Heikensten: Inflation and the interest rate
I should like to begin with a word of thanks for the invitation to talk about economic developments, inflation and the conditions for monetary policy. It is beginning to have the feel of a tradition.
Today I shall start by considering the course of events since we met here a year ago. Then I intend to take a look at the future and paint a picture of the outlook not just for 1997 but for the somewhat longer perspective that is relevant for monetary policy. Finally I shall draw some policy conclusions.
Growth but not employment
Before considering 1996, however, I shall look further back. Since the summer of 1993, when the dramatic fall in production was broken, the annual growth rate in the Swedish economy has averaged about 2.8 per cent. This is considerably higher than the rates we were accustomed to in the '70s and '80s. In that capacity utilisation was so low initially, it is hardly surprising that growth as such has been high. What is surprising - and very troublesome - is the weak development of employment.
Sweden is not alone in Europe in having problems with the creation of new jobs; rather the reverse. But in the United States the situation is very different and even in Europe there are exceptions; one is Holland in the past decade (Diagram 1). The differences from Sweden are appreciable.
Today in Sweden the issue of unemployment is being discussed more and more. It is important that the premises for this discussion are correct. At the Riksbank we believe that in the coming two years production in Sweden will pick up again, although growth is unlikely to reach the average level from the past three to four years. This should generate some increase in employment. But a radical improvement in employment calls for more than economic growth; above all it requires that the labour market and wage formation function better.
Markedly lower interest rates
When I spoke at this conference a year ago the repo rate was 8.66 per cent. In the course of 1996 we lowered the rate 25 times. At the end of 1996 the repo rate was 4.10 per cent as against 8.91 per cent at the beginning. In the industrialised world, the Riksbank's reduction of its instrumental rate is accordingly one of the largest and steepest. Moreover, the current level of the repo rate is the lowest in Sweden for more than thirty years.
The cuts have been gradual and as far as possible predictable. As the outlook for inflation improved and interest rates declined, market perceptions of how far down the repo rate might move were also revised (Diagram 2). Our intention was to bring the repo rate down without any loss of confidence in monetary policy, as expressed in such price variables as the exchange rate and long bond rates. It worked; at no time did the market react so wildly as to disturb the picture.
There are good reasons for underscoring that such a monetary policy realignment presupposes that many factors are favourable. That was the case in 1996. The conditions in international financial markets were stable and positive for Sweden. Economic activity abroad and in Sweden was comparatively weak and implied a good outlook for inflation. Sweden's fiscal policy was perceived as increasingly credible as the budget outcome improved and both the Government and the Riksdag (Sweden's parliament) demonstrated a determination to fulfil earlier commitments.
Better outlook for inflation
In Sweden the outlook for inflation improved on several counts:
- Economic activity became somewhat weaker. Export growth slackened as a result of the krona's appreciation and slowdowns abroad. This was accompanied by a reduction of industrial stocks. The domestic economy was relatively weak, though activity does seem to have picked up, particularly last autumn. GDP growth probably amounted to between 1 and 1.5 per cent (Diagram 3). In terms of long-term balance, the composition of growth was favourable, with strong increases for net exports and industrial investment (Diagram 4). There was also a large surplus on current account.
- The inflation propensity seems to have decreased. The conditions for persistently low inflation improved in a number of respects. The expected rate of inflation fell from survey to survey (Diagram 5). Bond pricing shows a similar picture. Most of the surveyed groups now expect a rate of inflation, even in the longer run, that is well in line with the Riksbank's target. In parts of the economy it also looks as though pricing behaviour has changed; conceivable explanations for this include increased competition as a result of deregulations and EU membership. These tendencies are difficult to confirm for certain but the indications are clear.
