Skip to main content

Advertisement

SpringerLink
Log in

German car buyers’ willingness to pay to reduce CO2 emissions

  • Published:
Climatic Change Aims and scope Submit manuscript

Abstract

Motorized individual transport strongly contributes to global CO2 emissions, due to its intensive usage of fossil fuels. Current political efforts addressing this issue (i.e. emission performance standards in the EU) are directed towards car manufacturers. This paper focuses on the demand side. It examines whether CO2 emissions per kilometer is a relevant attribute in car choices. Based on a choice experiment among potential car buyers from Germany, a mixed logit specification is estimated. In addition, distributions of willingness-to-pay measures for an abatement of CO2 emissions are obtained. The results suggest that the emissions performance of a car matters substantially, but its consideration varies heavily across the sampled population. In particular, some evidence on gender, age and education effects on climate concerns is provided.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

Notes

  1. EU-15 comprises the 15 Member States prior to the 2004 enlargement of the European Union: Austria, Belgium, Denmark, Finland, France, Germany, Greece, Ireland, Italy, Luxembourg, the Netherlands, Portugal, Spain, Sweden and the United Kingdom.

  2. Improving the emission performance (i.e. fuel efficiency) of cars that run on fossil fuels reduces car travel costs. Consumers’ likely response to this cost reduction is an increase in car travel demand (implying the so-called “rebound effect”). Frondel and Vance (2009) investigate the determinants of car travel for German households and quantify the effect of fuel prices. Their results suggest that “the logic of introducing fuel efficiency standards to reduce emissions is dubious”.

  3. It should be noted that Hsu et al. (2008) coupled the question of gasoline tax increase with income tax reductions.

  4. In January 2009, the German government introduced government-financed trade-in incentives in order to stimulate demand for new cars and thereby to modernize the existing car fleet. Drivers who scrapped their at least nine years old cars received €2,500 for a new car (regardless of its fuel efficiency). Since our data was collected roughly one year earlier, it is not biased by this context.

  5. Within the survey both individual (75%) and group (25%) responses were allowed, whereas one individual was always designated to be the decision maker. Hensher et al. (2011) recently investigated in a vehicle choice context whether it makes a difference if one household representative or a group of decision-making household members is interviewed in order to reveal the household’s preferences. This might also be an interesting aspect for future research based on this data.

  6. Note that, though there is no official data on the income distribution of the target population, it seems that high-income households are also somewhat over-represented in the sample. However, 18% of survey respondents did not indicate their household’s monthly net income.

  7. Purchase price, fuel costs and engine power are standard explanatory variables in vehicle choice models (e.g., Horne et al. 2005; Ewing and Sarigöllü 2000; Brownstone et al. 1996; Brownstone et al. 2000; McCarthy and Tay 1998; McCarthy 1996; Bunch et al. 1993; Manski and Sherman 1980). CO2 emissions and fuel availability are used in only a few surveys (Horne et al. 2005; Brownstone et al. 1996; Bunch et al. 1993).

  8. Based on our data, possible class switching behavior cannot be observed. As noted by a referee, for instance, “some respondents may opt out of larger cars into a smaller vehicle in the presence of overt information about emissions”. In their recent study, Axsen et al. (2009) allowed vehicle classes to vary in the choice experiment.

  9. Because of this correlation between fuel consumption and CO2 emissions, fuel taxes are indeed working like a carbon tax. In Germany, fuel taxes are relatively high. For one liter of gasoline, for example, car drivers have to pay approximately €0.65 fuel tax. Moreover, the value added tax (19%) is added to the sum of the net fuel price and the fuel tax. Hence, the existing incentives for car manufacturers to develop low-emission (i.e. fuel-efficient) cars for the German market are high.

  10. Since, in the long term, there is no end-of-pipe technology that may address vehicle CO2 emissions, this is reasonable. Only for non-fossil fuels (i.e. biofuel, hydrogen, electric) we included the attribute level “no emissions”—since their in-use emissions are effectively zero. Biofuels may be considered CO2 neutral if they are the product of an entirely natural process of growth. However, emissions emerge in the course of the process of fuel production. Therefore, we also allowed positive CO2 emissions for non-fossil fuels. Respondents were informed about this context at the beginning of the experiment.

  11. For instance, hybrids might run on biofuels or LPG, so that the level of fuel availability can be lower than for gasoline cars; and depending on how the power is generated, an electric car can—at least theoretically—account for more emissions than an efficient gasoline-fueled car. Note that respondents were asked to treat all hypothetical alternatives so as they would exist.

