Heikensten: How to promote and measure central bank efficiency
Talet finns endast på engelska
Let me begin by welcoming you all to Stockholm and Sveriges Riksbank. We have more or less made it a practice to organise an academic conference or workshop at least every second year on a topic which we think is relevant for central bankers. Last year the theme of the conference was Monetary Policy and Financial Markets in an Enlarged European Union. This time we have chosen the topic of Central Bank Efficiency.
To attain legitimacy - and keep it - central banks have to use taxpayers' money in an efficient way, not least in countries like Sweden where the central bank is now an independent institution. This is an issue of great importance to the bank, and it has occupied much of our time in recent years. We want to continue this work and develop our thinking in the area. Efficiency is also a very important issue from a European perspective and will be as important should we become a member of the Eurosystem.
Central bank efficiency is a fairly abstract concept that encapsulates several areas ranging from the roles that are appropriate for central banks to perform, to cost efficiency in a narrower sense. An investigation of the relevant literature shows that there has not been much research conducted on any of these issues. Through this workshop, we hope to stimulate discussion and future research in the field. Let me take the opportunity at this point to thank the BIS for their involvement in the initiation of this workshop. We have also discussed the possibility of following up the workshop with a larger conference on a later date.
In these introductory remarks, I would like to raise some issues that I believe are important for enhancing efficiency in central banks. Naturally, my views on this are influenced by my own experience after having been at the Riksbank for almost 8 years, first as deputy governor and now as governor. In addition, I would like to discuss some issues connected to the benchmarking of central banks. I am very much looking forward to hearing your views on these issues as well as on the other topics on the agenda both today and tomorrow.
Well-defined objectives and operational independence
It is important that the objectives of public authorities are well-defined. Only then can we achieve a clear focus that enables the institution to be made accountable for its actions. Both focus and accountability are very important for efficiency.
Central banks generally have several objectives. The responsibilities of Sveriges Riksbank, for instance, are laid down in legislation as maintaining price stability and promoting a safe and efficient payment system.
Economic research has indicated that accountability is crucial for achieving the goal of price stability. While accountability should produce similar benefits for the other objectives of central banks, this is still very much an unexplored issue. As regards financial stability, one probable reason for this is that it is much harder to define financial stability than price stability. Another may be that the central bank seldom has sole responsibility for financial stability, sharing responsibility instead with the Ministry of Finance. A clear division of tasks among the authorities is important for producing accountability.
In Sweden, a large part of the responsibility for financial stability has been delegated to the Swedish Financial Supervisory Authority. The Riksbank and the Swedish FSA recently signed a so-called Memorandum of Understanding which outlined the responsibilities of the two authorities in the area of financial stability. The aim of this is to facilitate co-operation and increase efficiency in the work of each authority towards the common objective of financial stability. According to the Riksbank, the three most important factors for financial stability are the soundness of the financial institutions, contagion, and the efficiency of the financial infrastructure. The main role of the FSA is to ensure that institutions operating in the system are sound. The Riksbank, on the other hand, is responsible for the oversight of systemic risks and the efficiency of the infrastructure. This framework, however, should be open to question. For example, it might have to be altered in order to accommodate changes in the financial system. It should also be developed as our knowledge of the functioning of the financial sector increases.
With financial companies operating in several countries, the division of responsibilities among authorities in different countries will become an important issue. In Europe, a process has begun to line up a division of labour between central banks and other authorities as well as between the national central banks and the ECB. For the new structure to be efficient and stable, it is important that the roles and objectives of different actors are well defined. There are risks inherent in this process, however. The new system may prove too bureaucratic, stressing stability at the cost of efficiency. Also, it is important to prevent turf wars from dominating rational discussions about new challenges.
Not only is it important that the objectives of central banks are clear. It is also vital that the necessary means are available and can be applied.
The advantages of delegating monetary policy to an independent central bank have been analysed fairly extensively in academic research. The main argument for delegation is that politicians can avoid the temptation to use monetary policy for short-term stimulation of output and employment. The collective experience of previous years - and from academic research - is that such stimulation often runs the risk of leading to wage-price spirals, which in the end can have severe implications for welfare in society.
A parallel argument can be made concerning policy for financial stability, although it is still a rather unexplored topic in economic literature. By delegating the lender of last resort function to the central bank, and making it clear that the central bank will not support insolvent banks, the politicians can avoid the temptation to use central bank funds to bail out banks in financial difficulty.
