Thank you for the invitation to come here! This is my second hearing at the Riksdag Committee on Finance to discuss monetary policy. I thus cannot claim that I have very much experience of these hearings. However, this also applies to some of you who are here for the first time, and I am convinced that it will not present any obstacles to an interesting and rewarding discussion.
It is often a good idea to begin with a backward glance to gain some perspective for the current situation, and this time, with so many new members, it may be particularly useful. After the look back, I shall look forward and describe the Riksbank’s views regarding the economic outlook and monetary policy.
During the backward glance I shall assert that the new regime introduced around a decade and a half ago has worked well. We have had an entirely different stability in the development of the economy than before, and at the same time growth has been relatively high. My most important message when looking ahead is that the prospects for the Swedish economy are also bright. We see ahead of us continued good growth in the economy over the coming years and a continued improvement in the labour market. At the same time, the rise in inflation is expected to be moderate, but this requires among other things a gradual increase in the interest rate.
A look further back…
It is now almost one and a half decades since we in Sweden were forced to abandon the fixed exchange rate. This was the starting point for the changeover in economic policy that led to both monetary policy and fiscal policy taking on different roles than they had in the 1970s and 1980s. Fiscal policy had previously often been overly expansionary and contributed to excessively high inflation. It was now given the main task of ensuring that public finances developed in a way that was sustainable in the long term. Monetary policy, on the other hand, was no longer tied to defending the exchange rate in that the krona was allowed to float. Its main task was instead to more directly maintain inflation at a low and stable level, by using interest rate changes to affect demand in the economy.
The new regime has lived up to expectations (Figure 1). Following the initial deep crisis, inflation has been much lower and fluctuated less than in the 1970s and 1980s. At the same time, growth in the economy has been higher and more stable than before. Real wages have also developed favourably. In many respects, developments in Sweden are good also from an international perspective, and this was certainly not the case during the 1970s and 1980s.
There are, of course, areas where developments have not been so positive. This applies in particular to employment. However, the insight that employment in the long term depends on how well the labour market functions is beginning to gain ground. This is where efforts should be made to obtain a long-term solution to the employment problem. The decisive factor is not how the Riksbank sets the interest rate or how expansive fiscal policy is conducted. The idea that there is no conflict in the long term between objectives for good employment and low inflation is otherwise one of the reasons why Edmund Phelps is this year being awarded the Riksbank’s prize in economic sciences in memory of Alfred Nobel.
Now, I do not wish to claim that everything positive that has happened is due to better thought out economic policy. There have also been significant changes in other areas that have contributed to the favourable developments. However, it can be observed that the changeover of monetary policy has given a steadiness and stability to economic policy that was previously lacking. This should be a fairly important factor.
…and a little closer
Let me now say something in more detail regarding the developments over the past 3-4 years, as these have been rather special in some ways. At least at first glance, growth, inflation and employment have developed in a slightly contradictory manner in relation to one another during this period.
Since the middle of 2003, inflation has been below the target of 2 per cent, and periodically far below this (Figure 2). It is really only during the past six months that inflation has once again begun to approach levels reasonably in line with the target.
At the same time, growth has been high. Since 2004, GDP has increased by slightly more than 3 per cent a year on average. This is a fairly remarkable figure, both in historical terms and in comparison with many other countries, in particular Europe. All in all, inflation has been remarkably low, while growth has, if anything, been higher than usual.
Despite the good growth, employment has developed weakly. The high demand for goods and services does not appear to have resulted in any great demand for labour. However, the labour market situation recently appears to have begun to improve.
So, how can inflation have become remarkably low and employment have developed as weakly as it has, despite economic activity chugging along nicely? The economy appears to have been affected by forces that have held back both inflation and employment. What can these forces be?
To begin with, it is worth noting that it is not only in Sweden that inflation has been unexpectedly low in recent years (Figure 3). In Norway, for instance, inflation fell heavily a few years ago in a similar manner to Sweden. In Finland inflation has been even lower than in Sweden. The forces that have kept down inflation thus appear to some extent to have been common to other small, open economies in our region.
One such common denominator is probably that prices of imported goods have developed weakly. The ongoing globalisation process is one reason for this. Low-cost countries, in particular China, have taken on an increasingly large role in the world economy and we have begun to import more and more goods from these countries. If one looks at price developments for consumer goods - a category that contains many imported goods - prices have actually fallen in recent years, both in Sweden and our Nordic neighbours (Figure 4).
