ISTANBUL – Central bankers want only a few things. To achieve any of them they usually seek to nudge inflation expectations, demonstrate the transparency of monetary policy, and establish their institutions’ credibility. To communicate their intentions simply and clearly, they may set an explicit target range in terms of a particular economic variable, or announce a forecast for the variable, or offer forward guidance by specifying a threshold value for it that must be met before changing interest rates.
But what should that one economic variable be? In the 1980s, major advanced countries tried the money supply. After the monetarist approach failed, some switched to targeting the inflation rate. But they repeatedly missed their targets.
Until the currency crashes of the 1990s, emerging and developing countries tended to target their exchange rates. Many then also switched to inflation targets; but they tend to miss these targets even more often than the advanced countries do.
The problem with these approaches to monetary-policy targeting is that even though a particular numerical target may be reasonable when it is set, subsequent unexpected developments often make the target hard to live with. The monetary authorities are then confronted with a harsh choice between violating their announced target, and thus undermining the credibility that was the point of the exercise, or setting policy too tight or too loose, thus doing unnecessary damage to the economy.