## Mr King looks at interest rates and the new Monetary Policy Committee in the United Kingdom
Lecture delivered by the Deputy Governor of the Bank of England, Mervyn King, at Que University in Belfast on 17 May 1999.
## Present at the creation
Two years ago, when it announced that the Bank of England would be granted independence n(N)ew Labour Government surprised everyone, not least the Bank of England. Since decisions on interest rates have been taken not by the Chancellor but by the Monetar Committee. Much has followed from that single bold decision to take interest rates out day politics. The transfer of the operational responsibility for setting interest r Monetary Policy Committee was a major policy, indeed constitutional, innovation. excitement of being involved in the construction of a new monetary institution for t Kingdom has both motivated and rewarded the extraordinary efforts of the Bank staff who the new process work, and I want publicly to thank them for that. In the words of Dean to be 'present at the creation' is an experience which few involved in policy are ever to share.
Inevitably there have been changes to the Bank. It has had to redirect its resources achieving and maintaining monetary stability in this country. Responsibility for supervision has been transferred from the Bank to the new Financial Services Authority. of managing public sector debt has passed to a new Debt Management Office. And the Trea relieved of its responsibility for setting interest rates, can now focus on fiscal polic to raise the long-run growth rate of the UK economy, as well as ensure that the fru growth are distributed fairly. Each body now has clearer responsibilities.
Of course, not all of our present monetary framework dates from 1997. It was following from the ERM in September 1992 that monetary policy was first directed to achieving an inflation target, initially a range of 1-4% and, subsequently 2½% or less. With a exchange rate, any anchor for the price level had to be based on a domestic nominal ta move from secrecy to transparency in the conduct of policy also dates back to 1992 when Chancellor asked the Bank to produce a regular independent Inflation Report in order 'to make the formation of policy more transparent and our decisions more accountable'. The first Inflation Report appeared in February 1993. Those changes led to a significant improvement i macroeconomic performance.
But many, especially elsewhere in Europe, doubted the UK's commitment to monetary stabil the absence of a willingness to remove operational decisions on interest rates from th arena. Long-term interest rates contained a risk premium to reflect the possibility tha and magnitude of interest rate changes might reflect political considerations. The 'Ken show' was probably nearing the end of its run. And both principals needed their own show
So when Gordon Brown announced radical changes to the Bank of England and the moneta policy framework two years ago they were widely welcomed because they offered the basis durable commitment to price stability.
What, then, is the substance of the main reforms announced in May 1997? First, there i and unambiguous objective for monetary policy - an inflation target of 2½%, set Government, with inflation defined as the increase in retail prices, excluding mortga
payments, over the previous twelve months - RPIX inflation. Second, decisions on interest r are taken by a Monetary Policy Committee, or MPC, comprised of nine individuals with expertis in monetary policy - five executive members of the Bank and four non-executives - who operat on the basis of one person one vote. Third, the process is characterised by a high degr transparency and accountability, both to provide the public with explanations of the MP actions and to hold members of the MPC individually accountable for their decisions. Have th reforms made a difference to the UK economy? My aim this evening is to pose and begin to answer three questions about the MPC two years on.
- (1) What is the track record of the MPC after two years?
- (2) What has the MPC learnt during that period?
- (3) What are the challenges for the next two years?
## What is the MPC's track record?
Because of the infamous 'long and variable' lags between changes in interest rates and t impact on inflation, the MPC does not yet have much of a track record in terms of infla outturns. But since the target given to the MPC is for inflation, let me start there. Betwe 1997 and March 1999 (the latest month for which inflation has been published) RPIX inflati averaged 2.7%. During that period, inflation ranged from a low of 2.4% to a high of 3.2%. O the past nine months it has remained within 0.2 percentage points of the target, a remar degree of stability.
It is most unlikely that inflation can remain, month-by-month, as close to the target a Unexpected movements in the exchange rate, oil or other commodity prices, indirect tax rates multitude of other influences could all lead to divergences of inflation from target. Because lags I mentioned earlier, it would take the MPC some time to bring inflation back to the level. What matters, however, is not relatively small month-to-month movements in the inflat rate, but the expectation that inflation will average 22% over a number of years.
