macroeconomics

HKS727 Case Number 1973.0

This case was written by Albert L Nichols, Adjunct Lecturer in Public Policy, and Anjani Datla, Case Writer, at the John F. Kenne- dy School of Government, Harvard University. HKS cases are developed solely as the basis for class discussion. Cases are not intended to serve as endorsements, sources of primary data, or illustrations of effective or ineffective management. (January 2012) Funding for this case was provided by the Joseph B. Tompkins, Jr. Fund for Case Study and Research. Copyright © 2012 President and Fellows of Harvard College. No part of this publication may be reproduced, revised, translated, stored in a retrieval system, used in a spreadsheet, or transmitted in any form or by any means without the express written consent of the Case Program. For orders and copyright permission information, please visit our website at http://www.ksgcase.harvard.edu/ or send a written request to Case Program, John F. Kennedy School of Government, Harvard University, 79 John F. Kennedy Street, Cambridge, MA 02138.

Corporate Average Fuel Economy Standards 2017-2025

Introduction

In December 2011, the United States National Highway Traffic Safety Administration (NHTSA) and the En-

vironmental Protection Agency (EPA) issued a joint proposal to dramatically increase fuel economy and reduce

greenhouse gas emissions of passenger cars and light trucks sold between Model Years (MY) 2017 and 2025. 1 The

joint proposal was an important step in implementing President Obama’s 2010 memorandum calling for both

agencies to build on a national program that would “produce a new generation of clean vehicles.” 2 The national

program, created in 2009, relied on both the Corporate Average Fuel Economy (CAFE) rules administered by the

NHTSA and EPA’s authority to regulate carbon dioxide emissions from vehicles and other sources.

The heightened activity on fuel economy standards reflected national concerns about dependence on im-

ported oil as well as global climate change. By 2010, Congress had rejected the creation of a cap-and-trade system

to reduce carbon dioxide emissions. But many observers saw increasing energy efficiency as a win-win alternative

that more than paid for itself in fuel savings. A New York Times article described hearings on the proposed MY

2017-2025 fuel economy standards held in Detroit in early 2012 as “more than 90 people who spoke… asserted

that the stricter fuel economy requirements would create jobs, reduce oil consumption, create cleaner air and save

drivers money, all while helping automakers increase their profits.” 3

As part of the national program, the NHTSA and EPA had promulgated new CAFE standards in 2010 and

2011 to substantially raise the combined average fuel economy of new cars and light trucks to 34 miles per gallon

1 74859 Federal Register / Vol. 76, No. 231/ Thursday, December 1, 2011.

2 “Presidential Memorandum Regarding Fuel Efficiency Standards,” May 21, 2010, http://www.whitehouse.gov/the-press- office/presidential-memorandum-regarding-fuel-efficiency-standards, (accessed August 14, 2012). Found under Whitehouse.gov, the Press Office. 3 Bunkley, N, “New Gas Economy Rules Generate Wide Support,” New York Times, January 17, 2012.

This document is authorized for use only by Hameeda Lamb in Business & Economic Policy Graduate Online Fall 2018-2019 at Northwood University, 2018.

HKS Case Program 2 of 16 Case Number 1973.0

(mpg) by MY 2016, and had, for the first time, imposed standards on heavy duty trucks. 4 Under the new joint pro-

posal, the NHTSA estimated that by MY 2025, new cars and light trucks would achieve even greater average fuel

economy of nearly 50 mpg. The NHTSA and EPA predicted that the improvement in fuel economy would reduce

national fuel consumption by 4 billion barrels of oil and cut greenhouse gas emissions by 2 billion metric tons of

carbon dioxide over the lifetimes of the new vehicles. 5 Overall, the NHTSA concluded that the national program

would generate substantial benefits—primarily in the form of savings in fuel costs for vehicle owners—amounting

to four times the cost. The present value of the estimated net benefits of the program was about $323 billion at a

discount rate of 3 percent, as shown in Table 1. 6

Environmental groups had long called for higher fuel economy standards and welcomed the government’s

proposed tightening of the standards. The Environment Defense Fund strongly endorsed the regulation and said it

would "help provide energy security, economic security and climate security.” 7 The Union of Concerned Scientists

believed the standards would “help spur investments in new automotive technology, which will create jobs and

sustain the recovery of the American auto industry.”8 Several consumer groups also hailed the proposal, with the

4 25324 Federal Register / Vol. 75, No. 88 / Friday, May 7, 2010; and 57106 Federal Register / Vol. 76, No. 179 / Thursday, Sep- tember 15, 2011. 5 “NHTSA and EPA Propose to Extend the National Program to Improve Fuel Economy and Greenhouse Gases for Passenger Cars

and Light Trucks,” NHTSA CAFE 2017-2025 Notice of Proposed Rulemaking Factsheet, http://www.nhtsa.gov/fuel- economy,(accessed July 2, 2011). Found under NHTSA.gov, CAFE-Fuel Economy. 6 The NHTSA estimated all costs and benefits in constant 2009 dollars. Based on Office of Management and Budget guidelines,

the NHTSA calculated present values of all costs and benefits at both 3 and 7 percent discount rates. This case reports figures based on a 3 percent discount rate, which the NHTSA appears to have favored. The NHTSA’s results with a discount rate of 7 percent were very similar in relative terms; e.g., at both discount rates, the ratio of benefits to costs was 4.0. See Appendix 1 for a brief introduction to discounting. 7 “Environmental Defense Fund Statement on U.S. Environmental Protection Agency and National Highway Transportation Safe-

ty Administration ‘Proposed Rule 2017- and Later Model Years Light Duty Vehicle Greenhouse Gas Emissions and Corporate Average Fuel Economy Standards’,” Mark MacLeod, Public Hearing, Philadelphia, Pennsylvania, January 19, 2012. 8 “Important Step Towards Cleaner More Fuel Efficient Cars and Trucks,” Union of Concerned Scientists, 2011,

http://www.ucsusa.org, (accessed August 13, 2012).

