I decided to further look at yield spreads and put a bit of a local spin on it by adding in the element of North Dakota oil production. I am trying to figure out if rates and rate movements impact the oil industry in North Dakota. The interesting thing here is that the cost structure of the firms is really different.
Once wells are sunk there are large elements of, well, sunk costs that remove themselves from consideration by the firm at the margin. What I do not know, and really could not figure out from the news articles, is the specific maturity structure to the debt loads that a “typical” ND oil company (if such a thing exists). Knowing the maturity structure would give insight into the appropriate yields to use. Without that I just used the monthly yields on the ten and two year US Treasury securities.
I used monthly spread calculations here instead of daily because the North Dakota oil data is available in a monthly series. The North Dakota oil data looks like this:
I used the year-over-year percentage changes remove seasonal fluctuations in the data and put it into percentage terms as well. Running a preliminary VAR it seems that the fifth and sixth lags of the spread variable exert statistically significant influences over the oil production variable.
The differing scales for the variables and the volatility differences make a two y-axis graph difficult and probably not the most informative. The cost issues with oil production also raise questions about the lag structure as far as the response to interest rates. So this post probably raised more questions than it answered but I think it is likely the case that there was a response on the part of oil companies to cheap rates as documented in numerous papers and recent books. The question then gets to whether this is sustainable going forward and what the risks are for the state in particular.