Stephen King: Left a bit ... right a bit ... the mysteries of inflation targeting and Bob Monkhouse's 'Golden Shot'

I can't quite remember watching the late Bob Monkhouse's The Golden Shot: I was probably too busy playing with my Corgi toys. I can, however, recall his anecdote - possibly apocryphal - about Bernie the Bolt.

As he explained it - during a speech at the end of a rather boozy dinner I happened to be attending, so I may not have captured all the relevant nuances - The Golden Shot, broadcast live to the nation, required viewers to phone in to the studio and, using the image on their televisions at home, to instruct Bernie the cameraman to aim his camera - with crossbow attached - at a target. "Left a bit, right a bit ... fire". On one occasion, though, the viewer on the end of the line was behaving a little strangely. Right a bit, right a bit ... right a bit more ... until the target has disappeared altogether. Monkhouse couldn't understand what was going on. It was only then that the contestant admitted that he was in a phone box and could only vaguely see through the window of the TV shop across the road.

All of which tells you that targets are not always easy to hit, either because the connection between instrument and target is a little vague or, alternatively, because the target is insufficiently well-defined. For economists, the obvious parallel with The Golden Shot is inflation targeting. Central bankers know in theory the inflation rate they're aiming at - either because they have selected the rate themselves, in line with the European Central Bank's approach, or because it's been imposed upon them from on high, as with the Bank of England - but they may not always know how to get there. And even if they do get there, they might wonder whether they really and truly have achieved their objective.

Part of the problem is Bernie the Bolt. Central banks are like the man in the phone box. They know what they want to achieve but they cannot be sure of how to achieve it. They are, if you like, dependent on the behaviour of others. And that behaviour, of course, changes. No one, least of all the Bank of England, thought that base rates would peak at 4.75 per cent in the latest interest rate cycle and then come down: they did, though, because the housing market responded in a very different way from previous episodes to changes in interest rates.

For central banks, though, there is a further problem. It's all very well defining a target but inflation targets are a lot less stable than those you attempt to hit with a crossbow. Indeed, central banks can't even agree with each other on the kind of target that should be hit. For the European Central Bank and the Bank of England, the inflation target is so-called "headline" inflation: it includes all those volatile components like food and energy. For the Federal Reserve, there is no formal target at all - although if Ben Bernanke, the incoming Fed chairman, is to have his way, a target might eventually emerge. His preference is for a "core" target that excludes things like food and energy.

At this point, your minds might be drifting towards related subjects - trains and anoraks, perhaps - but, during periods of sustained oil price increases, the difference between these two approaches to inflation targeting becomes a bit more interesting: higher oil prices will raise headline inflation relative to its target but need not raise core inflation relative to target unless, say, wages start to rise in response to the initial hikes in oil prices. In other words, the pressure on the ECB and Bank of England to raise interest rates might, in these circumstances, be greater than for the Federal Reserve: same shock, if you like, but a very different reaction.

Central bankers would probably object to this rather simple description. They would argue, instead, that the choice of headline or core inflation is a secondary consideration: what really matters is not so much the monthly inflation rate and its occasional oscillations but, rather, the stability or otherwise of inflation expectations. On this interpretation, monthly movements in either headline or core inflation should be regarded as no more than ciphers that provide clues about the likely evolution of inflationary expectations.

Of late, inflationary expectations have been remarkably low and stable. Central banks appear, therefore, to have been doing an excellent job. The central banks, though, don't seem to share this view. If their performance has been so good, and inflation expectations have been so well-contained, why have the Federal Reserve and, more recently, the European Central Bank chosen to raise interest rates? The answer comes in the form of an unobservable counterfactual: in the absence of interest rate increases, inflationary expectations might well have picked up, so we can all thank our lucky stars that our policymakers acted with such pre-emptive zeal.

This, though, shifts policymaking away from the known knowns, to use Donald Rumsfeld's take on epistemology, towards the known unknowns. We trust our central banks because they can deliver low inflation. The central banks, though, act on the basis of their expectations of how our inflation expectations will evolve on the back of a range of potential inflationary shocks, including increases in house prices, stock prices, commodity prices and, most obviously, oil prices. But what if their expectations of our expectations are wrong? For example, higher oil prices have not led to the traditional "second round" increase in wages that was a key feature of previous inflationary periods: if that second round effect does not materialise, then a trigger-happy central bank might appear, in hindsight, to have been mistaken.

How damaging might this mistake be? In a world of low and stable inflationary expectations, too aggressive a tightening can be reversed, as we saw with the Bank of England last year.

At the limit, however, it could be argued that the mistakes don't matter at all because changes in policy rates are increasingly irrelevant for the performance of the economy. The Federal Reserve is still unsure as to why, despite persistent increases in its key policy rate, there has been barely a ripple of reaction from long-term interest rates and from financial markets more generally. Might it be that our trust in central banks is now so great that they achieve their goals not so much through their own actions but, rather, through the hypnotic monetary trances that we've all succumbed to?

This is not to say that central banks should give up changing their policies. Rather, the danger lies in the inflation targeting framework itself. If the inflation target is always met, it becomes more difficult for central banks to justify changes in interest rates unless they admit that monetary policy has broader objectives than the simple control of inflation. But that admission would reduce the transparency and credibility of policy that central banks have worked so hard to achieve. The broader the list of objectives, the more difficult it is for a central bank to communicate its intentions to financial markets and to the population at large.

We're likely to see an example of exactly this problem later this week. Sweden's Riksbank is meeting and will announce its latest policy decision on Friday. Financial markets are convinced that Sweden's monetary masters will deliver at least a 0.25 percentage point increase in interest rates. Sweden has a 2 per cent inflation target. The chart, though, shows that inflation is currently only a whisker above 1 per cent. So why is the Riksbank raising rates? Doubtless it will argue that prospective inflation is in danger of being too high. But, like the Bank of England before it, I suspect that these words will ring rather hollow. Right a bit ... right a bit ... right a bit more ... oh, let's not bother with the target at all.

Stephen King is managing director of economics at HSBC

stephen.king@hsbcib.com