.
The statement "With the exception of externalities, there is little quantitative evidence of
the impact of these failures on consumers' choices of energy using durable goods" seems
hard to reconcile with the preceding discussion. Is there any better (peer-reviewed)
support than ACEEE (2007)?
This section also offers the opportunity to introduce other studies outside the economics
literatureare there any documented industry studies on their approach to consumer
payback expectations?
Section 3
BLP (1995) is a seminal paper, and should probably be presented first. Is the same critical eye
being given to them with respect to omitted variables. They used HP/weight to represent power,
but what if weight is valued separately, such as for safety concerns in that era?
Dasgupta et al. (2007): this description seems too curt. I had a hard time following the structure
of the model and the conversion of the results. (Also later in the discussion of McManus, where
the coefficient units are interpreted as discounted lifetime miles).
Bento et al. (2005): I believe the parameter estimates are provided in the Appendix. Are they in a
form that can be useful for comparison?
Feng et al. (2005): given the important question raised of the units, it seems worth asking the
authors directly.
Brownstone et al. (1996): did any of the identification for WTP for fuel economy come from the
demand for alternative fuel options? It wasn't clear in the discussion what was done with the
D-4
-------
information about alternatives, and whether that is reliable, given the lack of availability of these
technologies.
Cambridge Econometrics: if they included both fuel costs and fuel consumption rates, was there
some other heterogeneity to distinguish between the two?
Espey and Nair (2005) is referred to before it is discussed; perhaps the order should be reversed.
Also, since the gas guzzler tax is mentioned, one should perhaps point out that they
mischaracterized the policy in such a way as to affect their consumer price calculations (it is
actually paid by producers, not consumers), although this is unlikely to affect their results. The
lack of controlling for manufacturer may, however.
In discussing the differences with the Rosen method, the implications for how their results might
be biased were unclear.
Bhat and Sen (2006): Table 6 is difficult to interpret.
Busse et al.: p. 32 "These results do no necessarily imply that new car buyers undervalue fuel
economy, since the change in price also depends on the elasticities of supply and demand." It
seems they also depend on the formulation of price expectations.
Minutiae
p. 8 "Two recent analysis have quantified the [potential] impacts of uncertainty..."
p. 10 2nd para: market imperfection]
sales[-]mix (might be preferred)
p. 12 end of the Gramlich discussion is a bit confusing; it was not clear what his model
predicted for fuel economy in comparison to the EPA numbers. May just need to be clarified
which numbers are being discussed.
p. 12 [reverse order so that citation always begins the paragraph] BLP (1995), in a seminal
paper,...
"consumer I" should be "consumer /'"
p. 16 last full sentence ("Different influences... in econometric analyses of VMT, for
example") could use some citations.
p. 23 top para: "But this is not stated in the report[;] indeed,...
p. 25 figure says "Qin and Rubin" instead of "Fan and Rubin"
References please check working papers to update citations
p. 36 2nd para: remove the "," after "Allcott and Wozny"
e.g., Fischer et al. (2007) is Energy Journal28: 1-29.
D-5
-------
Review of:
"How Consumers Value Fuel Economy: A Literature Review"
by David L. Greene
Reviewed by Christopher R. Knittel
January 22, 2010
Dear Sir/Madam:
This report summarizes the literature regarding the value that consumers place on fuel economy
at the time of purchasing a vehicle. Understanding this is important for a variety of public policy
issues. How consumers trade-off paying more for fuel economy with future fuel savings defines
their discount rate. Therefore, statements such as "consumers undervalue fuel economy"
necessarily imply that the consumers' discount rate is below society's discount rate. If consumers
do not value fuel economy as much as society-at-large, then gasoline will be over-consumed and
policies may improve social welfare.
Equally important, though not analyzed in this report (it is discussed however), is why consumers
undervalue gasoline, relative society (if they do indeed undervalue fuel economy). This has
important implications for which policies are most effective at aligning discount rates. For
example, do consumers undervalue fuel economy because capital markets are imperfect?
Subsidized loans would likely be the most effective policy. Do they undervalue fuel economy
because of a lack of cognitive ability? Education policies and/or information campaigns are
likely to be most effective. Papers dealing with these issues are largely lacking, unfortunately.
Bottom line: This is an excellent review. This review will be an asset to policy makers. I agree
with the conclusion that the literature remains mixed as to whether consumers full value fuel
economy at reasonable discount rates. I provide some comments that may improve the review.
Is the goal to summarize, or analyze papers?:
It strikes me that there are two ways to conduct a review such as this. One is to report the results
of papers with little, or no, discussion of the quality each paper. The other is to both review and
critique each paper. I found a lack of uniformity in this regard. For many of the papers, the latter
D-6
-------
strategy is taken. The report discusses key weakness of these papers. Still many papers are
simply summarized, with little or no discussion of their strengths or weaknesses. This gives the
impression that the results in these papers are more accurate. This may actually be the case, but if
it is, this should be stated more explicitly.
For example, the first paper (Alcott and Wozny) is simply summarized with no editorializing.
One possibility is that there are no weaknesses in this paper. But, many of the issues raised later
for other papers also relate to this paper. For example, the report correctly states, when
discussing the Sawhill paper, that in order to get a consistent estimate of how consumers value
fuel economy, the author is required to correctly specify how consumers form expectations about
gas prices and drive their vehicles. If any of these are incorrectly specified, the estimate will be
biased. The same is true for the Alcott and Wozny paper. While I find Alcott and Wozny's
specification for how expectations are formed to likely be more accurate, ironically, they assume
consumers are hyper-rational in the sense that they form their gas price expectations based on
NYMEX futures prices. In some ways the myopia result is inconsistent with this key assumption.
