Rosenberg: The current monetary policy situation
The whole speech
Thank you for the invitation to come here and speak about monetary policy.
In recent months there have been many comments on what the Riksbank should decide with regard to the repo rate at the next monetary policy meeting on 5 February. We at the Riksbank have nothing against discussions on monetary policy arising. On the contrary, we find it useful as it forces us to sharpen our arguments and our reasoning and to work on improving our methods of analysis. However, to avoid any misunderstandings, it is important for us to inform others of how we work. I shall therefore begin with some general words on the Riksbank's inflation target and monetary policy strategy before going into the current situation.
The Riksbank's objectives and working methods
Our monetary policy objective is to maintain price stability, which the Riksbank has defined as an increase in the consumer price index of 2 per cent a year with a tolerated deviation interval of +/- 1 percentage point. When the inflation target was first introduced, it was rather controversial and many regarded it as unrealistic. Now there is general support for the inflation target and we have a good record of target fulfilment. There is also considerable confidence in the inflation target, judging from households' and firms' expectations of the future inflation rate. Of course, this does not prevent differing opinions being held, both within the Riksbank and by external analysts, as to how the repo rate should be managed in order to maintain price stability.
Monetary policy is based on how the inflation rate is expected to develop in the near future, primarily 1-2 years ahead. There are two reasons for this. The first is that monetary policy measures make their full impact on inflation during this time horizon. Thus, the Riksbank has little possibility to counteract shocks that affect inflation in the short term, but it can conduct a well-balanced monetary policy to ensure that inflation is maintained around an average rate of 2 per cent a couple of years ahead. The second argument in favour of forward-looking monetary policy is that this provides scope for taking into account developments in the real economy. If the repo rate were to be changed continually to try to bring inflation back to the target level as quickly as possible every time a shock occurred, it could lead to considerable, and undesirable, fluctuations in interest rates, production and employment. During the years that monetary policy has employed an inflation targeting strategy, we have developed a clear analytical framework for our work. The Riksbank has developed a simple rule to make monetary policy more easily comprehensible and to create stability when making monetary policy decisions. What this entails is that if the inflation rate 1-2 years ahead is expected to be below the 2 per cent target, the repo rate will usually be lowered, and vice versa. However, it is important to point out that this rule cannot be applied mechanically and nor has this been the case earlier. It is also important to emphasise that openness regarding our forecasts is a vital part of the monetary policy reasoning, as it enables discussion and assessment of our decisions. I will return to this issue later.
Monetary policy therefore focuses on expected future inflation and not the most recent CPI outcome. However, new results can of course affect the assessment of future events. The current economic situation is very important in that it comprises the foundation for our forecasting work and therefore affects the conclusions we will reach regarding future inflation. Nevertheless, we cannot use monetary policy measures to influence economic conditions in the very short term, unless we are prepared to make drastic changes in the interest rate.
The Riksbank's views on inflation prospects are described in detail four times a year in our Inflation Reports. In connection with the publication of two of these reports there is also a public hearing of the Riksbank Governor in parliament by the Riksdag Committee on Finance. The initiative to these meetings came from the Riksbank, as a way of creating further clarity and transparency with regard to monetary policy. A press release is published after each interest rate decision, containing a brief description of the background to the decision. A more detailed account of our discussion and the motivation for the decision can be obtained from the minutes of the monetary policy meetings, which are usually published approximately two weeks after the meetings.
Growth expectations appear to be correct
Our monetary policy decisions are thus based on the forecast of inflation 1-2 years ahead. The assessment of inflation prospects begins with an analysis of developments in the real economy. This has not been an easy task in recent years and the Riksbank, in common with many other analysts, has been forced to reconsider its forecasts of economic growth on a number of occasions.
Sweden, like most other countries, was affected by the economic slowdown in 2001. The slowdown was aggravated by the uncertainty following the terrorist attacks on 11 September. Growth in Sweden and in the OECD as a whole was then around 1 per cent. Economic policy was made more expansionary in many countries and at the beginning of 2002 there were signs that uncertainty had declined and that economic activity was about to improve. Growth improved slightly and assessments of future growth assumed that economic activity would continue to strengthen.
However, these expectations were not met. During the second half of 2002 economic prospects took a downturn once again, growth forecasts began to be adjusted downwards and the time for the economic upturn was postponed further. This applied both internationally and in Sweden. It was partly due to the confidence crisis suffered by the business sector as a result of corporate scandals in the United States and a number of large-scale bankruptcies. This contributed to a new stock market fall. The threat of war in Iraq and the ensuing outbreak of war for a time created greater uncertainty over the global economy than usual. The lack of clearly positive economic signals once the acute stage of the war was over, contributed to an increasingly pessimistic view of the economic situation during the first half of 2003. The weaker economic activity was also reflected in the Riksbank's inflation and growth forecasts, and the repo rate was cut by a total of one percentage point during spring and summer 2003.