Another factor that has held back inflation in Sweden, and has probably been even more important, is that productivity growth has been unexpectedly strong. This has kept down production costs in companies and subdued price increases on domestically-produced goods and services. As imported goods are processed and distributed in the country before being sold to consumers, productivity growth has also held back price increases on imported products (Figure 5).
It is not entirely clear why productivity improvements have been so rapid in recent years. One possibility is that the major investments in information technology made at the end of the 1990s have now begun to show results. It is also conceivable that the higher productivity growth is a result of companies in many sectors perceiving increased competition, not least as a consequence of globalisation. As a result, they have been forced to rationalise and make their operations more efficient. This could also explain why employment has developed so weakly. More efficient production processes have enabled companies to meet rising demand without needing to recruit new staff to the same extent as in previous economic upturns.
The fact that inflation has been below target for a good three years indicates that this was a development the Riksbank had not succeeded in predicting, although we did assume that inflation would fall when the previously rapid upturn in energy prices rebounded. Had we been able to foresee that inflation would be as low as it has been, we would possibly have maintained a lower interest rate than we did. However, we were far from the only ones to make inaccurate forecasts (Figure 6). On the whole, in recent years, analysts of the Swedish economy appear to have been surprised by both the strong growth and low inflation. For example, it was not until some way into 2005 that most analysts began to assume that inflation for the year would be below the Riksbank’s tolerated deviation interval. Of course, all analysts were aware that the economy was affected by, for instance, globalisation and the rapid developments in information technology. But it is a long step from there to being able to predict exactly how great the impact of these forces will be on, for instance, productivity and prices.
Like other central banks, we are constantly refining our forecasting tools to enable us to become better at predicting the future. Hopefully we will gradually increase the precision of our forecasts. However, at the end of the day it is not realistic to believe that we should be able to predict this type of event in its entirety.
With regard to developments in recent years, it is also an important insight that inflation has been held down by forces that are essentially favourable for the economy. To put it another way, developments have been dominated by positive changes in what is known as the supply side of the economy. The fact that prices are pressed down by increased competition and increased foreign trade is something we all can benefit from – as consumers. And the fact that productivity growth in the economy is high should benefit everyone in the long term, in the form of increased welfare.
Our monetary policy strategy
Before we look ahead, it may be useful if I explain how we reason in general when making our interest rate decisions.
The objective of monetary policy is price stability – in other words, low inflation. We have specified this as consumer prices, measured according to the consumer price index, increasing by two per cent a year.
As it takes time before interest rate changes have an effect, monetary policy must be based on forecasts of the inflation rate a couple of years ahead. These forecasts are, of course, uncertain and developments in recent years in particular have shown that events occur from time to time that are difficult to predict. It is therefore necessary to assume that inflation will during certain periods deviate from the target and sometimes deviate substantially.
The important thing in the long term is not the temporary fluctuations in inflation, but that expectations are anchored around the inflation target. The inflation target can then fulfil its important function as nominal target when prices and wages are set and investment decisions are made. And in this respect we appear to have succeeded well (Figure 7).
Let me in this context also briefly mention how we make our forecasts. Previously we made forecasts on the basis that the repo rate - our policy rate - would remain constant during the forecast period. If inflation was expected to be lower than two per cent, this was a signal that the interest rate needed to be cut, and if it was expected to be higher, the rate needed to be raised. However, the assumption that the repo rate would remain entirely unchanged over a long period of time is fairly unrealistic. Many times – for instance when everyone is expecting a long period of interest rate increases or interest rate cuts – this assumption of a constant repo rate makes it difficult for us to make reasonable forecasts.
Now our forecasts are instead based on an assumption that the repo rate will develop in line with market expectations. Our current forecast assumes, for instance, that there will be gradual increases in the repo rate over the coming period. This new assumption does not in itself have any significance for the policy conducted. However, it makes it easier to evaluate our forecasts as they become more realistic and it is easier to compare them with others’ forecasts. It is also important to know this assessment to understand our communication. If the forecast shows inflation as being lower than two per cent a couple of years ahead, this is no longer a reason to cut the repo rate. It may instead mean that there is reason to raise the interest rate at a slightly slower pace than the market is expecting.
A further step in our method of making forecasts could be to make our own forecasts for interest rates, instead of using market expectations as a basis. This would require a number of changes in working methods, both for those making forecasts and for us Executive Board members. Personally, I think this would be a natural progression and it is a possibility we are considering.
