This study assesses empirically the effects of oil price shocks on the real economic activity of the main industrialized countries. Multivariate VAR analysis is carried out using both linear and non-linear models. The latter category includes three approaches employed in the literature, namely, the asymmetric, scaled and net specifications. Evidence of a non-linear impact of oil prices on real GDP is found. In particular, oil price increases are found to have an impact on GDP growth of a larger magnitude than that of oil price declines, with the latter being statistically insignificant in most cases. Among oil importing countries, oil price increases are found to have a negative impact on economic activity in all cases but Japan. Moreover, the effect of oil shocks on GDP growth differs between the two oil exporting countries in the sample, with the UK being negatively affected by an oil price increase and Norway benefiting from it.

Oil and gold are used as investment assets and so they are closely related to the evolution of stock market indices, given that any influence on decisions about investment portfolios can affect stock market returns. Consequently, it is important for investors to analyze the direction of influence between crude oil, gold and stock markets when designing and implementing their investment strategies.Thus, this paper studies the direction of the causality among the three markets for the US case. In doing so, we apply linear and non-linear Granger causality tests for daily data from the Great Moderation onwards. Evidence is provided as to the importance of considering the possibility of nonlinear relationships between the three markets, a feature which cannot be revealed using conventional linear causality tests which would therefore lead to an information loss about the true link.The results for the full sample indicate that the causality goes in all directions, which implies that changes in the stock market returns may be monitored by observing changes in the returns of the two commodity markets considered (and vice versa). This may help to design substitution investment strategies. The results also indicate that the direction of influence between markets does not exhibit material differences between various subsamples, with the exception of the causality relationship between the two commodity markets. This may in part help explain the contradictory results and mixed conclusions found in previous related literature.

This paper presents evidence of a non-linear relationship between real GDP growth and oil price changes for the US economy. We also argue that this non-linearity is not merely due to the use of data from the mid-1980s onwards, as most authors, so far, seem to believe. In fact, we find the existence of non-linearity with the use of data earlier than 1984, and even before 1977. Furthermore, we question that the non-linear transformations of oil prices proposed in the literature are the most appropriate ones for reflecting such non-linearity.