As part of the arrangements for accountability, the Governor, on behalf of the MPC, is requir write an open letter to the Chancellor whenever inflation deviates by more than a percen point from the target. At the outset, the fan charts published in the Inflation Report showed that the odds on writing such a letter during the following two years were 3 to 1 on. In the eve letter was required. Although the probability of writing such a letter has diminished, it high. The latest Inflation Report published last Wednesday implies that the odds on our having to write a letter over the next two years are 2 to 1 against, or one in three.
Thirty years ago the UK embarked on an inflationary journey, and what a roller coaster rid turned out to be. During the 1970s inflation averaged no less than 13% a year, with a peak of 27% in August 1975. During the 1980s inflation averaged 7% a year, and it has fallen to during the 1990s. That period also contained two of the deepest recessions experienced by major industrial country in the post-war period. Since the inflation target was adopted in O 1992 inflation has averaged 2.8%. The improvement in inflation performance during the 1990 can be seen in Chart 1. 1 Inflation has been remarkably stable during the six years since an inflation target was introduced. Indeed, it has been more stable than in any period since m RPI figures were first collected in 1947.
1 Charts not included in this Review.
Moreover, the adoption of a monetary regime which produced low and stable inflation h damaged growth in output and employment. Since the end of 1992 output has grown at average annual rate of 2.8%, well above the post-war average. More significantly, gro more stable, with a standard deviation less than one-half that of output growth in earl decades. And unemployment, as measured by the Labour Force Survey, has fallen from 10.7 its peak in 1993 to a current level of 6.3%.
So since the inflation target was adopted inflation has been lower, and growth as high post-war decade, and both have been more stable. And total output in the economy ha increased for 27 consecutive quarters, the longest period of uninterrupted growth Second World War.
Part of the credit for this improvement in UK macroeconomic performance belongs to the framework (both pre- and post-MPC), and part was the result of favourable econ circumstances. From 1992 to 1996, the ability to grow at above trend rates without an i inflationary pressure was made possible by using up the margin of spare capacity creat deep recession of the early 1990s. After 1996, inflationary pressures were contained commodity and other world prices, and the impact of a sharp appreciation of sterlin developments have had a significant impact on inflation. And they were largely unpredic although monetary policy has faced some difficult choices over the past two years benefited from an unexpected downward movement in imported inflation. These exter influences have offset a much higher level of domestically generated inflation. Des inflation has been above the target more often than it has been below - in fact, it has the target in only one month so far since the MPC was set up. As the benign effects temporary external influences on inflation start to wear off, the challenges facing th become, if anything, even greater than those in its first two years.
If the time lags in the monetary transmission process mean that it is too early to jud on its inflation track record, then what other criteria for assessing its performan relevant? One measure would be expectations of future inflation. Following Gordon Br announcement of Bank independence on 6 May 1997, inflation expectations in financial ma fell sharply. Chart 2 shows the expected inflation rate at different horizons implied conventional and index-linked government bonds. These expectations fell by about percentage points on the day of the announcement. Since then inflation expectations ha further. Chart 3 shows the way in which two particular measures of inflation expectat evolved since May 1997. The first of these measures is the expectation of inflation ahead implied by bond market yields. The second is a Consensus Economics average of expectations of inflation five years ahead held by economic forecasters. Since 1997, expectations on both measures have steadily fallen towards the target, and the e forecasters' expectation has now become firmly anchored on the target. The bond market m has risen somewhat in recent months, although this reflects unusually low real yields linked debt following the introduction of the minimum funding requirement for UK pe funds.
Does this fall in inflation expectations matter? The simple answer is yes. If the MPC inflation expectations on the target, then unexpected movements in inflation will not i lead wage bargainers or price-setters to change their view of where inflation is headed will help make less likely the inflationary spirals of the past. Both inflation and out less volatile as a result.
But there is also another, albeit one-off, benefit from the new-found credibility of policy subsequent reduction in inflation expectations. Over the past two years, inflation has been c 2.5%. If inflation expectations had not themselves converged on the target, then inflation have been lower than expected. Why does this matter? Suppose wage bargains are struck on th assumption that inflation will be 3 per cent, but prices in fact rise by 2.5 per cent. Real be higher than either employers and employees expect, and employment - the demand for labour - will fall. An unexpected fall in inflation lowers both output and employment. So by brin inflation expectations down towards actual inflation, the new monetary policy framework h probably led to higher output and lower unemployment than might otherwise have occurred. We might call these benefits a 'credibility windfall'. It is extremely difficult, if not impo know how large that windfall might have been. But even a windfall of 0.1 percentage points GDP is a sizeable amount. Of course, somebody might very well point out that we could hav reaped this same 'windfall' by allowing inflation to rise to match the higher infla expectations. But that would be inconsistent with aiming to hit the inflation target, and mig have led to a further ratcheting-up of inflation expectations. It would have lost us all th that low and stable inflation brings.
