On 27 February the Kiwi dollar closed at a scarcely credible 81.66 US cents, a new post-float high, and there was talk of it soaring further into the stratosphere. Little wonder the latest National Bank business opinion survey showed sharp falls in business confidence, and in expectations for exporting, profits, activity and employment.
It's tempting to draw the conclusion the Kiwi is 'overvalued' and is way north of where it 'ought to be'. But there are two problems with that.
In a free market the exchange rate that emerges from all the supply and demand for a currency has to be in some sense the 'right' price. It's kind of hard to argue that the price the participants freely set for themselves ought to have been something else.
But let's suppose what we really mean is that for some reason - say, because of a 'bubble' or a panic - a currency's value can be temporarily driven away from some estimate of what it 'ought to be' on some reading of all the economic fundamentals. But the problem now is that we don't have a great handle on where it 'ought' to be.
It's true we can (more or less) figure out a rate that would make exporters' lives at least bearable (somewhere around the 60 US cent mark seems to be a reasonable guess). But exporters aren't the only participants in the forex markets. Investors in the US, for example, observe their local interest rates plummeting, but also notice Kiwi interest rates are steady and high: what's the 'right' exchange rate that takes into account their entirely rational decision to add some Kiwi dollars to their portfolios?
In principle, you could build a fancy econometric model to try and figure it out. You'd feed in all the things you think might affect the currency - overseas investors' risk tolerance for a flutter overseas, our commodity prices, interest rates, whatever - and you'd see what came out.
The answer, though, would still depend on the assumption that exchange rates are, more or less, determined by those economic 'fundamentals'. And that's where you'd be seriously wrong.
In recent years economists have been modelling in the laboratory how traders in markets behave. And every time the answer comes out the same: markets have funny internal dynamics of their own and don't track the fundamentals for long periods of time. They can, and will, generate 'bubbles', for example, where the price of what's being traded appears to take leave of its fundamental senses. And they generate the subsequent crashes, too.
The latest example comes from some research conducted on students at Emory University in Atlanta, nicely set up so the researchers could see everything that was going on. They knew the fundamental value of the asset being traded, what the traders' expectations were for its value, and the outcome for a whole series of trading experiments: they ran this 'market' four times, each time with multiple trading sessions.
This market generated 'bubbles'. Prices soared, especially in the first session, way above where they 'ought' to be. That was mostly because, once trading had started, traders didn't base their expectations on what the fundamental value of the asset was but on what has happening to its price. If it was going up, they were minded to believe it would go up more, so they traded accordingly, and it went up even more, further inflating the bubble. Ultimately there was a 'crash' back to the fundamental value.
Interestingly, though, the traders learned from the experience. In repeated markets they learned to be wary of the looming bubble, and it became smaller. Even then, however, their collective behaviour generated price movements not related to the fundamentals. For example, each thought they'd be smarter than the next fellow, and sell out just before the bubble peaked. Each of them thinking that way, however, burst the bubble in itself as each of them increased their selling and drove the price down. The bubbles burst before the traders thought they would.
So what, you ask? For one, I'd be less trusting of exchange rate forecasts based on 'fundamentals' and more ready to insure against the apparently unthinkable.
I think I'd also encourage the Reserve Bank to intervene again. Last time it did (July 2007) the currency was just over 80 US cents. Intervention produced an immediate 2 cent drop, and by mid-August had helped bring about a 10 cent drop. Now might be a good time to give the market another nudge downwards, which could set off another self-fulfilling decline as traders assume further falls. And then we might return to a level our exporters can survive on.