Table 1. Estimated Benefits and Costs of Proposed CAFE Standards for Model Years (MY) 2017-2025

Notes: Entries are in billions of 2009 dollars. Entries are sums over lifetimes of vehicles sold during model years 2017-2025, discounted to the start of MY 2017 at 3 percent real discount rate.

Source: Preliminary Regulatory Impact Analysis: Corporate Average Fuel Economy for MY 2017-2025 Passenger Cars and Light Trucks, Department of Transportation, National Highway Traffic Safety Administration, Table 13, http://www.nhtsa.gov/fuel-economy, (accessed May 1, 2012).

Item Present Value ($billions) Technology costs $109 Total benefits $432

Net benefits $323

This document is authorized for use only by Hameeda Lamb in Business & Economic Policy Graduate Online Fall 2018-2019 at Northwood University, 2018.

HKS Case Program 3 of 16 Case Number 1973.0

Consumer Federation of America calling the stringent targets “one of the most important consumer protection

measures to be adopted on the federal level in decades.” 9

A few skeptics, however, believed the proposed rule was fundamentally misguided. Although the majority

of auto manufacturers agreed to meet the higher fuel economy targets, the National Association of Auto Dealers

voiced opposition. The association stated that vehicle prices would go up more than $3,000 on average (signifi-

cantly more than NHTSA’s estimates) to achieve the higher CAFE standards and that "seven million licensed drivers

would be priced out of the new-car market.” It added that “real life fuel economy improvements cannot be

achieved, and related policy benefits cannot be realized, unless and until consumers actually buy and use those

new vehicles.” 10

History of the Corporate Average Fuel Economy Program

In 1973, the Arab oil embargo plunged the United States into a large-scale energy crisis. For almost six

months, between October 1973 and March 1974, the Arab members of the Organization of Petroleum Exporting

Countries (OPEC) shut down oil supply to the U.S. in retaliation for its military support to Israel during the Arab-

Israeli war. The embargo led to an unparalleled fuel shortage in the country, marked by sky-rocketing fuel prices

and long lines of cars waiting at gas stations. In the aftermath, oil shortages became a rallying cry for energy inde-

pendence.

Acting on the belief that improving the fuel economy of cars would go a long way toward lowering petro-

leum imports, Congress passed the Energy Policy and Conservation Act of 1975, which gave birth to the Corporate

Average Fuel Economy program—the country’s first ever fuel economy regulation for cars and trucks. The NHTSA,

under the Department of Transportation, was assigned responsibility for administering the CAFE program and im-

plementing annual vehicle fuel economy standards starting with MY 1978. In 1979, the NHTSA also began setting

CAFE standards for light trucks, separate from passenger cars. 11

The CAFE legislation specified fuel efficiency minimum standards for passenger cars that started at 18

mpg in MY 1978 and rose up to 27.5 mpg by MY 1985. With no further increases, fuel economy for passenger cars

and light trucks hovered at 27.5 mpg and 20 mpg respectively, for more than twenty years. 12 Although the stand-

ards remained constant, the overall average fuel efficiency of new vehicles declined as minivans, pickup trucks,

and Sport Utility Vehicles (SUVs)—most of which were classified as light trucks and thus subject to less stringent

CAFE limits—became increasingly popular.

9 “A Key Step to Ending America’s Oil Addiction: Policy Makers, Consumers and Automakers are Shifting New Vehicles to Higher Fuel Economy,” Mark Cooper, Jack Gillis, Consumer Federation of America, July 2012. 10 National Association of Auto Dealers comments letter, February 13, 2012, Re: 2017 and Later Model Year Light-Duty Vehicle Greenhouse Gas (GHG) Emissions and Corporate Average Fuel Economy (CAFE); 49 CFR Parts 523, 531, 533, 536, and 537 and 40 CFR Parts 85, 86 and 600; Doc. Nos. NHTSA–2010–0131 and EPA–HQ–OAR–2010–0799. 11 “History of Fuel Economy: One Decade of Innovation, Two Decades of Inaction,” The Pew Environment Group, April 2011. 12 Ibid.

This document is authorized for use only by Hameeda Lamb in Business & Economic Policy Graduate Online Fall 2018-2019 at Northwood University, 2018.

HKS Case Program 4 of 16 Case Number 1973.0

Under the CAFE program, auto manufacturers did not have to meet the minimum required fuel efficiency

target for each vehicle they sold, but could average across vehicles in a particular class. 13 Manufacturers also were

awarded credits if they over complied with the standard in a given year, which they could transfer or carry forward

or backward, with limited flexibility. Automakers could pay a penalty rather than meet the standard, but only a few

relatively small makers of imported luxury cars (such as Jaguar, Mercedes and Porsche) chose to pay the fines in-

stead of complying with fuel economy targets. 14

In 2007, Congress passed the Energy Independence and Security Act (EISA), which had numerous provi-

sions related to energy efficiency, including new CAFE targets for motor vehicles. EISA required that the NHTSA

increase the average fuel economy of cars and light trucks to 35 mpg by 2020. 15 In 2008, the NHTSA proposed

standards for cars and light trucks manufactured during MY 2011-2015. In 2004, California had legislated even

tighter fuel economy standards, in the form of limits on carbon dioxide emissions. 16 The EPA rejected California’s

request for a waiver to impose emissions rules in 2007, citing federal stipulations under the CAFE and EISA legisla-

tions, which precluded independent action by states. 17 But within days of assuming office in 2009, President Barack

Obama urged the EPA to reconsider California’s waiver request.

