Fuel taxes as an energy saving policy - potential misperceptions of
dynamic effects
Stian Blackstad Hackett
University of Bergen, Department of Geography
Annual Conference of the System Dynamics Society, St.G allen 2012
Abstract
A road fuel tax can be a cost-efficient policy instrument that internalizes some of the external costs of driving such as
air pollution, greenhouse gas emissions, congestion and noise, and may also stimulate conservation and gradually
improve the fuel economy of the car fleet. Fuel taxes are however often unpopular among the public, even when
there is popular support for the principle of governmental policy to mitigate pollution and improve energy efficiency.
This study is a first step in examining whether some of the opposition towards fuel taxes could be the result of
overlooking or misperceiving some of the time-dynamic effects in the fuel tax system. While principle-based tax
resistance may be unlikely to change quickly, resistance that is the result of misperceptions or overlooked factors
could potentially be reduced if the economic system of fuel taxation was better understood. First, in part 1, we
discuss the potential for overlooked factors and misperceptions. Second, in part 2 we use survey data from the
United Kingdom to test for one potentially overlooked factor in the fuel tax system directly, namely the effect of fuel
taxes on the average fuel economy of the car fleet. We find that the opposition towards fuel taxes is strong, and we
do not find evidence indicating that the effect of taxes on fuel economy is not considered. Respondents also provide
good estimates of the delay associated with changing the composition of the car stock. The results could be explained
by the finding that respondents do not seem to have a high willingness to pay for having a more fuel economic car
fleet as a country, even in a situation where high oil prices in the future are considered a risk. This indicates
principled opposition rather than misunderstandings or overlooked factors. Future studies could examine more of
the potential misperceptions and overlooked factors that are presented in part 1.
1. Introduction
While road fuel taxes are often favored by economists and policymakers, they are frequently
protested by the public. This conflict between policymakers and the general population is
potentially problematic when the public in principle supports government measures to reduce
greenhouse gas emissions and to use energy more efficiently (Kallbekken, Kroll et al. 2011), and
itis worthwhile to study why fuel taxes are opposed and how effective policies can be designed
in order to increase acceptance. As all environmental and energy-saving measures come with a
cost and fuel taxes are favored by economists for their cost efficiency, it is interesting to study
why they are so out of favor with the public.
This study has two parts. First, we use a theoretical model to show the potential misperceptions
of fuel tax effects in general, developing a general hypothesis that the hostility towards them may
partly be caused by misperceiving or not taking into account some of the effects of fuel taxes, so
that they are seen as more costly than they actually are. Second, we use survey data from a
sample of the adult population in the United Kingdom to test for one important potentially
overlooked factor, namely whether the effect of taxes on average fuel economy is considered.
Importantly, our study is focused on the energy-saving effects of fuel taxes and not their effects
on CO? emissions. While climate policy may be a strong motivation for fuel taxes, the belief that
emissions from human activity are influencing climate change may not have majority support. In
the United Kingdom for example, only 48% of those surveyed in a 2007/2008 Gallup poll said
they believed increasing temperature was caused by human activities (Gallup 2007/2008). The
figures for other European countries and the United States were in a similar range. When
examining what influences and may influence public opinion, it is therefore important to also
address the non-environmental effects of fuel taxes, even when the motive is emission reduction.
1.1.Opposition to fuel taxes - previous studies
Opposition towards fuel taxes has been extensively studied, and a large number of explanatory
factors have been presented. The variation in attitudes and perceptions is great among
individuals, but perhaps the most persistent factor, seen in multiple studies, is an apparent lack of
faith that the tax revenue will be fairly handled by the government, indicated by the fact that
earmarking tends to increase acceptance (Hsu, Walters et al. 2008; Kallbekken and Aasen 2010;
Kallbekken, Kroll et al. 2011; Salen and Kallbekken 2011). Dresner, Jackson et. al. (2006a) find
that the incentive effect of environmental taxes may be overlooked by the public in European
countries that have implemented environmental tax reform, although Kallbekken and A asen
(2010) find some evidence to the contrary in a focus group study in Norway. In Dresner, Jackson
et al. (2006a), where several European countries were compared, a common suspicion was that
the supposed environmental effects of a “green” tax reform were just being used to sell the policy
to the public, while the real motive was to raise more government revenue.
