The tail risk of a spike in oil prices

In judging risks surrounding the Israel-Iran conflict, the obvious key tail risk relates to a reduced supply of oil and a consequent spike in oil prices.

This is a tail risk from two well-known angles.

Firstly, Iran could attack shipping through the Strait of Hormuz, which accounts for about 20-25% of the world consumption of oil, much of it sourced from Saudi Arabia.

Secondly, regime change could limit Iranian oil production, at least for a time, where Iran accounts for about 5% of world output, nearly all bought by China.

Oil price shocks matter because they are commonly associated with deeper-than-normal recessions via their impact on spending and investment and with some central banks raising interest rates to combat higher oil-led inflation.

In a broad analysis of the characteristics of recessions across more than twenty advanced economies over most of the post-WW2 period, the IMF found that oil price shocks featured in about half of modern recessions.

The IMF’s analysis included a comparison of how key economic and financial indicators behaved across: (1) recessions without oil shocks; (2) recessions with oil price shocks (i.e., where there was a price rise of about 15% or more); and (3) recessions with a severe oil shock (i.e., a price rise of about 20% or more).

The IMF’s analysis is summarised in the charts below, which show the average %/pp/bp change in key indicators across the different types of recessions, where:

• Recessions with oil shocks are usually deeper than normal, with larger demand-led declines in output and larger rises in unemployment. The average decline in output is about 5½% in a recession with no shock, around 7½% in a recession with a shock, and about 9¼% in a recession with a significant shock.
  
  
• Oil price shocks are usually stagflationary, in that they feature both higher unemployment and higher inflation, whereas inflation typically falls in a recession without an oil shock. Unemployment increases by an average 1pp when there is a recession without a shock, 1¼pp when there is a shock, and about 1½pp in a severe shock. The comparable changes in inflation are a decline of about 1pp in a recession with no oil shock, a rise of about ¼-½pp with a shock, and an increase of ½pp in a recession with a significant shock.
  
  
• Central banks cut interest rates by broadly about the same amount across the different types of recessions – i.e., a range of about 75-100bp – but monetary policy is generally more stimulatory in recessions with oil shocks because real interest rates fall by more given higher inflation.
  
  
• Asset prices typically fall by more in recessions with oil price shocks. House prices drop by an average 4% in a recession without an oil shock, 5% when there is a shock, and about 6% when there is a severe shock. Stock prices fall by 4% in a recession regardless of whether there is an oil shock, but drop an average 6% when there is a significant shock.

If this tail risk is realised, then the IMF’s analysis provides a useful guide to the potential impacts, although with two key caveats.

One is that central banks might be reluctant to cut rates and could even raise rates if inflation expectations rose, as the disastrously high inflation of the 1970s was underpinned by oil shocks triggering higher inflation expectations and too loose monetary policy.

Another is that stock prices could fall by more than suggested by the IMF analysis given that long-term measures suggest equity prices remain significantly overvalued.