It is not only the inflation forecasts we examine when formulating monetary policy and considering how to set the interest rate in future. Our policy is flexible in the sense that we also look at how the economy as a whole is developing. We have the ambition, within the framework of our price stability task, of also contributing to stable development in production and employment. How do we do this?
Well, by considering how quickly we should aim to return inflation to target. If, for instance, inflation is below target while growth in the economy is high, as has been the case in Sweden in recent years, a monetary policy that aims to quickly push up inflation through very low interest rates could lead to demand being stronger than the economy can manage further ahead. We would then risk overheating, where prices and wages begin to increase rapidly and where inflation expectations would perhaps also be adjusted upwards. The Riksbank could then be forced to tighten monetary policy substantially. If we instead gradually begin to raise the interest rate earlier, there is a greater chance that we can avoid overheating. We can thereby also avoid the recessions that often follow overheating. This type of policy can thus mean that the fluctuations in the economy are smaller over time. This is where the two-year horizon that we work with comes in. It means that we can in normal cases allow up to two years before inflation attains the target.
What I have described so far with regard to the principles for monetary policy is currently an accepted view of how monetary policy with an inflation target should be conducted. Central banks around the world have chosen slightly different varieties, where the details of the monetary policy system may differ considerably. However, the overall strategy is still the same.
How to take into account house prices?
One difficult question recently has concerned the rapid increases in house prices. These have risen at an annual rate of around ten per cent in recent years (Figure 8). Households have been taking on new loans at almost an equally rapid rate.
There are several reasonable explanations as to why house prices have been rising so much for some time. For instance, there have been relatively few new homes built over the past decade. The high productivity growth may also have meant that households expect a more favourable development in income in the long term, compared with before and therefore take on more debts and buy homes. Furthermore, the credit markets are not as regulated as they were before. Together with the expansive monetary policy, this has contributed to high credit volumes in the economy.
However, one cannot rule out the possibility that there are also exaggerated elements in this development and that expectations have become unrealistic as to how house prices will develop in future. One contributing factor could be that the interest rate in this economic cycle has been cut to a historically low level and this has led to exaggerated optimism regarding interest rates. Bond rates have also fallen to historically low levels both in Sweden and in many other countries. This could have further fuelled the optimism regarding interest rates. Of course, it is not easy to know whether these are important mechanisms or what can be considered a reasonable interest rate level in the long term. However, the longer the time that passes without the rate of increase in house prices being subdued, the greater the unease that imbalances will build up in the housing market and in households’ debt/equity ratios.
So how should monetary policy take this type of risk into consideration? This is a question that has been much discussed in recent years, both between central banks and within the academic world, as house prices have also risen rapidly in many other countries. However, there is unfortunately no common view that we can lean on. We will probably have reason to return to these issues when our and other peoples’ experiences of the fluctuations in asset prices have increased and when progress is made in the academic research in this field. It will also be interesting to see the independent assessment of monetary policy ordered by the Committee on Finance and which is expected to be ready soon.
However, it is not possible for decision-makers to sit back and wait to become wiser and more experienced. At the Riksbank we do not consider it to be reasonable to entirely ignore the risks entailed in the rapid increase in borrowing and the house prices rises, although we cannot easily take these risks into account in our usual forecasts of economic developments over a couple of years. If it proves to be the case that economic imbalances are building up, they will sooner or later need to be corrected. When these corrections occur, the result could be weak development of the economy over a long period of time.
We have therefore acted to reduce the risk of this scenario to some extent and to contribute to a calmer adjustment in house prices. We have done this by highlighting the risks entailed in the rapid credit expansion and by beginning the interest rate increases that were already on the table slightly earlier than would otherwise have been justified. I would at the same time like to emphasise that this does not mean we have downgraded our inflation target. On the contrary! We believe that monetary policy that is active and forward-looking, but at the same time cautious, is the best means of safeguarding low inflation and a stable economy in the long term.
The current situation and future prospects
How do we regard future developments? Before I answer this question it may be worth a reminder that the forecasts we have made are based on the assumption that the repo rate will develop in line with market expectations (Figure 9). This means that the repo rate is expected to rise slightly further in future.
International economic activity has been good in recent years. There are numerous indications that international growth will be relatively strong in future, although it is expected to slow down slightly (Figure 10). Despite the good international economic activity, prices are expected to rise at a fairly moderate rate. This is partly due to the fact that international price pressure appears to be continuing.