Another reason for giving the central bank formal operational independence is that only then can the Executive Board of the central bank be made fully accountable for the bank's operations, and as I have said already, accountability is a crucial means of attaining efficiency.
Transparency
Accountability is intertwined with the issue of transparency, not only through clearly defined objectives, but also through actions. A central bank should be clear about which actions are being taken and for what reasons. Most people agree that central banks should do their best to minimise uncertainty over policy actions, as this will make policy more efficient. As my colleague Mervyn King said, if monetary policy becomes so undramatic and predictable that it is boring, then we will really have achieved a low level of policy uncertainty!
The issue that remains open to debate is which information should be published and when. At the Riksbank, we have chosen to publish our inflation forecasts four times per year and to publish edited minutes of the Executive Board meetings, where interest rate decisions are taken with a lag of about two weeks. While we strive to be as open as possible, there are of course limits. One of the most important is determined by what we know about the functioning of the economy and the effects of monetary policy.
Not everyone agrees, however, that transparency is a good thing for attaining financial stability. Many people argue that so-called constructive ambiguity is the best way to avoid problems of moral hazard in the banking industry. I believe, however, that transparency is important in this area as well. Banks that are "too big to fail" from a systemic point of view will in any case count on support from the government. Especially when it comes to countries like Sweden, which experienced a severe banking crisis not so long ago, it is likely that support to banks that encounter financial problems will be expected if nothing else is being communicated. Another risk associated with constructive ambiguity is that it could limit the motivation of central banks to seriously consider what to do in a crisis. In other words, it could provide an excuse for not thinking the issues through. However, if you are not prepared, you could easily become the bank's prisoner in a time of crisis. Through deeper analysis of the "too big to fail" problem and open communication of crisis strategies, it should be possible if anything to mitigate the moral hazard problem. If you can show that the banks do not need to be systemically important, it should make banks even more uncertain about the possibility of receiving support than any policy of constructive ambiguity.
At the Riksbank, we communicate our view on financial stability through different publications. A Financial Stability Report is published twice a year and we also publish articles on issues of financial stability in our journal, Economic Review. A new regulatory framework governing the way in which banking crises are to be managed and resolved has been proposed by a government inquiry that was set up after the Swedish banking crisis. The Riksbank strongly supports the implementation of such a framework as it would increase transparency and thereby the efficiency of crisis management even further. We have also sought to have these principles applied internationally.
Transparency is also important in the sense that it facilitates external evaluation of the operations of a central bank on the basis of relevant information. This type of evaluation will normally enhance efficiency, and it is especially important in cases where the central bank determines its own budget. The monetary policy of Sveriges Riksbank is reported to the Swedish parliament twice a year and the bank's overall activities are subjected to regular evaluation by the Parliamentary Auditors. They recently performed a comprehensive audit of the operations of the Riksbank in which they highlighted a number of areas where costs appear to have increased more than expected. We welcome this evaluation, which has given us fresh inspiration in our search for increased efficiency.
Finally, transparency stimulates improvement of a central bank's internal analysis and decision-making processes. When all-important arguments made internally have to be explained externally, the quality of the internal work rises. By being more transparent, we are now open to criticism that is both sharper and more specific, and this has helped us to improve our analysis. This particular effect of transparency is not discussed to a great extent in economic literature but is of great practical importance.
Concentration on core activities
One problem for many public institutions is that they lack focus. Research and practical experience show that a concentration on core activities is a prerequisite for good performance.
Central banks are involved in many different activities. The extent of this involvement should be a point for discussion, however. Some central bank tasks may have arisen by historical accident and then remained in the central bank domain without ever being questioned. This raises the issue of whether some services could be provided more efficiently by private entities.
For instance, the Riksbank owned a paper mill and a note-printing works in the town of Tumba, as well as a mint in the city of Eskilstuna, for three hundred years. These are largely manufacturing operations that differ greatly from the work involved in monetary policy and financial stability. Since our job is to concentrate primarily on monetary policy and the difficult systemic questions in the financial world, we decided to sell both the operations in Tumba and the mint in Eskilstuna. This has not only given us more time for our core activities, but I also believe that a more professional owner, with many years of experience in the trade, has increased the chances of success for the operations in Tumba.