  12. Hensher (2006): “As we increase the ‘number of alternatives’ to evaluate, ceteris paribus, the importance of considering more attributes increases, as a way of making it easier to differentiate between the alternatives. This is an important finding that runs counter to some views, for example, that individuals will tend to ignore increasing amounts of attribute information as the number of alternatives increases. Our evidence suggests that the processing strategy is dependent on the nature of the attribute information, and not strictly on the quantity.”

  13. Brownstone and Train (1999) and Brownstone et al. (1996) used data from a similar large choice matrix in their transportation studies.

  14. Note that the survey was conducted through personal interviews, which helped to guarantee the quality of the data. The personal interview situation motivated respondents to finalize the questionnaire (including the choice experiment) thoroughly and enabled respondents to avoid possible misunderstandings.

  15. Note that we assume β n to be constant over time for a given person n and, therefore, allow for correlation over time. This is reasonable since the repeated choices were all made within one interview.

  16. In a recent paper, Beck et al. (2011) analyzed data from a vehicle choice experiment among potential car buyers from the Sydney metropolitan area. Using also a mixed logit model, they found that less engine cylinders are preferred on average.

  17. McFadden and Train (2000), for example, specified fuel availability as normally distributed. During the model specification search for this paper, normal distributions for fuel availability and CO2 emissions have also been tried. The resulting share of sampled population with unexpected coefficient sign was between 10 and 15% for both. Hence, it is reasonable to assume that the unlike signs occurred purely by specification.

  18. Note that fuel costs and CO2 emissions each multiplied by minus one actually enter our models. This is due to the fact that a log-normally distributed coefficient has to be positive for all individuals. This conversion is undone after the estimation. See Hole (2007) for more details.

  19. During the model specification search, direct income effects on price sensitivity have also been tested; however, any significant income effect was not found.

  20. This value has been chosen since it is the sample mean and, in addition, almost the sample median.

  21. With HEEQ the general qualification for university entrance is meant. In Germany, the so called “Abitur” certificate can be received after 12 or 13 years at school (compared to other secondary school certificates after 10 or less years at school) and allows holders to attend university.

  22. During the survey interview respondents were asked to indicate the household’s monthly net income (possible ranges were: up to €1,000, between €1,000 and €2,000, between €2,000 and €4,000, or more than €4,000). Note that the gender (below |0.10|), age (below |0.15|), and education (below |0.20|) variables are only slightly correlated with the different income ranges—at least for those who did indicate their income level. Hence, there is no evidence that any identified gender, age, or educational effect might be some sort of income effect.

  23. According to Sælensminde (2006), lexicographic choices, which do not represent lexicographic preferences, may basically arise for two reasons: (1) simplification, as the choice task is too difficult, or (2) a study design using widely differing choice alternatives or attribute levels. Depending on the “sorting attribute” used by a respondent and the reason for his/her lexicographic choices, WTP estimates may be lower or higher than the real one.

  24. Note that for log-normally distributed variables (i.e. fuel costs, fuel availability and CO2 emissions) the presented estimates and standard errors for mean, median and standard deviation are computed after the estimation using Stata’s nlcom command, as described in Hole (2007). The Stata output after the mixlogit command gives the mean (b) and standard deviation (s) of the natural logarithm of log-normally distributed coefficients. The median, mean and standard deviation of the coefficient itself can be computed by \(\exp(b)\), \(\exp(b+s^2/2)\) and \(\exp(b+s^2/2)\times\sqrt{\exp(s^2)-1}\), respectively (Shimizu and Crow 1988). For fuel costs and CO2 emissions, actually, the median and the mean formulas have to be additionally multiplied by minus one. This is due to the sign change introduced in the estimation process (Hole 2007).

  25. Note that, in Germany, the annual vehicle tax for diesel-driven cars is higher than for gasoline-driven cars, while the tax on diesel fuel is lower than the tax on gasoline. Interestingly, Beck et al. (2011) report diesel to be the least preferred fuel type (behind gasoline and hybrid) in their Australian sample.

  26. If β~N(b,s), then (β − b)/s ~N(0,1). Thus, P(β < 0) = Φ( − b/s), where Φ is the cumulative standard normal distribution.