We know less about public attitudes towards fuel taxes as an instrument to improve fuel economy
and to reduce the general need for driving in society. In a poll by the New Y ork Times (2006)
somewhat confusing results were found; while a majority (85%) opposed a federal tax on
gasoline, 55% said they would support it if it would “... reduce the United States‘ dependence on
foreign oil”, and 59% would support it if it would “cut down on energy consumption and reduce
global warming”’. The willingness to use a higher gasoline tax to reduce income or payroll taxes
orto “pay for the war on terrorism" were significantly lower (28% and 24%’ respectively)
(Times 2006) This would seem to indicate that respondents did not believe a gasoline tax would
have much of an effect on gasoline use, oil dependence and the environment, which would be
consistent with the findings of Dresner, Jackson et al. (2006a). Possibly, this could be explained
by the perception that there are few possibilities to cut down on driving in the short term, and / or
that drivers are basically stuck with the car that they have for a long time. In other words, there
may be a conscious or unconscious preference for weighing short-term costs higher, while the
potential for cost reduction in the long term is either forgotten or highly discounted. This
hypothesis forms a basis for our study.
* An interesting finding when taking into account that this number may be higher than the percentage of people who
believe global warming is a man-made problem - a 2007/2008 Gallup poll found this to be 49% in the United States.
See Gallup (2007/2008). "Poll: Awareness, Opinions About Global Warming Vary Worldwide, see
http://www.gallup.com/poll/117772/A wareness-O pinions-G]obal-W arming-V ary-W orldwide.aspx #2."
2 When answering this question, respondents had by then been asked whether they would support a $2/ gallon tax
increase. While the question of willingness to pay for the war on terrorism did not present any particular tax level,
the answer could potentially be influenced by having brought a salient cost to mind in a previous question.
2. Fuel tax effects, potential misperceptions
In the following, we describe a theoretical model of some of the more complex effects of fuel
taxes and illustrate it by causal loop diagrams (CLDs). For an introduction to CLDs, see Sterman
(2000). With formalization, the model could be simulated with realistic parameter values, but this
is not done here. As the mechanisms are not overly complex the diagrams should be sufficient to
explain the model, and the purpose is simply to communicate the mechanisms, not to replicate
historical behavior or to make numerical predictions.
2.1. Effect on fuel economy, recycling of tax revenue
revenue recycling
3" {%, %} i
disposable income of + disposable income of
nom drivers (Ey) drivers (E/yr) _
%
revenue collected
(Eh)
fuel tax rate (£/1)
oil price (£/) +
retail fuel price _+
eh) “fel wed tne —
yo E
demand for fuel economic, wy cost () a
electric and AF vehicles
(carsiyr)
er Sw cost per L/S
; average fuel intensity (Efsam)
Cost of fuel intensity ~~ of car fleet (km) +
reduction (£/l/km/car)
. fed
The diagram shows how a higher retail fuel price will increase the average cost of driving per km
in the short term, but there is an offsetting effect from the slower, gradual transition to lower fuel
intensity in the car fleet. The effect of fuel prices on fuel economy follows from standard
economic theory and has empirical support (Clerides and Zachariadis 2008; Schipper 2008) The
magnitude of this effect depends on the costs of reducing fuel intensity - not only in direct costs,
but also by the decreased utility for car buyers who have preferences for larger cars, for example.
The initial reduced driving may also put some downward pressure on the fuel prices set by
retailers, although this effect may be uncertain. Expenses from fuel consumption drain the
budgets of drivers and balances driving, but their budgets are also, to an extent, replenished by
the recycling of the tax revenue. The “revenue recycling factor” broadly represents governmental
spending policy. It is represented as the portion of fuel taxes that are directly or indirectly
retumed to the economy, whether by direct payments, the reduction of other taxes or by
providing services that would otherwise have to be paid for privately. The factor also
incorporates the population distribution between drivers and non-drivers, allocating part of the
revenue to each group depending on government policy. Unless the redistribution is done by
direct cash payments or reductions of other taxes, the effect on disposable income will typically
be delayed, as it takes time for the government to establish services that can save voters money.