During the second half of 2003 the views of economic activity became slightly more optimistic again. It now appears that economic growth in Sweden and the United States during 2003 was largely as we had expected at the beginning of the year. In Sweden, investment showed a much poorer growth than expected during 2003, but this was counteracted by good growth in consumption, in both the private and public sector, and by favourable developments in foreign trade. One could therefore say that the Riksbank, in common with other analysts, became overly pessimistic in connection with the Iraq war, at least with regard to the view of economic growth in the United States and Sweden. With regard to developments in the euro area, the situation is slightly different. There, the growth forecast for 2003 has been revised down further during the second half of the year and it is only very recently that some positive economic signals could be distinguished. However, it is quite clear that growth in the euro countries was much poorer during 2003 than we believed it would be at the beginning of the year.
All in all, we have made an upward revision to our international growth forecast for not only 2003, but also 2004 and 2005. The assessment in the December Inflation Report was that growth in the OECD area would be around 2 per cent in 2003 and then rise to 2.9 per cent in 2004 and 2.7 per cent in 2005. The new information that has become available since December confirms the original picture of the global economy, as our next monetary policy meeting approaches.
The monetary policy decision in December was based on the assessment that the growth rate of the Swedish economy would gradually improve from 1.5 per cent in 2003 to 2.5 per cent in 2004 and 2005. So far, the new information received is consistent with this view.
The significance of the labour market for inflation forecasts
Developments in the labour market are very important for our inflation forecasts and we have had to revise our forecasts on several occasions over the past year. Employment measured as the number of people in work was on average as expected in 2003. Developments in the public sector were unexpectedly strong at the beginning of the year, which counteracted the fact that developments in the private sector were weaker than we had expected. The strained economic situation in the public sector indicates that employment there will develop weakly during 2004, while the economic upturn is still not expected to be reflected in any increase in the number of employees in the private sector. The fact that the labour market to some extent lags behind production development in the business sector is a normal part of the economic cycle, but developments in employment in the private sector have so far been slightly weaker than we had expected.
Monetary policy can only marginally influence the weak labour market situation in the coming months. However, the situation in the labour market is a central factor in the assessment of demand and thus for long-term assessments of inflation. Labour market developments can influence future inflation in several different ways. If the demand for labour remains weak during a longer period and unemployment soars, there is a risk that this will have negative repercussions on household consumption and thus on demand. Households have become more pessimistic during the autumn, both with regard to unemployment and with regard to their private finances. So far we have seen no signs that household demand is about to slacken, but this was a factor we included in the risk assessment in the December Inflation Report. More unutilised resources could subdue inflationary pressure and finally result in a lower inflation rate.
The labour market situation can also be an indicator of supply conditions in the business sector and thereby of how inflationary pressure will develop. The weaker demand for labour in the business sector during 2003 has not been due to weak production growth, but to companies being able to increase production without increasing the number of employees. Nor is it the case that the already employed have worked more. Instead, the average number of working hours has fallen more than we had anticipated. Production per hour worked, i.e. productivity, has thus shown a stronger development than expected. But it is still too early to determine the cause of this. It is therefore also uncertain whether the increase in productivity will endure. It could be an element of the jobless growth phenomenon now being discussed intensively in the United States, although productivity growth has not been as strong in Sweden as it was there. While the statistics are uncertain, it is still exciting to see that, according to the Swedish National Accounts, productivity growth has also become more rapid in the service industries. One possible explanation for the strong growth in productivity put forward by the Federal Reserve, for instance, is that increased use of IT is now beginning to make a mark outside of the IT sectors. The explanation for the time lag before this has become visible could be that it takes time to learn the new technology and to reorganise the company in order to benefit from the advantages in terms of increased productivity from earlier investments. If the strong productivity growth proves to be permanent, it means that the economy can grow at a more rapid rate without inflation accelerating.
One question concerns the consequences of the poor employment situation and strong productivity growth for the outcome of wage negotiations and wage developments as a whole. If the labour market is functioning efficiently, the weak labour market situation should result in modest wage increases. Unfortunately, we know that, historically, this has not always been the case. The good productivity growth may also make it difficult for firms to withstand higher wage demands. It could become particularly worrying if the wage negotiations assume that the improvement in productivity is permanent and it later begins to decline again.
