Causality between energy and output in the long-run

Abstract

Abstract:Though there is a very large literature examining whether energy use Granger causes economic output or vice versa this literature is fairly inconclusive. Almost all existing studies use relatively short time series or panels with a relatively small time dimension. Additionally, many recent papers continue to use what seem to be misspecified models. We apply Granger causality and cointegration techniques to a Swedish time series data set on energy and economic growth spanning 150 years to test whether increases in energy use and energy quality have driven economic growth. We show that these techniques are very sensitive to variable definition, choice of additional variables in the model, and sample periods. All of the following appear to make a finding that energy causes growth more likely: using multivariate models, defining variables to better reflect their theoretical definition, using larger samples, and including appropriate structural breaks. However, it is also possible that the relationship between energy and growth has changed over time and that results from recent smaller samples reflect this. Energy prices have a significant causal impact on both energy use and output.

Determinantes do risco de crédito rural no Brasil: uma crítica às renegociações da dívida rural

“…Nesse sentido, um passo comum em estudos de cointegração é se aplicar testes de causalidade de Granger para se investigar sobre o sentido de causalidade das variáveis (Stern & Enflo, 2013). De fato, diz-se que uma variável x Granger causa outra variável z se esta pode ser mais bem prevista se utilizarmos a história de x e z do que apenas a própria história de z .…”

“…De fato, diz-se que uma variável x Granger causa outra variável z se esta pode ser mais bem prevista se utilizarmos a história de x e z do que apenas a própria história de z . O conceito de causalidade de Granger se baseia na ideia de ordenação causal, segundo a qual duas variáveis podem ser contemporaneamente correlacionadas por acaso, mas é improvável que os valores passados de x sejam úteis para prever z se estiver se controlando para os valores passados de z , a menos que x realmente cause y (Stern & Enflo, 2013).…”

“…De fato, o pequeno poder do teste de causalidade de Granger pode ser agravado se o número de defasagens no VAR for mal especificado (Granger, 1988), número insuficiente de observações, omissão de outras variáveis importantes e presença de não linearidades (Stern & Enflo, 2013). Dessa forma, optamos por manter no modelo todas as variáveis em que as hipóteses nulas não foram rejeitadas em ambas as equações (7) e (8), no caso as variáveis end , end 2 , p , ic (vide resultados na Tabela 3).…”

Constitutive Explanations as a Methodological Framework for Integrating Thermodynamics and Economics

“…Accordingly, researchers who wish to expose the role of energy in growth do not use the market valuations but material measures of energy input, thus actually mixing up different measurement standards (in the corresponding production functions, capital is measured in monetary terms, labor in hours per annum, and energy in physical magnitudes, with different specifications, see e.g., [16]; [5]: 197ff, 334ff; [6]: 205ff, 262ff, who, however, proceeds with transformations into dimensionless magnitudes via referring to a base year). Interestingly, this raises similar questions for energy as have been raised for measuring labor, in particular with regard to the "quality" of energy (which compares with distinguishing educational levels, hence different qualities of human capital, in measuring labor input) [14].…”

“…As long as the specific causal mechanisms are not made explicit (in the case of money, mediated via the financial sector), econometric analysis cannot achieve unequivocal results. As for energy, this explains in a most simple way why releasing energetic constraints which became binding at a certain time may strengthen the econometric correlations, whereas under conditions of less binding constraints, such as resulting from technological improvements of energetic conversions and energy savings, these might become less significant for a considerable period of time [16]. Yet, this by no means implies that there are no causal mechanisms that link energy and growth in the latter case.…”