Monetary Policy: The End of History? Stephen Grenville Deputy Governor Monetary Authority of Singapore 30th Anniversary Conference on Central Banking in The New Economic and Financial Landscape Singapore - 20 July 2001

It would be hard to exaggerate the changes of monetary policy in the past thirty years or so. The question I want to ask, given these extraordinary historical changes, is whether we have now reached the "end of history" in the sense that Francis Fukuyama had in mind when he talked about political systems, which had evolved so that one successful system dominates and displaces others. Has this happened with monetary policy? Before I get into any detail here I want to record a summary answer — that monetary policy has reached a satisfactory state where it has a clearly defined "best-practice" framework which is performing well; but that it would be wrong to think we have reached some nirvana, from which no further perfecting is possible. So the short answer is "no" — there is more history to come.

Where we have come from

When I first studied economics, going back more than thirty years, my recollection is that we thought about monetary policy in the following terms:

• monetary and fiscal policy were near-perfect substitutes, used in tandem to achieve the same short-term goal — the smoothing of the business cycle. If the cycle could be smoothed, all would be well in the best of worlds, with inflation under control and growth as good as it could be;
• the analytical framework behind this was a non-vertical Phillips curve — i.e. there was some trade-off between real output and inflation, although we did not think much about whether this trade-off would remain in the longer term;
• monetary policy worked through regulation and quantitative controls (which were, it should be said, quite effective in this simple world);
• low interest rates were widely regarded as a "good thing", perhaps because the number of borrowers was greater than the number of rentiers, but more probably because we thought that growth and investment were in themselves good things, and that these would be encouraged by low, stable interest rates. It may also be relevant that governments were reluctant to inflict losses on bond-holders, and wanted to borrow more, cheaply;
• there was not too much concern about international integration, particularly in the capital market. To the extent that we worried about the external sector, it was always in terms of the current account and the pressure that might come on the (fixed) exchange rate;
• there were, of course, financial crises in those days, but they seemed to have a different nature. I was reminded of this when I read the recent quote of one banker who, when asked why his bank had come through the U.K. 1973 secondary banking crisis so well, replied that "it was simple: we only lent to people who had been to Eton". Things happened differently in that gentlemanly and old-fashioned world;
• the close working relationship between the central bank and the government might be illustrated by quoting H.C. Coombs’1 endorsement of the Montagu Norman view "that a central bank should be like a good wife. It should manage its household competently and quietly; it should stand ready to assist and advise; it can properly persuade and cajole and on occasions even nag; but in the end it should recognise that the government is the boss." (p. 63)2.

The transition

All this came to an end in the 1970s, with stagflation, the disappearance of the Phillips curve trade-off, exchange rate collapses and subsequent floats, and large and disruptive foreign capital flows around the world. There was a relatively brief flirtation with monetarism (to a large extent a direct response to the high inflation of the 1970s). Its basic tenet — a stable money demand function and control over money supply — could not be maintained in practice. Monetarism may well have provided some kind of "heat-shield" which, for example, helped the Americans get inflation down in the Volcker deflation period of 1979-1983. Its empirical foundations, however, were not stable enough to allow it to be the continuing basis of policy. But there were three lasting legacies of this period:

• monetary policy was seen as a separate, specialised function, with a comparative advantage clearly distinguishable from fiscal policy;
• it required a nominal anchor;
• the framework should constrain the decision-making process (i.e. it was not just a technical or analytical construct). Without this, the authorities might be led astray by siren voices. This was formalised in the idea of "time inconsistency", although much of the discussion in this literature provided a woefully inadequate representation of the motivation of central banks.

Meanwhile, the twin snowballs of financial deregulation and globalisation were rolling forward, gathering weight and altering the landscape as they went.

The current system

We took aboard the three legacies, while abandoning money targeting. For many of us, the alternative was to aim directly for the target of price stability. In the formalised version of this — inflation targeting — there are a variety of attractive characteristics:

• a properly specified inflation target is achievable (i.e. under the control of the authorities), unlike a money target;
• it addresses the issue of the political decision-making process, and its proclivity to be diverted by biases which will be inflationary in the long run. The central bank needs protection from these forces, in the form of independence;
• at the same time, it would be unreasonable to expect the democratic process to provide unelected officials with decision-making capacity of this nature without any guiding or limiting rules. Central to this paradigm (although by no means uniquely associated with inflation targeting) is a high degree of accountability and transparency on the part of the monetary authorities. They have to explain themselves not only to the Parliament (as the source of their ultimate authority), but also to the public at large and to financial markets in particular. Markets have become not only the transmission path of monetary policy, but are also the guardians and watch-dogs on monetary policy, ready to sound the alarm should the authorities stray from the straight and narrow;
• these inflation targets also provide an anchor for price expectations (thus addressing the problem of the shifting short-term Phillips curve).

This model was pioneered by New Zealand (who introduced it, not as a specific model tailored to the unique needs of monetary policy, but rather as part of a process of reforming governance and accountability, where the various arms of policy would be given clear goals and expected to achieve them). Many of us (in due course, the New Zealanders included) found that this narrow version of inflation targeting was unsuitable for one reason or another and we have adopted more sophisticated, less rigid or narrow versions, principally addressing the issue of how much variance in inflation is tolerated. While there may not be a long-term trade-off between inflation and output, there is a trade-off between inflation variability and output variability — to try to adhere too closely and too constantly to a narrow inflation target is likely to be unnecessarily disruptive of output. Various inflation targeters have addressed this problem with different specifications of the target, but it is notable just how similar most of us have become.