In March 2009, the NHTSA promulgated the final rule for MY 2011 and announced that it would propose

revised fuel efficiency standards for 2012-2016. 18 In May of the same year, President Obama directed the EPA and

NHTSA to work together to establish coordinated national fuel efficiency and carbon emissions standards, and to

move up fuel efficiency targets set in EISA from 2020 to 2016 (which would also be consistent with California’s

proposed carbon emissions standards). As part of the negotiations leading up to the directive, California agreed to

follow the federal standards for 2012-2016 and auto manufacturers agreed not to pursue further legal efforts to

block California from setting its own emissions standards for later years. The EPA issued California’s waiver in June

13

The average mpg for a group of vehicles is computed using a harmonic mean rather than a regular arithmetic mean. See Ap- pendix 2 for a discussion of the differences between the two types of means. 14 The level of the fine has varied over time and currently adjusts with inflation. At the time of the proposal, the rate was $5.50/0.1 mpg per vehicle (74900 Federal Register/ Vol. 76, No. 231/ Thursday, December 1, 2011). Thus, for example, if a manufacturer’s fleet averaged 5 mpg more than allowed, it would pay 5 x 10 x $5.50 = $275/vehicle. The EPA, however, had authority to charge dramatically higher penalties under the Clean Air Act (74902 Federal Register/ Vol. 76, No. 231/ Thursday, December 1, 2011). 15 “History of Fuel Economy: One Decade of Innovation, Two Decades of Inaction,” The Pew Environment Group, April 2011. 16 Under the Clean Air Act California was allowed to set its own tighter emission standards for vehicles in recognition of the fact that it had started setting standards long before the Act’s passage in 1970. The CAFE legislation, however, explicitly ruled out state standards. Standards for carbon dioxide emissions were nominally like standards for other pollutants, such as hydrocar- bons, stated in emissions per mile, but were effectively equivalent to fuel economy standards because the only practical way to reduce carbon dioxide emissions per mile meant reducing fuel consumption per mile. 17 See full timeline of California Greenhouse Gas Waiver Request at http://www.epa.gov/otaq/climate/ca-waiver.htm, (ac- cessed October 12, 2012). Found under EPA.gov, Transportation and Air Quality, Transportation and Climate, Regulations and Standards. 18 “DOT Posts New Fuel Economy Standards for Model Year 2011 Cars and Lights Trucks,” NHTSA Press Release, March 27, 2009, http://www.nhtsa.gov/About+NHTSA/Press+Releases/2009/DOT+Posts+New+Fuel+Economy+Standards+for+Model+Year+201 1+Cars+and+Light+Trucks, (accessed October 11, 2012). Found Under NHTSA.gov, About NHTSA, Press Releases.

This document is authorized for use only by Hameeda Lamb in Business & Economic Policy Graduate Online Fall 2018-2019 at Northwood University, 2018.

HKS Case Program 5 of 16 Case Number 1973.0

2009. In September 2009, the NHTSA and EPA formally proposed new rules to implement the negotiated targets,

which were made final in 2010. 19

In January 2011, the NHTSA, EPA, and the State of California announced their intent to develop coordi-

nated rules for MY 2017-2025. Negotiations among the two federal agencies, the State of California, automobile

manufacturers, auto industry unions, and environmental groups proceeded.

In the CAFE rulemakings of 2008 and earlier years, auto manufacturers had vociferously opposed tighter

limits and had submitted detailed technical and economic critiques against higher fuel economy standards. Having

lost their effort to block action by the State of California, however, auto manufacturers were deeply concerned

that without a national agreement that satisfied California, the state would set tighter standards, requiring the

production of different vehicle versions for different parts of the country. The potential challenges of complying

with heterogeneous emissions standards were compounded by the fact that several other states typically adopted

California’s requirements. In addition, two of the “big three” manufacturers (General Motors and Chrysler) were

under pressure to cooperate during negotiations because they had received “bailouts” from the federal govern-

ment. Under these circumstances and in light of widespread support for energy efficiency regulations in general,

and CAFE in particular, most automakers felt it necessary to agree to the more stringent fuel economy standards.

On July 29, 2011, President Barack Obama, accompanied by the heads of 13 major automobile manufac-

turers that accounted for 90 percent of all vehicles sold in the U.S., announced an “historic agreement” to increase

the average fuel efficiency standard of cars to 54.5 mpg by the year 2025. 20 To implement the agreement and in

keeping with federal rulemaking guidelines, the NHTSA and EPA published a “Notice of Proposed Rulemaking”

(NPRM) in the Federal Register and opened it up for public comments in December 2011. Following standard prac-

tice, the agencies would promulgate the final rule (possibly modified from the proposal) or withdraw the proposal

after analyzing public comments and possibly conducting additional analysis. Ultimately, in August 2012, the pro-

posed CAFE standards for MY 2017-2025 were adopted as the final rule. 21

Estimated Costs and Benefits of Proposed Standards for MY 2017-2025

As required by presidential orders dating back to the late 1970s, the NHTSA and EPA each issued a Regula-

tory Impact Analysis (RIA) to accompany the NPRM. 22 The NHTSA’s RIA, more than 800 pages long, provided an

19 25323-25728/ Federal Register / Vol. 75, No. 88 / Friday, May 7, 2010. 20

“President Obama Announces Historic 54.5 mpg Fuel Efficiency Standard,” White House Press Release, July 29, 2011, http://www.whitehouse.gov/the-press-office/2011/07/29/president-obama-announces-historic-545-mpg-fuel-efficiency- standard, (accessed July 2, 2012), found under Whitehouse.gov, the Press Office. Note that mpg figures are calculated under different sets of assumptions depending on the purpose (e.g., to measure compliance under the CAFE legislation as opposed to predicting actual performance). As a result, apparent differences in mpg figures may not reflect real differences. 21 “Obama Administration Finalizes Historic 54.5 mpg Fuel Efficiency Standards,” NHTSA Press Release, August 28, 2012, http://www.nhtsa.gov/About+NHTSA/Press+Releases/2012/Obama+Administration+Finalizes+Historic+54.5+mpg+Fuel+Efficien cy+Standards, (accessed October 12, 2012). Found Under NHTSA.gov, About NHTSA, Press Releases. 22

The estimates in the NPRM and the RIA were often slightly different because of differences in assumptions (and in some cas- es inadvertent errors). This case relies primarily on the RIA because NHTSA officials have informed the authors that its esti-

This document is authorized for use only by Hameeda Lamb in Business & Economic Policy Graduate Online Fall 2018-2019 at Northwood University, 2018.

