Nyberg: The Riksbank's monetary policy strategy
Introduction
Let me begin by thanking you for the invitation. It is a pleasure to be able to come here and talk about the Riksbank and our monetary policy.
When conducting modern monetary policy, it is important that the principles guiding interest rate decisions are known to all. Monetary policy largely has an effect through its influence on expectations of future inflation and interest rates. Therefore, we regularly describe our views on economic developments in our Inflation Reports and publish the minutes of our monetary policy meetings to enable everyone to understand the considerations we make. In addition, twice a year the Governor of the Riksbank holds a public speech for the Riksdag (the Swedish Parliament) Committee on Finance and responds to questions from members of parliament and journalists. Moreover, we Executive Board members hold a large number of speeches, in which we discuss various aspects of the Swedish economy and explain our interest rate decisions. I should say that a continuous debate on monetary policy and its terms is a necessary condition for the policy to function satisfactorily.
One difficulty with monetary policy is that it does not have an immediate impact. It is therefore necessary to base monetary policy decisions on forecasts of future inflation and developments in the real economy. Of course, these forecasts are uncertain and can also be made in different ways. One of the main purposes of the Inflation Reports is to explain how we make them, so that others can question us and discuss our methods. I shall return to the forecasts, as we have recently changed our methods in this area.
During the spring I and my colleagues on the Executive Board described in a number of speeches the principles that guide our monetary policy. In May we also published a special document where we described these principles. (1) One of the reasons why we considered it important to publish such a document was that there had been a lively debate on monetary policy for some time with disparate opinions. For instance, some people considered we should have more than one objective for monetary policy, in addition to the inflation target. Others claimed that we already had more than one objective, while some debaters pointed out that more than one objective risked creating confusion.
Although the debate has not been quite so lively since then – which may of course be connected with the fact that we have had a long, pleasant summer that may have turned people's thoughts to other matters than monetary policy - I see that monetary policy is still sometimes described in a way that I do not really recognise. Two questions in particular tend to arise: Do we only take into account inflation, or do we also care about the real economy, for instance, the employment situation? And what do we think of the rapid increase in household indebtedness and in property prices?
The aims of monetary policy
To understand the monetary policy decisions, it is of course essential that there is no uncertainty regarding the objective of monetary policy. There is only one statutory objective for monetary policy and that is to maintain price stability. We at the Riksbank have then chosen to specify this objective in terms of an explicit inflation target. This entails an annual change in the consumer price index (CPI) of 2 per cent with a tolerance interval of ± 1 percentage point.
One might ask why we allow a certain measure of inflation when Parliament has given us the task of maintaining price stability. One reason why we have chosen not to set the inflation target at zero is that an inflation target that is too low can under certain circumstances create adverse conditions for employment. There is an inbuilt resistance within the economy to nominal reductions in prices and wages and a certain measure of inflation can therefore make it easier to implement necessary changes in relative prices and wages. A consensus has developed among the countries with economic policy aimed at price stability that a reasonable rate of inflation is around 2 per cent. At the same time, I would like to point out that there are no scientific grounds for determining exactly how an inflation target should be formulated.
Of course, it is impossible to have the ambition that inflation should always be close to two per cent in all situations. The economy is exposed to unexpected shocks and these cannot be immediately counteracted as monetary policy’s impact comes after a time lag. An attempt to bring inflation back on target too quickly may also lead to unnecessarily large costs for the real economy. Let me explain what I mean here.
Monetary policy cares about the real economy
If inflation is for some reason above or below the inflation target, it is necessary to decide how quickly it should be brought back on target. In these situations we have good reason and good opportunity to take into account developments in the real economy and not merely in prices. If, for instance, an increase in the price of energy has brought inflation well above two per cent, it would be possible to use a large interest rate hike to quickly bring inflation back on target. However, this has a price in the form of slowing down economic growth. If high productivity growth and low import prices were instead to lead to inflation being well below two per cent, it might be possible to correct the problem quickly by cutting the interest rate. However, this would risk stimulating the economy in a way that would not always be desirable. By bringing inflation back on target at a slower rate, the real economy will develop in a more stable manner. We can thus contribute to stability, although monetary policy cannot affect the long-term growth rate in the economy.