  27. It should be noted, however, that on average the indicated upper price bounds increase with income in the sample.

  28. Since July 2009, the annual vehicle tax is partly based on a vehicle’s CO2 emissions: for each gram exceeding the predefined level of 120g per km, €2 have to be paid each year. However, this holds only for newly registered vehicles. Note that respondents maybe anticipated this tax reform when making their choices. Beck et al. (2011) provide evidence that individuals are more likely to choose fuel-efficient vehicles if (annual or variable) emissions charging is present.

  29. It is important to note that the given WTP measures are point estimates which are measured with uncertainty. We also have to take into account the standard errors.

  30. The standard deviation with respect to a higher UPB is €296.85 (standard error: 86.11), and with respect to a lower UPB €103.60 (30.80).

  31. If X ~Λ(b,s), then P(X < x) = Φ((ln (x) − b)/s), where x > 0 and Φ is the cumulative standard normal distribution (Shimizu and Crow 1988).

  32. Findings of Viscusi and Zeckhauser (2006) are interesting in this regard. In 2004, they surveyed over 250 Harvard students, thus a group of relatively well-educated individuals. On average, the students estimate the climate-change-induced temperature increase in Boston consistently with the Intergovernmental Panel on Climate Change (IPCC) estimate. In their paper, Viscusi and Zeckhauser provide a rough calculation that converts the students’ willingness to pay higher gasoline taxes to curb climate change into an amount of 1,500 dollars per year.

  33. Brouwer et al. (2008) give an overview of the still very limited literature on WTP estimates for climate policy based on stated preference methods. It should be noted that most studies cited therein survey the WTP for the use of a tonne of CO2 equivalent rather than for its abatement.

  34. Note that the reason for abandonment is not identified and that in 2007 some 20% of abandoned cars were licensed again abroad.

References

  • Algers S, Bergström P, Dahlberg M, Dillén JL (1998) Mixed logit estimation of the value of travel time. Working Paper, Uppsala University, Department of Economics, Uppsala

  • Andersen S, Harrison GW, Hole AR, Rutström EE (2009) Non-linear mixed logit and the characterization of individual heterogeneity. Working Paper 6-2009, CBS Copenhagen Business School, Department of Economics, Frederiksberg

  • Axsen J, Mountain DC, Jaccard M (2009) Combining stated and revealed choice research to simulate the neighbor effect: the case of hybrid-electric vehicles. Resour Energy Econ 31:221–238

    Article  Google Scholar 

  • Beck MJ, Rose JM, Hensher DA (2011) Behavioural responses to vehicle emissions charging. Transportation 38:445–463

    Article  Google Scholar 

  • BMU (Federal Ministry for the Environment) (2008) Environmental awareness 2008. Federal Ministry for the Environment, Berlin

  • Brouwer R, Brander L, Van Beukering P (2008) “A convenient truth”: air travel passengers’ willingness to pay to offset their CO2 emissions. Clim Change 90:299–313

    Article  Google Scholar 

  • Brownstone D, Train KE (1999) Forecasting new product penetration with flexible substitution patterns. J Econom 89:109–129

    Article  Google Scholar 

  • Brownstone D, Bunch DS, Golob TF, Ren W (1996) Transactions choice model for forecasting demand for alternative-fuel vehicles. Res Transp Econ 4:87–129

    Article  Google Scholar 

  • Brownstone D, Bunch DS, Train KE (2000) Joint mixed logit models of stated and revealed preferences for alternative-fuel vehicles. Transp Res Part B Methodol 34(5):315–338

    Article  Google Scholar 

  • Bunch D, Bradley M, Golob T, Kitamura K (1993) Demand for clean-fuel vehicles in California: a discrete-choice stated preferences pilot study. Transp Res Part A Policy Pract 27(3):237–253

    Article  Google Scholar 

  • Carlsson F, Martinsson P, (2008) How much is too much? An investigation of the effect of the number of choice sets, context dependence and the choice of bid vectors in choice experiments. Environ Resour Econ 40:165–176

    Article  Google Scholar 

  • Caussade S, Ortúzar JdD, Rizzi LI, Hensher DA (2005) Assessing the influence of design dimensions on stated choice experiment estimates. Transp Res Part B 39:621–640

    Article  Google Scholar 

  • Daziano RA, Bolduc D (2011) Incorporating pro-environmental preferences towards green automobile technologies through a Bayesian hybrid choice model. Transportmetrica. doi:10.1080/18128602.2010.524173

