“When people feel like spending – in other words when demand for goods and services exceeds supply – inflation tends to rise”. So says the Bank of England.
Recent facts, however, suggest this isn’t true. Next week’s figures are likely to show that the core rate of inflation (which excludes food, energy, alcohol and tobacco) has fallen sharply in the past 12 months. And wage inflation has been stable for three years in the face of a fall in unemployment to a 43-year low. All this is despite the fact that Bank of England governor Mark Carney said last year that demand would grow faster than supply.
Why, then, has inflation been insensitive to rising demand? One reason – discussed last year by the Bank’s chief economist Andy Haldane – is that workers have lost bargaining power and are unable to parlay a tight labour market into pay rises to the extent that they used to.
But there’s also an issue in product markets: the notion that supply constraints cause inflation to rise once demand reaches a certain point is deeply questionable. Igal Hendel and Yossi Spiegel, two US economists, have shown why. They studied a steel mill, and showed that as demand rose so too did output even without an increase in the plant’s size. Rather than cause inflation, higher demand led managers to tweak working practices to raise productivity: this, after all, should be the point of managers. “Capacity is not well defined, even in batch-oriented manufacturing,” they concluded.
This isn’t an idiosyncrasy. In the 1950s and 1960s inflation stayed low despite full employment, precisely because managers responded to booming demand as they did in that steel mill – by raising productivity.
If this is true in old-fashioned manufacturing, it is even more the case in the new economy. One feature of this, say Jonathan Haskel and Stian Westlake in their book Capitalism without Capital, is that companies are more scalable than they were in the old economy. They can increase production at almost no extra cost. Capacity constraints are weak or non-existent. Video game companies, music companies, newspapers selling digital subscriptions or dating sites can all supply additional customers at almost no extra cost. Yes, they must spend fortunes on product development and advertising. But these are fixed costs. Their marginal costs are minuscule. 'Capacity' is not meaningful.
The idea that inflation will rise when demand approaches capacity is therefore deeply dubious. The very fact that inflation has been stable since the early 1990s while demand has fluctuated a lot tells us this.
Although it isn’t obviously true, this idea is however very useful. A central bank that believes that inflation will fall when demand is weak and rise when it is strong will cut interest rates in a recession. It will therefore help to stabilise demand even though its job, formally, is to target inflation. This is very valuable, especially if governments have renounced the counter-cyclical use of fiscal policy in favour of debt stabilisation – as they have done since the 1990s except in the immediate aftermath of the 2008 crisis.
The Bank might be wrong about inflation, therefore. But that mistake has not been especially expensive – at least, not so far.
