Independence and transparency in monetary policy
I should like to begin with a word of thanks for the invitation to this Breakfast Meeting.
This morning, rather than discuss current monetary policy, I shall be considering some questions that have to do with monetary policy's long-term conditions.
Next year, a proposal that has received the preliminary approval of the Riksdag (Sweden's parliament) will make the Riksbank more independent. Here, Sweden is participating in a development that has been in progress in the industrialised countries for a couple of decades—monetary policy decisions are being delegated to the central banks. This has been criticised, both because monetary policy is considered to focus too one-sidedly on a single objective and because the increased independence is said to limit democratic influence unduly. It may therefore be appropriate to review the arguments behind the changes and comment on what they require of the Riksbank.
A discussion of the process towards more independent central banks—and the considerations that underlie this process—must take monetary policy's objective as a starting point. I shall therefore begin by discussing why low inflation is important. Then I shall consider the arguments for central bank independence and outline the changes that are in train in Sweden. Finally I shall put the case for the importance of monetary policy transparency.
Why low inflation?
From time to time in the debate, the importance of having low inflation as monetary policy's objective is called into question. The critics consider that the fight against inflation is overdone and results in an unnecessary restriction on growth and employment. So allow me to motivate the wisdom of focusing monetary policy just on low inflation.
The Riksbank's basic function is to provide Sweden with a stable and efficient payment system. In practice, this amounts to being accountable for maintaining a fixed value of money and ensuring that the banks provide money in satisfactory way. If the payment system is to function properly, however, it must have a solid reputation. If people believe, for example, that the value of money will be appreciably eroded, or will fluctuate markedly, they will be more uncertain and confidence in the payment system will be weakened. That can result in wrong decisions.
In other words, low inflation renders the economy more efficient and makes it easier for households and firms to arrive at long-term decisions. A look at the situation in Sweden in the 1980s will illustrate this. High inflation, recurrent cost crises and devaluation cycles made economic development erratic, with weak investment and productivity as a result. More and more misplaced investments were made as a “bubble” grew and when the property market subsequently collapsed, Swedish society was burdened by heavy adjustment costs.
Inflation also leaves its mark on the distribution of income and wealth. High inflation tends to widen existing differences in wealth because it varies in its impact on different types of earnings and saving. A person who has savings in a bank and earns a fixed nominal wage is a loser from inflation, whereas someone who can save in real assets, for instance as a home-owner; stands to gain from rising inflation. Once again, Sweden in the 1970s and '80s is a good example: prices rose sharply for houses and tenant-owned dwellings, while bank assets were eroded. Inflation widened the gaps.
So far, so good, most people would say. Few desire a return to the earlier inflationary and speculative economy. A knottier question is just what rate of inflation to aim for.
In Sweden we have chosen to focus policy on an annual CPI inflation rate of 2 per cent. As monetary policy is not capable of directing inflation exactly, when the target was established in 1993 it was surrounded by a tolerance interval of ±1 percentage point.
This target is well in line with the rate of inflation that other industrialised countries, not least in Europe, are aiming for, though there are some minor differences. It is also well in line with a statement some years ago by Mr. Alan Greenspan, head of the US Federal Reserve, to the effect that inflation should be so low that economic agents need not worry about it when making economic decisions. Perhaps that is as close to the truth as one can come.
Policy for low inflation
Low inflation has been the normal state of affairs in the Swedish economy. Over the past one hundred years the rate has fluctuated around two per cent. It is only in the post-war period, more exactly from the late 1960s onwards, that inflation has taken off. The same thing happened in other countries. Part of the reason seems to have been that economists and politicians changed their view of what monetary policy could achieve. At that time many people considered that monetary policy could induce permanently higher employment; more jobs could be brought about simply by letting inflation rise. What they overlooked was the central banks' primary function of safeguarding the payment system's stability.
History has also clearly demonstrated that this strategy does not work; in many countries, unemployment rose parallel with inflation. In Sweden the rate of inflation rose to two-digit figures, accompanied by the erratic economic development which meant that we dropped down the “growth league”. Then, when the inflation bubble burst, unemployment rose.
What the economists and politicians had not allowed for was how rising inflation would leave its mark on expectations of future inflation. The original idea—that higher inflation, via lower real wages, would result in higher demand and employment—did not hold in the longer run when firms and households realised that their elected representatives and central banks were fuelling inflation. In the original scenario, the higher inflation was supposed to come unexpectedly and therefore would not result in wage compensation. In practice, the increased inflation affected expectations of future inflation, for which employees demanded compensation. With the higher nominal wages, real wages rose again and labour demand fell back. A situation rose where prices and wages chased each other in an upward spiral. Unemployment remained in the best case at its initial level.