    Google Scholar 

  • DeShazo JR, Fermo G (2002) Designing choice sets for stated preference methods: the effects of complexity on choice consistency. J Environ Econ Manage 44:123–143

    Article  Google Scholar 

  • DIW (Deutsches Institut für Wirtschaftsforschung) (2008) Wochenbericht des DIW Berlin Nr. 50/2008

  • EEA (European Environment Agency) (2008) Annual European Community greenhouse gas inventory 1990–2006 and inventory report 2008. Copenhagen

  • EU (European Commission) (2008) 20 20 by 2020: Europe’s climate change opportunity. COM(2008) 13 final, European Commission, Brussels

  • Ewing GO, Sarigöllü E (1998) Car fuel-type choice under travel demand management and economic incentives. Transp Res Part D 3:429–444

    Article  Google Scholar 

  • Ewing GO, Sarigöllü E (2000) Assessing consumer preferences for clean-fuel vehicels: a discrete-choice experiment. J Public Policy Mark 19:106–118

    Article  Google Scholar 

  • Frondel M, Vance C (2009) Do high oil prices matter? Evidence on the mobility behavior of German households. Environ Resour Econ 43(1):81–94

    Article  Google Scholar 

  • Greene WH, Hensher DA (2003) A latent class model for discrete choice analysis: contrasts with mixed logit. Transp Res Part B 37:681–698

    Article  Google Scholar 

  • Hanley N, Adamowicz W, Wright RE, (2005) Price vector effects in choice experiments: an empirical test. Resour Energy Econ 27:227–234

    Article  Google Scholar 

  • Hensher DA (2006) How do respondents process stated choice experiments? Attribute consideration under varying information load. J Appl Econ 21:861–878

    Article  Google Scholar 

  • Hensher DA (2010) Hypothetical bias, choice experiments and willingness to pay. Transp Res Part B 44:735–752

    Article  Google Scholar 

  • Hensher DA, Rose JM, Greene WH (2005) Applied choice analysis: a primer. Cambridge University Press, Cambridge

    Book  Google Scholar 

  • Hensher DA, Beck MJ, Rose JM (2011) Accounting for preference and scale heterogeneity in establishing whether it matters who is interviewed to reveal household automobile purchase preferences. Environ Resour Econ 49:1–22

    Article  Google Scholar 

  • Hersch J, Viscusi WK (2006) The generational divide in support for environmental policies: European evidence. Clim Change 77:121–136

    Article  Google Scholar 

  • Hole AR (2007) Fitting mixed logit models by using maximum simulated likelihood. Stata J 7:388–401

    Google Scholar 

  • Horne M, Jaccard M, Tiedemann K (2005) Improving behavioral realism in hybrid energy-economy models using discrete choice studies of personal transportation decisions. Energy Econ 27:59–77

    Article  Google Scholar 

  • Hsu S-L, Walters J, Purgas A (2008) Pollution tax heuristics: an empirical study of willingness to pay higher gasoline taxes. Energy Policy 36:3612–3619

    Article  Google Scholar 

  • Hunter LM, Hatch A, Johnson A (2004) Cross-national gender variation in environmental behaviors. Soc Sci Q 85:677–694

    Article  Google Scholar 

  • IPCC (Intergovernmental Panel on Climate Change) (2007) Climate change 2007—the physical science basis. Contribution of Working Group I to the Fourth Assessment Report of the IPCC, Cambridge University Press, Cambridge

    Google Scholar 

  • KBA (Kraftfahrt-Bundesamt) (2008) Statistische Mitteilungen: Fahrzeugzulassungen—Bestand, Emissionen, Kraftstoffe, 1 Jan 2008. Flensburg

  • KBA (Kraftfahrt-Bundesamt) (2009) Fachartikel: Halter der Fahrzeuge, 2 Nov 2009. Flensburg

  • Kolstad CD (2000) Environmental economics. Oxford University Press, New York

    Google Scholar 

  • Ladenburg J, Olsen SB (2008) Gender-specific starting point bias in choice experiments: evidence from an empirical study. J Environ Econ Manage 56:275–285

    Article  Google Scholar 

  • Ladenburg J, Olsen SB (2010) Gender anomalies in choice experiments—are biases really gender dependent? Working Paper 2010/1, University of Copenhagen, Institute of Food and Resource Economics, Copenhagen

  • Louviere JJ, Hensher DA, Swait JD (2000) Stated choice methods—analysis and applications. Cambridge University Press, Cambridge