Differentiating papers by their focus:
The author should be commended for reviewing such a diverse set of papers. The report
categorizes papers by the data used in the analysis and empirical model (e.g., aggregate data v.
micro-level data and discrete choice v. hedonic pricing model). I propose two alternatives. Many
of the papers reviewed did not focus on measuring the implicit discount rate, or even on the issue
of how gas prices affect vehicle demand. If a paper did not focus on either of these two
questions, it is difficult to gauge the robustness of their results with respect to these questions,
the quality of the variation in gas prices in the data, etc. Perhaps a better method would be to first
focus on those papers where gas prices and vehicle choice are the central research question, and
those papers where this is more a tangential part of the analysis.
Differentiating papers by their publication status:
Many of the papers reviewed have not undergone the peer-review process. While I do not claim
that mistakes do not make it through the peer-review process, related to comment above, in some
ways a review of an unpublished paper requires a more thorough review of the paper itself. A
D-7
-------
second way to organize the review would be to focus first on published papers and then second
on unpublished papers. At the very least the fact that a number of the papers are unpublished
should be explicitly noted in the text. Many readers may never make it to the bibliography.
Comments related to discussion of issues:
The review has an excellent discussion of the issues related to how consumers value fuel
economy and the market for fuel economy more generally. I enjoyed reading this section. I have
a few comments related to it, however.
At the top of page 5, the author poses: "does the market value fuel economy improvements at the
discounted expected value of future fuel savings over the lifetime of the vehicle, or less, or
more?"
The author should use the phrase "at society's discounted expected value..."
In many ways questioning whether the market values fuel economy improvements at the
discounted expected value of future fuel savings is simply a tautology. There is always a
discount rate that equates how much consumers are willing to pay for fuel economy
improvements given the expected value of the future fuel savings. Therefore, any statement
about whether they are willing to pay too little, or too much, is actually a statement about how
the consumers' discount rates compare with society's discount rate.
On page 5, the author states, "On the other hand, if consumers are myopic and consider only the
first three years of fuel savings, for example, fuel economy standards can increase welfare even
based solely on private costs and benefits."
If the author is making a statement about the consumers' welfare, and not society's, then more
should be added here. If consumers are myopic because their utility function is myopic, then I do
not see how fuel economy standards make them better off. If my discount rate is actually 20%,
because these are my preferences, then I am better off by investing less in fuel economy
compared to someone whose discount rate is 10%. Forcing standards does not improve my
D-8
-------
welfare. Someone with a 10% discount rate may disagree with my choice, but there is nothing
here that makes my choice non-optimal for me. In contrast, if I behave myopically for reasons
outside of my utility function, such as a lack of access to credit or a lack of cognitive ability, then
fuel economy standards, as well as subsidized loans and/or information campaigns, can improve
my welfare. Alternatively, the author can also point to more behavioral economics reasons why
this market might fail, for example, because consumers have self-control issues. I find many of
these reasons to be quite compelling. My point is that the author should be more clear here. As it
currently reads, it implies that high discount rates, alone, imply fuel economy standards are
welfare improving. This is simply not the case.3
On page 5 the author states that it is surprising that there is no basic research on how consumers
consider fuel economy. This is a strange use of the term basic research, which I take to mean
research on pure science. Does the author equate basic science with interviews? I have never
heard basic science used in this manner.
Comments on specific paper reviews:
The reviews of the papers are very well done and the choice of papers very comprehensive. A
have a few comments that may improve this discussion, as well as a few papers to be added.
Busse, Knittel and Zettelmeyer
Unfortunately the discussion of Busse et al. on page 32 is incorrect. The report states that the
paper does not control for vehicle attributes. In fact, it controls for an extremely wide set of
vehicle attributes through the use of "fixed effects." In the price equations there are fixed effects
for vehicle type, defined as the interaction of make, model, model year, trim level, doors, body
type, displacement, cylinders, and transmission. Put in more layman's terms, the paper is
controlling for all vehicle attributes within this vehicle type. That is, the only way something like
horsepower would not be entirely controlled for is if there were two vehicles of the same make,
model, model year, trim level, doors, body type, displacement, cylinders, and transmission that
had different horsepower levels. So, for example, any attribute that does not vary within 2009
11 freely admit that the use of "can improve" helps.
D-9
-------
Honda Accord four-door EX-L, with a 3.5 liter, 6-cylinder and an automatic transmission
vehicles is completely controlled for. The only attributes that would not be entirely controlled for
are attributes that vary within this definition, such as color or dealer installed options. Variables
such as horsepower likely do not vary within the same make/model/engine size/number of
cylinders. Viewed in this way, perhaps this paper is best included in the Hedonic Models section.
In addition, the latest version (NBER Working Paper 15590) also controls for when in the
lifecycle of vehicles (both new and used) the transaction took place. This has cleared up a lot of
the issues that the author raises at the top of page 32.
The newest version of Busse et al. also is clearer on what is required, theoretically, to account for
the differences across new and used car markets. Market power in the new car market is not
required. As the author of this report notes, the results simply imply that the supply elasticity in
the new car market is more elastic than in the used car market (where a perfectly inelastic supply
is likely a good first-order approximation). Therefore, I would recommend that the author edit or
drop the sentence that states "The used car market was believed to function more efficiently."
Finally, Table 10 of the report should include the used car results for Busse et al. Without taking
into account the results regarding market shares in the new car specifications, the used car results
are a more accurate reflection of the shift in demand from increases in gas prices. So, if only one
of the sets of results is to be reported, the used car results should be reported. Reporting only the
new car results, without a discussion of the quantity effects, or the used car results, is misleading.