- Temporary effects have also been at work. The krona's appreciation accounted for approximately 0.4 percentage points of the fall in inflation from 2.6 per cent in December 1995 to -0.2 per cent in December 1996. For the CPI, the part played by decreased interest rates has been even greater; the downward effect on the inflation from falling house mortgage rates totalled 1.4 percentage points. It is because of the temporary effects that the average level of the inflation in 1996 is outside the Riksbank's target interval. As a rule, the interval surrounding the target figure should suffice to absorb changes of this type in the composition of inflation. That this was not the case in 1996 had to do with the marked reappraisal of the situation in Sweden. Both the rate of inflation generated domestically and the ways in which the Riksbank measures underlying inflation show a fall that is less dramatic (Diagram 6).
Improvement in government finance
Perhaps the most pronounced swing in the past year (actually since the summer of 1995) has occurred for government finance. Confidence grew in the ability of the Government and the Riksdag to reverse the crisis. A number of factors contributed to this.
- The budget outcome successively improved in relation to earlier official estimates (Diagram 7). As regards the budget deficit for 1997, current estimates suggest that Sweden will do well in a European context. And although Government debt is still high, here, too, the turn seems to be more distinct than expected.
- Another factor behind the swing is the Government's clear adherence to the convergence programme from the summer of 1995. When it seemed that the budget target was in danger in the spring of 1996, additional measures were presented to fulfil the ambitions. This had favourable effects on interest rates and credibility.
Developments in recent years prompt two distinct warnings. Building up the credibility which characterised the past year took a long time and was very costly. The announcement of extensive measures in the winter of 1994-95 did not suffice. It was not until the budget outcome started to turn that we could reap the positive effects on bond rates, etc. It must also be born in mind that a part of the budget's reinforcement is tied to the stronger exchange rate and the lower interest rates. Sustained confidence is therefore a sine qua non for budget consolidation.
In view of the political debate, there may be cause to underscore fiscal policy's importance for monetary policy. A tighter fiscal policy normally implies a better situation for inflation; it makes it easier for the Riksbank to fulfil the inflation target in conjunction with low interest rates. In addition, a fiscal policy that lacks credibility creates problems for monetary policy - as was the case during much of the period 1990-95. The radical lowering of interest rates during 1996 is something for which our thanks are very much due to fiscal policy.
Continued economic upswing
So much for history. In discussing tendencies for the coming years I shall start from our inflation report last December.
In the report, conditions for exports in the coming years were judged to be good. Nothing has occurred to modify that assessment. Industry's competitive position is favourable. To an increasing extent, export growth seems to be coming from rapidly expanding regions, not least Asia. The international upswing last autumn is expected to become somewhat stronger in both 1997 and 1998.
Judging from order inflows and other business tendency figures, the growth of exports is likely to be modest. The most recent figures may perhaps indicate that international developments will generate somewhat weaker growth than expected, but the difference would be marginal. This is countered, moreover, by an exchange rate that, at present at least, is weaker.
The conditions for increased demand for consumption were also judged to be favourable (Diagram 8). Neither has the picture in this respect changed at all substantially. Increased consumption is indicated by low interest rates and a level of household saving that, historically, is relatively high and is expected to show some fall. To this can be added an accumulated need to renew the stock of durable goods. The main factor in the opposite direction is unemployment, which is persistently high and even tended to rise in the autumn. Neither are the increments to household income in the period 1997-98 expected to be as large as in 1996. Then there is the uncertainty that stems from weak government finance and the changes that have been made and may be called for in the social security system, including pensions. All in all, however, it seems that the subdued but stable increase in consumption may continue.
In view of the above, we adhere to the growth scenario that featured in the last two inflation reports. This implies an accelerating growth rate that for 1996 and 1997 combined totals almost 4 per cent. Development in 1996, however, seems to have been marginally weaker than we counted on. Some further acceleration of growth is then foreseen in 1998. The conclusion in the inflation report therefore holds good: it is not likely that in the coming two years economic activity will rise above the level of potential output at which inflation is liable to pick up (Diagram 9).
Wage development a key factor
The wage increases during 1996 exceeded what the Swedish economy can manage in the longer run. The underlying rate of wage increases moved up to more than 5 per cent. The rate of wage increases on top of collective agreements also accelerated in manufacturing.
