Join our community of smart investors
OPINION

On heterogeneity

On heterogeneity
August 25, 2016
On heterogeneity

Take for example a recent finding by Bank of England agents that "a majority of companies spoken with did not expect a near-term impact from the result [of the EU referendum] on their investment or staff hiring plans". Does this mean that the referendum result won't hurt the economy?

Not at all. Economic downturns are about what happens at a minority of companies. They occur not because all companies cut spending but because a minority do so. In their classic study of the 1989-91 recession Paul Geroski and Paul Gregg found that "recessionary pressures are concentrated on a relatively small number of firms." In that recession, 10 per cent of companies accounted for 85 per cent of the fall in employment.

If only a minority of companies cut investment and hiring, this might be enough to cause a recession. And in fact, the Bank's agents found that a third of companies do expect to see weaker investment and hiring.

Some other Bank of England numbers remind us of the diversity of companies. They show that in the past 12 months non-financial companies' bank deposits have grown by 7.5 per cent while their borrowing has risen 3.7 per cent. This makes little sense if you think of 'the corporate sector' as a single entity: why borrow at a high rate to leave the cash on deposit at a negligible one? But it does make sense when you appreciate the diversity of companies: some need to borrow to expand, while others can't see how to grow and so are just building up cash piles.

The key thing here is that entrepreneurship is inherently idiosyncratic. Some people can see opportunities to make profits, some can't and others see opportunities where none in fact exists. It's the net balance of these perceptions that generates booms and slumps.

You might think all this is childishly obvious. But it conflicts with the sort of economics that students are tortured with. Standard textbooks use representative agent models in which there is a single company and a single consumer. And they typically pay little attention to the entrepreneur, preferring to see bosses as mathematicians who merely solve constrained optimisation problems.

Even if my perspective is obvious, it has some significant implications.

First, it helps explain why recessions have generally been unpredictable. It's because, as New York University's Xavier Gabaix has shown, they can originate in the failure of one or two firms. Whether such failures cause serious downturns or not depends upon the interlinkages between them and other firms: the failure of a hub such as a big bank causes a recession, but the failure of a spoke such as BHS doesn't.

Secondly, it explains why recessions matter. If a 1 per cent fall in GDP meant that everyone's income fell 1 per cent, nobody would care much. But that's not what happens. Recessions hit a few people very hard, causing emotional as well as financial damage.

Thirdly, this warns us that anecdotes are useless as a guide to the macroeconomy. Even in recessions, some firms thrive and expand; Geroski and Gregg found that two-fifths did so in 1989-91. And even in good times some fail: BHS did so despite strong overall high street spending. It's not just economists that can't tell you whether we're in a recession or not: nor can journalists.