In addition to the dozen or so countries which are card-carrying inflation targeters, there are a number of others which have systems generically quite similar, if less formalised (as is the case in the United States — see the recent speech by Larry Meyer3). And, of course, there are others who have chosen a different nominal anchor — Singapore, where the exchange rate provides the equivalent and analogous anchor, with the same ultimate objective.

Where are we going?

With so many countries having evolved to quite similar systems, have we reached the "end of history"? I think, for the moment, we have reached some kind of stable resting point: certainly in Australia we are very satisfied with the performance of this framework. We regard this system as having reached an advanced stage of evolution, with the formal exchange of letters between the Treasurer and the Governor in 1996, endorsing the framework which had then been in existence for some years. It can no doubt be tweaked in various operational ways, but we would regard these as minor modifications, rather than full upgrades. So there is no obvious or compelling logic which would move us to a subsequent stage of history.

Of course, "things happen", and they may happen so as to invalidate the inflation targeting framework. One way we might be knocked off this framework is if the problematic task of forecasting inflation proves too difficult - in effect, if the lags in monetary policy are longer than our ability to forecast inflation. Similarly (observing Japan at present) it is possible that an economy could find itself in a period of prolonged deflation, with monetary policy apparently not able to counter these forces.

However, I cannot help thinking that the real issue lies elsewhere. With continuing success, the feeling will arise that it is all a bit too easy: that the old inflation monster, now tamed, seems a bit of a pussy-cat and not, by itself, of such enormous importance to justify devoting monetary policy solely to its control. Just about the whole of the world (whether inflation targeters or not) has reduced inflation to manageable levels. It seems at least possible that people will begin to regard this task as so easy, and so assured of success, that monetary policy could be asked to do more. I can think of two directions in which we might be asked to do more.

First, we have established a wide consensus that monetary policy can do something to smooth the course of the business cycle, but this should not be the primary goal. But if inflation control begins to look easy, people may ask central banks to focus the power of monetary policy more fully in the direction of smoothing the cycle. I do not propose to say more about this, other than to observe that this can, at least in principle, be encompassed within the inflation targeting framework, recognising the legitimate trade-off between inflation variability and output variability. But we also need to remember how much more painful it is to have to wind inflation back down, once price expectations have been disturbed.

More interesting, in the current debate, is whether monetary policy should do more to achieve financial stability. A clear dichotomy or specialisation has developed, in which monetary policy is seen as the instrument for maintaining price stability, and prudential regulation is "assigned to" financial stability. This dichotomy is increasingly reflected in the vogue of separating the central bank from the prudential regulator.

But financial stability is under constant bombardment — reflected in the increasing complexity of the Basel II rules; the pressure to widen the reach of financial regulation; the increasing recognition that financial stability problems are coming not just from institutions, but from markets and their evolution; and the difficulty of staying ahead of the evolution of ever-more-complex financial instruments and products (driven by competition and sometimes by the objective of getting around prudential regulations). The LTCM experience is a reminder of how the leading-edge players will be constantly pushing beyond the regulators’ capacity to understand, let alone react effectively. On top of this we see the evolution of markets, often driven by the desire to minimise risk to individuals, taking actions which may make the overall system more vulnerable — the examples here are the various forms of equity insurance which were put in place prior to the 1987 equity correction, and (more recently) the increasing use of VaRs (value-at-risk models) and various risk management procedures which will tend to ensure correlation of changes of opinion (the herd changes direction simultaneously) and of mistakes. All this is happening in a world of enormously increased capital flows.

In this world, it will be surprising if we are not asked whether monetary policy can make a bigger contribution to maintaining financial stability. The obvious place to start would be with asset prices. Asset prices already have some role in inflation targeting, in as far as they impinge on the prospect for CPI inflation. But looking at the experience of Japan during the 1980s (and perhaps the United States more recently), it might be asked whether monetary policy should do more to directly influence asset prices which (in the case of Japan) were a central cause of financial instability and the subsequent decade-long stagnation. Asset "bubbles" are clearly enormously disruptive: the issue in debate is not the desirability of minimising them, but whether they can be identified beforehand and whether the instrument available is suited to the task. Without a clear mandate to do this, and an analytical framework for identifying a bubble, it would be a bold central bank which deliberately set out to burst the bubble, knowing that many in the economy were enjoying the euphoria.

How are these issues to be resolved? One hopes that the answer is: with analytical and intellectual endeavour; and with common sense not coloured or distorted by dogma. Given the importance of the issues at stake, it seems unlikely that the history of monetary policy has ended, with no further evolution likely or possible.

Footnotes

1. Coombs, H.C. (1971), Other People's Money, Australian National University Press, Canberra. (back to text)
2. Views about the appropriate relationship between governments and central banks have changed almost as much as those about men and women! (back to text)
3. "Inflation Targets and Inflation Targeting", remarks by Governor Laurence H. Meyer at the University of California at San Diego Economics Roundtable, San Diego, California, 17 July 2001 (back to text)