Rational consumers

In the day job, I point out that the ratio of consumer spending to house prices has done a reasonable job of forecasting medium-term GDP growth, and that right now it is pointing to weakish growth.

Granted, this might in part be because of the denominator. When house prices are high – a low ratio in my chart - they subsequently fall and this depresses GDP via collateral effects (pdf) if nothing else.

But there might be something else going on. Consumer spending should be forward-looking. If so, low spending will be a sign that consumers expect bad times. And it’s possible that, across millions of consumers, errors in expectations will roughly cancel out: unless there are strong peer effects, the individuals who are wrongly optimistic will be balanced out by those who are wrongly pessimistic. To the extent that this is the case, aggregate spending will predict economic growth.

This shouldn’t strike you as outlandish. In the same spirit, Sydney Ludvigson and Martin Lettau have shown that consumer spending can predict (pdf) equity returns in the US, and Bank of England economists have corroborated this in the UK (pdf).

In fact – given economists’ inability to foresee recessions – it’s possible that consumers, in aggregate, are better economic forecasters than the professionals. Perhaps aggregate spending does what Friedrich Hayek claimed the price system does: it collects the dispersed, fragmentary and partial knowledge of millions of people into meaningful signals. Just as the decentralized price system embodies more knowledge than any central planner, so aggregate spending knows more than any centralized forecaster.

I don’t say this to make an economic forecast – though it’s worth noting that the consumption-house price ratio is sending a similar message to our record current account deficit. Instead, I do so to question the “nudge” agenda. This presumes (pdf) that wise and rational policy-makers can correct the follies of irrational lay-people. What we have here, though, is evidence that lay-people might (on average and in aggregate) know more than experts.

This is not the only evidence. The field of behavioural political economy – and, I guess, world history – reminds us that elites are as prone to cognitive biases as everyone else. As William Easterly has said:

Experts are as prone to cognitive biases as the rest of us. Those at the top will be overly confident in their ability to predict the system-wide effects of paternalistic policy-making – and the combination of democratic politics and market economics is precisely the kind of complex and spontaneous order that does not lend itself to expert intuition.

This is not to defend the simple-minded assumption that market agents are always rational. Instead, I suspect that irrationality, like intelligence, is context-specific: rational consumers (in my very limited sense of the phrase) might be irrational voters. The trick – in our personal lives as well as in economic research and policy-making – is to know when and how particular environments generate cognitive biases and when they don’t.

Comments

[Matt Moore]

'rational consumers (in my very limited sense of the phrase) might be irrational voters.'

I don't see degree of rationality as being something we bring to a particular situation. Rather, it is partly created by the situation. Consumer spending carries pretty strong incentives to be rational - feedback is typically quick and direct, and your actions matter. In voting, feedback is weak and indirect and you actions don't matter on the margin (which is to say, they don't matter at all)

It's just not worth the effort of being super rational in the voting space - and that decision is rational.

[Jonathan da Silva]

When Concorde crashed and there had been 30 similar incidents in 25 years or so. Experts said oh no perfectly safe just bad luck. Their's was the cognitive bias that an outlier like Concorde and their professions integrity would not make such a deadly mistake.

IMO BrExit analysis seems a similar case falling back on Classical economics & non analogous case studies. Certainly neo Classical 'conomists should head to Port Talbot and explain these workers are now freed to do what they have a comparative advantage in. "Who's Ricardo" likely one of the nicer responses. As to Productivity loss it's almost funny to bring that up at a time like this.

[Raphshirley]

If one curve was a predictor of another curve shouldn't the predicted lag behind the predictor in time? Otherwise is it not just correlated?

[Deviation From The Mean]

The problem is that the price signal is dumb, in human terms 20 IQ points behind Joey Essex, it doesn't ask questions. The price signal cannot tell us what impacts there will be on the environment for example. You need a central body manipulating the price for that to happen.

In this same way consumers are equally dumb, under capitalism they are almost totally passive, do not ask questions, they consume without thinking. In other words they completely ignore very pertinent knowledge.

This is why capitalism does not speak of efficient consumption, other than what reduces the costs for capitalists.

[Ken Houghton]

Ralphshirley - The graphic has a 5-yr lag on the house price ratio.