Papers that solve for price from a discrete choice model
A number of papers in the literature, and among those reviewed here, argue that if consumers
value fuel economy "correctly," the price of a vehicle should rise and fall one-for-one with the
vehicle's lifetime fuel costs. If prices move less than one-for-one, this is taken as evidence that
consumers are myopic. These papers typically specify utility to be some variant of the logit
model (e.g., simple logit, nested logit, etc.) and then solve for the equilibrium price for a vehicle,
conditional on the assumed functional form for utility. The intuition behind this empirical
D-10
-------
strategy is best summarized by Alcott and Wozny's motivating example, where price is
expressed as:
where G is the discounted fuel cost and q the quantity of vehicles available. They note that: "To
keep this consumers indifferent between the vehicles and the most attractive outside option, the
overall product price p+G must stay the same as G changes." This statement requires that the
utility from the outside option does not also change with G. For those consumers where the "no
car ownership" option is not the most attractive outside option this restriction is unlikely to hold.
For example, imagine using this framework to test whether consumers correctly value fuel
economy using an ideal experiment and tracking Prius prices. In one universe gas prices are $2,
in another alternate universe gas prices are $3. Given the condition above, in order to keep
consumers from shifting demand from the Prius to the next best option, whose operating cost is
assumed not to have increased, the price of the Prius must be lower in the $3-universe by the
increase in fuel costs over the life of the vehicle. If the price of the Prius falls by less than this
amount, estimating the above equation would force you to conclude that consumers undervalue
fuel economy.
Now instead imagine that the fuel cost of the next best option for many Prius owners has
increased by more than the Prius, because the next best option is likely to be something like a
Camry. In this case, the demand for the Prius actually increases, and prices increase. Thus, we
have explained why we concluded that consumers are myopic in the above regression. Indeed,
this is precisely the reason why Busse et al. argues, and shows empirical support for, that the
demand, and therefore price, of more fuel efficient cars will increase when gas prices increase.
And, is the reason why these authors focus on the difference in price changes between high and
low fuel economy vehicles.
D-ll
-------
Of course across all vehicles, the price of some vehicles will decrease and the price of others will
increase. Therefore, we should not expect the coefficient associated with discounted fuel costs to
be negative one, since it will represent the average of these prices changes.
In principle, if one were to specify the correct utility function and demand system, then the above
empirical strategy would be a valid test of myopia. The most general demand system would
imply that the price of a given vehicle is a function of its fuel costs, as ~well as the fuel costs of all
competitors. This is often intractable. Instead, the profession has migrated to specific demand
models that are more parsimonious, such as logit demand models, since the implication of these
demand models is that the ratio of the demand for a particular vehicle to the demand for the
outside good is independent of the fuel costs of other vehicles. However, this is simply a by-
product of the functional form assumptions, and as the Prius example illustrates, is not a general
theoretical result, or one that is very intuitive. Insofar as these models are mis-specified, we will
be biased towards finding myopic consumers.
A conundrum that may be impossible to solve:
As noted by the author in his review of the Sawhill paper, the correct statistical test of consumer
myopia requires specifying the present discounted value of future fuel expenditures for each
vehicle. This requires assumptions regarding how consumers form expectations regarding future
gas prices. The available choices for this are limitless, ranging from assuming that consumers
view gas prices as a random walk (so today's price is sufficient to describe the expected value of
future gas prices) to models that assume consumers use some number of recent prices within an
autoregressive model, to a hyper-rational model where consumers use NYMEX futures on crude
oil and a regression model of how gas prices move with crude oil.
For ordinary least squares to be unbiased, the econometrician must have the correct model of
how consumers form expectations. Otherwise, the key right hand side variable is measured with
error and the estimated coefficient will be biased towards zero.4 This typically biases the
statistical test towards concluding consumers are myopic. Unfortunately, in the end there may
D-12
-------
not be a solution to this. Indeed, the irony is that some papers assume complex ways in which
consumers form expectations, but then conclude they are myopic. In some way this strikes me as
being inconsistent. If consumers are myopic, then they are also not likely to use NYMEX futures
to form their expectations. Therefore, using NYMEX futures to generate the variable of interest
is likely to bias a researcher towards concluding myopia.
Other papers to consider in the review:
The paper should consider reviewing the following papers:
Kilian, Lutz and Eric Sims (2006). The Effects of Real Gasoline Prices on Automobile Demand:
A Structural Analysis Using Micro Data, Working Paper, University of Michigan (April).
Klier, Thomas, and Joshua Linn (2008). "The Price of Gasoline and the Demand for Fuel
Efficiency: Evidence from Monthly New Vehicles Sales Data." Working Paper, University of
Illinois at Chicago (September).
Sallee, James, and Sarah West (2008). "Testing for Consumer Myopia: The Effect of Gasoline
Price Changes on the Demand for Fuel Economy in Used Vehicles." Working Paper, Macalester
College (December).
4 This holds when the measurement error is uncorrelated with the regressor. When the measurement error is correlated with the
regressor, the bias can go in either direction.
D-13
-------
Review of Greene, D.L. (2009) How Consumers Value Fuel Economy: A
Literature Review
Walter McManus, University of Michigan Transportation Research Institute
EPA intends to use vehicle choice models to inform the Agency's analyses of the impacts of
regulatory programs that affect vehicle fuel economy. To adequately predict the ultimate impacts
and judge the effectiveness of particular regulatory actions, EPA needs to understand how
consumers (and producers) are likely to change their market behavior in response to the
regulatory actions.