HKS Case Program 6 of 16 Case Number 1973.0

elaborate benefit-cost analysis of the proposed rule. The agency reported estimates of costs and benefits for nine

alternative sets of CAFE standards including options both less and more stringent than the proposed rule. This case

reports results for the “preferred alternative” (the proposed standard).

Vehicle Footprints and Fuel Economy Standards

The Energy Independence and Security Act of 2007 stipulated that the NHTSA would determine fuel

economy targets for new cars and trucks based on a mathematical function of vehicle “attributes”. After consider-

ing various alternatives, the NHTSA decided to use a vehicle’s “footprint” (length of the vehicle multiplied by its

width) as the sole attribute. The fuel economy target for a vehicle was calculated as a linear function of the foot-

print, with both lower and upper cutoffs. The functions differed for cars and light trucks and rose progressively

higher with each model year, thereby increasing fuel efficiency. Based on forecasts of sales of cars and light trucks

by footprint, the NHTSA estimated the average fuel efficiency that would be achieved by manufacturers’ fleets

each model year, as shown in Figure 1. Less stringent functions were specified for light trucks than for cars. The

NHTSA predicted that the proposed rule would raise average fuel economy of passenger cars from 37.5 MPG in MY

2016 to 52.7 mpg in MY 2025 and save 4 billion barrels of oil over the lifetimes of the cars and light trucks sold dur-

ing the CAFE program. 23 The average mpg ultimately realized for new vehicles, however, would depend on the mix

of vehicles actually sold.

Cost Estimates: Technology Costs

The largest cost of the CAFE program was for new technologies that would increase fuel efficiency in vehi-

cles manufactured between MY 2017 and MY 2025. To calculate these costs, the NHTSA identified the types of

mates are intended to represent “real-world” expectations, whereas the NPRM applies some assumptions about compliance specified in the legislation that are unlikely to occur. 23

These figures are for tested mpg and do not account for the 20-25 percent lower mpg achieved on-road.

Figure 1. Estimated Fuel Efficiency Achieved by LDVs under NHTSA's Proposal for MY 2017-2025

Note: Values are estimated mpg achieved by cars and light trucks combined based on the proposed functions and pro-

jected vehicle sales. Value for 2016 reflects existing CAFE rules while those for other years reflect proposed limits. Source: 74888 Federal Register/ Vol. 76, No. 231/ Thursday, December 1, 2011.

0

10

20

30

40

50

60

A ve

ra ge

m pg

Passenger Cars

Light trucks

Combined

This document is authorized for use only by Hameeda Lamb in Business & Economic Policy Graduate Online Fall 2018-2019 at Northwood University, 2018.

HKS Case Program 7 of 16 Case Number 1973.0

technologies manufacturers would likely use to improve the fuel economy of vehicles and how much they would

cost. The agency assumed that manufacturers would choose the least expensive combinations of technologies

needed to achieve the higher standards. The NHTSA’s cost estimates also assumed that manufacturers would not

reduce safety, comfort or performance in meeting the CAFE standards. In addition to direct costs, the NHTSA also

included associated indirect costs (such as research and development, corporate operations, and sales). 24 The

agency’s total predicted technology costs ranged from $1.7 billion for MY 2017 to $29.9 billion for MY 2025, when

the proposed standards would be most stringent. When the costs for all model years were discounted at 3 percent

back to the start of MY 2017, they totaled $109 billion, as shown in the top row of Table 2. 25

Benefit Estimates

Before estimating the overall benefits of the national program, the NHTSA had to account for the fact that

increasing fuel efficiency reduced the cost per mile of driving, which provided incentives for vehicle owners to

drive more. Econometric studies indicated that this increase in driving—known as the “rebound” effect—was sig- 24 The indirect cost multipliers applied to direct costs varied from 1.2 to 1.7 depending on the “complexity” of the technology and whether the analysis was for the short or long-term. 25 Table 2 numbers are based on discount rate of 3 percent. The NHTSA also provided calculations based on a discount rate of 7 percent.

Table 2. Estimated Present Values of Benefits and Costs of Proposed CAFE Standards for MY 2017-2025 Based on Discount Rate of 3 Percent

Notes: Entries are in billions of 2009 dollars.

Entries for individual model year are discounted (at 3 percent) to the beginning of the model year. The “Total” column is the sum of the values for individual model years discounted (at 3 percent) back to the start of MY 2017. Parentheses indicate negative benefits (costs rather than benefits from rule). “Lifetime fuel expenditures (pretax)” do not include a reduction in the present value total tax payments of about $39.2 billion.

Source: Adapted from Preliminary Regulatory Impact Analysis: Corporate Average Fuel Economy for MY 2017-2025 Passen- ger Cars and Light Trucks, Department of Transportation, National Highway Traffic Safety Administration, Table 13, http://www.nhtsa.gov/fuel-economy, (accessed May 1, 2012).

Component 2017 … 2025 Total Technology costs $1.7 $29.9 $108.9

Benefits (other costs) Li feti me fuel expendi tures (pretax) $7.1 $86.5 $345.3 Cons umer s urpl us from addi ti onal dri vi ng $0.1 $2.6 $7.4 Refuel i ng ti me val ue $0.4 $2.6 $12.8 Petrol eum market external i ti es $0.4 $4.4 $17.9 CO2 reducti ons $0.7 $9.7 $37.8 Reducti ons i n "conventi onal " pol l utants $0.3 $2.5 $11.1 Increas ed conges ti on, acci dents and noi s e ($0.8) ($9.2) ($37.2) Subtotal $8.1 $99.0 $395.0

Net total benefits $6.3 $69.1 $286.0

Present Values ($billions)

This document is authorized for use only by Hameeda Lamb in Business & Economic Policy Graduate Online Fall 2018-2019 at Northwood University, 2018.

