Determinants and Price Elasticity of Residential Natural Gas Demand in Belgium: A System GMM Approach

The global energy transition is changing how households use energy, making it important to understand how residential gas demand reacts to economic and policy shifts. Although heating decarbonisation has received growing attention, there is still limited evidence on household natural gas use based on large and detailed datasets. This paper helps to fill this gap by analysing residential natural gas demand in Wallonia, Belgium, using a unique panel dataset from the regional distribution system operator (RESA) covering over 250{,}000 supply points over 11 years (2009–2019). By combining these data with socio-economic, weather, and price information, we examine what drives household gas consumption and how it responds to price changes. To capture dynamic adjustments and control for unobserved differences between households, we apply a dynamic panel model using the Blundell–Bond system GMM estimator. The results show a short-run price elasticity of about –0.28 and a long-run elasticity of –2.22, suggesting that households adjust only slightly in the short term but much more strongly over time when faced with higher prices. We also find strong persistence in gas use and no clear differences in price responsiveness across income groups. Overall, the findings offer new insights into how households adjust their gas use over time and can help inform policies that manage the decline in residential gas demand in a gradual and socially equitable way, while supporting the broader transition to low-carbon energy systems.