While many Oilers love to argue that "every" modern recession has been preceded by significant YoY increases in the price of oil, as if this spiking price of oil has been the 100% culprit 100% of the time, this is truly not the case. But what is the case is that a significant percentage of modern expansions... particularly those already experiencing weakness... have been tripped up a good bit by persistent and/or sharp increases in the price of oil.
In the graph below (from FRED) we can clearly see how persistent and/or sharply rising oil does frequently precede NBER-identified recessions
Year of greatest YoY spike & Quickly or Concurrently Followed by Recession: Y or N
1973: Y
1980: Y
1987: N
1989: N
1990: Y
1996: N
2000: N
2003: N
2004: N
2008: Y
2010: ?
Modern recessions associated with a big spike in oil within the same 12 month period:
1973, 1980, 1990, 2007
Modern recessions -not- associated with big a spike in oil occurring in the same 12 month time frame:
1970, 1981, 2001
Number of times oil appreciably spiked from 1970 through 2009: 10
Number of spikes since 1970 that occurred right before or early into a recession: 4
Recessions to spikes: 40%*
Number of false alarms (Spike "suggesting" recession that never happened): 6
Oil spike-Recession false alarms: 60%*
These stats would even be lower/higher (even less suggestive of a causal relationship) if one considers that the
1990 &
2007 recessions actually began before those year's oil spikes really took off, making for an argument that there was actually no cause & effect relationship, at all, those times. Coincidental events, certainly. Making the case that spiking oil prices lead to recession in 1990 or 2007, not at all quite as clear.