International Choice Modelling Conference, International Choice Modelling Conference 2017

Font Size: 
The unintended consequences of carbon taxes on domestic fuel consumption
Flavio Freire Souza, Akshay Vij, Sangeeta Bansal

Last modified: 28 March 2017


Efforts to establish climate and energy policies that reduce greenhouse gas (GHG) emissions and set efficiency are growing globally. One such instrument widely discussed in global climate policy debates is carbon taxes, regarded as a cost effective instrument for mitigating GHGs. Carbon taxes impact fuel alternatives differently. Whereas the taxes can significantly reduce consumption of fossil fuels in regulated markets, the same cannot be said about biomass alternatives (e.g., firewood) due to their unregulated markets. Thus, on the one hand, an increase of fuel prices may lead to a reduction in total energy consumption. On the other, a reduction is likely to impact regulated energy markets, and substitution effects might nudge households toward more viable but less carbon efficient alternatives.

Empirical findings reveal that households store and use multiple fuel sources (not unlike menu choice), two on average (Gupta & Köhlin, 2006). Such findings stand in contrast with the energy ladder theory - which advocates for a shift from traditional biomasses towards cleaner and more efficient alternatives given an alleviation in the budgetary constraints of households. In fact, communities from different countries focus on different combinations of fuel sources. Heltberg (2004) points out that the most common combination in Guatemala, for instance, is firewood and LPG. In Vietnam, wood is complement with straw for most households (52%). Both firewood and kerosene are used for cooking in rural South Africa. Thus, it is important to understand how households select the set of energy fuels suitable to their needs.

There have been a number of attempts in the literature to model fuel choice behaviour of households. Rao and Reddy (2007) have used single discrete choice models to estimate the choice of primary fuel. While such an approach is convenient and tractable, it overlooks the fact that most households typically consume more than one fuel type. Gangopadhyay, Ramaswami, and Wadhwa (2003) relax this assumption by allowing households to choose different subsets of the full set of fuel types, but their approach does not account for differing levels of consumption of each fuel type.

In this study we investigate the suitability of a multiple discrete continuous framework to understand choices of fuel sources. In particular, we employ the framework extended by Bhat, Castro, and Pinjari (2015). It allows households to choose more than one fuel type and explicitly models the quantity consumed of each fuel. Complementarity and substitution effects between different fuels are captured by allowing the utility of each fuel type to be sensitive to the level of consumption of other fuel types. The impact of other variables on household fuel choices, such as income, religion and the amount of time women can spend at home, is captured through their effect on the baseline utility of each fuel type.

We use data from National Sample Survey (NSS) Reports of India. NSS is a nationally representative survey of over 100,000 households. The NSSO consumer expenditure surveys aim at generating estimates of household monthly per capita expenditure (MPCE) separately for the rural and urban sectors of the states and union territories and for different economic groups. The data has information on consumption of various fuels, household demographics and also total monthly per capita expenditure of households. We use the 66th - 68th rounds of NSSO surveys. This data set spans from 2009 -2012.

Understanding choices for energy solutions has important policy implications, particularly for families from developing countries. The indoor pollution produced from the consumption of traditional biomass fuels for cooking and heating can impact the health of families. Electricity boosts educational performance as families can study at night. The burden of collecting firewood and other materials often befalls on women, impacting on the amount of time they can spend with their children. They are also generally responsible for cooking and cleaning pots, activities that greatly benefit from the use of modern fuels. Additionally, fuel choices are inherently linked to their environmental impact, for example the deforestation of surrounding areas, the externalities generated from having multiple sources of air pollution impacting on the environment, and the accumulation of black carbon on snow and ice which results in accelerated melting.

India is one of the settings that would particularly benefit from a greater understanding of complementarity and substitution patterns of energy fuels. In alignment with the energy ladder theory, the Indian government has introduced a policy to alleviate the costs of cleaner fuel consumption for families from lower socio-economic background. Poorer families, more likely to resort to biomass alternatives, have the costs for the purchase of kerosene reduced. There is uncertainty, however, with regards to the ramifications from a reduction in the governmental budget allocated to this policy. On the one hand, households now familiar with the benefits from cleaner energy consumption could opt to continue enjoying such advantages. On the other, there is a chance they will fall back to firewood and other biomass fuels. This second scenario is also likely to take place if the government decides to follow through with establishing GHG taxes. Thus, to avoid these unintended consequences, it is crucial that demand models of fuel consumption correctly capture the underlying structure of household decision-making.



Bhat, C. R., Castro, M., & Pinjari, A. R. (2015). Allowing for complementarity and rich substitution patterns in multiple discrete–continuous models. Transportation Research Part B: Methodological, 81, 59-77.

Gangopadhyay, S., Ramaswami, B., & Wadhwa, W. (2003). India: Access of the Poor to Clean Household Fuels. Retrieved from

Gupta, G., & Köhlin, G. (2006). Preferences for domestic fuel: analysis with socio-economic factors and rankings in Kolkata, India. Ecological Economics, 57(1), 107-121.

Heltberg, R. (2004). Fuel switching: evidence from eight developing countries. Energy Economics, 26(5), 869-887.

Rao, M. N., & Reddy, B. S. (2007). Variations in energy use by Indian households: an analysis of micro level data. Energy, 32(2), 143-153.

Conference registration is required in order to view papers.