The great thing about a free market economy is that, as the Nobel laureate Friedrich Hayek pointed out, it is a decentralised problem-solving device. If the price of something rises, suppliers respond by raising output while customers use it more efficiently or switch to other things.
Of course, we cannot predict how specific people do this. But that’s because, as Hayek stressed, they have localised specific knowledge of production processes and opportunities that we don’t have. Which is why markets work better than central planning.
If this is correct, then inflation caused by localised shortages can only be a temporary phenomenon. Rising prices induce behavioural changes that cause prices to fall.
But is it correct? Three facts suggest so.
First, inflation in western economies has been stable for the past 25 years even in the face of big swings in demand, monetary growth and commodity prices. This tells us that people do indeed adjust to changes and so we avoid persistently high inflation.
Secondly, firms do indeed increase efficiency as shortages loom. Igal Hendel and Yossi Spiegel studied an old-fashioned US steel mill and found that as it approached full capacity managers discovered ways of eking out more production without much raising prices. This is not a localised finding. It helps explain why economists who have tried to forecast inflation by using gaps between actual and potential output have consistently over-predicted inflation – the ECB being a major offender here.
Thirdly, the link between output growth and commodity prices is stronger in the short term than in the long. Since 1985 the correlation between the GSCI index and OECD industrial production has been 0.53 for annual changes but only 0.17 for five-yearly changes. Why might this be? It’s because in the short run commodity producers cannot much increase output when prices rise and users cannot easily economise. In the longer run, however, both can adjust thereby dampening the response of prices to demand. There’s a reason why people talk about commodity price cycles.
All this has a simple implication. We can get inflation in the short run because suppliers and demanders can’t quickly respond to rising prices and shortages. But in the longer run they can do so, and so rising prices are a problem that the market solves itself.
From this perspective sustained high inflation is only possible if there is a policy failure – if central banks allow generalised inflation to get out of control. So far, though, there’s less than overwhelming evidence for this. In the US, five-year inflation expectations are at the same level they were in 2005. In the UK they are a little more out of whack, at the highest level since 1996. But in both cases, these numbers are more consistent with expectations of central banks tolerating inflation for a short while until economies are healed rather than with allowing it to rip.
Which brings us to a paradox. Those who expect sustained high inflation have little faith in our key institutions: free markets and independent central banks. Few of them, however, seem to draw any wider political inferences from their lack of trust.
