Speculation that Canadian Mark Carney, who took charge at the Bank of England last week, might radically change British monetary policy cooled in the months after his appointment was announced. No doubt, this softening of expectations was justified — the constraints of the office are tighter than the new governor and his admirers would like. It’s a pity though. Carney should stay ambitious. Monetary policy could stand some innovation, and not just in Britain.
The new governor is known to favour clearer "forward guidance" on policy, of the kind the Federal Reserve has been trying to use. The first policy announcement on his watch left interest rates and asset purchases unchanged, but told markets that interest rates might stay low for longer than they expect. The pound weakened (presumably as intended) as a result.
I’m not a great fan of Fed-style forward guidance, for reasons I’ll explain. If Carney wants to make his mark in Threadneedle Street, he should try to be bolder. Here’s how: Get the Bank of England to adopt a target for growth in total cash spending, or nominal gross domestic product.
This idea isn’t new — it has a long history — but in the past few years it’s attracted new interest. Credit for this goes partly to the recession and partly to Scott Sumner, a professor of economics at Bentley University in Massachusetts. Sumner and a group of blogging supporters have raised the idea to new prominence. Carney aroused excitement last year by expressing support for the idea before backtracking.
Modern advocates of a cash-spending target sometimes overstate their case. That’s all right — how else do you gain attention? They’re also pushing a version of the idea that, in my view, isn’t the most appealing. Nonetheless, their thinking contains a vital insight that central banks would do well to recognize.
The Bank of England, like most other central banks, uses an inflation target to guide — and to help it explain — its monetary policies. The basic idea is to anchor expectations of low inflation. If you can do that, the thinking goes, low inflation will become a partly self-fulfilling prophecy, and volatility in the real economy will be less. If you fail to do it, and expected inflation gets out of hand, actual inflation will follow and the central bank will eventually have to cause a recession to get it back down.
When the inflation-target model was first adopted in the 1990s, a big attraction was that it acknowledged the limits of monetary policy. In the long run, monetary stimulus can’t raise output — and because expectations are so important, central banks shouldn’t let themselves be suspected of trying. They should focus instead on keeping inflation low and stable. The inflation target is a commitment to do just that.
Yet in the short and medium term, monetary policy can and does affect the real economy, and central banks can’t pretend otherwise — least of all during a recession as deep as the one that started in 2008. As a result, the supposedly bracing clarity of a simple inflation target has by degrees been blurred.
The Bank of England calls its policy "flexible inflation-targeting," with the emphasis on flexible. In practice, U.K. inflation has persistently overshot the two per-cent target — a good thing too, because the recession would have been even worse otherwise. The U.S. Federal Reserve’s latest guidance to the markets has rebalanced its statutory "dual mandate," softening the short-term inflation commitment and promoting a specific measure of real economic activity — unemployment — as a semi-formal threshold for guiding monetary policy.
The result is a muddle, as the markets’ struggle to understand the Fed’s most recent policy announcement shows. The problem is complexity: The Fed is juggling too many variables. The advantage of a total-spending target is that it provides the simplicity that inflation targets promise but don’t deliver.
Central banks operate only indirectly on inflation. In the first place, they influence aggregate spending. That’s all they can do and all they should be asked to do.
Suppose the central bank announces a two-year target for growth in demand of, say, 4.5 per cent a year. Over the long-term, that would be consistent with inflation of two per cent and growth in output of 2.5 per cent. If output grew more slowly in the short-term, maintaining growth in demand would imply a temporarily higher inflation rate. Here’s the point: The total-spending target allows for that.
The central bank doesn’t need to weigh short-term fluctuations in unemployment, output growth and prices, then explain each month why it thinks the prospects for each of these variables conforms to an ever more elaborate forward plan (which the markets don’t believe anyway). If demand is lagging the target for growth in spending, the central bank stimulates (using interest rates when available or quantitative easing if not). Otherwise, it doesn’t.
There’s a catch, though. The presentational benefits of a total-spending target would be mostly lost, in my view, if central banks did what most NGDP advocates suggest and set a multi-year target for the level of total spending as opposed to its rate of growth. This seemingly small difference has big implications.
A long-term target for the level of demand doesn’t forgive previous overshoots or undershoots — it requires them to be clawed back. Potentially, this has one enormous advantage. If the promise to make good a cumulatively large shortfall in demand is believed, it will exert a correspondingly powerful stimulatory effect.
Demand in Britain is currently more than 15 per cent below its pre-crash trend — calling, on the face of it, for massive and sustained stimulus to put it back on track. The implication, though, is that in delivering this stimulus the central bank might tolerate high inflation even after the economy had begun to recover strongly. This, indeed, is the whole idea. It’s the promise of an inflation overshoot that provides the extra kick.
The trouble is, this target-the-level scenario puts the central bank back in the position of having an awful lot of explaining to do. Its estimate of the current gap between actual and potential output and its estimate of long-term potential growth acquire great significance for short-term monetary policy.
Also, of course, potential growth rates change. If potential growth has fallen, policy might stay too expansionary for too long. If potential growth has gone up, policy might stay too restrictive too long (a possibility some NGDP advocates tend to forget, partly because under current conditions the risk is remote). In any event, the benefits of firmly anchored expectations of inflation are put in jeopardy.
In practice, the choice between the growth-rate approach and the levels approach isn’t black and white. A lot depends on the details — and policy makers would probably end up with a hybrid of the two.
But if I were Carney, I’d lean toward modesty and simplicity and make the case for targeting growth in total spending at a rate consistent with the Bank of England’s long-term inflation mandate of two per cent. This would be much easier to explain than the current policy and, best of all, wouldn’t force the central bank to make promises it can’t keep.
It would also save Carney some time — which he’ll need to run the Bank of England’s hugely enlarged responsibilities as a financial regulator. Welcome to London, Governor Carney.
Clive Crook is a Bloomberg View columnist.