How much do fuel taxes really cut fuel use?
30 September 2025
By Ander Iraizoz (Oxford University Centre for Business Taxation) and José M. Labeaga (UNED)
Fuel taxes are one of the most widely used tools for tackling climate change. By raising the price of gasoline or diesel at the pump, governments hope to discourage consumption and shift consumers towards cleaner alternatives. Fuel taxes are relatively easy to implement and directly target the environmental externality of CO₂ emissions.
The effectiveness of fuel taxes depends crucially on how consumers respond to higher prices. Do drivers cut back significantly their consumption when prices rise? This responsiveness is best measured by the price elasticity of fuel demand. If demand is highly elastic, even a modest tax increase could have a large environmental payoff. If it is inelastic, taxes increase revenue but do little to change behaviour.
Correctly measuring this elasticity is crucial for designing climate policy. In our recent paper (Iraizoz and Labeaga, 2025), we show that cross-border shopping for fuel can seriously bias elasticity estimates. When fuel taxes differ across regions, drivers may not simply cut down on fuel use in response to higher prices; instead, they may just cross into the neighbouring region to fill up the tank. If we do not account for this, it looks as if fuel demand is far more responsive to taxes than it really is.
Estimating the elasticity using spatial tax variations
A common way to estimate the elasticity of fuel demand is to exploit regional variation in fuel taxes (Li et al., 2014; Rivers and Schaufele, 2015; Tiezzi and Verde, 2016). For example, if one region raises its fuel tax while a neighbouring region does not, researchers can compare the resulting changes in fuel sales. As the two regions face the same national oil market and broader economic conditions, this comparison provides an accurate indication of the impact of higher prices on fuel use.
However, this strategy has a potential problem. Precisely because these regions are so integrated, drivers can easily cross the border to refuel in the cheaper region. This means that observed changes in fuel sales reflect not only how much people drive less, but also how much fuel they are buying elsewhere. This distinction is extremely important for climate policy because shifting purchases across space does not reduce emissions.
Spanish fuel tax reforms as a natural experiment
Spain offers a particularly useful setting to gauge the empirical importance of this issue. Between 2002 and 2019, Spain’s autonomous communities (its regional governments) were allowed to adjust their own fuel excise taxes, up to 4.8 cents/litre. Some regions raised the tax, while others did not, and the timing of these changes often differed. These asymmetric changes created significant price differences at regional borders. Figure 1, for example, illustrates the regional fuel tax bands across Spanish provinces in 2007 and 2017. The figures illustrate (i) the sizeable cross-sectional tax differences across borders, and (ii) the difference in taxes between the two figures indicates substantial variations in fuel taxes over time.
Figure 1: Regional diesel taxes in 2007 and 2017
(a) 2007
(b) 2017
In our study, we compiled monthly data on diesel sales and prices in Spanish provinces from 2007 to 2020. By tracking how sales responded not only to own region prices but also to price differentials relative to neighbouring regions, we were able to disentangle genuine reductions in fuel consumption from spatial substitution.
Key results
Our results show the effects of cross-border fuel substitution on the estimation of the price elasticity of fuel demand:
- Strong evidence of cross-border substitution. On average, a 1% increase in the relative price of diesel in one region compared to its neighbour resulted in a 2.8% fall in local sales as drivers switched to the cheaper region. We further verify that this effect is particularly large close to borders. Within 25 kilometres of a border, the elasticity of substitution jumps to –14.4.
- Ignoring substitution overstates fuel demand elasticity. If we naïvely estimate the elasticity looking at the effect of local prices on local sales, we find a value around –3.0, which suggests highly responsive demand. However, once we correct for substitution, the elasticity decreases to –0.7, indicating that demand is much less elastic.
Implications for climate policy
Our results have important implications regarding coordination of climate policies.
- Uncoordinated fuel taxes reduce emissions less than expected. The price elasticity of fuel demand and CO2 emissions should guide climate policy. Ignoring spatial substitution would lead us to conclude that a 1% increase in diesel prices would cut emissions by 1.6 million tons of CO₂ per year. However, once we correct for substitution, this figure falls to just 0.4 million tons, which is only a quarter of the original estimate. This stark difference shows the importance of policymakers carefully considering spatial substitution possibilities when evaluating the environmental effectiveness of taxes
- Coordinated fuel taxes are key to unlocking emission cuts. Effective climate policy requires overcoming political acceptability concerns for coordinated climate policy. Spain’s regional diesel tax was introduced partly to make taxation more politically palatable, with the revenue earmarked for healthcare (hence its nickname, the “health cent”). However, as drivers could easily refuel elsewhere, regions with higher taxes lost both revenue and effectiveness. This highlights a broader tension that governments should overcome: while local environmental policies can help governments tailor policies to voters’ preferences, a lack of coordination can erode results when goods are mobile.
Conclusions
Our research shows that when drivers can easily substitute purchases across borders, fuel demand is much less responsive to prices than sales data suggests. For researchers, this means being cautious about using regional variation in taxes as an identification strategy. For policymakers, it highlights the importance of coordinated action.
The lessons learned extend beyond Spain. In federal systems such as those of the United States and Canada, or in the European Union, tax jurisdictions have considerable powers in setting fuel taxes. However, unless neighbouring jurisdictions coordinate, spatial substitution may undermine both climate and fiscal goals.
Ultimately, if governments want fuel taxes to deliver meaningful reductions in emissions, they should endeavour to ensure that policies are consistent across regions. Otherwise, the environmental benefits could be lost.
References
Iraizoz, A. & Labeaga, J. M. (2025). Tax-Induced Spatial Substitution and the Price Elasticity of Fuel Demand. CBT Working Paper 2025-03.
Li, S., Linn, J., & Muehlegger, E. (2014). Gasoline taxes and consumer behavior. American Economic Journal: Economic Policy, 6 (4), 302–342.
Rivers, N., & Schaufele, B. (2015). Salience of carbon taxes in the gasoline market. Journal of Environmental Economics and Management, 74 (100), 23–36.
Tiezzi, S., & Verde, S. F. (2016). Differential demand response to gasoline taxes and gasoline prices in the U.S. Resource and Energy Economics, 44 (100), 71–91.