The revenue recycling factor is a crucial element when addressing both the efficiency and the
acceptance of fuel taxes. If the government is wasteful with the tax revenue it collects, taxpayers
get little back, and there may be undesirable distribution policies. But it is hard to imagine an
economy where taxpayers get nothing back - the factor should at the least be larger than 0,
meaning that the actual cost of driving is not increased by an amount directly proportional to a
fuel tax increase. The delay associated with revenue recycling does however mean that there is
lag between paying the tax and getting a government service back. Even direct cash rebates or
reduction of other taxes may come with a delay, although probably shorter.
2.2. Domestic oil production
For countries with domestic oil production some of the price paid at the pump eventually comes
back to the population, so that for large producers, high oil prices do not necessarily increase the
price of fuel relative to income. A gain, however, this depends on whether the country’s oil
industry actually benefits the population (including drivers) in general, and the effect may be
weak is ownership is concentrated or government institutions are functioning badly. This is
illustrated by the “distribution factor for oil revenue”, which, if high, will recycle revenue from
oil production back to the population with a delay. In other words, seeing high prices at the pump
because of high oil prices should theoretically be seen in a less negative light in countries that are
large producers and where the oil industry benefits the population, whether through ownership,
employment or redistributed tax revenue from the industry.
distribution factor for
oil revenue (%)
+ revenue recycling
redistributed of! 4 factor (%, %} f
revenue (E/yr) a x
+ disposable income of
+. diposable income of
non-drivers (£/yr) rivers (Ei)
+, domestic income from ¥
oil production (E/yr)
domest of ro
production
| aor &
Sy \ fuel tax rate (EI)
"
cil price (EA) v a
P retal fel price ee
n (C) fuel used (yx)
demand for fuel economic, pelea
electric and AF vehicles Ea
(carsyn)
cost per km
Cost of fuel intensity average fuel intensity (E/kam)
erage fue intensity
reduction (EM/kmcat) ~~ of car Beet (ihm)
2.3. Large-market effects on oil prices
If a country’s market for fuel is large, fuel taxes may have a significant effect on the global price
of crude oil. Initially, an increase in the fuel tax rate would be expected to decrease fuel demand,
putting downward pressure on the global crude oil price as well as on fuel producers and retailers.
The longer-term expected effect would be an improvement in average fuel economy that could
reduce demand, and thus oil prices, further. This effect has been described in more detail by Stoft
(2008).
nye
4
fuel tax rate (£/)
~_oil le +
retail fuel price +
+ “fuel used (lyr)
demand for fuel economic,
electric and AF vehicles
(cars/yr)
wy
average fuel intensity
° of car fleet (/km) ———————
2.4. Large-market effects on automobile producers
Large automobile markets may have an effect on producers. If higher fuel prices result in
increased demand for more fuel economic cars, electric cars or alternative-fuel vehicles, large
markets should be expected to have a bigger influence. The effect is especially important if we
consider learning curves and the potential to reduce production costs with experience and
increased scale, which holds the potential for crossing a “tipping point” and thus tap into positive
feedback loops for diffusion (Struben and Sterman 2007).
distribution factor for
oil revenue (%)
+ revenue recycling
redistributed oil factor {%, mY Se S
pe +
reveme (Ef) — fp
+ disposable income Ao er err income of
nore drivers ( w drivers (Ei)
4, domestic income from
oil production (£/yr)
+ x ar
domestic oil
production a
° fuel tax rate (EM)
il price (EM) \:
oil price +
— diving (kevyn)
retail fuel price + hee
(EN) ~—~Tuel used (yn)
at
demand for fuel economic, total cost (E/yr)
electric and AF vehicles
Cost of fuel intensity (carsiyn)
redcsin Eanfan \ _ cost per km
?
f \ average fue intensity (Efkm)
| = * of car fleet (Vim),
| +
production cost of fuel Production scale
economic, electric and AF _——(carsiyr)
vehicles (E/car) ~~ |
+
Experience effect (% Ean )
cost reduction) _
For an introduction to learning curves and scale effects, see for example Baloff (1971), Argote
and Epple (1990) and Struben and Sterman (2007).
2.5. Summary
The causal model described above illustrates some of the more complex and dynamic effects of
fuel taxes. The complex relationships indicate that the potential for misperceptions is there,
especially if we consider the apparent difficulty people may have with understanding feedback
mechanisms and time delays (Sterman 1989b; Moxnes 1998). Several effects are beneficial,
indicating that the costs of a fuel tax increase should not be measured as a 1-to-1 relationship.
