Inflation and the interest rate
First a word of thanks for the invitation to speak to you today.
What I have to say is based in the main on the Riksbank's latest inflation report, which was published less than a month ago. In some respects, I shall be developing what we said there about the outlook for inflation, though our assessment of the situation is essentially unchanged. I shall begin, however, with some remarks about why monetary policy is constructed as it is, with the objective of safeguarding the value of money.
1. Monetary policy's foundations
Experience advocates a stable environment
The reason why the Riksbank's objective is to safeguard the value of money is basically simple: low, stable inflation contributes to a good environment for the growth of production and employment.
Support for this view is readily found in the pattern of economic developments in different industrialised countries. In those countries where inflation has been allowed to accelerate, growth has normally been weaker and more erratic. We have seen this in Sweden. In the first two-to-three post-war decades Swedish economic policy was characterised by stable rules and inflation was low. Economic growth in those years was good and new jobs were created to a greater extent than in the past two decades, when inflation shot up and there was a series of devaluation cycles.
There is really nothing odd about this. We all know that stable conditions make it easier to arrive at wise economic decisions. When wages and prices chase each other in an unpredictable way and both interest rates and the exchange rate fluctuate widely as a result of this, with a certain amount of luck or smartness it is sometimes possible to do good business. But most people surely realise that this is not the best setting for either long-term investment or for those who have to live on an earned income or a pension. The head of the US Federal Reserve, Alan Greenspan, has formulated his view of what monetary policy's objective should be in roughly these words: inflation should be so low that economic agents do not have to worry about it.
Limited room for employment policy
Why, then, are there those who still argue that higher inflation would be a good thing? I believe this is mainly because they perceive a conflict between price stability and low unemployment. This perception in turn has its roots in a short-run, cyclical perspective. When economic activity is rising or falling on account of transitory demand fluctuations around a long-term trend, this negative relationship does exist. Rising demand for goods and services generates increased demand for labour and lower unemployment; this tends to be accompanied by rising wages and higher inflation. The opposite also applies.
But this does not mean that higher inflation results in permanently higher employment and lastingly lower unemployment. On the contrary, as a result of the economic setback that normally follows a return to an inflation economy, unemployment is liable to rise even higher. Against this it might be argued that it is not a matter of giving inflation its head, but of allowing just a little more inflation. Nor is it a matter of doing this as a systematic policy, but just for once. Once again, the Swedish experience in the 1970s and '80s is both instructive and representative of what has happened in many other countries. A little more inflation over a short period can very easily become inflation that is high and persistent.
In any event, in the longer run the level of unemployment is determined by factors other than the demand situation, factors such as institutions, rules and conventions in the economy and above all in the labour market and wage formation. If the causes of unemployment were in fact solely or even mainly cyclical, then unemployment would not differ so greatly as it does between countries that are in the same cyclical phase and have had much the same growth rate for a long time.
Targeting inflation
The Riksbank has the objective of safeguarding the value of money. In practice, the Riksbank targets inflation. The target rate is 2 per cent (with a tolerance interval of *1 percentage point), assessed in annual terms. As there is a time lag of 1 to 2 years before an interest rate adjustment has its full effect on the rate of price increases, the Riksbank has to work with forecast inflation. When we have reason to fear that inflation is rising above the targeted rate, it will soon be time to raise the instrumental rate and vice versa.
Assessments of future inflation are presented by the Riksbank four times a year in the series of inflation reports. The assessments are published as a guide to economic agents in their own deliberations about what the future level of interest rates is likely to be. The reports also provide a basis for the discussion of monetary policy's construction that the Riksbank is interested in encouraging.
Simplifying somewhat, inflation is determined by two factors: the situation concerning demand and supply and economic agents' expectations of future inflation. The first factor is fairly straightforward; when demand tends to exceed what the economy is capable of producing, the resultant shortages and bottlenecks will normally push prices and wages up. The part played by expectations may be less evident. If firms and employees expect increased inflation, they will tend to raise prices and demand higher wages, respectively. This may be sufficient to start and continue an inflationary process. Another factor of importance for inflation is the value of the krona, both because it directly affects pricing and demand and because it can serve as an indication of expected inflation.
