The Swedish economy is now developing very favourably. GDP has increased on average by 3 per cent for the past three years and, in the Riksbank’s most recent assessment, it will continue to grow at between 3 and 4 per cent in the coming two years. This strong growth has brought about a fall in unemployment, which is now close to 4 per cent. While this is much higher than it was in the 1970s and 1980s, it is lower than what many people thought possible when the crisis in the Swedish economy was at its deepest in the early 1990s.
Lower central government debt reduces the future burden of interest
The central government debt, which dramatically increased in the years after the crisis, is now being paid off. At most, central government debt totalled 80 per cent of GDP in 1994. It is expected to fall to approximately 60 per cent of GDP this year. Sweden is thereby one of the most successful countries in Europe as regards tackling the problems of public sector finances. The requirements made on us to be able to participate in EMU have sometimes been highlighted in general discussion as the most important reason for the savings. The convergence requirements made in the Maastricht Treaty aim at reducing the risk for problems in the union with a single monetary policy. Reducing the central government debt to adapt to these requirements is naturally a sufficient reason for those countries that wish to take part in the union.
However, for the Swedish economy, a high central government debt creates major problems regardless of whether we participate in the monetary union or not. Weak central government finances lead to uncertainty about future tax burdens. The experiences of the mid 1990s show that uncertainty of this kind can lead to high interest rates, a weak exchange rate for the krona and substantial fluctuations in both variables. As a small country, Sweden is wholly dependent on international companies and the international financial markets. In order to safeguard welfare, we must carry out policies that create confidence in investors and financiers. We have no freedom of choice in this respect.
Reducing the interest burden for future generations is also important bearing in mind the future changes in population. The proportion of persons in the labour force will decrease considerably in the coming decades compared with those outside. Greater numbers of older people increase the costs of old age care and health care. At the same time, internationalisation means that it will be increasingly difficult to finance a welfare system by having higher taxes than elsewhere. In the light of this, it is firstly important that labour force participation and the number of hours worked increase in the economy so as to increase government revenue and payment capacity. Secondly, it is necessary that central government finances continue to be in surplus in the next 10-15 years, which can be used for amortisation. A lighter debt burden creates scope for future increases in costs. The Government’s expenditure ceiling and balance targets aim at creating such space.
At present, the prospects for growth in the economy are so good that the surplus in central government finances is expected to be sufficient to amortise central government debt at a rapid rate and to reduce income tax. The challenge lies in maintaining this ambition over a long period. It is therefore important that economic policy is focused on creating the conditions for growth. In this perspective, it would in my opinion be unfortunate to take decisions on a shortening of working hours since it would worsen the prospects for growth by reducing the supply of labour.
By taking responsibility for stable prices and financial stability, the Riksbank can make a contribution to creating favourable conditions for long-term good and even growth.
Stable inflationary prospects and free resources make possible low interest rates
The strong upswing that we now experiencing has been supported by the Riksbank being able to keep the interest rate at a relatively low level due to low inflationary pressure. Despite us being ahead of Europe in the cycle, we have an instrumental rate which is one per cent lower than in the euro area. Since inflation is at the same time lower, the differences in the real interest rate are less, however. An important reason for inflation being so low is the broad agreement on the price stability target that now seems to exist. An expression of this is that inflationary expectations remain at a stable level despite expectations on a continued very strong upswing. The Riksbank’s most important duty in future is to ensure that expectations remain at around 2 per cent inflation.
The very low inflation which is now registered has, however, surprised many people, including us at the Riksbank. An important explanation as to why GDP growth has now being able to exceed the growth in output that is possible in the long term for many years without threatening the inflation target is, of course, that the upswing started from a level of output that was greatly depressed after the years of crisis, where a lot of resources had been made redundant. When unemployment was at its peak, there were many who feared that it would persist permanently at a considerably higher level than we were previously used to. There was also opposition to the inflation target policy just for that reason. Examples were pointed to in Europe where unemployment had been over the Swedish for a long time, but where there had been greater success in combating inflation.