When it was realised that monetary policy could not generate permanently increased employment, politicians and people in general had to think again. The attempt to stimulate increased economic activity and employment by means of fiscal and monetary measures had failed. This line therefore had to give way to a policy that focused on permanently stable fundamentals for economic development. One of these fundamentals is low inflation; the central banks were in the best position to shoulder the task of safeguarding the value of money.
Sweden switched to the same track as other countries in the early 1990s. Policy was directed at price stability with the explicit ambition of thereby laying a good foundation for future growth and employment. The next few years were onerous for Sweden's economy. The bill for the excesses of the 1980s had to be paid at a time when global activity was weakening and Europe was hit by currency crises. Since the summer of 1993, however, events have taken a different turn. Growth above the rate in the boom years in the late 1980s has been combined with lastingly low inflation. Employment is now picking up but in order to push unemployment down towards the earlier, very low levels, changes are presumably needed in the workings of the labour market and wage formation.
Increased independence for the Riksbank
Today, therefore, most, though not all, economists and politicians are broadly agreed that monetary policy has a very limited potential for permanently altering the state of employment. But monetary policy is capable of influencing growth temporarily. So for those who decide economic policy, there is always the temptation to try to strengthen economic activity in the short run and push unemployment down. In order to restrict the scope for such a policy, governments and parliaments in most industrialised countries have taken the initiative and decided to make their central banks more independent. This is often combined with a clearer price stability objective. Some countries have also established procedures for assessing and exacting accountability.
Experience shows that the countries with a far-reaching delegation of monetary policy have had a favourable development of inflation. It is probably not just the formal framework that has resulted in low inflation. Central banks have been most successful in countries where the aversion to inflation is strongest and most widespread.
Central bank independence is thus, you might say, a straight-jacket that parliaments and governments have put on themselves. Paradoxically, it is just by tying their hands in this way that the legislators have actually given monetary policy more scope to pay some heed to cyclical considerations. This is so as a result of greater independence and because everyone can rest assured that inflation will continue to be low in the future. In actual fact, a credible inflation target strategy normally caters to cyclical factors. An economic slowdown tends to be accompanied by a weakening of inflationary pressure, making it possible to ease the monetary stance. The stance is then tightened again when inflationary pressure rises. In this way monetary policy can contribute to economic growth, without unnecessary upsets, at a rate that is consistent with the development of productivity, employment and so on.
A Government Bill proposing increased formal independence for the Riksbank as of 1999 was presented to the Riksdag in November 1997. Five political parties in the Riksdag lay behind the proposals and they have also given the constitutional amendments their preliminary approval.
The new legislation accordingly gives the Riksbank the constitutional function of conducting monetary policy. The constitutional provisions also state that the Riksbank is not to take instructions from authorities concerning the construction of monetary policy. The Riksbank's current obligation to hold consultations with the Government prior to important decisions of monetary and exchange rate policy is converted into an obligation to inform the Government. The price stability objective is incorporated in the Riksbank Act, which stipulates that the objective of monetary policy is to safeguard the value of money and promote a safe and efficient payment system.
The greater independence also involves changes in the Riksbank's management. Instead of being directed, as at present, by a Governing Board appointed by the Riksdag, there will be an Executive Board with six full-time members, chaired by the Riksbank Governor. It is the Executive Board that will manage the Riksbank and be accountable for all monetary policy decisions. Unlike the system in the United Kingdom, besides acting as a committee for monetary policy decisions, the Executive Board will be responsible for managing the whole of the Riksbank.
Even under the new law, however, the Riksbank will still have a Governing Board. It will consist of eleven members, instead of eight at present, and will have a more supervisory role. Its principal functions will be to monitor and control the activities of the Riksbank, appoint the Executive Board and, if necessary, dismiss any member of the Executive Board who no longer fulfils the conditions required for the performance of duties or is guilty of serious misconduct.
In this context I also want to say something about the timetable this autumn that the new legislation requires. As usual, our newly-elected Riksdag will appoint a new Governing Board in accordance with the existing procedure. As previously, these eight Delegates will direct the Riksbank and decide monetary policy, up to the turn of the year. When the constitutional amendments have been approved a second time, the Riksdag will appoint a Governing Board under the new order; this will probably be done in late November or early December. Before the end of 1998 the new Governing Board—which for a month or so will accordingly exist side by side with the current arrangement—must then appoint the members of the new Executive Board. The new Governing Board and the new Executive Board will then take over from 1 January 1999.
As there will be little time for the two new Boards to prepare their necessary initial decisions, preliminary work will have to be done this autumn. Matters that the new Governing Board will have to decide at its first meeting include the Riksbank's rules of procedure, while at more or less the same time the Executive Board will have to finalise the details of work at the Bank: which matters can be delegated further down the organisation, how matters should be drafted in the Bank and similar questions. At its first meeting, moreover, the Executive Board will have to decide the frequency of its meetings, which tasks its various members are to be responsible for and how the Board is to communicate with the general public.