The delay of the effects may however be long, and it may therefore he difficult for an individual
driver to identify what he “gets back” for the taxes, while the price at the pump is more salient.
The model would predict that fuel taxes, with all other factors equal, should be seen in a more
positive light in countries that
¢ Have well-functioning political institutions to ensure that the use of revenue is in
accordance with public opinion
¢ Have large enough fuel and automobile markets to influence the crude oil price and
automobile manufacturers
¢ Are net oil importers - especially if the market size indicates that oil prices could be
lowered by a fuel tax
The model could be expanded. For example, patterns of settlement or localization, the prevalence
of bike paths, public transportation and urban parking spaces may be influenced by long-term
fuel tax policies.
3. Survey design
A survey was conducted in the UK through the polling company Y ouGov. From their database of
respondents, a weighted sample of 2713 adults (>18 y) in the UK was drawn, and responded by
means of an online questionnaire on January 22" -24", 2012. The sample is designed and
weighted according to demographics” in order to form a close representation of the UK adult
population.
The respondents answered the following 5 questions:
1. The price, per litre, of petrol or diesel in Britain is about £1.35 at the pump. This price
includes a fuel duty (tax) of about 60 pence per litre. What do you think the British
government should set the fuel duty at, starting in 2012?
1. I think the government should set the fuel duty to ___ pence/litre
2. Don’t know
2. As well as a fuel duty of about 60 pence per litre, petrol and diesel prices also reflect the
costs of crude oil. Now, assume you were convinced that world crude oil production was
going to decline in the next ten years and that very high and lasting crude oil prices could
be expected from about 2020. What do you think the British government should set the
fuel duty at, starting in 2012?
1. I think the government should set the fuel duty to ___ pence/litre
2. Don’t know
3. A car’s fuel consumption per mile affects the costs of driving. The lower the average fuel
consumption, the less vulnerable the individual driver and the nation is to high crude oil
prices. The UK has had fuel duties on petrol and diesel for many decades. Do you think
these fuel duties have influenced people’s choices of cars? Fuel duties have made people
choose cars with:
Much lower fuel consumption per mile
Lower fuel consumption per mile
Same fuel consumption per mile
Higher fuel consumption per mile
Much higher fuel consumption per mile
Don’t know
Sc oie
3 See http://www.yougov.co.uk/about/about-QA .asp for details about the weighting procedure.
4. Please assume that because of high fuel prices, all people looking for a brand new car buy
a model with low fuel consumption per mile. If this was the case, how long, in years, do
you think it would be before 4 out of 5 of today’s cars were replaced with these new low
consumption models? If you are unsure, please make your best guess.
1. About years
2. Don’t know
5. The last two questions asked you about the effect fuel duties have on fuel consumption
and the time it would take for most drivers to switch to low consumption cars. Now that
you have had a chance to think about these issues, would you reconsider your answer to
the earlier question on the fuel duty? As well as a fuel duty of about 60 pence per litre,
petrol and diesel prices also reflect the costs of crude oil. Now, assume you were
convinced that world crude oil production was going to decline in the next ten years and
that very high and lasting crude oil prices could be expected from about 2020. Taking this
into account, what do you think the British government should set the fuel duty at, starting
in 2012?
1. I think the government should set the fuel duty to ___ pence/litre
2. Don’t know
3.1. Null hypotheses
Essentially, the same question is asked in Q1, Q2 and Q5 but with changed conditions: In Q2 and
Q5, a scenario of future crude oil scarcity is described, implying that the value of fuel economy
will increase. This question is included in order to induce motivation for those who may not
consider fuel economy to be relevant. Q5 differs from Q2 in that when answering Q5 the
respondent has answered question Q3 and Q4. These questions, although they do not provide new
information, are hypothesized to activate thought processes that could influence the respondent's
answer. For example, if a clear change was seen in Q5, we would have evidence to suggest that
the effect on fuel economy was not considered when answering Q1 or Q2. In all three questions,
respondents are informed that the current fuel duty is about 60 pence per liter*. This acts as a first
benchmark in each respondent's series of answers.