Market choices by producers and consumers of vehicles and fuels have direct negative impacts
on the nation's air quality and the stratospheric ozone layer. EPA, which is responsible for
protecting and improving air quality and the ozone layer, can influence but cannot control the
behavior of consumers and producers. Market models, including models of consumer vehicle
choice and models of producer technology and product decisions, could help EPA anticipate
market responses to regulatory actions.
This report is a peer review of EPA's comparison of recent mainstream economic estimates of
the value of additional fuel economy to consumers. The documents that I received from EPA (or
RTI International) a memo containing the charge questions, the draft report by Greene (2009),
and three unpublished papers that are covered in the draft report. The unpublished papers are
Alcott and Wozny (2009); Busse, Knittel, and Zettelmeyer (2009); and Sallee, West, and Fan
(2009). I reviewed all of these documents in developing my expert opinion as reported here.
Overall approach and methodology of the study
Greene (2009) adopts a conventional "literature review" approach. The question of how
consumers value fuel economy in vehicle choice has been the subject of a lively conversation on
and off for decades. One appealing feature of the paper is that the same author conducted a
similar review in Greene (1983). Consumers of research face many temptations to undervalue the
retrospective approach, and it is encouraging that EPA's choice of authors reveals the Agency's
rationality as a consumer of research.
The motivation, objectives, and structure of the paper are described in two paragraphs.
"Recent historically high fuel prices, combined with renewed interest in fuel economy and
greenhouse gas emissions standards for automobiles have engendered a number of new
assessments, many specifically aimed at understanding the effects of fuel prices and fuel
economy on consumers' vehicle choices. This paper reviews those studies, published and
unpublished, with the objectives of determining whether a consensus now exists on the value
consumers place on fuel economy, and of gleaning insights into how consumers use fuel
economy information in their car buying decisions.
"This paper is organized as follows. Section 2 summarizes views of how the market for fuel
economy functions, including both supply and demand. The body of the report is Section 3 which
reviews recent empirical estimates of the value of fuel economy based on aggregate and
disaggregate data, discrete choice models and hedonic demand analyses. Section 4 discusses the
implication of those estimates and Section 5 contains concluding observations." (page 4)
D-14
-------
The overall approach and methodology are within the mainstream of critical review practice in
economics, and the execution in unexceptionable. However, both the conventional approach to
reviewing economic research and the conventional approach to economic research itself do not
give much help to decision makers who are understandably made uncomfortable by the widely
differing expert opinions about parameters that are crucial in predicting the impacts of
regulations and in choosing between alternatives. An overall critique of conventional practice is
beyond the scope of this review. However, I do have some suggestions for improvements in the
methodology for comparing studies and for dealing with the lack of consensus in expert
opinions. (The References section below lists some papers by Edward E. Learner, a professor at
UCLA, in which a critique of conventional practice is developed along with alternative practices
that would enhance the usefulness of economic research to policy makers.)
Data analysis conducted for the study
The experts' papers differ in several dimensions that could be responsible for some of the
differences in opinions between experts. Greene (2009) helpfully provides a side-by-side
comparison for econometric model type, data structure, and estimation period. In the text he
covers other issues such as the econometric challenges each author faced and the strategies
chosen to address the challenges. It is difficult for the reader to verify that the same standards are
being used to assess each paper, especially since the amount of discussion by paper varies
greatly. This leads the reader to have less confidence in the reviewer's opinions of the individual
studies (that this or that one had plausible or implausible assumption, methodologies, or results).
Recommendations for alternate data and/or analyses
It would be useful to expand Table 10 (or to create a set of tables) to show other differences that
could be important sources of differences in estimates. The items that could be included would
show differences in:
1. Maintained assumptions reported (assumptions that are maintained across any alternative
specifications or sensitivities),
2. Reported sensitivity analyses (is one done? which assumptions are varied? how widely?),
3. Econometric challenges and approaches used to address them (Greene (2009) appears to
discuss these comprehensively in the text.), and
4. Out-of-sample predictive performance (are predictions made? about what? are the predictions
compared to actual outcomes? how accurate are the predictions?).
Adding these additional side-by-side comparisons would increase the reader's confidence that
the same standards are being used to form the reviewer's opinions about the relative usefulness
of each paper to inform decision making by EPA.
My opinion: The additional elements are responsible for more of the variation in the estimates of
consumer value for fuel economy than are the elements already included in Table 10.
To quantitatively assess whether we have made progress toward a consensus, we need a measure
of researcher/specification uncertainty and disagreement. The 1983 and 2009 results are not
compared quantitatively. Greene (1983) compared estimates of the consumer discount rate, while
Greene (2009) compared estimates of consumer willingness to pay for improvements in fuel
D-15
-------
economy. The idea is to compare the range of parameter estimates to the parameter value
predicted by mainstream theory. For the consumer discount rate the theoretical value is the real
discount rate, which is assumed to be 7% the following. For the willingness to pay version, the
theoretical value is 100% (or willing to pay $1 for future fuel cost savings with a discounted
present value of $1).
The metric, which I call the Multiple Extreme Estimates Test (MEET), is defined as the ratio of
the value predicted by theory to the difference between the two extreme estimates.
, ^^ lvalue Predicted by Theory! V
MEET = ! =
[Largest Estimate - Smallest Estimate] R
limMEET = 0 and Vim MEET = oo
fl-><» fl->0
Numerically, MEET measures the predicted value as a share of the range of expert opinion. The
range of expert opinion, R, is a crude and partial measure of the uncertainty or fragility of the
estimates due to differences in assumptions, specification, and opinions between researchers. The
metric MEET views this fragility compared to the theoretical value of the parameter, V.