Importance of transparency
This brings me to the degree of transparency in monetary policy. In recent years the Riksbank has worked to make monetary policy more transparent, with good communication with market players and the general public.
A basic reason behind our drive for transparency is the need for clarity. The idea is that clarity in monetary policy will make the policy more effective. If market agents and others understand how we are reasoning, there will be less risk of unnecessary fluctuations in interest rates and the exchange rate. That in turn should help to make general economic development more stable. In drastic terms one could say that we aim to make the markets so familiar with our analytical framework and our deliberations that the available information would enable them to set the interest rate on their own.
Another reason for transparency has to do with the Riksbank's social function. This responsibility is augmented in that the increased independence has the force of law. It is then all the more important to develop and communicate the arguments behind the Riksbank's actions. This is a prerequisite for a sound discussion and evaluation of our policy and for holding the Riksbank accountable if the policy is not conducted in a reasonable way.
More generally, moreover, the Riksbank should continue to work for increased transparency in order to promote a widespread understanding of what we are doing. In the current discussion, for instance, there is some criticism of the low-inflation target. The Riksbank is said to be unduly focused on combating inflation at the expense of growth and employment. We therefore need to persist with an open dialogue to explain that we, too, desire and strive for high economic growth but that the best thing monetary policy can do to this end is just establish a stable framework for other components of economic policy.
The internal work at the Riksbank also calls for external transparency. Doing away with a hush-hush attitude to outsiders makes a frank internal discussion easier. Clear policy goals are also a help in focusing internal work and analysis on the relevant issues.
The open presentation of a clear line of thought simplifies the educational task. The Riksbank's line consists of an explicit inflation target and an account of the factors that we consider affect inflation; it also includes how monetary policy measures are constructed in the light of the inflation assessment. Current forecasts and assessments are presented in the quarterly Inflation Reports. In between we try to comment, in speeches and other ways, on any changes in our appraisal of inflation.
Unfortunately, a soundly constructed line of thought does not always suffice. From time to time we are highly uncertain about how the available information should be interpreted. That is something which we and all economic agents must learn to live with—an exact statement may not be possible in such situations.
One matter connected with the independence and transparency of the Riksbank is whether or not the Riksbank should publish minutes of its discussions and/or show how decision-makers have voted. One reason why this issue has attracted so much attention recently is no doubt that the new Bank of England legislation stresses the monitoring of how individual members have voted.
The Riksbank's new Executive Board will clearly have to adopt a position on this matter. So permit me to raise a couple of relevant aspects. It seems to me that a distinction can be made between requiring accountability on the one hand and clarity or distinctness of policy on the other.
As regards accountability, it is clearly important how individual members vote. The Governing Board's long-term influence over Riksbank policy will, after all, be exerted primarily through the composition of the Executive Board. It follows that the Governing Board—presumably the general public as well—needs to be in a position to assess how different members of the Executive Board have acted. The knowledge that voting is made public no doubt also makes every member of the Executive Board all the more keen to arrive at a personal conclusion about the difficult issues that have to be resolved.
As regards transparency, on the other hand, what matters most in my opinion is to have a clear line of thought, as I outlined earlier, and on this basis find ways of communicating how assessments change over time. The latter can be done in Inflation Reports and speeches, as at present, as well as in communiqués after board meetings, etc. In this context, information about how the individual board members have voted is not so self-evidently important.
There will doubtless be reason to return to this issue. The conclusion from such a discussion is not automatic. A good combination of the two perspectives above is probably feasible. A further consideration is the need to create an Executive Board that functions properly not only when it comes to making wise monetary policy decisions but also in the Riksbank's effective management.
Concluding remarks
Since the early 1990s the Riksbank has clearly had the task of establishing price stability in Sweden. Strong confidence in the Riksbank means that monetary policy is able to achieve the desired results without resorting to emphatic interest-rate signals. In an environment with low inflation and high credibility, monetary policy may be in a better position to smooth fluctuations in growth and employment.
Confidence in the low-inflation policy does seem to have been established. Inflation has been low and inflation expectations are in line with the inflation target. A tight fiscal policy is contributing to this but so, of course, is a monetary policy that focuses consistently on low inflation. The credibility of policy is, however, a continuous question. The process will be facilitated by the Riksbank's statutory independence but the Riksbank must also, for example, work continuously to communicate its policy deliberations clearly and distinctly.
The conditions for policy are affected by many factors. A labour market and wage formation that function better make it possible to combine low inflation with higher employment. Fiscal policy is another central matter; it needs to be sustainable in the long run and preferably also help to counter any signs of incipient overheating.
If some component of economic policy fails to promote the inflation target, the burden on monetary policy will be heavier. Stronger measures by the Riksbank will be needed as a clear demonstration of the determination to achieve low inflation. Insufficient support for the inflation target and policies that are not permanently sustainable would then be a burden not just for the Riksbank but for the economy in general.
