So I was reading the less interesting bits of the paper and looked at a report on the price of oil which, as is pretty much routine, contained the suggestion that a shortage of refinery capacity was contributing to the high price of crude. Turning my brain on for a moment, it struck me that economics 101 suggests the exact opposite. Refineries and crude oil are complements in production, so a shortage of refinery capacity should lower the price of crude, while increasing the price of refined products such as petrol and distillate (easy thought experiment to show this: if there were no refineries at all, crude oil would be worthless).
A quick Google shows that I’m not the first person to notice this point, but the erroneous claim keeps on getting repeated.
How does an error like this get started. The obvious reason is that if the price of oil is driven by demand fluctuations (as at present) high prices for oil will be correlated with perceived shortages of refinery capacity. It will be easy to find specific instances to support an apparent causation going the other way. For a refinery working at maximum capacity, a breakdown will cause a supply interruption and will be very obvious. The same breakdown, occurring at a slack time, might be dealt with by rearranging operations and rescheduling maintenance and never reported to customers.