Additionally, we hypothesize that respondents who believe fuel taxes to have had a strong effect
on fuel economy will tend to view them more favorably, either in the “base” scenario (Q1), the
“scarcity” scenario (Q2) or in both.
Null hypotheses:
H1p: Responses to question Q1 and Q2 will not differ for the individual respondent.
H2o: Responses to question Q2 and Q5 will not differ for the individual respondent.
* At the time of the survey, the fuel duty was 58 pence per litre (See HM Revenue and Customs
http://www. businesslink.gov.uk/bdotg/action/detail ?itemId=1085465445&type=RESOURCES). Although this more
accurate number could be used in the study, we believe an approximation is sufficient for its purpose, as the answer
to Q1 is not its focus. Moreover, if people have a tendency to round off numbers, a benchmark of 58 p could mean
that some respondents who are uninterested or whose opinion would be best described as “no change” would answer
60 p (more likely than 56 p, for example), introducing upward bias.
H3o: Responses to Q1 and Q3 will be positively correlated.
H4o: Responses to Q2 and Q3 will be positively correlated.
The structure of Q1, Q2 and Q5 can be thought of as a “within- subject” design, in which the
interesting measure is how the individual respondent adjusts his or her answer when conditions
change. While a comparison of the mean responses gives a good first indication of the effect of
the introduced condition, important information may be overlooked if the magnitude of the
adjustment between two questions is differs between those who adjust their answer up and those
who adjust it down. A majority of respondents who adjust in a given direction may be canceled
out by a smaller group who make larger adjustments in the other direction. In terms of testing
systemic understanding, we consider the proportion of respondents that chooses a certain
direction of adjustment to be more interesting than the magnitude of adjustment, although
magnitude should not be overlooked either. For H3p and H4p, we use weighted least-squared
regression.
4. Results
In the following, entries refer to all valid responses except “don’t know” out of 2713 respondents.
A (w) denotes a weighted figure.
4.1. Responses Q1, Q2 and Q5
Mean (p/litre) Mean (w) (p/litre)__ Entries _Entries (w)
Ql 44,224 44,509 2091 2090,72
Q2 44,269 44,417 1824 1811,15
Q5 45,357 45,387 1893 1890,11
The outlier (ID=2326, Q1= 3000) was removed.
4.2. Responses Q3
Q3 asks respondents to evaluate the effect fuel taxes in the UK on the fuel economy of cars on
the road.
Fuel duties have made people choose cars % (w) % of entries (w)
with:
Much lower fuel consumption per mile 21 26
Lower fuel consumption per mile 44 54
Same fuel consumption per mile 12 15
Higher fuel consumption per mile 3 4
Much higher fuel consumption per mile il 1
Don't know 19
There is a clear tendency towards the opinion that fuel duties have influenced people’s choice of
cars. 65% (w) of respondents answered that the fuel duty has contributed to improve fuel
economy, and if “don’t know” answers are excluded, the figure is 80% (w). Whether these
numbers should be considered high or low is open to interpretation, but it seems clear that only a
small minority (4% and 5%) believe that average fuel economy has been negatively affected by
the fuel duty.
4.3 Responses Q4
Q4 asks respondents to make their best guess about how long it would take to replace 80% of the
car fleet. The outlier (ID =1064, Q4=1000) was removed.
Time to replace 80% of cars on road % (w)
Less than 5 years 7
5-9 years 20
10-14 years 27
15-19 years 8
20 years or more 12
Don't know 26
The mean response (weighted and unweighted) is 11,6 years. 8 respondents answered 100 years
or more; these could potentially be unserious entries or entries based on misunderstandings. If
these are considered outliers and removed, the mean response is 11,2 unweighted and 11,1
weighted. The weighted median when removing only ID 1064 is 8,2 years and 8,155 when
removing entries of 100 years or more.
There is a clear tendency for respondents to enter “round numbers” (5, 10, 15 etc.) in this
question, suggesting that intuition and guesswork may be prevalent. This would be unsurprising.
The most common response (including “don’t know”) is 10 years.
4.4. Change between Q1 and Q2
In total 1803 of 2713 respondents made valid entries to both Q1 and Q2. The outlier (ID=2326,
Q1=3000) was removed.