## What has the MPC learnt in the past two years?
The first two years of the MPC have been a learning process for all concerned, both the memb of the MPC and those outside who observe and comment on its actions. We have learnt from experience and made changes where necessary. For example, the minutes of MPC meetings are now published after two rather than six weeks. The additional delay created unnecessa speculation about the issues on the minds of MPC members, and made it difficult for us explain our actions - for example, to the Treasury Select Committee - when our decisions h moved on from those described in the most recent minutes. So we decided to accelerat publication.
I would like to focus on two main lessons from the first two years. The first is the importance of a symmetric inflation target. The second concerns how to manage a process i which the group has a common objective but in which individual members are held personally accountable for their decisions.
## Symmetry of the target
The inflation target which the Chancellor has given to the MPC is a simple point target for inflation of 22%. Prior to May 1997, the inflation target was 22% or less. The meaning of t words 'or less' was never entirely clear. Rarely in the history of economic policy have so words caused so much confusion to so many economic commentators. Moving to a symmetric point target has removed any ambiguity as to the objectives of the MPC. That is importa because the rationale for handing operational responsibility for setting interest rates to th that it is better qualified to make those decisions than elected politicians, whereas politicians have the democratic legitimacy to choose the target.
A symmetric inflation target should also make it easier for people to understand how the MPC likely to react to the different types of shock that may hit the economy from time to time. technical language, the MPC's policy reaction function should be predictable. When a unexpected shock moves inflation away from the target, it may not always be sensible, or ind feasible, for the MPC to return inflation to the target level immediately. Shocks stay twelve-month increase of the RPIX index for at least twelve months. To offset the impact o shock on RPIX inflation in less than a year or so might imply undesirable volatility in outpu
Hence the MPC has been given 'constrained discretion' concerning the horizon over whi tries to bring inflation back to the target level. The appropriate horizon depends, in g type and persistence of the shock. There is no simple rule to determine the appropriate depends on the nature of the shocks. That is why the MPC spends so much time trying to recent economic history. On average, since shocks may take several months to have th effect, a horizon of about two years is a reasonable one over which to bring inflation target. But if shocks are sufficiently large - in either direction - then it may be sen bring inflation back to the target level over a somewhat different horizon. In such cir the MPC would make that clear both in the minutes of its meetings and in the Inflation Report , and, if required, in an open letter to the Chancellor.
## Common objectives versus individual accountability
During its first two years, the MPC has had to develop a modus vivendi to manage the p tension between operating as a team with a common objective, on the one hand, an requirement for individual accountability of its members, on the other. The Committee as needs to explain its view of where the economy is heading. The Inflation Report has been the main vehicle for expressing the collective view of the Committee. Together with the mi the monthly meetings, it provides an explanation of past actions which should enable understand how we are likely to behave in the future.
At the same time, the voting records of individual members of the MPC are in the public That disclosure improves the quality of decisions, as well as the accountability of members, because there is no better incentive to cast one's vote for the policy most l the inflation target than the prospect of having to defend that voting record in publi all members of the MPC are well aware that there is no certainty about the precise interest rates which is desirable in any given month. That is why we discuss with eac great length the appropriate course of action before each member makes up his or her ow For such a system to work it is crucial that MPC members give their best judgment and d to reach an artificial consensus. How, then, should we balance the need to give a clear how the Committee sees economic developments and the requirement of individu accountability?