    Book  Google Scholar 

  • Manski C, Sherman L (1980) An empirical analysis of household choice among motor vehicles. Transp Res Part A Gen 14(5–6):349–366

    Article  Google Scholar 

  • Mau P, Eyzaguirre J, Jaccard M, Collins-Dodd C, Tiedemann K (2008) The ‘neighbor effect’: stimulating dynamics in consumer preferences for new vehicle technologies. Ecol Econ 68:504–516

    Article  Google Scholar 

  • McCarthy P (1996) Market price and income elasticities of new vehicle demands. Rev Econ Stat 78(3):543–547

    Article  Google Scholar 

  • McCarthy P, Tay R (1998) New vehicle consumption and fuel efficiency: a nested logit approach. Transp Res Part E Logist Trans Rev 34(1):39–51

    Article  Google Scholar 

  • McFadden DL, Train KE (2000) Mixed MNL models for discrete response. J Appl Econ 15(5):447–470

    Article  Google Scholar 

  • MiD (2010) Mobilität in Deutschland 2008: Tabellenband. Bonn and Berlin

  • Moore WL, Holbrook MB (1990) Conjoint analysis on objects with environmentally correlated attributes: the questionable importance of representative design. J Consum Res 16:490–497

    Article  Google Scholar 

  • Murphy JJ, Allen PG, Stevens TH, Weatherhead D (2005) A meta-analysis of hypothetical bias in stated preference valuation. Environ Resour Econ 30:313–325

    Article  Google Scholar 

  • Requate T (2005) Dynamic incentives by environmental policy—a survey. Ecol Econ 54:175–195

    Article  Google Scholar 

  • Revelt D, Train KE (1998) Mixed logit with repeated choices: households’ choices of appliance efficiency level. Rev Econ Stat LXXX(4):647–657

    Google Scholar 

  • Revelt D, Train KE (2000) Customer-specific taste parameters and mixed logit. Working Paper, University of California, Berkley

  • Sælensminde K (2006) Causes and consequences of lexicographic choices in stated choice studies. Ecol Econ 59:331–340

    Article  Google Scholar 

  • Sampei Y, Aoyagi-Usui M (2009) Mass-media coverage, its influence on public awareness of climate-change issues, and implications for Japan’s national campaign to reduce greenhouse gas emissions. Glob Environ Change 19(2):203–212

    Article  Google Scholar 

  • Shimizu K, Crow EL (1988) Lognormal distributions: theory and applications. Dekker, New York

    Google Scholar 

  • Stern N (2006) The economics of climate change: the stern review. Cambridge University Press, Cambridge

    Google Scholar 

  • Torgler B, García-Valiñas MA (2007) The determinants of individuals’ attitudes towards preventing environmental damage. Ecol Econ 63:536–552

    Article  Google Scholar 

  • Torgler B, García-Valiñas MA, Macintyre A (2008) Differences in preferences towards the environment: the impact of a gender, age and parental effect. Working Paper 2008.18, Fondazione Eni Enrico Mattei (FEEM), Milan

  • Train KE (2003) Discrete choice methods with simulation. Cambridge University Press, Cambridge

    Book  Google Scholar 

  • T&E (Transport & Environment) (2009) Reducing CO2 emissions from new cars: a study of major car manufacturers’ progress in 2008. Brussels

  • Viscusi WK, Zeckhauser RJ (2006) The perception and valuation of the risks of climate change: a rational and behavioral blend. Clim Change 77:151–177

    Article  Google Scholar 

  • Zelezny LC, Chua PP, Aldrich C (2000) Elaborating on gender differences in environmentalism. J Soc Issues 56:443–457

    Article  Google Scholar 

Download references

Acknowledgements

I would like to thank Georg Bühler, Bodo Sturm, and four anonymous referees for valuable comments and suggestions. Funding from the German Federal Ministry of Education and Research is gratefully acknowledged (Förderkennzeichen: 07WIN16A).

Author information

Authors and Affiliations

  1. Centre for European Economic Research (ZEW), L7, 1, 68161, Mannheim, Germany

    Martin Achtnicht

Authors
  1. Martin Achtnicht
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Martin Achtnicht.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Achtnicht, M. German car buyers’ willingness to pay to reduce CO2 emissions. Climatic Change 113, 679–697 (2012). https://doi.org/10.1007/s10584-011-0362-8

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10584-011-0362-8

Keywords

Search

Navigation