If the span of disagreement between experts is large, then MEET is small. The limiting case,
infinite disagreement, produces a MEET of 0; there is not a meeting of the minds. If the span of
disagreement is very small (compared to the theoretical value of the parameter), then MEET is
large. Complete agreement implies an infinite MEET.
Multiple Extreme Estimates Test (MEET)
(The ratio of the parameter value predicted by theory to the range of
extreme parameter estimates.)
_ ,. ,, , n. Greene Greene
Estimates used to compute MEET
All studies* 0.002 0.055
All but one "outlier" ** 0.096 0.080
Greene (1983) "plausible" *** 0.194
* 22/22 in 2009 & 8/8 in 1983
** The 1983 range 0% to 73% is used, based on Table 1. In
the text Greene (2009) reports the range for 1983 as 2% to
73%. The MEET for the narrower range is 0.099.
*** In my opinion, this represents Greene's opinion that the
plausible range is 4% to 40%.
Do the 1983-2009 changes in MEET provide any evidence that a meeting of the minds is closer
today than it was in 1983? The table shows several alternative calculations of MEET for Greene
(1983) and Greene (2009). The alternatives vary by which of the studies in each year are actually
used in computing the metric for that year (based on my interpretation of comparisons made by
Greene to which are added alternatives that include all studies in each year).
The first thing to observe is that all of the MEET values are very low, indicating that the
disagreement between experts is very large. Using all studies each year in the computation of
MEET shows an increase from 0.002 to 0.055. This result is driven by the extreme outliersthe
1983 outlier (-164% to 2840% range for consumer discount rate compared to 7% predicted by
D-16
-------
theory) is more of an outlier than is the 2009 outlier (-360% to 1410% range for willingness to
pay compared to the 100% predicted by theory).
Dropping the outliers (not something I would be comfortable doing if the authors of the 30
studies had reported more informative and thorough sensitivity analyses) would result in a drop
in MEET from 0.096 in 1983 to 0.080 in 2009. We are getting farther away from a consensus,
but the backtrack is small in absolute terms. If the meeting of the minds is in Ann Arbor, and
there was only one holdout and he was located in Davis (CA) in 1983, instead of beginning the
2,300-mile hike to Ann Arbor (according to Google), he wandered 35 miles in the opposite
direction.
Appropriateness of the datasets and any other inputs
It is not clear how the studies were selected for inclusion in Greene (2009). There needs to be
some explanation.
Appropriateness of the conclusions drawn
There are two fundamental conclusions made by Greene (2009). Firstly, expert opinion remains
widely divided about the value of fuel economy to consumers. Despite the improvements in the
quality of data about consumer behavior and in econometric tools, opinion is as widely divided
today as it was three decades ago. I agree that this is the appropriate, discouraging conclusion to
draw.
The second fundamental conclusion drawn by Greene (2009) concerns the causes of the wide
divisions in expert opinion and what should be done about it.
"Although the methodologies or model formulations of a few of the studies are questionable,
there do not appear to be clear associations among methods or data sources and the resulting
inferences. It is suggested that such conflicting results may be attributable to incorrect models of
consumer decision making about fuel economy, the statistical problems caused by omitted
variables, errors in variables and correlated variables, and the complexity of consumers' vehicle
choice decisions. Additional, empirical behavioral research appears to be needed to resolve the
issue." (abstract)
I am not convinced that the differences in assumptions, opinions, and methods have been
sufficiently examined to support even the weak proposition that "there do not appear to be clear
associations among methods or data sources and the resulting inferences." See my
recommendations on pp. 2-4 above. I am not sure exactly what Greene means by "incorrect
models" of consumer decision making about fuel economy, but if he means the efficient markets
hypothesis and economic rationality, then I agree. Assuming that it is this deficiency in our
models of consumer behavior that would be the subject of "behavioral research," I agree.
Overall clarity of the presentation
The overall clarity of the presentation was very high.
D-17
-------
References
Learner, E. E. (1983), Let's take the con out of econometrics, American Economic Review, 73,
31-43.
Learner, E. E. (1985), Sensitivity analyses would help, American Economic Review, 75, 308-
O 1 O
313.
Learner, E. E. (1987), Econometric metaphors, in T. F. Bewley (ed.), Advances in Econometrics,
Cambridge University Press, Cambridge, pp. 1-29.
Learner, E. E. (1989), Planning, Criticism, and Revision, Journal of Applied Econometrics, 4,
Supplement, S5-S27.
Learner, E. E. (1992), Bayesian Elicitation Diagnostics, Econometrica, 60, 919-942.
D-18
-------
March 30, 2010
MEMORANDUM
SUBJECT: EPA Response to Comments on David Greene paper, "How Consumers Value
Fuel Economy: A Literature Review," by peer reviewers Drs. Carolyn Fischer,
Christopher Knittel, and Walter McManus
FROM: Dr. Gloria Helfand, Assessment and Standards Division
Dr. Carolyn Fischer of Resources for the Future, Dr. Christopher Knittel of the Department of
Economics of the University of California at Davis, and Dr. Walter McManus of the University
of Michigan Transportation Research Institute reviewed David Greene's paper, "How
Consumers Value Fuel Economy: A Literature Review," dated December 29, 2009.
This memo includes a summary of comments prepared by RTI International, and responses and
actions in response to those comments from Dr. David Greene of Oak Ridge National Laboratory
and EPA.