There have been two main criticisms of the MPC in this respect. First, that the MPC has open about differences within it. Second, that the Committee has been not open enough. start with the first criticism. From its first meeting in June 1997 to the end of Committee was unanimous on all six occasions. Output appeared to be above trend, the market was continuing to tighten, and there was a need to slow the growth of domestic de order to reduce pressure on supply capacity. Interest rates were raised at each meeti June and August, and again in November. At that point the Committee was accused of acti a 'politburo'. But in January 1998 the first disagreement occurred. The Committee voted votes to three (the ninth member, John Vickers, did not join the MPC until June of la hold interest rates constant. In February and March the Committee was evenly divided, Governor used his casting vote in favour of an unchanged level of interest rates. Disa continued and in June the Committee voted 8 to 1 to raise interest rates for, as it tu last time in that interest rate cycle. By now, many commentators were disillusioned evidence of disagreements within the Committee. If I single out Philip Stephens f Financial Times it is only because his description, written in July 1998, of unhappine Committee was more vivid and articulate than most. He wrote:
'The Committee has thus far lacked both leadership and predictability. … On the evidence of previous six months, the MPC bears a closer resemblance to a post-graduate seminar than to forum for strategic decision-making. …To put [several] economists in the same room is to inv what one commentator has called paralysis by analysis. As a result the Committee's conclusi are entirely unpredictable.'
Since the beginning of 1998, the Committee has been unanimous at only one out of sixtee meetings. That was in July of last year when the Committee voted to hold interest rates con at 7.5%. Disagreements have become the norm, and are much less remarked upon by the press than they were before. Moreover, disagreement is now more widely accepted as natural. It wou be strange if the uncertainties facing the Committee were not reflected in small differen opinion. And views of the appropriate level of rates have never differed by more than percentage points. Nor have these technical disagreements led to 'paralysis by analysis'. In the Committee has reduced interest rates by 23 percentage points since October of last yea was precisely the analysis that led to the action.
The willingness of the MPC to change interest rates either up or down in order to meet symmetric inflation target has altered commentator's perceptions of the Committee itself. Ear its history, commentators were unable to resist labelling members of the Committee as eit 'hawks' or 'doves'. I argued then that it made no sense to use these descriptions because member of the Committee had the same objective. Members of the MPC cannot entertain closet views about their desired inflation rate because they will be held personally accountable fo judgments about the level of interest rates necessary to meet the inflation target. No ma good argument will never beat a good headline. Hawks and doves it was, though there were al references to ostriches, foxes and hedgehogs, and headless chickens. No doubt there were e less flattering comparisons made in private. Still, unlike England football managers, we hav - yet - been compared with root vegetables. And we have made progress. As Chart 4 shows, press references to 'hawks and doves' reached a peak in mid-1998 and have declined markedly since. As I predicted in a speech in May last year, 'as circumstances change, it is easy to that the 'hawks' shall be 'doves' and the 'doves' shall be 'hawks''.
But we were also criticised for not being open enough about differences of view. The ea minutes were thought to be too bland and to conceal expressions of individual points of v Much of that criticism was indeed fair. And we have tried to respond by making the minutes more accurate expression of the range of points of view that are expressed at the meeting. should stress that the discussion of the MPC focuses on the relative merits of different p explanations of what is happening in the economy, in a spirit of mutual enquiry, rather th exchange of fixed and conflicting views. A second criticism, made by the IMF team during t recent Article IV consultation, was that, whereas the minutes explained in some detail diffe of view among Committee members, the Inflation Report contained only a single projection for the inflation outlook. Again, there was some justice in this criticism, and we have responde In fact, for three out of the past four Inflation Reports , there have been explicit descriptions of differences of view among Committee members about prospects for inflation.
## What are the challenges for the next two years?
If monetary stability is to be a permanent feature of British economic life, the MPC will h overcome several challenges in the years ahead. Two of these I would like to discuss bri First, how will the Committee deal with expectations that have been raised by the remarka
stability of inflation during its first two years? Second, how can the MPC build a cons low inflation?
Despite the openness and transparency of the new monetary framework, the British, ind world, economy often moves in mysterious ways. Monetary policy is decision-making un uncertainty. That is why MPC projections are published as fan charts, not as point esti lags between a change in interest rates and its effects on output, demand and, ultimate mean that rough weather and occasional storms in the world economy are likely to blow course, at least for a time. A true test of the MPC is not whether it hits the target calm, but how it reacts to the storms, or economic shocks, that will inevitably arise.
The most immediate challenge for the MPC is that, although inflation may fall below th over the next year, before long the inflation prospect further ahead will depend more o inflationary pressures than it has over the past year or so, when external factors ma contribution in holding down retail price inflation. Lags mean that the MPC needs to lo The test of the MPC will be how it responds to the inflation prospect.