Unweighted Weighted Proportion (w)
#of respondents Q2 >Q1 279 310,74 0,17
#of respondents Q2 =Q1 1287 1247,31 0,70
#of respondents Q2 <Q1 237 236,27 0,13
Sum (n) 1803 1792,97 1,00
4.5. Change between Q5 and Q2
In total 1704 of 2713 respondents made entries to both Q2 and Q5. No outliers were removed.
Unweighted Weighted Proportion (w)
#of respondents Q5 > Q2 210 217,16 0,13
#of respondents Q5 =Q2 1347 1318,01 0,78
#of respondents Q5 <Q2 147 153,59 0,09
Sum (n) 1704 1688,76 1,00
4.6. Regression Q1 and Q3
Figure 4a shows the plot of Q1 vs. Q5.
Ql vs.Q3
250 *
200 rs
150 o 3 @ Qlvs.Q3
100 * —Linear (Q1 vs. 03)
50
0
0 1 2 3 4 5 6
Using weighted least squares regression, we find that the slope parameter is small and statistically
insignificant:
Standard error t p-value
R 0,019
R? <0,001
Standard error of the 26,95
estimate
Intercept 43,76 1,1661 26,353 <0,001
Slope 0,63 0,777 0,808 0,419
4.7. Regression Q2 and Q3
Q2vs.Q3
600
500 °e
400
300 @ Q2vsQ3
3 $ —Linear (Q2 vs Q3)
Using weighted least squares regression, we find that the slope parameter is small and statistically
insignificant:
Standard error t p-value
R 0,027
R? 0,001
Standard error of the 28,92
estimate
Intercept 42,94 1,899 22,613 <0,001
Slope 0,97 0,887 1,087 0,277
4.8. Results summary
A clear majority of respondents answer in Q1 that they would prefer a lower fuel duty than today,
and the mean preference for the duty level is considerably lower than the current fuel duty. When
the future “scarcity” scenario is introduced in Q2, the most common response is to not adjust the
desired fuel duty in any direction (70% (w) of entries), and the percentage of unchanged answers
between Q2 and Q5 is even higher (78% (w)). For those who do adjust, the tendency is upwards
adjustment both from Q1 to Q2 and from Q2 to Q5, and both are significant at the 0,01 level if
only entries that make adjustments are considered (using a normal binomial sign test which
excludes neutral entries). However, the more interesting result is the low percentage of change,
meaning that on average, the prospect of future oil scarcity does not change opinions of the level
of fuel taxes today. The low rate of change between Q2 and Q5 would indicate that the effect on
fuel economy probably was considered when answering Q1 and Q2.
Responses to Q3 and Q4 show that most respondents believe fuel taxes contribute to fuel
economy in the car fleet, and that it takes time to replace it. The average scrapping age of cars in
the UK is approximately 13,2 years, according to the British newspaper The Telegraph (2011).
This would suggest that the respondents’ mean estimate of 11,1 years (w) and mode of 10 years
to replace 80% of the fleet seems fairly accurate.
5. Discussion
The majority of respondents believes fuel taxes have contributed to improving the fuel economy
of the car fleet and are aware that the process may take a long time, but do not see a scenario of
future oil scarcity as an argument for neither a lower nor a higher fuel tax today. The findings
indicate that it is unlikely that voters are simply not thinking about the effect on fuel economy, or
that they believe the car fleet could quickly be replaced if prices should soar. This rules out two
potential misunderstandings.
We are still left with many potential contributing factors to the strong opposition towards fuel
taxes as a Pigouvian instrument for energy policy. A useful distinction may be made between
misperceptions and principled opposition - while clearing up a misperception has the potential to
change opinions, opposition based on principles is unlikely to change as easily. For example, if
one takes the normative “libertarian” view that government should not interfere in the price-
setting of any product unless there are externalities, even when it believes people are acting
against their own interests, only a clear demonstration of externalities can be convincing. And in
the case of burning fuel, only about half of the British population believes in its main potential
extemality - climatic change (Gallup 2007/2008). Still, it should not be overlooked that the
survey found little support for abolishing the fuel duty altogether and overnight, with a mean
desired level of ca. 45 p/litre from 2012. This would make fuel in the UK relatively cheap
compared with many OECD countries, but still clearly more expensive than in the United States.
