Elsevier

Energy Policy

Volume 87, December 2015, Pages 370-381
Energy Policy

Effects of policy characteristics and justifications on acceptance of a gasoline tax increase

https://doi.org/10.1016/j.enpol.2015.08.037Get rights and content

Highlights

  • Phasing in the tax increase modestly raised support.

  • Making the tax increase revenue-neutral increased support.

  • Using the extra revenue for energy efficiency increased support.

  • Information on high gasoline prices elsewhere did not influence support.

  • Variation in actual fuel prices did not influence support.

Abstract

Many economists argue that increasing the gasoline tax is an effective way to reduce fuel consumption. Yet, public support for such a tax increase has been rather low among US residents. We performed eight survey experiments (total N approximately 3000) to examine how selected policy characteristics and persuasive messages influence support for a gasoline tax increase. Several policy characteristics significantly increased support for a gasoline tax increase. Having the increase phased in over five years modestly increased support. Compared with giving the extra revenue to the US Treasury’s General Fund, both refunding the extra revenue equally to all American families and having this revenue used for energy efficient transportation strongly increased support. Support for a gasoline tax increase was not affected by the nature of the mechanism to achieve revenue neutrality. Most people supported a 20 cent per gallon tax increase to repair roads and bridges. Explaining how the gasoline tax increase would reduce fuel consumption slightly increased support for a gasoline tax increase, but neither being informed of the high gasoline prices in other advanced industrial countries nor the actual pump price of gasoline at the time of the experiment influenced support for a gasoline tax increase.

Introduction

Since the late 1990s, many nationwide polls and surveys document that majorities of Americans support governmental action to deal with climate change (e.g., McCright et al., 2013, McCright et al., 2014; Nisbet and Myers, 2007; Program on International Policy Attitudes, 1998). This is so despite the sustained conservative assault on climate science and policy (McCright and Dunlap, 2010, Oreskes and Conway, 2010) and the recent drop in belief in and concern about climate change among conservatives and Republicans (McCright and Dunlap, 2011). Public support is strongest for regulations requiring emission limits on industry and automobiles (e.g., Dietz et al., 2007; McCright, 2009; Nisbet and Myers, 2007; O’Connor, Bord, and Fisher, 1999), even when the costs of such limits are mentioned in the survey question. Americans also strongly support policies to develop renewable energy sources and improve energy efficiency (e.g., McCright, 2009; O’Connor et al., 1999). This is especially true for proposals shifting subsidies from fossil fuels towards sustainable energy sources—such as solar and wind energy (e.g., Dietz et al., 2007; McCright, 2009; Nisbet and Myers, 2007).

While 80-90% of Americans are willing to pay significantly higher prices for automobiles and electricity that utilize renewable energy sources, large majorities of Americans (65-70%) oppose higher taxes on gasoline and electricity (Dietz et al., 2007, Nisbet and Myers, 2007, O’Connor et al., 1999, O’Connor et al., 2002). Increases in fuel taxes are consistently the least popular mechanism for reducing fossil fuels consumption (e.g., Dietz et al., 2007; O’Connor et al., 1999, O’Connor et al., 2002; Zahran et al., 2006). Indeed, no major presidential candidate has proposed a substantial increase in the gasoline tax since Independent John Anderson in 1980. Not even President Obama, who aims to reduce fossil fuel consumption and improve infrastructure, has proposed such an increase (National Public Radio, 2015).

Yet, many economists (e.g., Porter, 2012; Whaples, 2006)—including some from the George W. Bush administration (e.g., Mankiw, 2009)—agree that the most effective way to reduce fossil fuel consumption is to increase taxes on them (see also U.S. National Research Council, 2010). Porter (2012) argues that increasing the gasoline tax is much more effective than fuel economy standards in encouraging fuel conservation. The problem with the latter is that, other things equal, more fuel efficient vehicles make each mile driven less expensive—thus encouraging more driving (see York, 2006). By contrast, increasing the price of fuel encourages less driving.

Despite evidence of opposition to such a tax increase, these economic arguments suggest it is worth investigating whether the acceptability of a gasoline tax increase would be influenced by different policy characteristics and/or different persuasive messages about or information relevant to such a policy. To this end, we employed insights from the fear appeal literature and existing scholarship on decision-making to perform eight experiments investigating how support for a gasoline tax increase among U.S. residents is influenced by different policy characteristics and/or justifications.

