Wednesday, May 4, 2011

Energy Prices and US Recessions (II)

Click for larger version in new window

Yesterday, I explored the changes in energy prices before and during US recessions from 1926 on using the Producer Price Index for energy goods (PPIENG).  One obvious problem with that analysis is that it confounds changes in energy prices with changes in the general price level, and the latter were larger in the 1970s than they have been since.

So to address that issue, and just to illuminate the subject from a different angle, I tried dividing out the PPIENG by a measure of broad inflation.  One obvious possibility is to use the consumer price index for all goods and services exclusive of energy (since we ideally don't want energy price changes in both the numerator and the denominator).  However, that series only goes back to 1957.  So before 1957, I use the regular CPI.  Just to give you a feeling that this is reasonable, here's the two from Fred again.

So then I take the ratio of the PPIENG energy index to the broad CPI.  That basically tells us the level of energy prices relative to overall prices of goods and services in the economy, and is the black line in this picture:

I really encourage you to click on this for the big version, as it packs a ton of information. The blue bars are recessions.  You can immediately see that the trend break after 1970 that I talked about yesterday is a very real feature of the black curve.  Before 1973, energy prices were not terribly volatile, were generally declining, and seem to have nothing to do with recessions.  After 1973, energy prices are much more volatile, and there is a strong association between energy price spikes and recessions, with all recessions having an associated spike just before or in the early stages of the recession, and only one prominent spike lacking an associated recession (that in 2005).

The most obvious cause of the regime shift is the peaking of US domestic oil supply (the green curve on the right scale), causing a shift to the US being heavily dependent on importing oil, and therefore at the mercy of OPEC control of the global oil supply.


bordoe said...

You may want to look at a chart like this:

Basically shows, to me, that the manufacturing sector had a very difficult time dealing with higher commodity prices starting around 2005, and it made sense to start laying people off due to margin squeeze.

Then obviously, there was a point when this became a very large move, but it was years after the 'real goods manufacturing' number turned negative.

Unknown said...

A quick googling netted the following paper, that seems to argue that there have been no energy price induced recessions since 1986, and that claims to explain why there was none in 2005.

Myself, I'm just not sufficiently versed in the matter to really comment. But from looking at the pretty pictures, it would seem they get rid of 1990 and 2001 by simply increasing the averaging interval till they go away.

Anonymous said...

This is a really super graphic. Nice job.

The really striking thing to me is the increased (and, I think, increasing) volatility. Such volatility makes planning for energy expenditures difficult.

CPI has undergone substantial redefinition over the past 30 years. I wonder what would happen if you used the ShadowStats series. Generally, the changes have tended to lower the official CPI number compared to previous methodologies. That should result in an increased denominator for the CPI beginning in the 80's. The larger denominator would tend to reduce the values of the series, so I guess both the scale and volatility would be somewhat lessened.

Interesting stuff.

Stuart Staniford said...

Bordoe - yeah, I agree, and it's similar in the GDP numbers. A growth slowdown starts in 2005, and worsens after that, but doesn't get to the stage of outright recession until 2007/2008.

Stuart Staniford said...


Thanks! I'm rather proud of it myself :-)

Stuart Staniford said...

Note that Jim Hamilton has a discussion today of the implications of current energy prices for a recession. He is always worth reading.

bordoe said...

Looks like we've got another recession rolling in:

Unknown said...

Looking at the longer trend I wouldn't read too much into the spike:
That's not to say there is no recession rolling in, or that the arguments made to explain away the spike ain't hollow...