The Bank of England has had a formal inflation target since the early 1990s. Is it the time for it to target something else?
That's the debate rumbling among economists these days - a debate that is now getting more attention thanks to a speech this week by the next governor of the Bank of England, Mark Carney, and the Federal Reserve's move to target unemployment. (See my last blog.)
In his speech, Mr Carney said it might make more sense in today's circumstances to target not the growth of prices (inflation) but the growth in the cash value of economic output: nominal GDP.
Academics and policymakers have been drawn to the idea of targeting nominal GDP, off and on, for decades, but they've hardly ever ended up wanting to do it in practice.
Will this time be any different? I wonder. But the liveliness of the debate shows you how concerned many now are about the prospects for economic growth.
Remember how we got inflation targets in the first place: we got them because academic opinion suggested that central banks could not really affect the underlying growth rate. The most that they could do was to provide a stable environment for growth, by controlling the rate of inflation.
In the 1980s we thought the best way to do that was by targeting the growth rate of money, but they could never find a measure of the money supply that had a reliable relationship with inflation. So they decided, in the 1990s, to cut out the middle man and just target inflation.
That was clear and easily understood by the public. It also seemed to work, which is why nearly every major central bank in the world (except the Federal Reserve) ended up with some form of inflation target.
We now know there was at least one big problem with this narrow approach: it encouraged the Bank of England to ignore a lot of other important stuff such as rising asset prices and the build-up of huge debts within the banking system, which weren't causing inflation but turned out to pose a massive threat to our economic stability.
That problem, in theory, is now being addressed, with the creation of the new Financial Policy Committee at the Bank for example. But the economy has done so badly over the past few years that some say the inflation target is leading the Bank to miss something else that is equally important: economic growth.
If the Bank targeted the cash value of GDP then the Bank would become responsible for the overall level of economic activity, not just the annual change in the consumer prices index.
So, to give a concrete example, you might say, instead of 2% inflation the Bank should achieve 4.5% growth in cash GDP every year for 10 years.
Of that 4.5%, you might hope that roughly 2.5% would be real growth in national output and the rest inflation. But under a strict nominal GDP target, the breakdown doesn't matter. You just have to get 4.5% nominal growth.
'Bygones are not bygones'
The other key feature of this approach is that if you fail to achieve the target you have to try to make up the difference in the years that come after.
So, as Mark Carney said in his speech, "bygones are not bygones": past failures affect future policy.
That is a big difference from inflation targeting, which always looks forward and, in effect, treats every monetary policy committee meeting as if it were the first. It is also why monetary "activists" like the idea of a switch, because it would put the Bank under pressure to do more to stimulate growth right now.
Since 2007, nominal GDP growth has averaged just 2.6%. To make up for that lost growth in the next five years, the Bank would need to target growth in nominal GDP of well over 6% a year between 2013 and 2017.
The current OBR forecast has cash GDP growing by just under 4.3% on average over this period. The forecast for 2013 is for nominal growth of 3.3%.
So, a big short-term reason why people like this idea is it gives the Bank license - indeed forces it - to do more, in an environment in which interest rates are already at rock bottom and central banks are finding it hard to persuade businesses that normal rates of growth are going to come back.
In effect, moving to nominal GDP targets would send a signal that the Bank was determined to get back the nominal growth in the economy that has been lost, even if it is at the cost of pushing inflation above 2% for a sustained period of time.
A longer-term reason why some prefer nominal GDP targets is that they would make it easier for the Bank to respond to supply shocks, such as a rise in the oil price, in a way that would not add to the short-term damage to the economy.
In 2008, for example, the European Central Bank raised interest rates just months before the financial crisis because of rising commodity prices and their likely effect on inflation, even though the European economy was already weakening sharply. Rate-setters in the UK considered doing the same as late as August that year. Under a nominal GDP regime, that wouldn't have happened. They would probably have cut rates.
But critics can also some raise some pretty powerful arguments against a change of target, the strongest being that the Bank of England would be giving the mistaken impression that it can control economic growth when it can't.
In the current environment, the change of target might also send a message that the Bank was not so concerned about keeping inflation low. In fact, for the next few years at least, it would be saying that it was not very concerned about inflation at all, if that was what it would take to deliver 6% plus nominal growth.
There are also practical arguments against a new target: such as the fact that nominal GDP is not calculated in a timely fashion and - unlike the CPI - is subject to large and frequent revisions. Also, most of the public don't have a clue what nominal GDP is, which is not a small objection.
But, as I said on Thursday, the biggest argument against that you hear from people at the Bank is that our "flexible inflation targeting" regime already gives you everything you might have wanted from a nominal GDP regime. After all, it's not as if the Bank has stuck slavishly to its inflation target over the past few years. Far from it.
The MPC has "looked through" the supply shocks that have hit the economy and pushed up inflation over the past few years - just as it might have done under a nominal GDP regime. It has done this without ever formally changing its mandate or raising questions about its long-term commitment to low inflation. That is surely preferable, say Bank officials, to a highly visible and possibly counterproductive regime shift.
Perhaps, but what opponents of nominal GDP targets seem to be saying is: "why change the inflation target, when you can just ignore it?" Some of them also seem to be saying that the Bank has actually been following a nominal GDP target for some time, it just hasn't told anyone.
The fact that people don't understand what nominal GDP is seems to me to be a really strong argument against making it a target. The backward-looking nature of this approach is also troubling, even to potential fans of the approach such as Mark Carney. There's a risk that the Bank will always be fighting the last war and thus - possibly - paving the way for the next one.
So, the debate will continue, and the defenders of the current regime have some strong lines of defence. What does not seem to me to be the best argument for the inflation target, given everything that has happened, is the confident claim that the people at the Bank will always be clever enough to know when to ignore it.