It is not stated in the law that the Riksbank should take developments in the real economy into account in addition to the price stability target. On the other hand, it was stated in the preparatory works for the Sveriges Riksbank Act that the Riksbank, without prejudice to the price stability target, should support the goals of general economic policy with a view to maintaining a sustainable level of growth and employment.
Not having the ambition of bringing inflation back on target as quickly as possible thus creates flexibility in formulating monetary policy that provides scope to take into account developments in the real economy. At the same time, it is important that this flexibility should not give rise to doubts about maintaining the inflation target in the long term. This is why we have chosen to specify a time horizon, which states that monetary policy should normally be aimed at achieving the inflation target within two years. The two-year horizon should be interpreted as a restriction as to how much consideration can normally be given to real economic developments. It is a restriction that we have imposed upon ourselves – like the specified inflation target – to create credibility for the price stability target. Similar reasoning lies behind the set tolerance interval we have chosen around the target. While the tolerance interval provides an interval within which inflation can normally be expected to fluctuate as a result of unexpected shocks and insufficient knowledge of exactly how monetary policy affects the economy, it provides a limit as to the size of the deviations that can be tolerated.
However, let me point out that there can be certain situations where the shocks to the economy are so great that there is reason to allow even longer than two years to bring inflation back on target. If our assessment was that such a situation had arisen, we would make this clear when publishing our monetary policy decisions. One condition is that we do not find ourselves in a situation where we would risk undermining confidence in the inflation target or in monetary policy.
These thoughts on how to take into account developments in the real economy are of course not unique to the Riksbank. Although the guidelines for inflation-targeting monetary policy vary slightly from one country to another, all central banks with inflation targets conduct monetary policy that to some extent takes into account the real economy. This is usually termed flexible inflation-targeting. In other words, we are not “inflation nutters”, who want to bring inflation back on target as quickly as possible, whatever the cost.
The importance of inflation forecasts for interest rate decisions
Let me move on to talk about the assumption for the repo rate that we use when making forecasts in our Inflation Report.
The forecasts in the main scenario in the Riksbank’s Inflation Report have been based over the years on the assumption that the repo rate is held unchanged during the forecast period. This assumption had the advantage of providing a very simple and clear policy rule for monetary policy: If the forecast for inflation at the end of the forecast period was below two per cent, the repo rate would normally be cut, and if it was above two per cent the interest rate would be raised. If the forecast was very close to two per cent the repo rate would remain unchanged. Of course, this rule could not capture all of the nuances in the monetary policy discussion, but it provided a rough explanation of the monetary policy decisions.
However, there were several disadvantages with the assumption of a constant repo rate. In situations where the market expected clearly rising or falling interest rates, it was difficult to compare our forecasts based on the assumption of a constant repo rate with those made by other forecasters. One reason was that the assumption of a constant repo rate often resulted in our forecasts for inflation differing from those of other forecasters; being higher when inflation was rising and lower when inflation was declining. This caused confusion in the discussions on monetary policy.
Since October last year, our forecasts are instead based on the technical assumption that the repo rate will develop in line with market expectations regarding the future repo rate, as expressed in the implied forward rates. This has made it easier to compare ourselves with other inflation forecasters and also to assess our monetary policy. (2)
The new interest rate assumption we now apply means that we can no longer use our old policy rule. If the forecast with a constant repo rate was close to two per cent, the repo rate would be held unchanged according to the old policy rule. If the forecast according to the new outlook shows inflation close to two per cent, the repo rate will normally develop in the way reflected by market expectations. This could mean that the repo rate, given the underlying inflation forecast, should be raised or lowered in a number of stages over the coming two-year period.