The labour market is performing better than expected
It is difficult to determine the rate at which GDP can increase in the long term. However, the Riksbank makes the assessment that an annual rate of increase of 2-2.5 per cent is reasonable. In the short term, output can continue to increase more rapidly as long as there are free resources. An important issue is the extent to which unemployment can fall before the situation in the labour market becomes so strained that wages and prices start to increase so much as to threaten the inflation target. The most recent demands in pay bargaining indicate that the new contracts will be somewhat higher than their predecessors, which probably reflects the fact that we are now in a period of increasing resource use. However, developments during the recent period show that large groups who have previously been unemployed have been able to return to the labour market. The increased use of resources has not led to a rise in nominal wage increases, which indicates that flexibility in the labour market is higher than we previously believed. The labour market policy carried out by the Government can have made it easier in various ways for people to move from unemployment to work although the importance of labour market policy is difficult to measure. However, it is reasonable to assume that re-entry into the labour market is facilitated by unemployed people being given the opportunity via various measures of maintaining normal routines. Employability can also have increased by the labour market training offered.
Stable inflation expectations and new arrangements for collective agreements have probably also reduced wage inflation. At the same time, real wages have increased strongly in recent years despite low nominal pay increases. This has probably increased support for the inflation target and contributed to holding down the nominal wage increases.
The weak development of inflation and the moderate pay increases despite strong growth in the economy have led the Riksbank to modify its assessment of the amount of free resources in the economy. Developments in the labour market indicate that it is possible to increase output more than we previously thought without threatening the inflation target. If we include those in labour market policy measures, unemployment still amounts to almost 7 per cent, which indicates about 3 per cent more unemployed now than in the 1970s and 1980s.
Change of inflation regime expensive in the short-term
In retrospect, we can note that the changeover to low-inflation economy after several decades with high and varying inflation required high unemployment and depressed output for a rather long period of adaptation. High inflation expectations despite low registered inflation and abundant free resources resulted in the Riksbank considering it necessary to hold interest rates up quite a long time with the good intention of maintaining stable prices in the long term. Experiences from this period clearly indicate then that the macro-economic costs can be very great if inflationary expectations are allowed to increase. At the same time, with high interest rates, fiscal policy was focused on reducing the budget deficits. In the light of this, it is not surprising that criticism of the Riksbank’s approach was strong at times. The discussion on making the Riksbank more independent aroused strong feelings not least among many social-democrats and in parts of the trade union movement. There are obviously different views of the extent to which it is necessary with such a stringent monetary policy during the years of adaptation. We can never obtain an answer to this but we can at any rate note that we now seem to have achieved the stable prices we aimed at. To date, it is too early to say whether price stability can be combined with as high employment as previously, but what we are now seeing gives rise to optimism.
Price stability requires some independence for the Riksbank
The strong pessimism on the future prospects for the Swedish economy that affected the general mood after the crisis in 1992 has then in recent years been transformed into an equally strong optimism! We can in now draw the conclusion that the pessimism was exaggerated but we know nothing yet about the situation with respect to today’s optimism. It would seem to be a fact that there has been over-optimism about developments in certain IT-related industries, although the question is how we should regard the more long-term growth of the Swedish economy.
Many structural reforms have been carried out. The financial markets have been deregulated, competition on the goods markets has also increased and many state monopolies have been abolished. Information technology has developed rapidly and even if the stock markets reacted too optimistically, the technological development should be favourable for rather more long-term growth. Productivity has also improved in recent years, which may also be sign that we are on our way into a higher long-term growth track.
The approach of economic policy has then been successful. Support and understanding also seems to have been attained for the price stability target taking precedence over other ambitions and demands. By giving central banks independent responsibility for the price stability target, an attempt has been made to avoid suspicions being created that the price stability target would be set aside for short-term employment considerations. In the long-term, a demand-stimulating economic policy would only be expressed in inflation when full utilisation of resources is reached. For central banks that, after a long period with a poor inflation record, are to establish credibility for price stability, independence can facilitate this transition.
When credibility and support for price stability has been achieved, there is less risk of monetary policy coming into conflict with short-term employment objectives. I would first like to briefly describe the basis of the monetary policy decisions made by the Riksbank.