3.1 OVERALL APPROACH AND METHODOLOGY OF THE STUDY
The reviewers are supportive of the overall approach and methodology.
Walter: "The overall approach and methodology are within the mainstream of critical review
practice in economics, and the execution in unexceptionable." He does comment on the fact that
"Greene (2009) adopts a conventional 'literature review' approach" and "both the conventional
approach to reviewing economic research and the conventional approach to economic research
itself do not give much help to decision makers who are understandably made uncomfortable by
the widely differing expert opinions about parameters that are crucial in predicting the impacts of
regulations and in choosing between alternatives. An overall critique of conventional practice is
beyond the scope of this review. However, I do have some suggestions for improvements in the
methodology for comparing studies and for dealing with the lack of consensus in expert
opinions. (The References section below lists some papers by Edward E. Learner, a professor at
UCLA, in which a critique of conventional practice is developed along with alternative practices
that would enhance the usefulness of economic research to policy makers.)"
Chris comments that the study does not discuss "why consumers undervalue gasoline, relative
society (if they do indeed undervalue fuel economy). This has important implications for which
policies are most effective at aligning discount rates." However, he recognizes that papers
dealing with such issues are lacking.
Response:
EPA is satis/led with the "conventional" literature review approach that Dr. Greene has used.
If there is a lack of consensus about model parameters, this is important information for EPA as
it pur sues its modeling activities.
-------
EPA did not request that Dr. Greene discuss why consumers may undervalue - or, more
generally, misestimate the value of gasoline. EPA 's initial interest is learning how consumers
consider fuel economy when they are purchasing vehicles. Greene includes some discussion of
this topic on p. 8 as well as in the Executive Summary and the Conclusion. We agree that
understanding why consumers may misestimate this value is an important topic.
3.2 APPROPRIATENESS OF THE DATASETS AND ANY OTHER
INPUTS
One reviewer commented that the choice of the papers was comprehensive. Another reviewer
recommended including some explanation of how the studies were selected for inclusion in
Greene (2009).
Response:
The paper includes a discussion of its selection of papers on p. 2. In particular, it states Dr.
Greene's intent to be comprehensive. The literature review includes some of the most well
known papers in the field (such as papers by Berry, Levinsohn, andPakes and Goldberg) as well
as other less well known works and papers in progress. Dr. Greene added additional papers to
the review between the initial draft and the final report, in response to reviewers' suggestions.
Dr. Greene notes that he felt it important to include studies not (yet) published in peer reviewed
literature because he considered them of publishable quality. EPA 's assessment is that, even if
the review may have missed some papers, there does not appear to be any systematic bias in
inclusion or exclusion. In addition, we support the inclusion of gray literature, because it is the
most current research being conducted.
3.3 DATA ANALYSIS CONDUCTED FOR THE STUDY
Two reviewers comment on the fact that the type of analysis done for the different papers varies
greatly.
Chris: "It strikes me that there are two ways to conduct a review such as this. One is to report the
results of papers with little, or no, discussion of the quality each paper. The other is to both
review and critique each paper. I found a lack of uniformity in this regard. For many of the
papers, the latter strategy is taken. The report discusses key weakness of these papers. Still many
papers are simply summarized, with little or no discussion of their strengths or weaknesses. This
gives the impression that the results in these papers are more accurate. This may actually be the
case, but if it is, this should be stated more explicitly.
For example, the first paper (Alcott and Wozny) is simply summarized with no editorializing.
One possibility is that there are no weaknesses in this paper. But, many of the issues raised later
for other papers also relate to this paper. For example, the report correctly states, when
discussing the Sawhill paper, that in order to get a consistent estimate of how consumers value
fuel economy, the author is required to correctly specify how consumers form expectations about
gas prices and drive their vehicles. If any of these are incorrectly specified, the estimate will be
biased. The same is true for the Alcott and Wozny paper. While I find Alcott and Wozny's
specification for how expectations are formed to likely be more accurate, ironically, they assume
-------
consumers are hyper-rational in the sense that they form their gas price expectations based on
NYMEX futures prices. In some ways the myopia result is inconsistent with this key
assumption."
Walter: "It is difficult for the reader to verify that the same standards are being used to assess
each paper, especially since the amount of discussion by paper varies greatly. This leads the
reader to have less confidence in the reviewer's opinions of the individual studies (that this or
that one had plausible or implausible assumption, methodologies, or results)."
Response:
As Dr. Greene discusses on p. 2 of the paper, "This review attempts to compare the inferences of
different studies on a consistent basis: a typical consumer's willingness to pay for a reduction in
the present value of fuel costs through improved fuel economy. " The comments on individual
studies are valuable in identifying issues and difficulties that arise in conducting these studies.
These critiques are, to EPA, nevertheless less important than a comparison of the estimates of
the willingness to pay for fuel economy across studies. Table 12 of the paper (pp. 50-51)
provides that comparison.
3.4 APPROPRIATENESS OF THE CONCLUSIONS DRAWN
There was general consensus among the reviewers on the major conclusion of the document: that
the literature remains mixed as to whether consumers undervalue, overvalue, or approximately
value fuel economy.
One reviewer raised doubts that differences in assumptions, opinions, and methods have been
sufficiently examined to support even the weak proposition that "there do not appear to be clear
associations among methods or data sources and the resulting inferences."
One reviewer expressed doubts about what Greene means by "incorrect models" of consumer
decision making about fuel economy but suggested efficient markets hypothesis and economic
rationality to be a possible interpretation of this phrase.
Response:
EPA accepts the general conclusion that the literature remains mixed and unsettled on how
consumers consider fuel economy when buying vehicles.