It now looks as though recent wage developments may have been calmer than expected. In the December Inflation Report we made a downward revision to the forecast for unit labour costs in the business sector for all three years covered by the forecast. This was because of the slightly lower rate of wage increases and a slightly more favourable development in productivity. The question of whether there is motivation for a further adjustment in the same direction is one that will be discussed at our monetary policy meeting in February.
Inflation prospects
The average inflation rate in 2003 was roughly as we had forecast at the beginning of the year, although the composition was different. At the beginning of 2003, electricity and oil prices were soaring and inflation was above 3 per cent. We forecast this to be a temporary situation and that energy prices would fall again. This has also been the case, but on the whole consumer prices for energy have increased more than we anticipated. Nevertheless, inflation has fallen to the level we expected, as price pressure in the rest of the economy was weaker than we had assumed. One important reason for this is a larger-than-expected fall in prices of imported goods. This in turn was connected with the appreciation of the krona and the weaker economic activity in Europe. The strong productivity growth in the business sector probably also contributed. Now, as we look ahead, it is important that the more lasting effects on inflation from the business cycle are distinguished from price fluctuations caused by disturbances in the energy markets and other more temporary factors.
Monetary policy
The views on inflation prospects, which in turn are based on the economic situation, employment, productivity and wage developments, held by the Riksbank and other analysts have not differed significantly. However, even minor differences between forecasts can lead to differing conclusions with regard to monetary policy. Consequently, it is not surprising that the Riksbank and other analysts sometimes reach different conclusions on what can be considered a suitable change in the repo rate. However, according to Swedish law, independent responsibility for monetary policy has been delegated to the Riksbank. Naturally, we listen to all well-founded analyses but in each situation we must ultimately rely on our own assessment of the right monetary policy measure for achieving the inflation target. It is nevertheless important, and of course a very reasonable request, that we are able to explain and motivate our assessments and decisions. We achieve this through the publication of our Inflation Reports, press releases, speeches and the minutes of the monetary policy meetings.
If we are open and clear with regard to our forecasts and assessments, it is easier for others to evaluate our monetary policy. This is particularly important as the effects of a repo rate decision are felt after a time lag. A public inflation forecast makes it possible to assess and evaluate the monetary policy stance at an early stage. An open discussion also contributes to increasing understanding of the inflation target. It is part of the strength of the monetary policy regulations that there has sometimes been rather lively discussion, without this having an effect on confidence in our main objectives. Another advantage of openness is that it gives forecasters, both inside and outside the Riksbank, the opportunity to make comparisons that help to sharpen our forecasting. We must not forget that inflation forecasts are made under great uncertainty and that inflation is affected by unpredictable factors. Nevertheless, during the 10 years we have employed inflation targeting, the average inflation rate has remained very close to the target level.
To summarise, we can observe that the growth rate for 2003 was weaker than we had expected in 2002. The inflation rate, however, was more or less as expected. It is possible that we became slightly too pessimistic with regard to economic activity in 2003. When we made our monetary policy decision in December the international upturn appeared to have been confirmed, and there were also signs that the Swedish economy was strengthening. Inflation had developed roughly as we had calculated in earlier forecasts. Although energy prices did not fall as far as we had calculated, domestic inflationary pressure excluding energy prices proved to be lower than expected, probably because of the weak labour market development and good productivity growth. The falling import prices had an even more significant effect on inflation. We assumed in our inflation forecast for the coming years that import prices would cease falling and instead begin to rise gently as international economic activity improved. The forecast for domestic cost pressure was adjusted slightly downwards as this was expected to show only a slight increase during the forecast period. The conclusion was that underlying inflation excluding energy, which was assessed as the best mirror of cyclically-related inflationary pressure, was expected to be in line with the target 1-2 years ahead, and this was why the repo rate was left unchanged.
The information received so far since December does not appear to significantly alter the picture of a recovery in economic growth. However, there may be a risk that the weak labour market situation will last longer than anticipated and have repercussions for growth and inflation through weaker private consumption. Now that we are in the midst of compiling the data on which to base our decision at the February meeting, it looks as though the new information received since December gives scope for a further slight downward adjustment to the forecast of domestic cost pressure. This is because it seems that labour market growth will be somewhat weaker than expected, and thereby also the rate of wage increases, and the fact that productivity growth during the first three quarters appears to have been somewhat more favourable than we had anticipated. The risk of contagion effects via energy prices also appears to have declined. However, as we said, we are in the midst of our forecasting work and not all of the information is complete as yet. Our monetary policy meeting will be held on 5 February and this is when the final decision on the repo rate will be made.
Thank you.