For price formation in 1996, the consequences of these tendencies were limited. As long as domestic demand remains subdued, firms have difficulty in passing through the increased wage costs to prices. The outcome is more likely to be rationalisation, increased productivity and, in certain cases, decreased employment.
The starting point for the Riksbank's assessment of the future rate of wage increases is that a downward adjustment will be achieved. At the time of the settlements that were reached in much of the labour market in 1994-95 the expected level of future inflation was considerably higher than it is today (Diagram 10). The present credibility of the inflation target did not exist. Moreover, employment was rising relatively fast in parts of the economy and profits were higher than they are today.
The downward adjustment we foresee is, however, a gradual process and slower than would be desirable. As far as inflation in the next two years is concerned, the slow adjustment is not seen as a problem, at least as long as the situation agrees with the Riksbank's main scenario, where the growth of consumption is subdued. For employment the problem is greater. Lower wage increases might be accompanied by a faster increase in labour demand.
The new information about the rate of wage increases since the December report supports the assessment that some retardation should be feasible during 1997. The wage statistics for the most recent months show signs of some improvement. Future wage formation is being discussed by the labour market organisations. Apart from this, lower wage increases in future agreements are indicated by a number of factors. Profit levels in export companies are lower and there is little possibility of passing through costs to consumers. To date in the 1997 negotiations, wage demands have been lower than in the previous round, though still high in relation to the low inflation expectations.
Inflation in line with the target
Over the last six months, markets have tended to focus on the latest figures for inflation. In a way, this is hardly surprising - the figures during the autumn have been unexpectedly low. It should be emphasised, however, that the Riksbank's policy refers to a longer period. Figures for "yesterday's" inflation are therefore relevant only in so far as they say something about the future. To some extent that has been the case in the past six months; the figures have suggested a somewhat lower inflation propensity. In the main, however, they have reflected the strength of the transient effects.
Inflation expectations, as measured in surveys and forecasts, have been stable, in some cases with a fall during the autumn and winter. The expected rate of inflation one year ahead is around 1.5 per cent and the average rate in the next five years is around 2 per cent.
What about the inflation forecasts for the coming years? There are many indications that the 12-month change figure for the CPI reached a low in December, with the prospect of gradually rising figures during the spring as the transient effects that pushed price increases down last year come to an end. At the beginning of 1997, moreover, prices have also been affected by, for example, changes to indirect taxes and subsidies; these changes are expected to raise the price level by about 0.5 per cent. On the other hand, house mortgage interest costs should continue to have a downward effect on prices, at least during 1997.
An upward tendency in inflation is therefore something that the Riksbank counts on in the next two years. Consequently it will not come as a surprise, at least as long as it is moderate. In the inflation report we envisaged that during 1997 the CPI would rise by an average of just under 1.5 per cent. For 1988 it was considered that inflation would be somewhat higher, around 2 per cent, which included an increase in the price level of about 0.5 per cent from further changes in taxes and subsidies (Diagram 11). The assessment accordingly pointed to a rate of inflation that would be in line with the Riksbank's target.
Same conclusion
The picture I have outlined does not warrant any change in the conclusions presented in the December inflation report. Our main scenario suggests that at present the monetary stance is relatively well balanced.
The new information about economic activity and price movements points to a picture of inflation in 1997 that is marginally more positive. In the opposite direction, the exchange rate has been somewhat weaker than we counted on. While this does not stem from what are commonly referred to as economic fundamentals, the fact remains that a weaker exchange rate - which Sweden has now had for a quarter or so - has effects on the picture of demand and inflation.
The Riksbank will continue to evaluate the effects of the interest rate cuts that were implemented during 1996. As pointed out on many occasions in the past year, we will also be keeping a close watch on the future development of economic activity. The ongoing construction of monetary policy will depend on our assessment of inflation's future course, for instance in the light of incoming information.
