What do devaluations achieve?

Perhaps unwisely, over the week-end I queried a bit of logic I saw on Twitter which had argued that a lower domestic currency makes a trade deficit worse.

The reply that came back to my query said that a lower dollar makes imports more expensive, QED.

And although I tried to point out that other things start happening, too, to my surprise I got further pushback in support of the the lower-dollar-makes-a-deficit-worse theory.

Maybe it's a generational thing. Back when I was starting out in economics, exchange rates were typically fixed or managed, currency changes were an actively deployed policy tactic, and the conventional thinking was that you reached for a devaluation to improve your trade deficit.  These days, exchange rates typically float, and the effects of devaluation as an option have, maybe, been forgotten about.

So here's a numerical example of what I always thought was the orthodox way of looking at it. It starts with New Zealand running a small trade deficit - our export revenue from wine doesn't quite match our import bill for oil - and it traces what happens when the New Zealand dollar drops against the US dollar. For dramatic effect I've made it a big fall, from parity with the US dollar to only 50 US cents.

The first panel is the starting point. The second panel shows the immediate impact of the fall in the Kiwi dollar. As the people on Twitter rightly feel, the first impact is of course to make imports more expensive, and the trade deficit does indeed get worse. What we're seeing at that point is the downward, or worsening, part of what used to be called the 'J curve' effect, and these days would probably be called a 'hockey stick' or 'Nike swoosh' effect - things get worse before they get better.

If that's all that the Twitter people were saying, fair enough, though I get the clear feeling that they also believe that the lower dollar will lead to a permanent worsening of the deficit, and that things won't actually get better later on.

But get better, they do, thanks to two responses to the lower dollar.

Panel three shows adjustment on the import side. Oil used to cost NZ$60; now it costs NZ$120. That's a powerful incentive to use less of it - by car pooling, by trading down to smaller car engines, by putting in solar panels, by using more energy-efficient public transport, whatever.

One respondent on Twitter argued that "How do you propose we "cut back" on imports without local industry to meet the demand? Should we just tell people to have less goods and medicines because they've become more expensive?". Well, the answer to that is that people routinely change their patterns of consumption to big price changes. If tickets to the big match get too pricey, you do something else for the weekend. If avocados are $7 each (as they were), you don't make guacamole. But I'll come back to that behavioural response at the end.

In panel three I've assumed that a variety of energy-saving measures in response to much more expensive oil lead to a reduction in demand from 400 barrels to 300. And we saw precisely that response in the real world to successive jack-ups in prices by OPEC, though it took quite a while for people to reorganise their affairs (eg by junking gas guzzlers and designing more fuel-effective engines).

That alone starts to eat into the initially higher trade deficit, which comes down from NZ$28,000 to NZ$16,000. The exact numbers don't matter: what does matter is that we've started going up the swoosh.

But there's also a response on the export side, which I've shown in the bottom panel. Before, the American buyer of our wine was paying US$200 a case. After the devaluation, he's only paying US$100 - it's a complete steal. He could well increase his order significantly - if he was profitably shifting New Zealand wine at US$200 a case, he'll have them flying off the shelves at US$100.

Or it's possible that the winemakers will raise their prices a bit. In the panel I've assumed they raised the Kiwi dollar price from NZ$200 to NZ$250, which for the US buyer means he's still paying a lot less (US$125) than the US$200 he used to pay. I've assumed that there's a mixture of more wine sold, as they are now a lot cheaper in US$ for US buyers, and a somewhat higher NZ$ price.

And hey presto, the overall outcome after both imports and exports have adjusted is a small trade surplus.

"Hey presto" may get you thinking this is all a sleight of hand, and indeed there is an assumption in the background here that those behavioural responses actually happen, and happen strongly enough, to turn things round. And there's a bit of economics (the 'Marshall-Lerner condition') that's figured out exactly how strong those responses need to be.

If you're what used to be called an 'elasticity pessimist', you don't believe the responses will in fact be large, and maybe that's where my Twitter correspondents are. Maybe, on the import side, we're always going to have to buy those medicines, and we can't skimp; maybe, on the export side, some wowser wine-shunning import monopoly won't order more cases then it used to. There's nothing logically wrong with that line of thinking, and at the end of the day the actual effect will turn on the size of the responses. Me - I'm more of an elasticity optimist, and especially over the longer run.

Before leaving what will be ancient history for some and the bleeding obvious for others, I'd just add that back in the day I was never a great fan of devaluations as a policy option. Can they improve the trade deficit? Yes, over time. But should you go there?

Mostly no. Devaluations can come with unpleasant side effects, including the ever-higher-inflation ever-lower-dollar spiral that New Zealand actually fell into. They can degenerate into tit for tat zero sum games (everybody can't simultaneously devalue against everyone else). And they encourage - or at the very least perpetuate - the idea that price is the best way to compete in world markets. So, at most devaluation makes an expedient policy tactic, but it's a poor economic strategy.