The collapse of the UK's inflation target

Gordon Brown's much-lauded economic reforms had, on my count, seven key elements:

1. Operational independence of the Bank of England
2. A modification of the previous inflation target to include a specific value as well as a band
3. The use of various "fiscal rules" to guide fiscal policy
4. The concentration of all financial regulation into one super-regulator, the Financial Services Authority (FSA)
5. A Tripartite framework for promoting financial stability including the Bank of England, the FSA, and the Treasury
6. A "taper relief" system for capital gains tax whereby investments kept in a business for a longer period were subject to less tax that those extracted quickly, as soon as capital gains were made
7. The abolition of the previous "surplus Advance Corporation Tax" regime, a reform intended to encourage longer-term investment and discourage over-payment of dividends

Of these:

• Reform (3) has been long discredited, particularly by the Treasury's continuous changing of the definition of the economic cycle and the fact that in 2005 we had the ridiculous spectacle of the UK's macroeconomic credibility depending on the accounting definition of roads maintenance
• Reform (4) was never considered particularly useful, merely convenient (or even just "tidy"), and is totally discredited in one particular: the removal from the Bank of England of prudential supervision of banks, a reform seen to have been an important (bad) contributor to the Northern Rock debacle
• Reform (5) was exposed as totally useless and unworkable by the Northern Rock debacle
• Reform (6) was clearly not regarded as valuable by the government itself, since it decided (without consultation and over a weekend) to reverse it in the 2007 pre-Budget Report as a way of funding their inheritance tax gimmick.
• Reform (7) has long-since been discredited as little more than a £5bn per year raid on the UK's pension funds, and an important contributor to the collapse of the UK's previously excellent pensions savings reserve

So, that just leaves Reforms (1) and (2).  I never thought that (1) was a reform that we should endorse for ourselves or consider a great achievement.  I shall explain why in a partner posting later this week.  The only one of these seven key reforms that I considered unambiguously good and worthy of our endorsing was (2).  And the government's behaviour of the past year has ruined that.  Last week's inflation report represents the collapse of the UK's inflation target.  I shall explain.

Inflation targeting was the monetary policy regime introduced in the UK in 1992, copying the system employed by the first inflation targeter, New Zealand, from 1990.  There are many different possible monetary policy regimes.  One is to use a fixed exchange rate.  For many centuries the UK used a fixed exchange rate for the pound, fixing its rate of exchange into gold.  Later, after the Second World War, the pound had a fixed rate of exchange to the dollar.  And under the ERM arrangement the pound had a rate of exchange to the deutschemark with a small tolerance band.  As we all know, the UK's ERM arrangement collapsed in September 1992, on Black/White Wednesday.  It was to replace this that inflation targeting was introduced.

There are many other possible monetary policies.  The US Federal Reserve employs what economists call "discretion" - that is to say, its policy setting has very general welfare- and stability-promotion goals, and is not specifically constrained.  In the past, the Federal Reserve has been a money supply growth targeter - it was mandated to pursue certain money supply growth rules.  Another possibility would be to pursue a price-level target, about which I have written many times (and to which I shall return later in this article).

As I have discussed in more detail previously, in the early period of inflation targeting there were two schools of thought as to how it should be done.  One model, the New Zealand model, employed a band of discretion (e.g. that inflation should fall in the range 0-2%).  On another model there would be a point target (e.g. 2.5%).  But with a band the tendency is for inflation to slip to the top of the band so the scope for discretion turns out to be an illusion (a 0-2% target in practice means a 1.8-2% target), whilst with a point target one misses almost all the time and it's not clear how big a miss is a Bad Thing.

The UK's initial inflation target was 1-4% RPIX inflation, with an ambition to get to the bottom half of the band by the end of the 1992 Parliament.  This was achieved.  Inflation in 1992-97 never fell below 1% or rose above 4%, and the Parliament finished with inflation at 2.5%.  The 1997 framework made things simpler and more straightforward.  There was to be a point target and a band of discretion, with a preference for the point and symmetry of the band of discretion.  The development this represented of the 1992-97 framework was very slight.  We moved from a target that might be written:

1-4%, with 2.5% or less by 1997

to

2.5% always, and not above 1.5% or below 3.5%.

This was obviously a very small development of the previous inflation target.  But I thought it was a good one.

Now in every other country that did inflation targeting, the target was often missed.  So when the 1997 target was introduced, it was anticipated that the UK would, likewise, often miss its target.  For that reason the target statement specified that if the target were missed then a letter would have to be written explaining why.

Be under no illusion.  This is what the letter-writing provision was included.  Many commentators have tried to convince us that having to write a letter should not be construed as meaning that the target was missed, because "early on, it was anticipated that there would be many letters written."  But the reason many letters were expected to be written was because it was expected that the target would be missed often, not because having to write a letter doesn't mean one has missed the target.