The good international economic activity is one reason why economic prospects remain favourable here in Sweden (Figure 11). However, this is also due to the expectation that growth in domestic demand will be strong. This in turn is due to continued good productivity growth and that the labour market situation is continuing to improve. Household consumption is also being stimulated by the fact that interest rates are relatively low. Little by little economic activity in Sweden will enter a calmer phase. One reason is that international growth will slow down and another is that monetary policy will become less expansionary. We are also assuming that productivity growth will in future slow down somewhat, which will mean that the economy approaches more normal growth figures in the long term.
With the development now forecast, it is reasonable to expect that inflation will rise (Figure 12). Domestic production costs are expected to increase more quickly as the economic upturn begins to make a greater impact on the labour market and thereby the rate of wage increase. The fact that productivity growth will gradually slow down also means that higher wages will have a greater impact on companies’ production costs than has been the case in previous years.
However, inflation is nevertheless expected to be relatively low during the greater part of the forecast period. A number of factors contribute to this. The continued international price pressure, combined with an expected appreciation in the krona mean that import price increases will remain low throughout the forecast period. The high productivity growth will also contribute to keeping down companies’ labour costs and thereby inflation (Figure 13). However, these inflation-dampening forces are thus expected to lose strength further ahead.
All in all, one can say that we see a favourable and not overly dramatic development of the economy before us. Essentially we see an economy where the wheels are turning at a good pace and things are generally going well. It is natural in such an economy for monetary policy to become less expansionary to ensure the good development continues for a long period.
It is difficult to say how quickly the interest rate increases should be made. The inflation forecast we publish today shows that inflation is expected to be slightly below target two years ahead. This is partly due to the fact that energy prices are expected to fall slightly and thereby hold down inflation. When the restraining effect of the energy prices abates, inflation will rise. At the same time, the economy is expected to develop strongly during the forecast period. To ensure that inflation is close to the target and to contribute to a balanced development of the real economy, we decided yesterday to raise the repo rate by 0.25 percentage points.
It may in conclusion be worth emphasising that forecasts are always uncertain and that many things may occur to change this picture. For instance, international developments may be weaker than expected if the US economy slows down more suddenly than expected. With regard to economic activity in Sweden, we instead see a risk that we are underestimating domestic demand. The way energy prices will develop in future is another genuinely uncertain factor in our forecasts.
Let me finally, before I summarise what I have already said, say a few words about our views on fiscal policy.
As I mentioned at the beginning, the economic policy regime we have had in Sweden over the past decade has functioned well. It has created a stability in the economy that we have not experienced in decades. An absolute condition for this regime has been that fiscal policy is conducted in a way that is compatible with long-term sustainable public finances. This condition would certainly appear to be met.
Having said that, one may observe that the formulation of fiscal policy in the short term affects supply and demand, employment and inflation. It thereby becomes a factor that needs to be taken into consideration when formulating monetary policy (Figure 14). The new budget entails a slightly more expansionary fiscal policy and thereby slightly stronger consumer demand than we had earlier assumed. However, above all the budget contained measures aimed at improving the labour market. In qualitative terms it appears reasonable to assume that these measures will increase the labour supply and gradually also employment. It is difficult to know how quickly this type of measure can have an effect, or how substantial the effect will be. There is thus considerable uncertainty in assessments of this nature. The overall assessment we make today is that the new government’s economic policy will have an impact on both supply and demand in the economy, but that the effects on inflation will overall be slight. However, we will have to follow developments closely and regularly evaluate our forecasts. In short, just as we always do.
Summary
Let me in conclusion summarise the most important points I have taken up today. My main message from the backward glance is as follows:
- The changeover in economic policy that began almost one and a half decades ago has given good results.
- In an efficiently function regime inflation will from time to time deviate somewhat from the target. One reason is that various changes in the economy occur now and then, as they have done in recent years, and the effects of these can be difficult to predict.
With regard to future developments, my most important points can be summarised as follows (Figure 15):
- The prospects for the Swedish economy look favourable. We are expecting growth to be good during the forecast period and the labour market situation to continue improving. At the same time, inflation is expected to rise gradually on the assumption that the repo rate will be increased roughly in line with market expectations.
- Falling energy prices contribute to inflation being expected to be slightly below target two years ahead. However, inflation will then rise.
- Monetary policy is expansionary. To ensure an inflation rate close to the target and to contribute to balanced development of the real economy, we decided yesterday to raise the repo rate by 0.25 percentage points to 2.75 per cent.