The Executive Summary, the Conclusion, and Table 13 of the report provide new, further
identification and assessment of the effects of different factors on modeling results. As Greene
notes on p. 55, "There does not appear to be an obvious explanation for the widely divergent
results. Neither model type, formulation of the variable representing fuel economy, data type,
time period, nor any other readily identifiable factor shows a strong association with inferences
about the values consumers place on fuel economy (Table 13). "
About the comment that there has been insufficient examination to support the proposition that
"there do not appear to be clear associations among methods or data sources and the resulting
inferences:" it is possible that further examination may reveal associations between results and
-------
underlying characteristics of the studies, and Greene 's statement leaves open that possibility. In
the meantime, Greene was not able to find such patterns.
Greene has modified the "incorrect models " language to refer to "the likelihood that the
rational economic consumer model does not adequately describe the decision-making of
consumers in the real world" (p. xix, the Abstract; seep. 9 for similarly revised language). This
new language appears to correspond to the reviewer's interpretation.
3.5 RECOMMENDATIONS FOR ALTERNATE DATA AND/OR
ANALYSES
3.5.1 Additional Data
Two of the reviewers suggested several other studies that are relevant for this work. A list of
these studies is provided below.
Dreyfus, Mark K. and W. Kip Viscusi (1995). "Rates of Time Preference and Consumer
Valuations of Automobile Safety and Fuel Efficiency." Journal of Law and Economics
38: 79-98.
Hughes, Jonathan E., Christopher R. Knittel and Daniel Sperling (2008). "Evidence of a Shift in
the Short-Run Price Elasticity of Gasoline Demand." Energy Journal'.
Kilian, Lutz and Eric Sims (2006). The Effects of Real Gasoline Prices on Automobile Demand:
A Structural Analysis Using Micro Data, Working Paper, University of Michigan (April).
Klier and Linn (2010 working paper) "The Price of Gasoline and New Vehicle Fuel Economy:
Evidence from Monthly Sales Data"
Klier, Thomas, and Joshua Linn (2008). "The Price of Gasoline and the Demand for Fuel
Efficiency: Evidence from Monthly New Vehicles Sales Data." Working Paper,
University of Illinois at Chicago (September).
Sallee and West (2009 working paper)
Sallee, James, and Sarah West (2008). "Testing for Consumer Myopia: The Effect of Gasoline
Price Changes on the Demand for Fuel Economy in Used Vehicles." Working Paper,
Macalester College (December).
West, S. (2008). "The Effect of Gasoline Prices on the Demand for Sport Utility Vehicles."
Working Paper. Macalester College.
Response:
Greene has added 5 papers to the review since the draft that the reviewers saw: Kilian and Sims
(2006), Klier and Linn (2008), Sallee, West, and Fan (2010), Vance andMehlin (2009), and
Fifer andBunn (2009). Time limitations affected his ability to incorporate additional papers.
3.5.2 Additional Analysis
The reviewers suggested several additional analyses to strengthen the study.
3.5.ii a. The author should take advantage of the summary section to indicate whether
some approaches reviewed in the study are likely to be better than others.
-------
Response:
In the Executive Summary, Greene suggests convening researchers to investigate jointly why
results differ so greatly. He also suggests investigation of alternatives to the rational economic
TM/^si/yl /ire* n f//ir»*/7nrr rtmmt T/^T^ //?/? /Tri/i/~ir/i?c
-------
1. Maintained assumptions reported (assumptions that are maintained across any alternative
specifications or sensitivities),
2. Reported sensitivity analyses (is one done? which assumptions are varied? how widely?),
3. Econometric challenges and approaches used to address them (Greene (2009) appears to
discuss these comprehensively in the text.), and
4. Out-of-sample predictive performance (are predictions made? about what? are the
predictions compared to actual outcomes? how accurate are the predictions?).
Adding these additional side-by-side comparisons would increase the reader's confidence that
the same standards are being used to form the reviewer's opinions about the relative usefulness
of each paper to inform decision making by EPA.
My opinion: The additional elements are responsible for more of the variation in the estimates of
consumer value for fuel economy than are the elements already included in Table 10."
He also suggested a measure of researcher/specification uncertainty and disagreement to
quantitatively assess whether we have made progress toward a consensus. He also provides
illustrative suggestions on ways to use and interpret this measure. "The idea is to compare the
range of parameter estimates to the parameter value predicted by mainstream theory."
The metric is called Multiple Extreme Estimates Test (MEET) and is defined as the ratio of the
value predicted by theory to the difference between the two extreme estimates.
|Value Predicted by Theory] V
|Largest Estimate - Smallest Estimate] R
\\mMEET 0 and Mm MEET
R R 0
"Numerically, MEET measures the predicted value as a share of the range of expert opinion. The
range of expert opinion, R, is a crude and partial measure of the uncertainty or fragility of the
estimates due to differences in assumptions, specification, and opinions between researchers. The
metric MEET views this fragility compared to the theoretical value of the parameter, V."
Response:
The Executive Summary, Conclusion, and Table 13 include further identification and assessment
of the characteristics that might, in principle, be useful in a meta-analysis: the publication
status of the study; the kind of model used; the dependent variable; the type of data; the time
period covered; the fuel economy measure; the form of price expectations; whether actual
transactions prices were used; whether the model allowed for heterogeneous tastes; whether the
model included both the demand and the supply side; and whether fuel economy standards were
included. This list does not include all the variables that the reviewers cite, but it includes many
of them. The discussions of individual papers include some of the other factors, such as some of
the assumptions of the models. As the Executive Summary and Conclusion indicate, there are no
obvious connections among these factors and the model results.