The new outlook thus means that the monetary policy discussion is based on an interest rate path throughout the entire forecast period, while the old policy rule only took into account what should be done at the time the decision was made. I have seen several examples where the old policy rule is used in discussions concerning the new interest rate path. This is fatal, as it can lead to entirely wrong expectations of monetary policy. The old policy rule is dead and the market now needs to bury it.
Even with the new outlook we care about developments in the real economy when making monetary policy decision. If, for instance, the forecast inflation rate were to be close to target two years ahead, but showed a rapid acceleration at the same time as growth in the real economy was high, it is possible that the implied forward rate path could not be expected to reflect a desirable development in the repo rate. In this type of situation it is possible that the repo rate would need to be raised more quickly at the beginning of the forecast period. When we make our decisions regarding the repo rate, we thus do not merely look to see whether inflation is close to target two years ahead, we take into account the entire development paths for inflation and the real economy. A desirable monetary policy is characterised by inflation under normal circumstances being close to the inflation target in a two-year time perspective while at the same time the paths for inflation and the real economy do not exhibit excessively large fluctuations.
For the record, I would like to point out here that the forecast assumption that the repo rate will follow implied forward rates does not of course entail any commitment from the Riksbank that this will be the case. We have no unique information about the future. We cannot in advance commit monetary policy to a particular sequence of future interest rate changes, as future economic developments are uncertain. The stand we take regarding the development of the repo rate at our monetary policy meetings is based on the information available at the time the decision is made. Our considerations can and should change if the economy develops in a manner different to that we had expected.
Managing risks in monetary policy decisions
As monetary policy must be based on uncertain forecasts of the future, there is a need to manage the risk that the economy will develop in a way different to what was expected by the central bank. The former Federal Reserve Governor Alan Greenspan called this the risk management approach to policy. Given this way of thinking, our Inflation Reports contain a special section where we discuss alternative scenarios for inflation and the real economy and the probability that these will materialise. It is the forecast of inflation and the real economy taking into account the risks of alternative scenarios that forms the basis of our monetary policy decisions.
Of course, it is not appropriate for a risk analysis to take into account all imaginable risks; not even all those that would have dramatic consequences if they were to materialise. For instance, it is not reasonable, in my opinion, to allow monetary policy decisions to be governed by the risk of global imbalances, as we are unable to affect this development through our own monetary policy. (3) On the other hand, I consider that it is important to always take into account risks that arise from the monetary policy conducted, even if it is sometimes difficult to quantify these risks and calculate their consequences.
One problem faced by many central banks with inflation targets, and which relates to this, is the risk of unfavourable developments in property prices and household indebtedness as a result of a long period of expansionary monetary policy. Sweden is no exception here.
Both cyclical and structural factors justify the growth rate in household debts and property prices being high over a period of time in Sweden and there is thus reason to take this into account in the normal analysis and forecasting work. Rising house prices lead to wealth effects that influence decisions regarding consumption and saving. Developments in debt affect the impact of interest rate changes on consumption. This does not necessarily entail any dramatic consequences. However, when the rates of increase remain at a high level for a long period of time, there is an increased risk of imbalances in the build up of debt and in house prices. This could in a worst case scenario lead to large fluctuations in both inflation and the real economy further ahead. The question is how to manage the risk of this type of imbalance occurring.
One way of managing this and similar risks is to make a change in the repo rate sooner or later than would otherwise be assessed as the most appropriate timing, given the forecasts for inflation and the real economy. Let me illustrate this with a brief summary of our reasoning in connection with the monetary policy decision made in February this year.