2. Economic activity and inflation expectations
Now for the assessment in our latest inflation report. First let me say something about the Riksbank's appraisal of the economic situation.
Demand and supply
The statistics from recent months clearly point to an upswing in the greater part of the economy. This largely confirms the Riksbank's assessments last spring. Developments to date in 1997 have been more balanced than before. Demand growth is now coming not only from exports and investment but to an increasing extent from private consumption as well.
The growth of household income has been held back for some years by the programme for fiscal consolidation. Activity has been further depressed by the need to cut spending on public activities. The fiscal withdrawal of household purchasing power is continuing both this year and next, so disposable income is largely unchanged in 1997 but will rise in 1998. After that, when the consolidation programme has been completed, fiscal policy is not expected to be as restrictive on households. Some increase in public activities can also be expected in the coming two years.
After rising for some years at a rather modest rate, private consumption has picked up in the past year. Households' expectations concerning their own as well as the national economy improved in September after a minor setback in August. Households are also more optimistic about the outlook in the labour market. Moreover, the August survey by the National Institute of Economic Research shows that firms in both everyday and infrequent trade are optimistic about sales trends in the near future.
It is not just disposable income that determines private consumption. The appreciation of wealth in the past year is probably important, too. Since the turn of 1996, house prices have risen 9 per cent and share prices 28 per cent. This suggests that in 1997 households could realise capital gains that may reach the equivalent of about 25 per cent of disposable income. It would be surprising if such an increase in wealth did not contribute to stronger consumer demand.
To sum up, most signs point to a good development of demand in the years ahead. In an expanding world market, export growth should continue to be relatively high. The Riksbank also counts on an accelerating increase in private consumption and at least an end to the fall in public consumption.
The broad increase in all the major demand components naturally exerts some pressure on the available resources. This implies rising capacity utilisation and a decreased availability of unutilised resources. The business tendency surveys for manufacturing, which provide information about capacity there, consistently indicate that unutilised resources fell from the first to the second quarter. An increased proportion of firms reported that output was being restricted by production factors rather than demand. This was accompanied by rising shortages of skilled workers and salaried technicians.
An important, though elusive, indicator of inflationary pressure is the gap between total demand and total output. Most indicators suggest that this gap has become somewhat smaller. In the inflation report, however, the Riksbank judges that the supply of unutilised resources is still relatively good. It is not least the high level of total unemployment that indicates this. A high rate of investment and rapid increases in productivity have contributed, at least to date, to an expansion of potential output to meet the growing demand. This has eased inflationary pressure during the upswing.
Inflation expectations
Another important indicator of inflationary pressure is the expectations of economic agents about future inflation. Along with actual inflation, these expectations have declined in recent years. They also show a tendency to converge around the 2 per cent inflation target. If this is the case, it is a good sign because it can be seen as increased confidence in the inflation target. This means that to a greater extent, economic agents are basing their own plans on expectations of low inflation, which provides better conditions for monetary policy to fulfil the inflation target.
In recent months the expectations of future inflation have tended to move up but this seems to be mainly an adjustment to the increase in registered inflation. The tendency, however, must be closely watched; a clear increase in inflation expectations above the inflation target would give cause for concern.
3. Inflation in the coming years
In the course of 1996 CPI inflation decreased markedly and the annual level was somewhat below the lower limit of the tolerance interval. Transitory factors partly explain the drop. Falling house mortgage interest costs, occasioned in part by the Riksbank's own interest rate cuts, played a large part in the low inflation figures. The trend has now turned and the rate of inflation is beginning to move up. But the main reason for this is that the transient downward factors are no longer at work to the same degree; the increase in underlying inflation is therefore more subdued.
In the light of the assessments of economic activity and other factors that affect inflation (e.g. inflation expectations), in the inflation report the Riksbank forecast that annual CPI inflation would be around 1 per cent in 1997 and around 2 per cent in both 1998 and 1999. Some acceleration was envisaged in the latter two years.