Simple decision-making rule as a guide
The new Riksbank Act has now been in force for almost two years. This act stipulates that the objective of the Riksbank is to maintain price stability and to safeguard a reliable and efficient payment system. When the Executive Board met for the first time, it clarified the approach to be adopted in monetary policy. It was already the case that inflation measured by CPI was to amount to 2 per cent with a tolerance interval of plus/minus 1 percentage point. The Riksbank had also previously announced that the target horizon would be 1-2 years in normal cases. We clarified that we can decide to disregard certain factors affecting prices which are not related to resource use in the economy and which only have a temporary effect on inflation. However, the prerequisite is that the effect on prices is temporary and should thus not affect inflation expectations. Subsequently, we have also said that there may be reasons for taking into account developments before and after the target horizon. When we depart from the main rule, we have undertaken to say that we are doing so and give reasons for why we are departing from the rule.
The monetary policy decisions are based on forward-looking assessments of inflation measured as CPI. This assessment is based on the interest rate being unchanged during the forecast period. An important factor for determining inflation is the degree of resource use in the economy, the output gap. The thinking underlying this is that inflation will be stabilised if the increase in output is in accord with the long-term growth track. If there are free resources in the economy in the initial situation, GDP can increase more quickly than the long-term sustainable growth without resources being under so much strain that inflation exceeds the target and vice-versa in a situation where resources are already under strain. This is conditional on inflation expectations being around 2 per cent. However, CPI is also affected by price changes that have other basic causes than the demand situation. Examples of such factors are changes in indirect taxes and subsidies, housing costs and oil price changes caused by supply disturbances. Inflation expectations also take this into account.
However, forward-looking assessments include a large measure of uncertainty. In the Riksbank’s inflation report, we describe the scenario thought to be most probable. If we make the assessment that the risks of higher inflation are greater than the risks of lower inflation or vice-versa, this is normally taken into consideration in the conduct of monetary policy.
Macroeconomic costs can motivate departures
When the credibility of the inflation target has been established, the probability is less that inflation target policy will lead to conflict with the desire to stabilise output and employment. In the long-term, monetary policy cannot affect growth and employment other than by creating stable expectations about future inflation. However, if monetary policy is very strictly focused on always stabilising CPI inflation at 2 per cent, this can cause major fluctuations in output. Among other things, this has to do with the time it takes for monetary policy to have an effect on inflation. It has also to do with the fact that CPI being affected by price changes in components that cannot be related to the state of demand. This has been taken into consideration in the design of the main rule and in clarifications and expansions. There has been an endeavour to design the rules in such a way as to avoid unnecessarily large fluctuations in employment and output. However, there must never be any doubt that monetary policy is always governed by the inflation target, although different strategies to achieve this objective in practice can produce different economic costs in the form of fluctuations in output and employment.
The fact that the target horizon is normally stated as 1-2 years is justified by the major fluctuations in output that can result from attempts to influence inflation in a shorter time perspective. Interest rates affect inflation with a time lag through effects on demand or the output gap and attain their greatest effect after 1-2 years. In order to reduce inflation in the short term, the interest rate would therefore have to be greatly increased with major negative effects on growth.
If CPI is affected by factors that are unrelated to the demand situation, a monetary policy that strictly complies with the main rule can also lead to unnecessary variations in output and employment. In order to counteract, for instance, that certain changes in charges lead to CPI exceeding the target, we would be obliged to carry out a monetary policy that would either lead to the use of resources in the economy being too low or too high. Provided that inflationary expectations are not affected, we can, to avoid unnecessary fluctuations in output, therefore decide in certain cases to conduct a monetary policy that leads to the CPI target not being achieved in a 1-2 year time perspective. For instance, we have normally decided to disregard certain factors that affect CPI by focusing monetary policy on UND1X – a price measure that disregards housing costs and indirect taxes and fees.
Concern for macroeconomic stability can in certain cases also justify our extending or shortening the time horizon. It is, for instance, conceivable that it may be justified to extend the target horizon if we make the assessment that there is a risk of a substantial increase in inflation immediately after the normal horizon. If, in such a situation, we wait to increase the interest rate and thus comply with the main rule, we could subsequently end up in a situation where we were forced to increase the interest rate very sharply to avoid inflation expectations accelerating and persisting when inflation rises rapidly. However, a large interest rate increase would have negative effects for output and employment in the short term. To avoid ending up in a such a dilemma in future, we can decide to take into account inflation developments after the target horizon.