We then raised the repo rate, despite the fact that the inflation forecast in itself could have warranted waiting some time before implementing an increase. Our assessment was that there were other circumstances that indicated the repo rate should be increased in February, but that it could thereafter be raised at a slightly slower rate. The very expansionary monetary policy conducted in recent years, together with the weaker krona, has stimulated the economy considerably. It has contributed to our now being in a situation where household indebtedness and house prices have increased at a more rapid rate than can be regarded as sustainable in the long term. Our assessment at the monetary policy meeting in February was that a slightly earlier increase in the repo rate would reduce the risk of an abrupt adjustment in house prices and household indebtedness further ahead, which might have adverse consequences for both inflation and the real economy. One could say that, bearing this risk in mind, we discussed the appropriate timing of an interest rate increase that we anyway considered necessary with regard to the inflation target.
I and my colleagues on the Executive Board of the Riksbank have pointed out in several different speeches that this procedure does not mean that we have introduced other objectives for monetary policy in addition to the price stability target and the aim to contribute to subduing fluctuations in the real economy. We expressed this quite clearly in the special document on monetary policy strategy we published prior to the summer. It therefore surprises me when I still see monetary policy described in some contests in a way that implies that we have other objectives besides price stability when making decisions on the repo rate.
I would also like to emphasise that our management of the risk of an adverse development in house prices and household indebtedness does not in any way mean that we consider that these aspects have greater relevance for what is regarded as desirable monetary policy than, for instance, the situation in the labour market. From time to time situations will arise where we will need to take into account, in a different way than the usual, the risks of adverse economic developments that are difficult to quantify. Recently this has been house prices and household indebtedness, but in the future it could be other variables. This is an important dimension of the flexibility in our inflation-targeting policy.
The current situation
Let me conclude with a few comments on our most recent monetary policy decision. A more detailed account is provided in the separate minutes of the monetary policy meeting, which were published the day before yesterday.
We chose at our monetary policy meeting on 29 August to raise the repo rate by 0.25 percentage points. Now that we have chosen to conduct less expansionary monetary policy, this is because we have made the assessment that it is necessary to avoid inflation being above target a couple of years ahead, while at the same time it contributes to a balanced development in the real economy. The increase in the repo rate is in line with the repo rate path expected by the market and which was used as a basis for the inflation forecast.
On the one hand, the high GDP growth indicates that inflation could rise more quickly in future than we assumed in June. But on the other hand, productivity has been slightly stronger than expected, which has contributed to holding back cost developments and the rate of increase in unit labour costs. This has a dampening effect on inflation.
However, there are indications that the labour market is beginning to strengthen, which should contribute to rising cost pressures. Our overall assessment was that inflation will increase and will be in line with the target a couple of years ahead, given that the repo rate is increased approximately in line with market expectations.
Our assessment at the monetary policy meeting was that the upside risks and downside risks were roughly equal. Among the upside risks we mentioned were higher energy, oil and electricity prices, with a risk of contagion effects and a higher rate of wage increase following on from good growth in the economy. The downside risks included the continued strong domestic productivity growth and international price pressures. With regard to household indebtedness and house prices, we expect that the growth rates will slow down, but we have not yet seen any certain signs of this.
The picture painted earlier in the year still remains largely unchanged. Our assessment throughout the year has been that the outlook for inflation and the economy over the coming years indicates that it is reasonable to assume the repo rate will need to be gradually increased. In accordance with this view, we have raised the repo rate four times so far this year and my assessment is that the repo rate will need to be raised further in future. However, as I mentioned earlier, we both can and should change our assessment if the economy develops in a manner we had not expected.
Thank you!
Notes
1) See “Monetary policy in Sweden", Sveriges Riksbank (this document can be downloaded from the Riksbank’s website, www.riksbank.com, or ordered as a booklet).
2) This is further described in the box entitled, ”Changes in the Riksbank’s forecasting methods”, Inflation Report 2005:1, in Irma Rosenberg’s speech ”The Riksbank and monetary policy” at Danske Bank, 29 September 2005 and in Lars Heikensten’s speech, ”Thoughts on how to develop the Riksbank’s monetary policy work”, at the Swedish Economics Association on 22 February 2005.
3) See also Tucker, P., (2006), "Reflections on operating inflation targeting”, Bank of England Quarterly Bulletin: Summer 2006.
