What happened? Lately, a chorus of analysts, dismayed by the weakness of the euro and (in Britain) by the strength of sterling, has been calling for exchange-market intervention to put matters straight. Most of these pro-intervention analysts, for all we know, still believe that exchange-rate targets are a mug’s game. Yet here they are demanding intervention—a policy which presupposes that governments (a) know what the exchange rate should be, and (b) can get their way by signalling that knowledge to the markets.

If intervention to correct currency misalignments is feasible, then the case for exchange-rate targets is stronger. Why? Because if intervention works, then an extra instrument of economic policy, in addition to monetary policy, is available—making it possible both to manage demand and, at least to some extent, to stabilise currencies. Conversely, the view that exchange-rate targets are unworkable implies that intervention is a waste of time. The current mood, which sneers at targets but calls for intervention, is confused.

As previously argued on this page, the theory of the vacant middle is in fact wrong. Governments, especially of small open economies, are right to pay regard to the value of the currency, and right to try to influence it under certain circumstances even if they do not go all the way and fix it once and for all (which has drawbacks of its own). Moreover, the evidence does show that under certain circumstances intervention can indeed work.

The distinction between sterilised and unsterilised intervention needs to be understood. The unsterilised kind happens when a government buys or sells its currency in the market and allows the transaction to change the money supply. Suppose Britain intervened in support of the euro, buying euros with pounds, and then allowed the resulting increase in the supply of pounds to stand. This is a clear relaxation of monetary policy. Nobody doubts this would lower the pound. But unsterilised intervention is the same as cutting interest rates. If you are willing to do that, you don’t need to intervene in the first place.

That is why by “intervention” economists usually mean sterilised intervention. The government neutralises the change in the money supply—in the example above, by selling debt to absorb the extra pounds. This kind of sterilisation is in theory an independent instrument of policy because it leaves monetary policy (defined in terms of the money supply) unchanged.

The standard reference on sterilised intervention is “Does Foreign Exchange Intervention Work?”, by Kathryn Dominguez and Jeffrey Frankel, published by the Institute for International Economics in 1993. Their answer to the question, overturning the prevailing view, was: Yes, sometimes. But it is more likely to succeed when public, when undertaken in concert, and when it conveys new information about the future of monetary policy.

In the case of the euro today, “in concert” is going to be difficult: with inflation rising, America hardly wants the dollar to fall. Some economists have argued nonetheless that Britain should intervene against the pound, which would help British exporters, supposing it worked, and would also do something to spare the euro’s blushes, even if it left the euro-dollar rate more or less unchanged.

The problem with this idea is not that intervention is bound to fail or that, if it did fail, the costs would be big: neither of these things is true. Given that the euro really is greatly undervalued against sterling, a long-term position in euros acquired at today’s prices would almost certainly be a money-making investment for the British government. Oddly enough, the problem arises if the policy succeeds, and the pound falls against the euro as intended.

If intervention pushed sterling down, the stance of monetary policy as the Bank of England understands it—notwithstanding the fact that the intervention had been sterilised—would be easier. If interest rates are now correctly set to keep inflation in line with the Bank’s statutory target, then they would presumably need to rise if the pound moved lower. And if the Bank raised rates, of course, the pound would be likely to appreciate again. At the very least, this would be awkward. Monetary policy would be, or anyway seem, in disarray—undesirable at the best of times, but especially with a newly independent central bank of uncertain credibility.

This problem would not arise if interest rates were currently based on a Bank forecast that sterling will very soon be a lot lower anyway: in that case, intervention to advance the schedule a bit would not constitute much of a loosening of policy. This week, in fact, the pound fell sharply of its own accord: the Bank’s forecasts assume it will fall no further. Successful intervention would contribute an additional loosening. In other words, even sterilised intervention is monetary policy. If intervention reflects a change in the central bank’s desired path for inflation, it has a good chance of working. Otherwise, it probably won’t.