Gas demand in the Swiss household sector

ABSTRACT

In this article, we analyse the gas demand of Swiss households and provide an estimate of the own price elasticity. We use household-level panel data from 2010 to 2014 for 958 Swiss households while controlling for several socioeconomic characteristics and dwelling attributes. The results report the own price elasticity of gas demand to be around – 0.73. An inelastic demand is expected as the gas demand among Swiss households originates primarily for space heating and water heating purposes. Policy implications are discussed.

KEYWORDS:

• Natural gas
• energy demand
• panel data
• households
• Switzerland

JEL CLASSIFICATION:

• D12
• D13
• Q41
• Q48

Acknowledgments

We would like to thank the Swiss Federal Office of Energy (SFOE) and all the cooperating utilities. This research is also part of the activities of SCCER CREST, which is financially supported by the Swiss Commission for Technology and Innovation (CTI). Comments from an anonymous referee as well as from participants of the 2020 Empirical Methods in Energy Economics (EMEE) workshop are gratefully acknowledged. All omissions and remaining errors are our responsibility.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ The levy was CHF 12/tCO${ }_2$ in 2008 and is currently at CHF 96/tCO${ }_2$. The federal council is allowed to increase the levy up to CHF 120/tCO${ }_2$ if the emission reduction target is not met.

² This distinction is generally possible using dynamic partial adjustment models, such as those used in Alberini and Filippini (2011) and Alberini, Gans, and Velez-Lopez (2011).

³ This discussion pertains to the gas price variable. For the electricity price, we make use of an average price variable that has been constructed as a weighted sum of the individual peak and off-peak marginal electricity prices faced by the household according to the utility’s tariff sheets. As weights, the overall mean time-of-use share of all households was used (the mean peak time-of-use was 0.73). For households that do not have a time-of-use-based pricing, the marginal price of electricity was directly used.

⁴ This implies that if we have $M$ households living within a postcode area, we use the average value over the $M-1$ other households in the same postcode area. This price measure is exogenous to the household but represents the average price of the area. We also explored a very similar approach at the municipality level, instead of the postcode level. The sign and magnitudes of the estimated coefficients are found to similar, although the price elasticity of gas demand is found to be slightly lower ($-0.628$) compared to the results reported at the postcode level.

⁵ Households with unreasonably low (below $50$ kWh) and high (above $100,000$ kWh) annual gas consumption were filtered out from the final sample.

⁶ Blasch et al. (2018) a general comparison of the survey data with some of the available national and regional statistics and at the end finds the dataset to be largely representative of the urban and semi-urban regions of Switzerland in terms of household composition, dwelling size and age distribution. With respect to household gas demand, a reference statistic is not available for comparison.

⁷ The F-test for the instruments in the first stage confirms the validity of the instruments.

⁸ With increasing quantity of gas consumed, the average gas price decreases because the fixed-fee component is spread-out. For most private residences in Switzerland, in general, the marginal gas price per kWh is not tiered and not directly dependent on the quantity.