The fear appeal literature (e.g., Griffin et al., 1999; Witte, 1994) suggests that for people to take action against a threat, it is not sufficient that they believe that the threat is severe and that they are susceptible to its consequences. They also must believe that there are practical ways of protecting themselves against the threat. This latter belief has two components. First, people must believe that there are actions that can solve this problem (i.e., response efficacy), and, second, they must feel self-efficacy in taking the proposed actions. Consistent with the response efficacy component, Dietz et al. (2009) find that people deliberating on climate policies often evaluate them by how well they would achieve their stated goals. Hence, one possible reason the general public so strongly opposes a gasoline tax increase is that many people are convinced that there is no reasonable alternative to driving and that people will drive as much as before the tax increase. Moreover, to have self-efficacy to act against a threat, people also must believe that such action is not excessively costly in terms of time, convenience, and money (e.g., Janz and Becker, 1984).

The above principles suggest two factors that may influence public support for a gasoline tax increase. To focus on response efficacy (i.e., that a gasoline tax increase will reduce fuel consumption), we study the effect of (a) providing reasoned arguments for how increasing the price of gasoline will reduce consumption and (b) presenting a policy such that revenue generated by the tax increase can help people decrease the amount of fuel necessary for transportation (i.e., improving public transportation and other forms of energy efficient transportation). To focus on reducing perceived costs of the tax increase, we examine the effect of (c) presenting a policy such that the extra revenue would be refunded to the American people, independent of the gasoline tax that they would pay.1 We also investigate the effect of (d) having the tax increase phased in gradually rather than implemented immediately.

Existing scholarship in social psychology points to the importance of comparisons to relative baselines or references groups (rather than to some absolute state) in understanding attitudes and behaviors (see e.g., Cash et al., 1983; Easterlin, 1974; Kahneman and Tversky, 1979). Further, as people adapt to changing situations over time, improvement or deterioration relative to this adaptation level produces positive or negative emotions, respectively (e.g., Helson, 1964; Kahneman, 1999; Parducci, 1995).

This research suggests two additional factors that may influence willingness to accept a gasoline tax increase. One factor relates to the relative baseline of comparison, and the other relates to the reference group that individuals consider when weighing alternatives. Using a natural experiment, created by a shift in prices over time, we examine how acceptance of a gasoline tax increase (e) varies with a change in the actual pump price of gasoline. Also, we investigate how such acceptance is influenced by (f) receiving information about the much higher gasoline prices in other advanced industrial countries.

Finally, two additional characteristics of a potential gasoline tax increase policy warrant attention. America’s civil engineers (American Society of Civil Engineers, 2013) and citizens (Agrawal and Nixon, 2015) are acutely aware of the poor condition of much of our nation’s transportation infrastructure.2 Given this and the fact that politicians sometimes propose legislation to improve this infrastructure, we examine how acceptance of a gasoline tax increase is influenced by (g) using the extra revenue for repairs in the transportation infrastructure versus encouraging fuel conservation in a revenue neutral manner. We further study how such acceptance is influenced by (h) presenting reasoned arguments about the need to increase funding for road and bridge repair. Also, given the continuing popularity of revenue neutrality in tax reform (e.g., Hatch, 2014; The White House 2015), we examine how acceptance of a gasoline tax increase (i) varies across two mechanisms for achieving revenue neutrality.

Between August 2012 and January 2015, we conducted eight online survey experiments to investigate how the policy characteristics and justifications mentioned above influence support for and acceptance of a gasoline tax increase. Table 1 provides information on the experimental stimuli, outcome variables, sample, and administration details for each experiment. Except for Study 2, which used a large nationally representative sample of nearly 1700 subjects, and Study 3, which used a small convenience samples of 41 subjects, the other six experiments used diverse convenience samples of between 237 (Study 1) and 540 (Study 8) subjects.

