Working Paper: NBER ID: w24295
Authors: Maximilian Auffhammer; Edward Rubin
Abstract: Half of American households heat their homes with natural gas furnaces and 43% use it to heat their water. Hence, understanding residential natural gas consumption behavior has become a first-order problem. In this paper, we provide the first ever causally identified, microdata-based estimates of residential natural gas demand elasticities using a panel of approximately 300 million bills in California. To overcome multiple sources of endogeneity, we employ a two-pronged empirical strategy: (1) we exploit a discontinuity along the border between two major natural-gas utilities in conjunction with (2) an instrumental variables strategy based upon the differences in the utilities’ rules/behaviors for internalizing changes in the upstream natural gas spot market. We estimate that the elasticity of demand for residential natural gas is between -0.23 and -0.17. We also provide evidence of significant seasonal and income-based heterogeneity in this elasticity. This heterogeneity suggests unexplored policy avenues that may be simultaneously efficiency-enhancing–in the absence of first best pricing—and pro-poor.
Keywords: No keywords provided
JEL Codes: O13; Q41
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| PGE and SoCalGas price elasticity of demand for residential natural gas (L97) | price elasticity of demand for residential natural gas (Q41) |
| lower-income households elasticity during winter months (H31) | elasticity of demand for residential natural gas (Q41) |
| higher-income households elasticity during winter months (H31) | elasticity of demand for residential natural gas (Q41) |
| households respond to lagged prices (D19) | elasticity of demand for residential natural gas (Q41) |
| seasonal and income-based heterogeneity in elasticity (H31) | policies to enhance efficiency while being pro-poor (D78) |