Even in the absence of unexpected shocks there are major uncertainties. Trying to un changes in the way the economy behaves is a crucial part of the MPC's work. In recent y have seen a combination of lower inflation, lower earnings growth and faster output gro previous relationships would have led us to expect. The reasons for this benign combi growth with low inflation are not easy to diagnose. Some have talked about a new era o paradigm', in which productivity growth has permanently risen. We should be cautious those who speak of new paradigms. Paradigm is a word too often used by those who would l have a new idea but cannot think of one.
Optimism about growth rates has been greatest in the United States, which is hardly s given the recent performance of the US economy. But wiser heads counsel caution. Greenspan has argued that, although there is evidence of a structural shift in th productivity, the laws of supply and demand have not been repealed. Eventually t relationships will return.
If the UK economy moves in mysterious ways, it does not always perform wonders. comparison between the US and UK is instructive. In both countries earnings growth ha lower, given the level of unemployment, than would have been predicted on the basis relationships. Inflation has not picked up despite a tight labour market. In the US one for this benign outcome is that costs have been held down by an increase in productivit Over the past three years labour productivity (output per person hour) in the US has g average annual rate of around 22%, well above the average rate of 1-12% over the pas years; and over the past year productivity growth rose to an annualised rate of 3% or m the same three years productivity growth in the UK has been 12% a year, well below the year average of 2-22% a year. In the US the application of new technology, esp information technology, is often held responsible for this improved performance. explanations should, in principle, apply to most industrialised countries, at least to But, whatever similarities there are in the recent economic performance of the US productivity growth is not one of them. Indeed, since 1990, UK productivity growth h exactly in line with the thirty-year average. So it would be premature at this stage to there has been a permanent shift in productivity behaviour in the UK, although further the recent differences in productivity growth across countries and how they might differences in the way information technology has affected working practices is importan
If not productivity, what other changes might help to explain the recent benign behaviou inflation? A more plausible explanation for the lower-than-expected growth rates of earnings, hence prices, is, in my view, a fall in inflation expectations. If that is the case, then improved performance of the last two or three years can be attributed to a combination of 'credibility windfall' resulting from the new monetary policy framework and the benign impa on inflation of the higher exchange rate and lower commodity and import prices.
In order to assess whether economic relationships have changed, it is vital that the MPC access to timely and accurate information. Official statistics comprise the bulk of infor available to the MPC. Following the recent review of data on average earnings, it has b decided to put the relationship between the Bank and the Office for National Statistics on a formal basis with a Service Level Agreement between the two organisations. Good progress ha been made on drafting that Agreement. But the MPC cannot live by official statistics alone, even on a diet of business and household surveys. In addition, the Bank has a network of tw Regional Agents. Their main task is to ask local people how they view the current state o economy. Each year the Bank's Agents make, in total, around 7,000 visits to business conta across the country. The attraction of this information is that it is timely and focussed Committee's needs. Other data are often published with a considerable lag. And there a frequently gaps and puzzles in those data which the Agents can help to resolve. Each month MPC decides on a 'question of the month' which the Agents then put to their contacts. Rece questions have covered trends in employment, the behaviour of earnings, the reaction of exp to the rise in sterling, and the likely effects of the Millennium on investment expenditure. a couple of weeks, the responses from a sample of about 150 companies are available to the MP More generally, the views of businesses around the country about the state of trade are p directly to the MPC through the intermediary of the Regional Agents who attend the monthly pr MPC briefing meetings. This may not be quite 'One 2 One' but it is a rapid and dire communication.
The Bank of England has never had an Agency office in Northern Ireland. It is time to put right. I am very pleased, therefore, to announce that in January next year the Bank of Englan open an Agency in Belfast. It will increase our contacts with, and knowledge of, the economy people of Northern Ireland. Nigel Falls, one of our most experienced Agents, will be the Agent for Northern Ireland. He is here tonight, and you will be seeing a great deal of h future.
