A Time-Series Analysis of the Aggregate Income-Energy Consumption Nexus: The Case of Italy

Abstract

We empirically analyze the nexus between GDP and energy consumption in the 1970-2009 years for Italy, using a time-series approach. After a brief introduction, we present the discussion of the data. Stationarity tests reveal that both series are non-stationary, or I(1). Moreover, since both series show the presence of a structural break, a Gregory and Hansen cointegration test has been performed. The results evidence the presence of a long-run relationship. Causality tests reveal that the “neutrality hypothesis” emerges, insomuch as the absence of a causal relationship between energy consumption and real GDP is discovered. The IRFs analysis evidences that a shock to the energy consumption affects GDP for one period, but dies out very quickly. While shocks to GDP create a smaller but significant response in the energy consumption, although it falls to zero in few periods. Finally, we calculate with an ECM that the long-run multiplier is 0.70. The energy consumption will increase to correct the disequilibrium, with 11% of the (remaining) deviation corrected in each subsequent time period. In addition, a one-unit increase in the GDP immediately produces a 1.07 unit increase in the energy consumption.

Keywords: Energy policies, energy consumption, GDP, time series, Italy

JEL Classification: B22, C22, N54, Q43