From what I have said until now, it is evident that our monetary policy decisions are normally based on the main rule that says that we should aim at keeping CPI inflation at two per cent 1-2 years in future. However, on every occasion when decisions are made, we must consider how to relate to the guidance provided by this simple decision-making rule. This assessment is based on the extent to which the main rule leads to unacceptable macroeconomic costs in the form of fluctuations in employment and output. However, departures from the rule must not damage credibility in the inflation target or lead to inflation expectations being affected. As we have gained experience and increased our knowledge, we have clarified how we intend to apply our decision-making rule. We will in all probability, continue to obtain new insights and be confronted by unexpected events. This means that we must continuously evaluate and review our strategies.
Independence requires openness
A very important part of our strategy, which I have not taken up before, is openness and transparency as to how monetary policy is conducted. This part can hardly be reconsidered.
The Riksbank’s independence creates good conditions for achieving and maintaining credibility for the price stability target. In order for the Riksbank’s legitimacy to be maintained, it is required that we show great openness on the grounds on which the monetary policy decisions are made. The conceptual framework and the decision-making rules that are guiding for the monetary policy decisions make up the basic framework for our communications with the surrounding world. We describe how we conduct monetary policy in the inflation reports, minutes and speeches.
However skilful we are, we cannot guarantee that we will always meet the target. The economic correlations change and unexpected events take place. We must therefore be evaluated on the basis of the quality of the assessments of growth and inflation that have served as the basis for the monetary policy decisions and the balance struck when we have arrived at decisions. It is therefore very important that we always state when we depart from the main rule I have described and the reasons for the departures. In this way, we will also avoid misunderstandings about our intentions.
A high degree of transparency sharpens and improves the internal analysis as well since we are then exposed to continual review by experts outside the Riksbank in the academic world and in the financial markets at home and abroad.
Clarity about the grounds on which we make our decisions creates good conditions for players in the financial markets to assess how the repo rate will develop in future. The more certain assessments are about future interest rate developments, the lower risk premiums that investors will require to purchase short-term interest-bearing treasury paper, for instance, or the banks to lend money to businesses and households. Great uncertainty leads to macro-economic costs such as lower investments and higher costs for central government debt. There is therefore a wish not only on the part of the participants in the financial market but also from the Riksbank to be as clear as possible about the factors underlying monetary policy decisions. However, absolutely most important is to make the inflation target credible since that controls the long-term interest rates. However, it is important to be as clear as possible about the conduct of monetary policy in the short term to avoid unnecessary uncertainty.
But this does not mean that we can remove the uncertainty that actually exists about future developments. We continuously evaluate our previous assessments when new information and statistics are received. And we continuously develop our methods and models and may, when we have received new information, be obliged to re-evaluate earlier conclusions. This was the case, for instance, in the spring when we revised our assessments about the output gap. The conclusion of that is that the interest rate path cannot be announced in advance and even less that we can undertake to supply the interest rate decisions that the market participants sometimes consider that we have given indications of.
The inflation target is necessary
Transparency stimulates debate both about the monetary policy strategy we have adopted and about the way in which the Riksbank carries it out. We have recently been criticised for not slavishly complying with the simple decision-making rule that I have just described. However, debate has also taken place on the suitability of having an inflation target as a guide for monetary policy. The alternative would be a monetary policy based on discretionary decisions resting on assessments of the monetary policy that would best favour long-term stable growth.
As will be evident from what I have said, while monetary policy is guided by a simple decision-making rule, the rule serves as a guide and not a straitjacket. We can depart from the rule if we make the assessment that the costs of complying with it will lead to unnecessary fluctuations in output and employment provided, that inflation expectations and the credibility of monetary policy are not affected. Monetary policy decision-makers are always confronted to some extent by having to strike a balance between price stability and output variability in the short term. The balances must be struck so as to maintain confidence in price stability. It is easy to see that a central bank with a good reputation and many years of successful combating of inflation has more freedom than a central bank in a country with many years of negative experiences of high inflation. On our part, it is undoubtedly the case that the inflation target is necessary as a lodestar in order for monetary policy to be able to contribute to creating good conditions for long-term growth.
