Energy consumption and economic growth in Italy: A wavelet analysis

This paper investigates the relationship between energy consumption and economic growth with over eighty decades of Italian dataset. The wavelet analysis is applied to decompose series into different time scales whereas the frequency domain technique is used to examine time-specific shocks. Results of both unit root and stationarity tests indicate all series are integrated of order one, however, no evidence of long-run relationship is reported between energy consumption and economic development. We observe that the causal flow from economic growth to energy consumption becomes dominant at lower scales (up to 4 years), while at higher scales the strength of causality from energy use to growth declines. Therefore, the influence of energy consumption on economic growth can significantly be detected only at lower scales. If only original series and lower scales are considered, causal findings lean towards the feedback mechanism, with bidirectional causal relationship. This bidirectional causality is reinforced at all frequency bands, thus, causality from energy consumption to economic growth is observed only at frequencies between 1.3–1.8 (3.49–4.83 years) and 2.2–2.4 (2.61–2.85 years). However, when higher scales are considered, the causality test results are in line with the conservation hypothesis. More precisely, causality from economic growth to energy consumption is reinforced by frequency technique at higher time scales (8–32 years) but only at a frequency more than 0.6 (more than 10.47 years). The differences in the applied results provide alternative policy implications, justifying the use of wavelet approach to decompose time series into various time scales.