Posted by John Irons at March 14, 2000 02:29 PM
Will higher prices mean a recession?

Economists tend to start biting their nails when they see the price of oil start to rise. One way to remember the timing of recessions in the US economy over the past 30 years is to think back to what are called “oil shocks” – a steep rise in oil prices. The most substantial shocks occurred in 1973-74, 1979-80, and 1991, and roughly correspond to the post-1970 recessions.

Since the price of oil has recently risen to over 30$ a barrel, a natural question to ask is if another recession is in the pipeline.

The first shock occurred at about the time of a long-run slowdown in productivity in the U.S. and much of the western world. The second shock came just before the severe recession in the early 80’s, and the final short-lived shock in 1991 occurred at about the same time as the short-lived 1991 recession.

Some economists are arguing that it may already be too late, the seeds of recession have already been sown. (For example, see A. Oswald's article (.pdf).)
 

Some data

Let's first look at the historical experience. The graph below looks at the behavior of oil prices and GDP growth over the past 40 years. While the relation is far from perfect, it does look as if rising oil prices may tend to lead to low real GDP growth, especially in the recession years as noted above.

A better way to examine this data is to look at a cross-plot of GDP growth versus the change in oil prices. The following graph looks at such a graph using Real GDP growth (vertical axis) and the percent change in oil prices (horizontal axis) from the prior year - since it takes some time for oil prices to fully impact the economy. The graph also contains a regression line which shows the relation between prices and output (standard deviation in parenthesis):

Interpretation

Given the fitted line, we see that a 10% increase in the price of oil would correspond to a substantial (and significant) 1/2 percentage point reduction in GDP growth.  Of course, I would like to warn everyone that these results are merely suggestive and in no way come even close to a full and careful econometric analysis - do not use this number as a serious quantitative measure of the effect of oil prices on the economy!

We would have to, at the very least, include in our analysis other factors that may impact GDP growth as well as to fully model the dynamics and allow for non-linear effects. Also, the above data is only annual. See papers by A. Oswald and J. Hamilton for more careful analyses.

The above analysis has also assumed that there is a causal relationship running from energy price increases to output declines. However, we might have the causality reversed. Growth tends to increase demand for energy, and hence higher GDP growth may push up oil prices. There may also be a “political economy” component, since the price of oil is affected by the production decisions of the OPEC countries and these OPEC countries may time their price increases in reaction to world economic conditions – thus complicating attempts to measure the effect of oil prices.

Recession? Not necessarily...

While there has perhaps been a relation between rising prices and GDP growth, we may not have to worry as much as in the past. Oil costs have been a shrinking part of the US economy in recent years. Since 1970, energy use as a faction of GDP has fallen by around 35%. (See graph)

If indeed the link between macroeconomy and oil prices were to weaken - it would be rather fitting: as soon as we seem to understand a macroeconomic relation, it tends to go away.

More Economics Features
Will oil end the boom? Post in the Forum.

Links From the Web

GDP Data
Oil Data - Monthly Energy Review from the Department of Energy.
Energy Outlook Annual
A. Oswald's home page
J. Hamilton home page

Posted by John Irons at March 14, 2000 02:29 PM