The assessment carried some reservations. Matters discussed in the report included the risks associated with rising activity, the difficulties in knowing how the supply side would react and the coming round of wage negotiations. There may be reasons for elaborating on that discussion in some respects.
Output gap closing by degrees
In the inflation report the Riksbank considers that Sweden's gross domestic product will grow by around 2 per cent in 1997 and around 3 per cent in 1998. In 1999 the GDP growth rate is expected to be rather more than 3 per cent.
This implies that in 1998 and 1999 growth will be stronger than the increase in potential output, for which the annual trend is estimated to be just over 2 per cent. That means that today's surplus capacity will be activated by degrees. It is difficult to tell how quickly this will happen. The Riksbank considers that the output gap - a way of attempting to gauge unutilised capacity in the economy as a whole - is currently about 2 per cent of GDP. With the growth rates envisaged above, the gap would then close some time in 1999.
Other information with a bearing on when inflation might be likely to accelerate was also included in the inflation report. Estimates of structural unemployment and thereby, indirectly, of the level of unemployment that can be combined with stable, low inflation, suggest that risks of inflation also exist less than two years ahead. The rate at which capacity utilisation rises may likewise play a part. A rapid reduction of unutilised resources might generate inflationary impulses even though there is still a capacity surplus at total level.
Risks in the wage negotiations
Wage formation is particularly relevant at present because the settlements that are due in the coming six months will cover a large part of the labour market. Labour shortages are not yet being reported from more than a few sectors but experience shows that the domino effects can be substantial.
The Riksbank's assessment of inflation in the latest report is based on an assumption that the rate of wage increases will be lower than in recent years. Support for this assumption can be found, for example, in recent wage statistics. The report, however, does not envisage a downward shift to, or even below, the level that has prevailed in the European Union in recent years, a shift that would make it easier for Sweden to tackle unemployment.
Wage formation is primarily a matter for the labour market organisations to arrange. The level of wage increases that different segments of the economy can cope with depends, for example, on the initial profit level together with the development of that segment's productivity, costs and prices. For the economy as a whole it is difficult to avoid the conclusion that, if an effective reduction of unemployment is to be feasible, a rate of wage increases which equals the inflation target plus expected long-term productivity growth represents a ceiling. Lower wage increases would make labour cheaper relative to capital investment and would thereby stimulate employment.
It is important to understand today's rules for economic policy. In keeping with the situation a quarter of a century ago, when Sweden was participating in the Bretton-Woods system, which was a functional and efficient arrangement with fixed exchange rates, for some years now the rules of the game have been clear. If the Riksbank judges that inflation will rise above the 2 per cent target, it is up to us to raise interest rates. Consequently, excessively high wage increases will not lead, except possibly in the very short run, to higher inflation. Instead they will result in weaker economic activity and lower employment in the years ahead. In the earlier situation with a fixed exchange rate, excessively high wage increases primarily affected manufacturing and other parts of the economy that are exposed to international competition. Today the effects will be felt in particular by sectors that are sensitive to interest rates, such as distributive trades and a debt-burdened public sector.
4. Conclusions for monetary policy
The overall assessment in the inflation report is that the monetary stance is well balanced. Today there is no reason to modify that conclusion. In the past month there has been no appreciable change in the outlook for inflation.
After the interest rate increases in Germany and other countries almost a week ago, many people have asked how this affects the Riksbank's actions. As stated already on several occasions, the answer is: not much as long as the conditions for inflation have not changed. The Riksbank targets inflation and it is this target that guides policy.
In view of what has happened in the past month, there is reason, finally, to highlight a couple of points in the report. The first is that monetary policy is being conducted against the background of an economic upswing. A gradual increase in capacity utilisation is therefore probable. Sweden does not differ from Germany in that particular respect. The other point is the diminished risk of a development that is appreciably weaker than in the main scenario; this assessment is supported by the latest statistics.
In the light of the rising activity, the main difficulty in assessing future inflation currently lies in an uncertainty about developments on the supply side, wage formation in particular. It is difficult to foresee how the Swedish economy will react to a broad economic upswing. It is therefore important, not least on that account, to analyse incoming information.