With one exception, we administered our survey experiments via SurveyMonkey (Studies 3-8) or Survey Gizmo (Study 1) to subjects recruited via Amazon Mechanical Turk (MT), a crowdsourcing website where “requesters” solicit “workers” to perform “human intelligence tasks” (HITs) for pay. MT has emerged as a practical way for recruiting a large number of subjects from a reasonably wide cross-section of the general public for experimental participation (e.g., Berinsky et al., 2012; Paolacci and Chandler, 2014; Weinberg et al., 2014). In Study 2, we recruited a nationally representative sample of subjects and administered our experiment via the GfK Group (formerly Knowledge Networks) as a project sponsored by the NSF-funded Time-sharing Experiments for the Social Sciences (TESS) Program. In all experiments, we limited participation to adults at least 18 years old who were residing in the USA.

All survey experiments were structured similarly. Each opened with some background information, such as the following: “Some have proposed that in order to limit global warming, Americans need to substantially reduce our use of petroleum and other fossil fuels. One proposal for doing so is to increase the tax on gasoline by $1 per gallon.” This was followed by the experimental stimuli (the information about different proposals to increase the gasoline tax), questions to assess attention to and comprehension of the messages and information about the proposals, and items to measure our outcome variables. The final sets of questions measured other attitudes and beliefs and typical demographic, social, and political characteristics.

To use only high quality data, we eliminated from our analyses data from subjects who met one or more of the following conditions:

  • subjects answering too few attention or comprehension check question correctly (who presumably paid insufficient attention to the proposals);

  • subjects with substantially incomplete surveys;

  • subjects who participated in one of our prior experiments, despite clear instructions prohibiting so;

  • subjects who were not residing in the USA at the time of their participation.

Thus, with the exception of Study 3, the sample size we used in our analyses was smaller than the total number of subjects who participated in the experiment.

Across our eight experiments we employed several measures of acceptance of a gasoline tax increase. In Studies 1–4, we used two measures. One item, support for $1 per gallon gasoline tax increase, measured how much subjects opposed or supported (“totally oppose”=1 to “totally support”=11) a $1 per gallon gasoline tax increase. A second item, maximum gasoline tax increase supported, asked subjects to indicate in integers the maximum gasoline tax increase (cents per gallon) they would support. In Studies 5–8, we used a five-item measure, support for a gasoline tax increase. We asked subjects how much they opposed or supported (“totally oppose”=1 to “totally support”=11) each of five proposed per gallon gasoline tax increases: $0.20, $0.40, $0.60, $0.80, and $1.00.

Section snippets

Methods

Study 1 was a pilot for Study 2. As mentioned in Section 1.2 earlier, the latter experiment, which was sponsored by TESS and administered by GfK, was embedded within a survey administered to a nationally representative sample of US adults. GfK reported a survey completion rate of 64.5%.

Study 1 employed a 2×3×2 between-subjects factorial design. One factor featured a persuasive message, while two factors dealt with policy characteristics. Subjects either received or did not receive a message

Methods

We designed Study 3 (N=41) to learn why subjects had not responded more positively to proposals that would refund the extra tax revenue to each American family. As such, it was not only a follow-up to Studies 1 and 2, but also a pilot for the studies discussed in Section 4. Study 3 employed a 3-condition within-subject design with randomized order of presentation. Subjects were asked about three different proposals, each of which would “increase the tax on gasoline by 20 cents per gallon per

Methods

We designed Studies 4, 5, and 6 to re-examine the relationship between revenue use and support for the gasoline tax increase after we clarified the description of the revenue refund. These three experiments each employed the same 3-condition within-subject design with randomized order of presentation. Subjects were asked about three different proposals, each of which would “increase the tax on gasoline by 20 cents per gallon per year, for five years, leading to a total increase of $1 per gallon

Awareness of high gasoline prices in other countries: study 7

The experimental social psychological literature finds that responses may be influenced by varying the reference groups or baselines of comparison (Cash et al., 1983, Easterlin, 1974, Fink et al., 1983, Kahneman and Tversky, 1979). We designed Study 7 to examine whether receiving information about the higher gasoline prices in other advanced industrial countries may influence support for a gasoline tax increase. It is plausible that Americans who regard $3.50 per gallon gasoline as excessively