Now, as it transpired, no letter had to be written until 2007.  The 2007 letter incident seemed to me to be a straightforward case of negligence/complacency.  In August 2006 the Bank of England's inflation report predicted that there was a significant chance that, in the first quarter of 2007 inflation might just exceed 3%.  This could have been prevented by enacting fairly rapid interest rate rises in the late summer and autumn of 2006 (which I voted for on the Shadow MPC) but the MPC preferred to raise rates more slowly and gamble that inflation would not go above 3%.  In the end, the MPC was unlucky.

What happened next was, in my view, highly retrograde.  Instead of accepting that it has missed its target and saying it would try to avoid doing so again, the MPC attempted to redefine its target so as to argue that it had not missed, and was allowed by the Chancellor to do so.  The MPC's case was that, all along, we had been deluded.  The UK had never had a point target with a band of discretion.  Instead, it had simply had a point target, and all that the +/-1% band was about was that inside that range it did not need to account to the Chancellor for its actions.  I urged vigorously at the time that this was the wrong way to interpret matters, but since inflation was only 3.1% for one month, it was possible to write off the discussion.

Earlier this year, it became clear that inflation was once again going to exceed 3%.  I urged that the Chancellor at that point had to tell us whether he accepted the Bank of England's or my own understanding of the UK's inflation target.  And I also urged that since it appeared that above 3% was going to be considered acceptable during 2008, it would be better to change the inflation target for 2008 to 3%.

The time to have done this was at the Budget.  Unfortunately this was not done.  Instead Chancellor Darling did not clarify the meaning of the target and still committed to retain the target at 2%.  Doubtless a number of my readers thought my suggestions eccentric, but I see now that other commentators are urging the inflation target to be raised.  As I explained in January, my preference (if we were to keep inflation targeting at all) was to keep the inflation target as I had understood it pre-2007, but considered it a priority to stay within the target band.  But a raised target would be better than an extended period of missing the target and that being regarded as "okay".

Inflation is now forecast to peak at 3.7% or even more (perhaps even exceeding 4%), and to stay above 3% for much of a year.  (How is that supposed to be compatible with the inflation target, which only even specifies what should happen when inflation exceeds 3% for up to three months!  Being above 3% for a year isn't even conceived of!)  Those urging for the inflation target to be raised, or urging that it does not matter that inflation is going so high tell us that it is all driven by food and oil price rises imported from abroad, and so should not be a concern.  Interest rates should not rise to offset such "real cost shocks".  Indeed, some even urge that we remove food and fuel costs from our policy target.

So the claim is that when there is a "real cost shock" policy should ignore it.  But real cost shocks occur all the time, and are, on average, deflationary - i.e. over time the real costs of production fall. I don't recall many people urging, through the late 1990s and early 2000s that we should delete imported Chinese deflation from our policy target, or keep interest rates high because imported deflation was irrelevant for policy purposes.  Similarly with internet-related deflation, or deflation from improved computer technology.  Did we ignore all of these for policy purposes, or make sure that we allowed only their "first-round effects"?

Insofar as there is an argument relating to real cost shocks it would be this: deflationary real cost shocks tend to be spread out over time, whilst positive shocks such as oil price rises are sometimes very concentrated.  It is one thing for there to be ten years of 0.2% real cost deflation and quite another for there to be one year of 2% real cost inflation.  But because inflation targeting focuses on each year's individual inflation, that means that the apparent effect of the real cost-raising shock looks much worse (because it is much more concentrated).

The way to deal with this issue is by not focusing solely on each year's inflation rate in isolation, but instead focusing on the average inflation rate path over time.  Then ten years of real cost deflation of 0.2% will build up a margin below target that will allow the absorption, within the targeting regime, of a 2% real cost-raising shock.  At the end of this period, the price-level path (and hence the average inflation rate) will still be on target.  This is precisely how price-level targeting works (a regime about which I have written many times before).

It seems to me that our current inflation targeting regime has, in all essentials, collapsed.  Inflation will reach nearly 4%, and no-one knows whether that is or is not thought too high by the Chancellor.  Do you know?  I don't.  Do you know of any number he would consider too high for the next year?  Suppose oil prices rise to $200/barrel.  Will 5% be too high?  10%?  It seems to me that our monetary policy regime no longer imposes any particular constraint on policy.  We are operating something very close to discretion.  Rather like the Federal Reserve, the Bank of England is pursuing, in a general sense, a policy that it thinks good for us.

Now discretion isn't necessarily a terrible regime.  The Fed has used discretion for years.  But it isn't inflation targeting.  Meanwhile the Chancellor clings, like an object of pathos, to his renewal of the 2% CPI target.  Is this really the best he can do??