Oil Price

Whether oil prices matter for the real and financial sectors of the economy is a multi-decade long question. The common wisdom on the street is that it does matter. Media appears to have helped solidify this perception; so much so that many (myself included) considered it a foregone conclusion. Research in the recent times appears to gainsay this; at least for the developed economies.

I ran an asymmetric copula test with time-varying parameters to capture how the extreme oil movements sync up with stock prices. The test’s interpretation is as follows: if the value is large, either or both of variables have predictive values for the other. If the sign is positive, they share a common direction. If not, there may be capital flight implications. The graph below is a representation of my early finding. Check the first panel for stocks. I only used S&P 500 here. Will add FTSE-100 and DAX results later and perhaps some emerging indices if time permits.

My result appears to align with recent efforts of researchers. Ordinarily, this would have been the end of the story. However, some brilliant researchers have been churning several indices of late capturing various dimensions of risk. One such effort I came across recently is by Prof. Sydney Ludvigson. When I plugged in the same codes for three types of uncertainty she quantified–Financial, Real Economic, and Macro–the results are starkly different. The co-movements in level and shocks are much stronger compared to stocks. Furthermore, if you notice the density plots and histograms below, the dispersion of the dependence parameters suggests that although oil prices may matter less for the stock markets (at least in terms of prediction or arbitrage value) it still matters as a conduit of investor sentiment. It’s also a lot more uncertain. Check the boxplots for clarity.

I haven’t yet looked at how these dependencies themselves co-move, or whether there are seasonal/cyclical elements in there–frankly, I suspect there is. Neither have I controlled for business cycle variables (yet). Once I do, a richer picture of the oil-finance relationship is likely to surface.I will update this post if/when I find something worthwhile.