Revenue neutrality and transportation infrastructure: study 8

Study 8 had two objectives, each focusing on potential characteristics of a gasoline tax increase policy. One was comparing support for a revenue-neutral tax increase for discouraging fuel consumption with support for a tax increase where the extra revenue is used to repair roads and bridges. Experts (American Society of Civil Engineers, 2013) and laypeople (Agrawal and Nixon, 2015) alike largely perceive our nation’s transportation infrastructure as in serious need of significant repair. It

Conclusion and policy implications

Large majorities of Americans oppose higher taxes on gasoline (e.g., Dietz et al., 2007; Nisbet and Myers, 2007; O’Connor et al., 1999, O’Connor et al., 2002). Yet, many economists argue that increasing the gasoline tax is the most effective way to reduce fuel consumption (e.g., Mankiw, 2009; Porter, 2012; Whaples, 2006). Given this, we drew upon insights from the fear appeal literature and existing scholarship on decision-making to conduct eight experiments investigating how support for a

Acknowledgments

Research for this article was supported by a Sustainability grant to the first author, from Michigan State University and by Time-sharing Experiments in the Social Sciences (TESS). While TESS gave us access to a population-based survey, neither TESS, nor the Sustainability grant, had any influence on our data analysis, writing, or conclusions. We thank Thomas Dietz for helpful comments and Calvin Coplai, Shaun Piwowar, and Renumah Newaz for research assistance.

References (48)

  • M. Brons et al.

    A meta-analysis of the price elasticity of gasoline demand. A SUR approach

    Energy Econ.

    (2008)
  • R.J. Griffin et al.

    Proposed model of the relationship of risk information seeking and processing to the development of preventive behaviors

    Environ. Res.

    (1999)
  • Agrawal, A.W., Nixon, H., 2015. How do Americans feel about taxes and fees to fund transportation? Top-line results...
  • American Society of Civil Engineers. 2013. Report card for America’s infrastructure. Reston, VA: American Society of...
  • J. Bassili

    Attitude strength

  • A.J. Berinsky et al.

    Evaluating online labor markets for experimental research: Amazon.com’s Mechanical Turk

    Polit. Anal.

    (2012)
  • Cantwell, M., 2009. Cantwell, Collins propose carbon auction: bill reduces emissions, returns revenue to consumers....
  • T.F. Cash et al.

    Mirror mirror, on the wall…? Contrast effects and self-evaluations of physical attractiveness

    Personal. Soc. Psychol. Bull.

    (1983)
  • J. Cohen

    Statistical Power Analysis for the Behavioral Sciences

    (1988)
  • T. Dietz et al.

    Support for climate change policy: social psychological and social structural influences

    Rural Sociol.

    (2007)
  • T. Dietz et al.

    How deliberation affects stated willingness to pay for mitigation of carbon dioxide emissions

    Land Econ.

    (2009)
  • R.A. Easterlin

    Does economic growth improve the human lot? Some empirical evidence

  • E.L. Fink et al.

    Positional discrepancy, psychological discrepancy, and attitude change: experimental tests of some mathematical models

    Commun. Monogr.

    (1983)
  • P. Goodwin et al.

    Elasticities of road traffic and fuel consumption with respect to price and income: a review

    Transp. Rev.

    (2004)
  • Hatch, Senator Orrin, 2014. Hatch outlines seven principles for comprehensive tax reform. Press Release. December 16....
  • H. Helson

    Adaptation-Level Theory: an Experimental and Systematic Approach to Behavior

    (1964)
  • N.K. Janz et al.

    The health belief model: a decade later

    Health Educ. Q.

    (1984)
  • D. Kahneman et al.

    Prospect theory: an analysis of decision under risk

    Econometrica

    (1979)
  • D. Kahneman

    Objective happiness

  • KPMG Peat Marwick, 1990. Changes in the progressivity of the federal tax system, 1980 to 1990. Prepared for the...
  • J.A. Krosnick et al.

    Attitude strength: an overview

  • N.G. Mankiw

    Smart taxes

    Eastern Econ. J.

    (2009)
  • A.M. McCright

    The social bases of climate change knowledge, concern, and policy support in the US general public

    Hofstra Law Rev.

    (2009)
  • A.M. McCright et al.

    Anti-reflexivity: the American conservative movement’s success in undermining climate science and policy

    Theor. Cult. Soc.

    (2010)
  • Cited by (0)

    View full text