Stephen King: Why deflation may yet destabilise the monetary policy regime

Asset prices can play havoc with the ability of a central bank to hit its inflation target over time

If you want a robust defence of current central bank practices, you could always help yourself to a few choice paragraphs from Christopher Allsopp's speech ("Macroeconomic Policy Rules in Theory and in Practice") presented to the conference on Policy Rules – the Next Steps at Cambridge University last week. Mr Allsopp, a member of the Monetary Policy Committee of the Bank of England, provides a series of powerful academic arguments in favour of current practices in Threadneedle Street.

At this point, I should warn you about the perils of cronyism. Chris Allsopp was my economics tutor at Oxford a couple of decades ago. So you might think that I'm going to be rather too sympathetic to Mr Allsopp's views. Well, I do have a lot of sympathy with what he has to say. But I also think that there are quite a few things that he has left unsaid. These could prove to be rather more challenging for policy makers in the years ahead.

Central to his argument, and, indeed, to the consensus within the economics community, is his observation that there is no long run trade-off between nominal developments (inflation) and the real economy. There appears to be nothing particularly controversial about this view. However, this observation may be more relevant for inflation than deflation. Deflation, after all, tends to leave the monetary authority a bit powerless for the simple reason that real interest rates tend to rise (in other words, that nominal interest rates cannot fall below zero). If those increases are beyond the control of the central bank, it would seem reasonable to argue that they could have an impact on the real economy.

Of course, deflation might be considered to be a peculiar special case that has little relevance in the real world. However, given Japan's recent experience and concerns about the relationship between asset price deflation and deflation for an economy more broadly, this seems to me to be an increasingly important issue. The underlying assumption throughout Mr Allsopp's speech is that monetary policy works. Maybe it's worth thinking about how to spot – and, hence, steer away from, situations where monetary policy is in danger of failing. To be fair, Mr Allsopp does mention this issue in passing, but I suspect that we're going to see a lot more attention paid to it in the years ahead.

Mr Allsopp puts in a stout defence of inflation-targeting regimes. He argues that an appropriate, transparent and credible policy will lead to "the anticipation of longer-term economic growth at potential rates and the understanding that deviations of output from potential and inflation from target will ... be as small as possible given the shocks hitting the economy".

Nothing wrong with all this and Mr Allsopp provides plenty of encouraging evidence to support these views. Although he admits that it may be too soon to tell, Mr Allsopp demonstrates that inflation has been remarkably stable and close to target since the MPC was brought into existence in 1997. He notes that, at the same time, financial market expectations of inflation have stabilised, suggesting that the regime carries a lot of credibility. In a later section of the speech, Mr Allsopp goes even further, suggesting that markets anticipate potential changes in monetary policy to such a degree that monetary conditions are altered even without the central banks doing anything themselves. He argues that the recent decline in market interest rates, across the yield curve, should be interpreted as an easing of monetary policy "despite no change in policy rates".

I'm not so sure about these conclusions. It's true that inflation has been remarkably well-behaved in recent years and I think the shift in financial market expectations is a particularly convincing vote of confidence in the new monetary arrangements. It is less obvious, however, that the low rate of inflation is solely the work of the MPC. Low inflation is, after all, a characteristic of virtually all major economies around the world. And, within all this, there are common factors, namely the collapse in tradable goods prices and, more recently, massive over-capacity in a whole series of global industries. To the extent that the UK has shared in these global phenomena, it is difficult to argue that inflation success has been all down to the good intentions of the MPC.

Moreover, declines in bond yields cannot be interpreted unambiguously as a loosening of monetary policy. Obviously, if the declines reflect an increase in capital availability - through a greater desire to take risk, through banking deregulation or whatever – then they would seem to point to looser monetary conditions. If, on the other hand, they reflect a drop in the demand for funds from businesses, or alternatively reflect a switch away from holdings of risky assets like equities and corporate bonds into nice, safe, government bonds and bank accounts, then I'm much less sure that falls in bond yields can be interpreted as an easing of monetary policy.

Put another way, market interest rates may come down a long way but that may not reduce the need for aggressive action from the central bank. Japan's experience in the 1990s – see my first chart – would appear to be a good example of where falls in market interest rates were in no way a sign of better things to come in terms of economic activity.

On asset prices, Mr Allsopp subscribes to the "received wisdom" that "generally it is neither necessary nor desirable for interest rates to respond to asset prices except to the extent that they contribute to inflationary or deflationary pressures". He does admit that there could be occasions when central banks might have to do something about asset prices but would prefer other policies (regulatory, fiscal and so on) to be used to deal with, say, house prices or share prices.

In my view, there may be a case for exploring the monetary linkages through to asset prices more closely. First, the environments that modern central banks have tried to create may be more conducive to bubbles. Central banks profess to hit inflation targets over the medium term and, without prejudice to that overall goal, to minimise deviations of output. If I know that, and you know that, and everyone else believes that, there is a danger that asset prices rise too far, based on the belief that the world is a more perfect place than it really is.

Second, and more importantly, asset prices can play havoc with the ability of a central bank to hit its inflation target over time. There is clearly no one-for-one relationship between asset price inflation and inflation of the price level. So, if in the short-term, inflation of the price level is close to target, it might be reasonable to ignore asset price inflation. Let's say, though, that, later on, the asset price bubble bursts, leaving people with unwanted amounts of debt (which, in a low inflation environment, tend to hang around for a long time). The process of debt consolidation could then lead to significant downward pressure on demand and inflation – so much so, in fact, that deflation becomes a possibility. Under these circumstances, real interest rates are in danger of rising – the opposite effect of that seen in the 1970s (see second chart).

In essence, the new monetary regime works well in preventing inflation. It is less obvious that it is flexible enough to guard against deflation, particularly if balance sheets are allowed to inflate in a potentially unsustainable way. Mr Allsopp provides convincing arguments to suggest that we have now got an insurance policy against yesterday's problems. It is less clear that we yet have a solution for tomorrow's potential challenges.

Stephen King is managing director of economics at HSBC