In other cases, a different public authority could be the best supplier. Like most other central banks, Sveriges Riksbank is responsible for money and banking statistics. Statistics are important for users in Sweden and abroad, and the Riksbank's statistics provide a basis for analysing the development of the financial markets. In Sweden, however, we already have a public authority whose core activities involve the collection and publishing of statistics, Statistics Sweden. We decided that transferring certain statistical tasks to Statistics Sweden would not only allow us to focus on our core activities, but it would also make the production work more efficient. As a result, Statistics Sweden has been collecting and compiling money and banking statistics on behalf of the Riksbank since last month.
Furthermore, we have transformed our system for distributing notes and coins into a company with the aim of selling it. This has proved to be less straightforward than the two other cases, however. One reason is that we have tried to do two things at the same time, namely to improve the efficiency of the market for cash, and also to rationalise our own cash management. Also, it has been difficult to draw a clear line between what the actual responsibilities of a central bank are, and what represents business that the bank has carried on for historical reasons. Today, we have managed to change the market structure in a direction that we believe will benefit society through lower total costs for handling cash. However, we have not yet found a good solution for the cash handling firm we created.
Measuring central bank efficiency
As I mentioned earlier, central bank efficiency involves many different aspects, none of which have been the subject of much research. Unfortunately, there have not been very many attempts to actually measure the efficiency of central banks. Some recent studies have focused on monetary policy efficiency, and these have provided some valuable insights. However, the achievements of the central bank are typically not related to its costs. On the other hand, a number of attempts have been made to measure the cost efficiency of central banks; these have typically focused on activities for which output is measurable, such as the production of notes and coins.
Measuring the costs of central banks is relatively easy. For example, one can look at the size of the bank's staff. The problem, however, is quantifying output.
Diagram 1 shows the number of central bank employees per 100 000 inhabitants in different countries. From this, we can see that the number of central bank employees varies greatly from country to country. This would also apply if the number of employees was related to some other measure of country size, such as GDP. Not all central bank functions have anything to do with the size of the economy, however, which is probably the main reason why Luxembourg fares worst.
One probable explanation for the large differences here is that the tasks of central banks differ from country to country. This raises the question, however, whether it is necessary for central banks to carry out all the activities that they perform. As I mentioned earlier, I believe that central banks should assess the scope of their operations on a regular basis. By focusing on core activities at Sveriges Riksbank, it has been possible to cut the number of employees from about 850 in 1995 to around 450 today.
In addition, one could go on to compare the costs of certain activities. We perform benchmarking in several areas, and I know that this is something that other central banks have been doing as well. One should also compare the costs of in-house operations with those of outsourcing. Such comparisons have led us to outsource some activities, such as cleaning services and our staff restaurant.
One potential problem when comparing activities separately is that it presupposes that objectives are not interrelated, and that there are no joint costs associated with the different activities. Recently, questions regarding the connection between price stability and financial stability have attracted increasing attention and have given rise to a debate on whether there may be situations where the two goals conflict with each other. In normal circumstances, however, we do not perceive any conflict among our objectives.
When measuring costs, it is important not to count the costs of the central bank only, but also of other parties that are important to performance. For instance, in some countries the central bank is responsible for bank supervision, while in others this is the task of a special financial supervisory authority. Countries also differ with respect to the private sector's involvement in fulfilling certain objectives. In some countries the private sector provides services that in other countries are provided by the central bank.
Here, I believe that more empirical work could be done. The BIS has a natural role to play in this respect, and that was also the reason that I first contacted the BIS when I got the idea for this workshop. I also believe that more comparisons could be made within the European System of Central Banks to help us all improve our efficiency levels.
Final remarks
Let me conclude by referring back to my opening remarks. Questions of efficiency have been at the top of our agenda for a long period of time, and there are good reasons for central banks to pay extra attention to their costs. It is true that central banks provide public goods and that minimising costs is therefore not always the best solution. However, we should be aware that the pressure to attain efficiency is weak, as central banks are protected from competition. Private enterprises that compete with each other are forced to be efficient and profitable in order to stay in business and satisfy their shareholders. Central banks generally lack such pressure. As a result, we need to work hard with these issues and invite external evaluation. Cost-consciousness is especially important in cases where a central bank determines its own budget, as hard-won legitimacy can be quickly lost.
At Sveriges Riksbank, we will be continuing our search for efficiency gains, and we will so do whether or not we join the Eurosystem. To this end, we need new influences and new ideas, which I hope to get over the coming two days. Once again, I would like to extend you all a very warm welcome to Sveriges Riksbank and this workshop.