HKS Case Program 8 of 16 Case Number 1973.0

nificant, though its magnitude was uncertain. 26 The NHTSA, based on research presented in these studies, assumed

that the elasticity of miles driven with respect to the cost of fuel per mile was -0.1; for example, if fuel economy

doubled, thus reducing fuel cost per mile by 50 percent, vehicles would be driven 5 percent more miles per year.

Savings in Fuel Expenditures

The single biggest estimated benefit of the national program was reduction in fuel costs. The NHTSA’s cal-

culations of fuel savings reflected both the rebound effect and the fact that real-world fuel economy was approxi-

mately 20 percent less than the official stated mpg. The agency used estimates of miles traveled and survival rates

by vehicles of different ages to determine fuel savings over the lifetimes of vehicles sold in each model year. 27 Us-

ing projected fuel prices from the Department of Energy, the NHTSA valued the reductions in fuel use and dis-

counted the savings back to the time of original sale. 28As shown in Table 2, the NHTSA estimated that vehicle own-

ers would save nearly $345 billion in the present value of fuel expenditures over the lifetime of vehicles purchased

between 2017 and 2025—more than three times the estimated technology costs. Moreover, the reduction in fuel

costs shown did not include savings of approximately $39.2 billion in fuel taxes because that additional benefit to

vehicle owners would be offset by equal losses in tax revenues to federal, state, and local governments.

Other Benefits for Vehicle Owners

The NHTSA used the concept of consumer surplus to gauge how much value drivers would gain from the

extra miles driven as a result of the rebound effect. The personal benefit of additional driving was estimated at

$7.5 billion. In addition, with higher fuel economy, vehicles would be able to travel further on the same fuel-tank

capacity. The agency predicted that fewer refuels would save drivers time and the cost of driving off course to find

a fuel station. The cumulative savings for consumers both while searching for fuel stations and at the pump would

amount to $13 billion.

Petroleum Market Impacts

The CAFE program was expected to have wider social consequences than those experienced by vehicle

owners and manufacturers. The NHTSA believed that higher fuel efficiency would lower consumption of oil and

decrease U.S. petroleum imports. Given that the U.S. imported nearly half of the petroleum it consumed in 2010,

the proposed CAFE standard could help make the country more resilient to unexpected price hikes in international

petroleum markets, which disrupted the macroeconomy. 29 Based on an updated analysis by a team of researchers

at Oak Ridge National Laboratory, the NHTSA used a benefit of approximately $13 per barrel reduction in imported

26 “Preliminary Regulatory Impact Analysis: Corporate Average Fuel Economy for MY 2017-2025 Passenger Cars and Light Trucks,” Department of Transportation, National Highway Traffic Safety Administration, pp. 613-620. 27 As vehicles age, they are driven fewer miles and more and more of them are junked or destroyed in accidents, resulting in declining survival rates. 28 For passenger cars, the NHTSA assumed a maximum lifetime of 26 years so that the fuel savings for cars manufactured in MY 2025 included savings till 2051. Because light trucks typically lasted longer, the NHTSA conducted the analysis for a lifetime of 42 years, or to 2077 for trucks cars manufactured in MY 2025 29 “Draft Joint Technical Support Document,” Proposed Rulemaking for 2017-2025 Light Duty Vehicle Greenhouse Gas Emission Standards and Corporate Average Fuel Economy Standards, EPA and NHTSA, pp. 4-29.

This document is authorized for use only by Hameeda Lamb in Business & Economic Policy Graduate Online Fall 2018-2019 at Northwood University, 2018.

HKS Case Program 9 of 16 Case Number 1973.0

oil.30 The total petroleum market external benefits for the duration of the program were estimated at approxi-

mately $18 billion.

Environmental Benefits

Lower fuel consumption also reduced environmental pollution, which in turn would yield health and eco-

logical benefits. The NHTSA estimated reductions in damages from two types of pollutants, carbon dioxide and

“criteria” pollutants. For carbon dioxide, the agency used findings from an interagency working group that had

been explicitly created to develop estimates of the value of reductions in carbon dioxide emissions for use by fed-

eral regulatory agencies. 31 The working group’s estimates varied depending on certain assumptions but had a “cen-

tral value” of $22 per metric ton in 2009 dollars. This value used a 3 percent discount rate and included damages

worldwide. 32 The program’s estimated benefit in reduced carbon pollution was nearly $38 billion.

The proposed rule was also expected to reduce emissions of “criteria” pollutants associated with ozone,

fine particles and other ambient pollutants regulated by air-quality standards under the Clean Air Act. Higher fuel

economy would reduce demand for fuel, which would in turn reduce emissions associated with fuel production

and distribution (for example, combustion of oil to generate heat used in refining processes). The fuel economy

standard would not affect tailpipe emissions of criteria pollutants from vehicles directly because vehicle emission

standards were stated per mile, not per gallon of fuel. The rebound effect, however, would increase miles driven

and thereby increase tailpipe emissions, eating into the net reductions in criteria pollutants. The NHTSA valued the

net reductions using the same estimates of damages per ton adopted by the EPA in rulemakings concerning these

criteria pollutants. Virtually all of the monetized benefits were for reduced fatalities associated with fine particles. 33