The second challenge for the future is to build a constituency for low inflation. There is I a good deal of support in this country for monetary stability. Those who experienced the hig unpredictable rates of inflation of the 1970s and 1980s - the 'inflation generation' - have e that all major political parties are now committed to stability. But there have been times past when low and stable inflation was abandoned by those who succumbed to the temptations o monetary laxity. Some of you here tonight are, I hope, taking a break from revision for your examinations. Thirty years ago this month I was in your position. As part of my revision, I s the inflation rates since the Second World War for several different measures of inflation. were all at or very close to 3% a year. No-one in Cambridge at that time - and certainly n students that were to become part of the inflation generation - heeded the late Richard K warning that creeping inflation was about to accelerate. The rest, as they say, is monetary I would like those of you revising for your final exams this month to be able to look back in years time and recall that the year in which you graduated was when we finally laid to res idea that the UK was an inflation-prone economy. Inflation was the norm for most of my adu life. It should not be in yours.
Why then, you may ask, do we need to build a constituency for low inflation? There reasons. First, it is important that support for low inflation not wane as the experienc unstable inflation of the past recedes. It would be a counsel of despair to belie inflationary excesses of the recent past would have to be repeated in order to retain low inflation. Yet, in a recent survey of households in Germany, Robert Shiller 2 found that support for stability was much higher among those who had experienced hyperinflation. O born before 1940 as many as 90% agreed with the statement that 'the control of inflation the most important missions of German economic policy', whereas only 51% of those born 1950 agreed with the statement. Those who value health most are often those who experienced sickness. There is a lesson here. The past decade has been good for central central bankers. But that success should not blind us to the realisation that if public price stability is based on no more than recent experience of the opposite, then the legitimacy of central banks requires a more solid foundation.
So how can we build the constituency for low inflation? Openness and transparency of th and explanations of our actions, are an essential part of our task. But we can do more past few months I have been chairing a working party in the Bank designed to answ question of how we can build the constituency for low inflation. We have been exploring educational initiatives to increase public understanding of the benefits of monetary st we have also been considering the way in which public opinion polls might be used to about the degree of public understanding of the inflation target, the role of the MPC i it, and, more generally, about the inflation process itself. I hope we shall be able to results of those deliberations before long. The aim is to ensure that the inflation replaced by a new generation, a generation for whom economic success will depend more on own efforts than on the accidental consequences of high and unstable inflation for the d of wealth - whether in housing, pensions or financial savings.
## Conclusions
So what should we make of the Monetary Policy Committee after two years? As all parents second birthdays are often the prelude to a difficult period: the 'terrible twos'. But I that, whatever difficulties lie ahead, the Chancellor will be able to take some p offspring. Philip Ziegler has described the British constitution as an accumulation o invented precedents'. Indeed, there is a tradition in the UK of making radical reforms way that it is impossible, after a few years, to imagine that things were done any other predict will be true of the MPC.
The whole monetary policy process is today systematic and professional in a way th unimaginable less than ten years ago. Then, meetings between the Chancellor and Gover discuss interest rates were often called at only a few hours notice. Now, analysis, mee MPC and the publication of both the decision and the reasons for it proceed on a pre-a timetable. The change has significantly reduced the uncertainty facing financial mar enhanced the quality of the decisions themselves. It is a massive improvement, and one w earned the UK a good deal of respect on the international stage, as judged by the A reports from the IMF and the convergence reports of the European Union.
Shortly after Robert Rubin announced last week his impending departure as US Trea Secretary, he gave the Commencement Address at New York University. In it he stated th principles for decision-making that had guided him during his career. They encapsulate
2 Shiller, R, 'Why do people dislike inflation?' NBER Working Paper No.5539, April 1996
the philosophy of the MPC. First, the only certainty is that there is no certainty. Second decision is a matter of weighing probabilities, or the balance of risks, as we say. Third, uncertainty we have to decide and act. Fourth, decisions should be judged not only on the re but also on how they were made. The first three principles guide every meeting of the MPC. A the fourth is one I commend to all those who wish to make their own judgment about the MPC two years on.
Institutions matter. The Bank of England is an old institution. But its success has been ba its willingness to change and adapt. In its 300-year history, probably no change has bee significant as operational independence and the creation of the Monetary Policy Committee. T Bank aims to be at the forefront of the theory and practice of monetary policy. But we canno on our laurels. There is still much to learn and to change. Economic theory moves forward economic behaviour is never constant. The practice of monetary policy must keep up. But wh we will not change is our commitment to monetary stability and to the achievement of t Government's inflation target. It is that commitment which can deliver economic stability to country.