The main policy excitement in the Budget is unlikely to concern fiscal matters this time. The personal allowance may rise to £10,000 and an aspiration to raise it further declared; corporate tax may be cut; pension tax relief may be reduced; a little extra capital spending on housing may come. The next Spending Review may see some hot debates about ringfencing. But in the Budget itself, the most important measure is likely to involve either George Osborne changing the inflation target or announcing a short consultation period about a potential change, with the change itself being introduced to coincide with the arrival of Mark Carney as the new Bank of England governor.
The inflation target has changed three times before:
- first in 1997, when it was switched from the original 1-4% range for RPIX (with a preference for the bottom half of the range by April 1997) to a 2.5%+/-1% range with preferred centrepoint;
- then in 2003 when it was switched to a 2%+/-1% range for CPI with preferred centrepoint;
- and most recently (less formally) in 2007 when the outer bounds of the range were abandoned after the upper bound of the previous inflation target was missed, and it simply because a 2% CPI target with no restriction on how far it was permitted to deviate from target.
I have argued for other changes:
- I wanted the target switched from inflation to average inflation (in what is called "price level targeting") from the late 1990s;
- I wanted the target raised in 2008;
- I wanted the target raised and tolerance bands introduced in 2011.
So there is no particular problem with changing the target.
Unfortunately, though, the ways floated in the press for changing the target all seem counter-productive to me, and unlikely to be regarded as anything more than an excuse for missing the target even more and achieving even higher inflation.
One alternative target proposed is the annual growth rate of nominal GDP instead of annual inflation. This is an old idea that used to be put forward regularly by Samuel Brittan but for which Vince Cable's SpAd Giles Wilkes has in recent years become the key public proponent. [Edited for infelicity: If the underlying growth rate of GDP is fairly stable, such that almost all economic shocks are demand shocks not supply shocks, then a nominal GDP growth rule would imply that policy is automatically looser in years when there is a negative demand shock and so low output, reflating our way out of recession, and tighter in periods when growth is above trend, cooling off booms.] In itself it is perfectly sensible, but subject to two key problems:
- First, nominal GDP data is much slower and less reliable than inflation data, and much more subject to revision. It would be much more likely for the framework to involve errors by its own standard – i.e. that in circumstances where actual nominal GDP growth were rapid, policymakers would cut because they wrongly believed nominal GDP growth were slow; and vice versa. Inflation data is not perfectly reliable, and of course inflation targeting, if operating properly, is really a framework to allow macroeconomic fine-tuning of real GDP (which is, almost by definition, measured less reliably than nominal GDP), so data challenges are not unique to nominal GDP targeting, but it is a practical weakness.
- Second, if the economy is subject to significant supply shocks, a nominal GDP target can produce seriously sub-optimal policy responses – e.g. if supply falls then maintaining nominal GDP growth constant will involve boosting inflation without any positive impact on real GDP.
I am unconvinced.
Another proposal is "flexible" inflation targeting. Well, it's not clear how much more "flexible" the UK's inflation target could be than the desire to meet the target not having motivated any interest rate change between late 2006 and early 2013 (which I regard as being the case), or inflation being almost continuously well above target for years at a time. What policymakers fret about is that, as the economy recovers, there will come a point at which interest rates will need to rise if an inflation target is in any way to be respected, and in fact if the target were to be met they would have to rise very aggressively. Anticipating such an interest rate rise in the upswing, it may be unattractive for potential investors to borrow to invest now, so the economy becomes mired in a low-investment / low-growth quasi-equilibrium. That is indeed a weakness of inflation targeting, and precisely one of the arguments I have used since 2002 to favour the use of price-level targeting over inflation targeting in escaping from a liquidity trap.
But policymakers seem to ignore a key problem in all this. They worry that under inflation targeting they cannot credibly commit to keep interest rates low even in the recovery phase. Well, under inflation targeting or nominal GDP growth targeting they are never going to want to do that, whether it is flexible or not. Only a levels target (e.g. a price-level target or a nominal GDP level target) could allow that (since it would be the meeting of a level, not a change, that would be crucial). But more pertinently than that, at this stage, the Bank of England has no credibility with anything, let alone with promises about complex situations in the future, because it has blatantly and consistently failed to keep its promises regarding the inflation target. If it has made no attempt to keep its promises in the past, why would anyone believe its promises about the future? Only once credibility is restored (i.e. only once people believe the Bank of England will keep its promises, regardless of whether it seems attractive at the time to do so) can we even contemplate its being credible in making complex future promises.