30 "Preliminary Regulatory Impact Analysis: Corporate Average Fuel Economy for MY 2017-2025 Passenger Cars and Light Trucks,” Department of Transportation, National Highway Traffic Safety Administration, p. 647. The NHTSA did not include two other types of potential benefits associated with reduced oil imports. The first of these was based on the notion that reduced oil imports would reduce military expenditures. However, the NHTSA, like other agencies, judged the evidence to support such an effect to be inadequate. The second type of benefit the NHTSA rejected was the “monopsony” effect. Based on the notion that if the United States, a large enough consumer, reduced its demand, the world price of oil would be driven down. This ef- fect had been included in some other federal analyses. Gains to net importers (such as the United States), however, would be offset by losses to net exporters. In its analysis of damages from carbon dioxide emissions, the NHTSA had adopted a global perspective that weighted damages the same regardless of where they occurred. It would have been inconsistent to count mo- nopsony benefits in the United States but not offsetting losses elsewhere in the world. 31 “Technical Support Document: Social Cost of Carbon for Regulatory Impact Analysis – Under Executive Order 12866,” Inter- agency Working Group on Social Cost of Carbon, United States Government, February 2010, http://www.epa.gov/oms/climate/regulations/scc-tsd.pdf, (accessed May 22, 2012). 32 If the value were limited to effects in the United States, which is common practice in benefit-cost analyses of proposed feder- al regulations, it would be only 7 to 23 percent as large, depending on the method used to scale down the global damages to U.S.-only values. Ibid., p. 11. 33 Fine particles are emitted directly and are formed in the atmosphere from reactions involving other pollutants (such as sulfur and nitrogen oxides).

This document is authorized for use only by Hameeda Lamb in Business & Economic Policy Graduate Online Fall 2018-2019 at Northwood University, 2018.

HKS Case Program 10 of 16 Case Number 1973.0

The EPA valued changes in direct emissions of fine particles at $300,000 per ton. 34 The estimated total benefit from

lowered emissions of criteria pollutants was $11 billion.

Costs of Increased Accidents, Congestion and Noise due to Rebound Effect

Finally, the NHTSA also recognized that the increased driving associated with the rebound effect would have negative impacts such as damages associated with greater accidents, congestion, and noise, all of which were

functions of miles driven rather than fuel consumption. 35 The costs of increased accidents, congestion, and noise

were valued at $37 billion.

Summary of Social Costs and Benefits

The NHTSA’s detailed benefit-cost analysis indicated that between 2017 and 2025, the $432 billion in

benefits accrued from the new fuel economy standard would dwarf the $109 billion in technology costs by more

than $323 billion (as shown in Table 2). The estimated net benefits were dominated by savings in fuel costs. For

the consumer, these savings would vastly outweigh the anticipated costs in higher vehicle prices.

Why Does the Market Fail to Provide More Fuel Efficient Cars?

Following common practice in benefit-cost analyses, the NHTSA’s primary analysis did not consider the

distribution of costs and benefits among different parties. In a relatively short section toward the end of the RIA,

however, the NHTSA reported on an analysis of costs and benefits from the perspective of vehicle owners. Table 3

summarizes that analysis. Under the program, vehicle owners would receive savings in fuel expenditures (including

lower payments of fuel taxes, which were not included in the social benefit-cost analysis), savings in refueling time,

and the consumer surplus from additional driving. Vehicle owners would also bear the technology costs. The

NHTSA estimated that those costs were only slightly more than one-quarter of the private benefits, leaving large

net benefits from the perspective of individual vehicle owners.

NHTSA analysts recognized that these findings presented an apparent paradox that merited

some discussion. If there were large net gains for vehicle owners (and for manufacturers who better met

buyers’ preferences) from vehicles with higher mpg, why was the market not generating more fuel-

efficient vehicles on its own? The NHTSA asked the following questions in the regulatory impact analysis:

Assuming these comparisons [of private costs and benefits] are accurate, they raise the

question of why current vehicle purchasing patterns do not result in average fuel econ-

omy levels approaching those that this rule would require, and why stricter CAFE stand-

ards should be necessary to increase the fuel economy of new cars and light trucks.…

More specifically, why would potential buyers of new vehicles hesitate to make invest- 34 “Preliminary Regulatory Impact Analysis: Corporate Average Fuel Economy for MY 2017-2025 Passenger Cars and Light Trucks,” Department of Transportation, National Highway Traffic Safety Administration, Table VIII-16. 35 The NHTSA’s estimates for accidents were limited to damages and injuries suffered by the occupants of other vehicles or pedestrians under the assumption that driving decisions by individual owners already reflect the costs borne by the vehicle’s own occupants.

This document is authorized for use only by Hameeda Lamb in Business & Economic Policy Graduate Online Fall 2018-2019 at Northwood University, 2018.

HKS Case Program 11 of 16 Case Number 1973.0

ments in vehicles with higher fuel economy that would produce the substantial econom-

ic returns? And why would manufacturers voluntarily forego opportunities to increase

the attractiveness, value, and competitive positioning of their car and light truck models

by improving their fuel economy? 36

The NHTSA proceeded to offer several potential reasons for this paradox, including what it called “stand-

ard market failures” and inconsistencies in consumer behavior patterns that could be explained by behavioral eco-

nomics. 37

1. The agency acknowledged that the underlying assumptions it used to estimate the economic impacts of the program might not be in line with consumers’ evaluations of the costs and benefits of higher mpg. For

instance, vehicle buyers might discount future fuel savings at a greater rate than used by the agency. In its

explanation, the NHTSA noted that, “combinations of a shorter time horizon or higher discount rate could

reduce or even eliminate the difference between the value of fuel savings and the agency’s estimates of

increase in vehicle prices.”

2. The NHTSA recognized the possibility that its estimates of the costs of incorporating fuel efficient tech- nologies were too low, and did not adequately reflect the full costs of preserving other vehicle features

(such as safety or performance) while improving fuel efficiency.

36 “Preliminary Regulatory Impact Analysis: Corporate Average Fuel Economy for MY 2017-2025 Passenger Cars and Light Trucks,” Department of Transportation, National Highway Traffic Safety Administration, p. 699. 37 The summaries below are derived from discussion in ibid., pp. 699-723.

Table 3. Benefits and Costs from the Perspective of Vehicle Owners

Notes: Entries are in billions of 2009 dollars.

Entries for individual model year are discounted sum (at 3 percent) over lifetimes of vehicles sold in that model year. Entries for “Total” are further discounted to MY 2017 and summed across model years. Parentheses indicate negative values (costs rather than benefits from rule).

