How should we think about the weak pound? One view is that it gives a welcome boost to UK exports. This idea, though, was countered last week by Ben Broadbent, deputy governor of the Bank of England. He said: "The most plausible explanation for the depreciation is that, in the eyes of the market, leaving the EU will make exporting harder and more costly. To help compensate the currency needs to be cheaper."
This, he said, implies that exporters are now in what is only a temporary sweet spot: "If the currency market is right the UK's future trading relationships will be less favourable; if it's too pessimistic sterling is likely to rebound." Either way, exporters will find life tougher in future than it is today.
I'd express this a little differently. I'd start from a big fact about exchange rates first pointed out back in 1983 by Richard Meese and Ken Rogoff and largely supported by later developments. This is that exchange rate moves aren't closely tied to observable macroeconomic factors. Instead, exchange rates, like other asset prices, are more volatile than underlying fundamentals.
There's a reason for this. When investors buy or sell assets, they don't (and certainly shouldn't) think of the future merely as a single possible outcome. Instead, they consider the range of possibilities. An asset price will therefore be the probability-weighted average of pay-offs in all those possible states of the world. This means that even if observed economic variables don't change, prices can do so if investors' change their opinion about the chances of a boom or crisis.
Exchange rates, like other asset prices, are more volatile than underlying fundamentals
Patrick Minford and colleagues have shown how such changes can explain stock market volatility. The same thing happens with exchange rates. These can change not because of changing observed reality but because of changing probabilities about the future: this is the so-called peso problem. This helps explain why exports and capital spending are usually not very sensitive to exchange rate moves: the same fear of disaster that drives the currency down also deters companies from expanding.
This, I suspect, is what's happened to sterling. Its weakness is due perhaps not to a dogmatic certainty that we'll get a hard Brexit in which exporters suffer increased tariff and non-tariff barriers but to traders believing that there's an increased probability of that scenario.
Now, in the past FX investors have often overestimated the probability of disaster. Two things make me say this. One is that carry trades (borrowing a low-yield currency to buy a high-yield one) have yielded profits on average in part as a reward for taking on crash risk. The other, documented in a paper by Martin Eichenbaum and Sergio Rebelo at Northwestern University and Benjamin Johannsen at the Federal Reserve is that low real exchange rates have usually led to rising nominal rates - which tells us that, on average, currencies have fallen too far perhaps because the risks traders feared didn't in fact materialise.
For practical purposes, all this means I agree with Mr Broadbent. It is possible that sterling will rise if those fears of a hard Brexit prove exaggerated. There are countless precedents for markets' worst fears to be proved wrong. More likely, we could see continued volatility as sterling rises and falls with every actual or expected twist and turn in our negotiations with the EU.
My difference with him is more one of theory. The FX market - like the stock market - tells us not what will happen, but what might happen. If sterling does rise, therefore, it'll not mean the market was irrational, but merely that it was impossible to attach a correct probability to events that are impossible to predict.