Thus, although certain changes to the Bank of England's target would be attractive (specifically, a switch to a price-level or average inflation target), the changes actually proposed (nominal GDP targeting or "flexible" inflation targeting) will be regarded by everybody as precisely what they are: an excuse for ignoring the inflation target even more blatantly. Only once the Bank of England is set a target that the government actually wants it to meet and that the government then enforces, regardless of whether it seemed attractive to meet the target at the time, can credibility be restored to UK monetary policy. I fear that that time is still some way off.
The main policy excitement in the Budget is unlikely to concern fiscal matters this time. The personal allowance may rise to £10,000 and an aspiration to raise it further declared; corporate tax may be cut; pension tax relief may be reduced; a little extra capital spending on housing may come. The next Spending Review may see some hot debates about ringfencing. But in the Budget itself, the most important measure is likely to involve either George Osborne changing the inflation target or announcing a short consultation period about a potential change, with the change itself being introduced to coincide with the arrival of Mark Carney as the new Bank of England governor.
The inflation target has changed three times before:
I have argued for other changes:
So there is no particular problem with changing the target.
Unfortunately, though, the ways floated in the press for changing the target all seem counter-productive to me, and unlikely to be regarded as anything more than an excuse for missing the target even more and achieving even higher inflation.
One alternative target proposed is the annual growth rate of nominal GDP instead of annual inflation. This is an old idea that used to be put forward regularly by Samuel Brittan but for which Vince Cable's SpAd Giles Wilkes has in recent years become the key public proponent. [Edited for infelicity: If the underlying growth rate of GDP is fairly stable, such that almost all economic shocks are demand shocks not supply shocks, then a nominal GDP growth rule would imply that policy is automatically looser in years when there is a negative demand shock and so low output, reflating our way out of recession, and tighter in periods when growth is above trend, cooling off booms.] In itself it is perfectly sensible, but subject to two key problems:
I am unconvinced.
Another proposal is "flexible" inflation targeting. Well, it's not clear how much more "flexible" the UK's inflation target could be than the desire to meet the target not having motivated any interest rate change between late 2006 and early 2013 (which I regard as being the case), or inflation being almost continuously well above target for years at a time. What policymakers fret about is that, as the economy recovers, there will come a point at which interest rates will need to rise if an inflation target is in any way to be respected, and in fact if the target were to be met they would have to rise very aggressively. Anticipating such an interest rate rise in the upswing, it may be unattractive for potential investors to borrow to invest now, so the economy becomes mired in a low-investment / low-growth quasi-equilibrium. That is indeed a weakness of inflation targeting, and precisely one of the arguments I have used since 2002 to favour the use of price-level targeting over inflation targeting in escaping from a liquidity trap.
But policymakers seem to ignore a key problem in all this. They worry that under inflation targeting they cannot credibly commit to keep interest rates low even in the recovery phase. Well, under inflation targeting or nominal GDP growth targeting they are never going to want to do that, whether it is flexible or not. Only a levels target (e.g. a price-level target or a nominal GDP level target) could allow that (since it would be the meeting of a level, not a change, that would be crucial). But more pertinently than that, at this stage, the Bank of England has no credibility with anything, let alone with promises about complex situations in the future, because it has blatantly and consistently failed to keep its promises regarding the inflation target. If it has made no attempt to keep its promises in the past, why would anyone believe its promises about the future? Only once credibility is restored (i.e. only once people believe the Bank of England will keep its promises, regardless of whether it seems attractive at the time to do so) can we even contemplate its being credible in making complex future promises.
Thus, although certain changes to the Bank of England's target would be attractive (specifically, a switch to a price-level or average inflation target), the changes actually proposed (nominal GDP targeting or "flexible" inflation targeting) will be regarded by everybody as precisely what they are: an excuse for ignoring the inflation target even more blatantly. Only once the Bank of England is set a target that the government actually wants it to meet and that the government then enforces, regardless of whether it seemed attractive to meet the target at the time, can credibility be restored to UK monetary policy. I fear that that time is still some way off.