We analyze the dynamics in the global crude oil market based on a structural vector autoregressive model. We identify the model by presuming that reduced form residuals can be traced back to structural shocks that are independently distributed over the cross equation dimension. The resulting point estimates of the impulse response functions allow for a direct comparison with the outcomes of more conventional identification approaches. Our results are remarkably similar to the results regarding oil market dynamics in Kilian and Murphy (2012) and Inoue and Kilian (2013) even though they rely on statistical arguments instead of a set of theory-based a priori restrictions.
Based on the results from our statistical approach, we investigate the cumulative contributions of different oil shocks on the rapid fall in oil prices at the end of 2008 and 2014, as well as the effects of different oil shocks on macroeconomic aggregates in the US, the euro area, and China.