Source: Values for individual components from Table 2. Classification of “private costs and benefits” adapted from Prelimi- nary Regulatory Impact Analysis: Corporate Average Fuel Economy for MY 2017-2025 Passenger Cars and Light Trucks, Department of Transportation, National Highway Traffic Safety Administration, Table VIII-27.

Component 2017 … 2025 Total Technology costs $1.7 $29.9 $108.9

Benefits (other costs) Li feti me fuel expendi tures (i ncl udi ng tax) $8.0 … $96.1 $384.5 Cons umer s urpl us from addi ti onal dri vi ng $0.1 $2.6 $7.4 Refuel i ng ti me val ue $0.4 $2.6 $12.8 Subtotal benefi ts (other cos ts ) $8.4 $101.2 $404.7

Net total benefits $6.7 $71.3 $295.7

Present Values ($billions)

This document is authorized for use only by Hameeda Lamb in Business & Economic Policy Graduate Online Fall 2018-2019 at Northwood University, 2018.

HKS Case Program 12 of 16 Case Number 1973.0

3. Improvements in fuel efficiency could be used to increase MPG or to yield improvements in other attrib- utes, such as better performance or larger vehicles. Consumers may prefer cars that use technology im-

provements to achieve better performance (e.g. engines with higher horsepower) or more room, not just

higher MPG. 38

4. In the face of uncertainty about future fuel prices, the NHTSA suggested that buyers might exhibit “loss aversion,” a tendency to give much more weight to potential losses than to potential gains. Faced with

uncertainty about future fuel prices and hence savings, vehicle buyers might give too much weight to the

scenarios in which low prices led to net losses from spending more on fuel-efficient vehicles. 39

5. Imperfect competition among manufacturers could reduce their incentives to supply vehicle attributes for which consumers were willing to pay. For example, the NHTSA considered the likelihood that manufac-

turers intentionally did not invest in fuel-efficient technologies for larger vehicles such as minivans and

SUVs because they believed that prospective consumers were unlikely to pay more for higher fuel econ-

omy in this vehicle range.

6. The NHTSA also suggested that the market for vehicle fuel economy might not work efficiently because buyers lacked sufficient information on the fuel efficiency of vehicles, or had difficulty comparing ongoing

fuel-cost savings with higher upfront prices, reducing their willingness to pay for greater mpg. 40

After discussing these possible explanations, the agency concluded that:

[NHTSA]- has been unable to reach a conclusive answer to the question of why the ap-

parently large differences between its estimates of benefits from requiring higher fuel

economy and the costs of supplying it do not result in higher average fuel economy for

new cars and light trucks. 41

38 An article published in 2011 estimated that if horsepower, weight and torque had remained at their 1980 levels, fuel econo- my would have increased by 60 percent, rather than the 15 percent actually experienced between 1980 and 2006. Christopher R. Knittel, “Automobiles on Steroids: Product Attribute Trade-Offs and Technological Progress in the Automobile Sector,” Amer- ican Economic Review, 101, (December 2011). 39 Studies by behavioral economists indicate that individuals generally give much higher weight to potential losses than to po- tential gains. 40 Manufacturers had been required from the start of CAFE to post fuel-economy information on labels placed in the windows of all new cars. The EPA had revised the specifications of those labels for MY 2008 and had already established new require- ments to start with MY 2013 (see Exhibit A). 41 “Preliminary Regulatory Impact Analysis: Corporate Average Fuel Economy for MY 2017-2025 Passenger Cars and Light Trucks,” Department of Transportation, National Highway Traffic Safety Administration, p. 711.

This document is authorized for use only by Hameeda Lamb in Business & Economic Policy Graduate Online Fall 2018-2019 at Northwood University, 2018.

HKS Case Program 13 of 16 Case Number 1973.0

Exhibit A. Comparison of Current and New EPA-Required Window Stickers for New Cars Note: Labels must be displayed in the windows of new cars. Actual size of label is a minimum of 7” x 4.5”(40 CFR 600.307-08, “Fuel economy label format requirements”). Sticker for Model Years 2008-2012

Sticker for Model Years 2013-:

Source: 2008-2012 sticker from http://www.fueleconomy.gov/feg/ratingsnewsticker.shtml, (accessed October 10, 2012), found under Fueleconomy.gov, About EPA Ratings. 2013 and beyond sticker from http://www.fueleconomy.gov/feg/Find.do?action=bt1, (accessed October 10, 2012), found under Fuelecono- my.gov, About EPA Ratings.

This document is authorized for use only by Hameeda Lamb in Business & Economic Policy Graduate Online Fall 2018-2019 at Northwood University, 2018.

HKS Case Program 14 of 16 Case Number 1973.0

Appendix 1. Discounting and Present Values

When costs and benefits occur over multiple years, it is standard practice to “discount” values for the var-

ious years to a common year (often the start of the project in question). The sum of these discounted values is

their present value. For readers of the case who are not familiar with discounting, this appendix provides a very

brief overview of the mechanics of and rationales for discounting.

Rationale for and Mechanics of Discounting

The basic rationale for discounting is that benefits received now are more valuable than those received

later. Conversely, it is generally better to incur costs later rather than earlier. That is because there is an opportuni-

ty cost to spending money now, rather than investing it for later. Even when the funds would not otherwise be

invested, but rather would be consumed now, there is an opportunity cost because people generally prefer con-

sumption now to consumption later (as witness the many people who take on debt through credit cards and other

forms of borrowing).

Suppose the opportunity cost of funds is that it could be invested at an interest rate of r. Thus, $1 invest-

ed today will grow to $(1+r) next year, $(1+r)2 in two years, and so forth. Alternatively, if you needed to spend $1

in two years, you would have to invest only $1/(1+r)2 today, because it would grow to $1 in two years.