The fact that the average fuel economy of the car fleet has steadily improved in Britain since
1970 (Schipper 2008) could suggest a higher “tolerance” for taxes, and the effect of localization
and settlement patterns, public transportation and other infrastructure on the need for driving may
also be significant. There may be a reinforcing feedback loop at work which is stronger in the
United States than in Britain, similar to the one described in Hsu, Walters et al. (2008): As long
as fuel taxes and prices are low (as they have long been in North A merica), infrastructure and
average fuel economy slowly adapts to a state that requires more fuel per unit of economic
welfare produced, which leads to strong political opposition towards raising fuel taxes. The result
is a lock-in situation with a transportation sector that may be vulnerable to rapid oil price
increases, but where it is politically infeasible to raise taxes as a policy instrument.
Drawing on the model in section 2, we would expect that the potential effect on the oil market
and automobile producers would be less for the UK than for a larger market, such as the United
States. In theory, this could explain some of the opposition in the UK and in smaller countries,
but it is unknown whether these arguments are considered at all by the general public.
6. Conclusion
The study presents a theoretical model which summarizes some of the dynamic effects of road
fuel taxes, drawing on standard economic theory. The model’s purpose is to show the potential
for misperceptions and overlooked factors that could lead to a view of fuel taxes as more
expensive than they actually are. We then test public opinion for attitudes towards using fuel
taxes as a policy instrument to improve fuel economy of the car fleet, and find that while the
effect is considered, opposition is still strong and unchanged.
References:
Argote, L. and D. Epple (1990). "Learning Curves in Manufacturing." Science 247(4945): 920-924.
Baloff, N. (1971). "Extension of the Learning Curve -- Some Empirical Results." Operational Research Quarterly
1970-1977) 22(4): 329-340.
Clerides, S. and T. Zachariadis (2008). "The effect of standards and fuel prices on automobile fuel economy: An
intemational analysis." Energy Economics 30: 2657-2672.
Dresner, S., T. Jackson, et al. (2006a). "Social and political responses to environmental tax reform in Europe: an
introduction to the special issue." Energy Policy 34(8): 895-904.
Gallup (2007/2008). "Poll: Awareness, Opinions A bout Global Warming Vary Worldwide, see
http://www.gallup.com/poll/117772/A wareness-O pinions-Global-W arming-V ary-W orldwide.aspx#2."
Hsu, S.-L., J. Walters, et al. (2008). "Pollution tax heuristics: An empirical study of willingness to pay higher
gasoline taxes." Energy Policy 36: 3612-3619.
Kallbekken, S. and M. Aasen (2010). "The demand for earmarking: Results from a focus group study." Ecological
Economics 69: 2183-2190.
Kallbekken, S., S. Kroll, et al. (2011). "Do you not like Pigou, or do you not understand him? Tax aversion and
revenue recycling in the lab." Journal of Environmental Economics and Management 62: 53-64.
Moxnes, E. (1998). "Overexploitation of Renewable Resources: The Role of Misperceptions." Journal of Economic
Behavior and Organization 37(1): 1234-1248.
Schipper, L. (2008). Automobile Fuel; Economy and CO2 Emissions in Industrialized Countries: Troubling Trends
through 2005/6.
Sterman, J. D. (1989b). "Modeling managerial behavior: Misperceptions of feedback in a dynamic decision making
experiment." Management Science 35(3): 321-335.
Sterman, J. D. (2000). Business Dynamics - Systems Thinking and Modeling for a Complex World, McGraw-Hill.
Stoft, S. (2008). Carbonomics - How to fix the climate and charge it to OPEC, Diamond Press.
Struben, J. and J. D. Sterman (2007). "Transition Challenges for Alternative Fuel Vehicle and Transportation
Systems." MIT Sloan Working Paper No.4587-06. Available at SSRN: http://ssrn.com/abstract=881800 or
http://dx.doi.org/10.2139/ssm.881800.
Seelen, H. and S. Kallbekken (2011). "A choice experiment on fuel taxation and earmarking in Norway." Ecological
Economics 70: 2181-2190.
Telegraph, T. (2011). Ten facts about the UK car industry. The Telegraph July 6th 2011.
Times, N. Y. (2006). New Y ork Times Poll, February 22-26, 2006. . New Y ork Times. Available at
/http://www.nytimes.com/packages/pdf/national/20060228_
poll_results.pdf.