-------
With so many variables and not very many studies, a statistical meta-analysis of these results is
unlikely to produce statistically valid results.
The MEET criterion is an interesting way to think about the variation in study results. EPA at
this time is satisfied with a qualitative discussion of the variation of these results.
3.5.ii.c. Though the issue of consumer heterogeneity has been mentioned in several
places, this may be important area for additional analysis, both to explain differences and
to suggest future research directions.
Response:
The issue of modeling consumer heterogeneity is likely to be important in these models. At this
time it is a source of distinction among models: some approaches, especially mixed logit,
incorporate this feature. The review at this time does not extend to include a discussion of the
"dream " analysis - what would be the ideal data sets, methods, etc. Assessing the state of the
literature is a first step to get to that discussion, and this review has focused on that step.
3.6 OVERALL CLARITY OF THE PRESENTATION
The reviewers agreed that the overall clarity of the document was high.
Reviewers had several suggestions about organizing the paper.
i. One reviewer suggested that since the articles are organized logically by class of data and
methods, it would be helpful to review those methods at the beginning of each section
"including the advantages and drawbacks, as well as key factors or assumptions that can
affect the results."
Response:
Greene has not included this review of methods. EPA agrees that it would be useful information,
but such review information is available elsewhere.
ii. It has also been suggested that it would be helpful to begin with the seminal papers in
each category, "so as to understand the evolution and improvements made subsequently."
Response:
Greene has reorganized the presentation of the papers. For instance, the section on discrete
choice models with aggregate data now begins with the paper by Berry, Levinsohn, andPakes
(1995), a widely cited paper.
iii. Two reviewers agreed that the publication status of papers should be recognized and one
reviewer suggested that the review should "focus first on published papers and then
second on unpublished papers."
Response:
-------
Greene discusses, on p. 2, his reasons for including unpublished papers - in particular, his
opinion that "they are ofpublishable quality. " The new Table 13 has a column identifying the
publication status of all the studies included.
iv. Chris Knittel: "Many of the papers reviewed did not focus on measuring the implicit
discount rate, or even on the issue of how gas prices affect vehicle demand. If a paper did
not focus on either of these two questions, it is difficult to gauge the robustness of their
results with respect to these questions, the quality of the variation in gas prices in the
data, etc. Perhaps a better method would be to first focus on those papers where gas
prices and vehicle choice are the central research question, and those papers where this is
more a tangential part of the analysis."
Response:
Perhaps the major concern with studies that do not focus on fuel economy decisions is that there
may not have been adequate attention to specification of the variable, or problems with omitted
or collinear variables. For those reasons, the studies may not produce robust estimates of the
implicit value of fuel economy.
On the other hand, studies that focus on these variables may suffer from the attention as well: if
the results come out against expectations, there can be a temptation to revise the analysis until
the results meet expectations. Only if results to all specifications are reported is it possible to
see whether the reported results are robust.
As Greene discusses in the conclusion (p. 55), omitted variables, errors in variables, and
correlated variables are likely to be problems that may be difficult to overcome in the vehicle
choice context. Fixed effects may help, but may not solve the problem.
On a related note, another reviewer commented on studies that do not explicitly estimate the
value of fuel economy: "Since few econometric studies have explicitly estimated the value of
fuel economy, a significant contribution of the paper is to translate a wide variety of results into
more consistent indicators. Yet, they could still be more consistent. In Table 10, can all of the
results be converted into the same WTP metric, preferably using the same assumptions about
vehicle lifespan, VMT, discounting, etc.?"
Response:
Table 12 seeks to translate estimates from all the papers into both a willingness to pay for fuel
economy as a per cent of the discounted present value of fuel savings, and an implied annual
discount rate. As Greene discusses in the paper, it is not possible with some of the studies to find
the common metric; with others, he has questions about some of the values reported that are
necessary for the metric. EPA hopes that identifying these difficulties may stimulate future
researchers to report these values more consistently.
Another comment on the presentation of the paper was that "the author's baseline assumptions
need to be stated clearly early in the paper, so careful attention can be given throughout to
assumptions that deviate from that baseline. Indeed, the statement made in the Allcott and
Wozny analysis (' Calculating the discounted present value of fuel costs requires a number of
-------
assumptions...') should be made earlier in a general context, so that the variety and range of
assumptions can be understood and explored."
Response:
Greene includes his assumptions for a baseline analysis in the Appendix to the paper and
discusses these assumptions on p. 2.
3.7 OTHER CLARIFICATION COMMENTS
One reviewer provided the following comments to clarify or rephrase certain statements.
1. The authors should "use the phrase 'at society's discounted expected value...' (top of
page 5) instead of 'does the market value fuel economy improvements at the discounted
expected value of future fuel savings over the lifetime of the vehicle, or less, or more?'"
Response:
This change has been made.
2. The statement (on page 5) that "On the other hand, if consumers are myopic and consider
only the first three years of fuel savings, for example, fuel economy standards can
increase welfare even based solely on private costs and benefits" should be made clearer.
Response:
This language continues in its original form.
3. One of the reviewers comment on the statement on page 5 that "it is surprising that there
is no basic research on how consumers consider fuel economy." "This is a strange use of
the term basic research, which I take to mean research on pure science. Does the author
equate basic science with interviews? I have never heard basic science used in this
manner."
Response:
The word "basic " has been changed to "behavioral. "
Two of the reviewers also provided several detailed comments on specific paper reviews, Section
2 of the document, and editorial comments and corrections to typographical errors.
Response:
Greene has incorporated most of these comments into the revision.
------- |