Demand for alternative-fuel vehicles when registration taxes are high

https://doi.org/10.1016/j.trd.2010.11.001Get rights and content

Abstract

This paper investigates the potential futures for alternative-fuel vehicles in Denmark, where the vehicle registration tax is very high and large tax rebates can be given. A large stated choice dataset has been collected concerning vehicle choice among conventional, hydrogen, hybrid, bio-diesel, and electric vehicles. We estimate a mixed logit model that improves on previous contributions by controlling for reference dependence and allowing for correlation of random effects. Both improvements are found to be important. An application of the model shows that alternative-fuel vehicles with present technology could obtain fairly high market shares given tax regulations possible in the present high-tax vehicle market.

Research highlights

► Alternative-fuel vehicles have potential in high-tax vehicle markets. ► Correlated random effects are important in vehicle choice. ► Controlling for reference-dependent preferences is important in vehicle choice. ► The hypothesis of loss averse individuals is confirmed in vehicle choice.

Introduction

Traffic growth makes the transport sector the main contributor to the growth in green house gas (GHG) emissions in Europe. One approach to diminish GHG emissions is to induce consumers to buy climate-friendly alternative-fuel vehicles (AFVs) more frequently, for instance electric, bio-fuelled, or hydrogen-fuelled vehicles.

Studies by Ewing and Sarigöllü, 1998, Batley et al., 2004 and others have concluded that the performance of AFVs must be improved to resemble that of conventional vehicles for AFVs to obtain significant market shares. AFV market shares may alternatively be increased through price reductions. The government can achieve this, e.g. by giving rebates in the registration tax for AFVs. In markets where the registration tax is not so high, it is limited how much the sale of AFVs can be stimulated in this way without actually subsidising AFVs.

This paper considers the potential market shares for AFVs in a Danish context. This is important because of the very high Danish registration tax. First, the high Danish car prices allow us to design a stated choice experiment in which the price of new AFVs can be considerably lower than the price of conventional cars, without such scenarios being rejected as implausible by respondents. Second, given the high prices we may investigate whether consumer preferences seem to be affected by these.

The stated choice experiments are pivoted around an actual vehicle purchase. The framing gives credibility to our choice experiment as respondents have recently faced a choice situation involving vehicles. Furthermore it provides an opportunity for testing whether the attributes describing the recent vehicle purchase set a frame for the stated choice experiment.1

Section snippets

Stated choice data

A stated choice survey was conducted because the analysis concerns vehicles with market shares close to zero. To add realism, the experiment was designed around a reference vehicle that the respondent had recently purchased. Besides adding realism to the choice task this allows the respondents to assume that alternatives are similar to the respondents’ reference vehicle with respect to characteristics not listed as attributes in the experiment, e.g. vehicle class.

The survey considered five

Modelling

We employ a mixed logit model with random effects, taking the panel nature of data into account (Train, 2003). The random effects follow a general multivariate normal distribution. We allow for reference dependence in the price coefficient. In the context of vehicle choice it seems reasonable to hypothesize that individuals develop a reference given their own vehicle purchase history. Here we assume that the reference is the most recent vehicle purchase. In a conventional model, the marginal

Discussion

Next, we compare our results to earlier studies. A reason why our results may differ from some earlier studies is that their experimental design makes the AFVs more expensive on average than the conventional vehicles, e.g. Batley et al. (2004). This means that these studies tend to measure the willingness to pay for AFV characteristics but not the reference-free valuations that we derive from our model. Given our results we would expect the reference-dependent willingness-to-pay measures from a

Conclusion

We have estimated and evaluated preferences for AFVs in Denmark. Using a mixed logit model, we describe individuals facing the choice among conventional vehicles and four types of AFVs. Improving on earlier studies, we control for reference dependence in the price coefficient and allow for correlation of random effects. The estimations give significant and positive fuel-specific constants showing that everything else equal individuals would be more likely to choose environmentally friendly AFVs

Acknowledgements

This paper is based on the research project “Market potential for renewable fuels passenger cars” funded by the Danish Council for Strategic Research. The authors would like to thank Katrine Hjort for her valuable comments on an earlier version of the paper.

References (12)

There are more references available in the full text version of this article.

Cited by (0)

View full text