Assume we have a stream of benefits and costs, where Bt are Ct are the values in year t. The net benefits

in a given year are Nt = Bt – Ct. If the benefits and/or costs happen “now” and then over a period of T years, the Net

Present Value (NPV) of that stream is:

T T

r N

r N

r N

NNPV )1(

... )1(1 2

21 0

+ ++

+ +

+ += .

If the analysis is done in constant rather than nominal dollars, as most benefit-costs analyses are, the discount rate

used should be a real one (net of inflation). For example, if the relevant nominal rate is 10 percent and inflation is

expected to be 2 percent, the corresponding real discount rate is a little less than 8 percent. 42

For example, suppose the extra cost of a car under a proposed rule is $500, but it yields savings of $70 per

year over the next 10 years. Thus, the annual cash flow is -$500 in year 0 and +$70 for each of the next 10 years.

The total savings will be $700 compared to costs of $500, but the costs are incurred now and the benefits are re-

ceived later. Simply adding up the costs and benefits over time implicitly uses a zero discount rate. With a positive

discount rate, the present value of benefits may not exceed costs.

42 Strictly speaking, if the relevant nominal interest rate is n and the inflation rate is i, the corresponding real rate is 1

1 1

− + +

= i n

r .

For n = 10% and i =2%, r = 0.078. = 7.8%, slightly less than the 8% we get by subtracting the inflation rate from the nominal in- terest rate.

This document is authorized for use only by Hameeda Lamb in Business & Economic Policy Graduate Online Fall 2018-2019 at Northwood University, 2018.

HKS Case Program 15 of 16 Case Number 1973.0

Choice of Discount Rate

If the temporal distributions of costs and benefits differ substantially, the discount rate used to compute

present values can have a very large impact on whether a project yields positive or negative estimated net bene-

fits. For example, at a discount rate of 7 percent, $1 received each year for 10 years is equivalent to $7.02 received

now. At a discount rate of 3%, the present value is significantly higher at $8.58. However, at 12 percent, the pre-

sent value is only $5.65.

The discount rate(s) used to calculate present values for government projects or regulations have been

the subject of considerable controversy over many years. The Office of Management and Budget (OMB), which

reviews RIAs and issues guidelines for conducting them, recommends that agencies report results for rates of both

7 percent and 3 percent. 43 According to OMB, the higher rate reflects estimates of the long-term average real re-

turn on private capital in the U.S. economy (before taxes). The lower rate reflects the lower after-tax average re-

turns that investors have received over the long-term. 44

43 “Circular A-4 Regulatory Analysis,” Guidelines and Discount Rates for Benefit-Cost Analysis of Federal Programs,” Office of Management and Budget, September 17, 2003, http://www.whitehouse.gov/sites/default/files/omb/assets/regulatory_matters_pdf/a-4.pdf, (accessed August 24, 2012). 44 After-tax rates of return are reduced from pre-tax rates by corporate and individual income taxes at the federal, state, and local levels.

This document is authorized for use only by Hameeda Lamb in Business & Economic Policy Graduate Online Fall 2018-2019 at Northwood University, 2018.

HKS Case Program 16 of 16 Case Number 1973.0

Appendix 2. Harmonic Means Used to Compute Average MPG

The average mpg for a group of vehicles is computed using a harmonic mean rather than a regular arith-

metic mean. The arithmetic mean of mpgs implicitly assumes that each vehicle uses the same amount of gasoline

per year. It is much more plausible to think that vehicles will drive the same numbers of miles per year (before

accounting for the rebound effect).

Suppose we have two cars, one that gets m1 mpg and one that gets m2 mpg. The arithmetic mean of those

two numbers is:

. 2

21 mmmeanArithmetic +=

In contrast, the harmonic mean is:

  



  

   

   

 += 2/

11 /1

21 mm meanHarmonic

To see the difference, consider an example where the two cars get 10 and 40 mpg, respectively. The arithmetic

mean is 25 mpg. However, the harmonic mean is much lower: 16]2/)40/110/1/[(1 =+ . To see why the harmonic

mean is a more useful measure in this context, suppose each car travels 10,000 miles per year. Total consumption

of fuel will be 10,000/10 + 10,000/40 = 1,000 + 250 = 1,250 gallons. Because they travel a combined 20,000 miles,

their average mpg will be 20,000/1,250 = 16 mpg, the harmonic mean of 10 and 40 mpg.

In Europe, fuel efficiency is measured in liters/100 km and a smaller number means greater fuel efficien-

cy. When the measure is fuel volume per distance traveled, a simple sales-weighted average gives the correct

overall average for a fleet of vehicles. For example, if we converted our mpg figures to gallons/100 miles they

would be 10 gallons per 100 miles and 2.5 gallons per 100 miles. Their simple average would be (10 + 2.5)/2 = 6.25

gallons per 100 miles, which is equivalent to 100/6.25 = 16 mpg.

The use of mpg can also confuse one’s intuition about the relative impacts of changes in fuel economy.

For example, suppose we increase fuel economy for vehicle 1 (a minivan) from 15 mpg to 20 mpg and for vehicle 2

(a small size sedan) from 30 to 60 mpg. Which change saves more fuel? It is tempting to say vehicle 2, where we

have increase fuel economy by 30 mpg compared to only 5 mpg for vehicle 1. Moreover, for vehicle 1 the percent-

age improvement is only 33 percent, compared to 100 percent for vehicle 2. However, if we work out the actual

savings in fuel consumption if each vehicle travels 10,000 miles per year, they are exactly the same:

Vehicle 1: 10,000 miles/15mpg – 10,000 miles/20mpg = 667 gallons - 500 gallons = 167 gallons

Vehicle 2: 10,000 miles/30mpg – 10,000 miles/60mpg = 333 gallons – 167 gallons = 167 gallons.

As this example illustrates, increasing mpg is much more valuable when the base is low than when it is relatively

high.

This document is authorized for use only by Hameeda Lamb in Business & Economic Policy Graduate Online Fall 2018-2019 at Northwood University, 2018.