Bäckström: The current situation for monetary policy

In my opening remarks to your Committee six months ago I noted that the Swedish economy was developing impressively. Growth was stronger than many had counted on and inflationary pressure was moderate. New jobs were being created and unemployment was falling.

Since then the economy has definitely continued to impress. Forecasters have had occasion to revise their assessments of activity successively upwards.

GDP growth in 1999 amounted to 3.8 per cent and the number of new jobs totalled almost 90,000 – the highest figure in thirty years. This has been accompanied by surpluses on the current account and government finances. Moreover, the upward revisions refer mainly to growth and employment. Forecast inflation has not been adjusted upwards to the same extent, which is a testimony to the good circle at present in the Swedish economy.

Inflation in the past three years

Measured in terms of the consumer price index (CPI), inflation has in fact been exceptionally low for three years; the average annual level has been 0.5 per cent. This level for the period as a whole is made up of CPI increases of 0.9 per cent for 1997, 0.4 per cent for 1998 and 0.3 per cent for 1999.

This means that the rising demand pressure in the Swedish economy in recent years has not shown up in higher CPI inflation. Contrary effects have come from other factors. To clarify this, we can take a look at inflation’s constituents.

From Diagram 1 it will be seen that in the past three years, underlying domestic inflation (UNDINHX)(1) has been almost continuously in the vicinity of the 2 per cent inflation target.

The pressure from domestic demand has been countered, however, by a weak international price trend. This is evident from the path of UND1X inflation, which also includes imported prices. The main explanation for the weak development of import prices in this period is the crisis that began in Asia in the autumn of 1997. It can be mentioned that towards the end of 1998, just the prices of coffee, petrol and domestic heating oil lowered the price level in Sweden by almost half a percentage point.

Because of the long lag before monetary policy changes elicit effects, parrying this impact from import prices would have required the Riksbank to identify the Asian crisis’ damping consequences for commodity prices long before they actually occurred.

But the weak development of import prices is not the only explanation. Inflation in this period was lowered still further by changes in interest expenditure as well as in indirect taxes and subsidies.

On several occasions I have shown the Committee how monetary policy exerts direct effects on CPI inflation through house mortgage interest expenditure. Reducing the interest rate, in order to stimulate demand and thereby hold inflation around the target, has the initial effect of lowering the CPI because house mortgage interest expenditure decreases. This effect continues until after households have renewed the loans at the new, lower interest rate. The greater part of the difference between the CPI and UND1X in Diagram 1 is due to this interest rate effect.

The Riksbank has considered the possibility of using some other index for targeting monetary policy but we concluded that there are grounds for continuing with the CPI as the overriding policy target. Among other things, the index is well known and relatively comprehensive.

In the light of the problems I mentioned and the ensuing discussion, the Riksbank has chosen to present a clearer description of how the CPI target is to be interpreted. Deviations should be permissible if the CPI change figure is influenced by transitory factors (for example, changes in indirect taxes and subsidies or household interest expenditure) or if bringing inflation promptly back to the targeted level would be costly for the real economy. The magnitude of these factors is to be stated in advance by the Riksbank. In this way, the members of your Committee, as well as other observers, will be in a better position to assess the Riksbank and the policy we implement. For each decision it is clear which transitory factors the Riksbank has chosen to disregard in the formation of policy.

These clarifications are essentially a codification of the approach that the Riksbank has applied to the formation of monetary policy in recent years, although any deviations from the target were then discussed less precisely. In the period from 1997 to 1999 the path of inflation was affected to a large extent by transitory price movements that the Riksbank considers should be of secondary importance for monetary policy. Policy has therefore focused on the underlying or core rate of inflation as measured by UND1X; this averaged 1.2 per cent, with constituent annual rates in these three years of 1.4, 0.9 and 1.4 per cent.

The Report that has been submitted to the Committee today contains an appendix that is intended to serve as a foundation for the Committee’s evaluation of the implementation of monetary policy and its outcome in the period 1997–99. This is something that the Committee requested earlier.

Let me now consider the implementation of monetary policy in the course of 1999.

Monetary policy in the past year

The early months of 1999 were marked by continued effects of the Asian crisis on the world economy, with weaker international demand. In order to prevent this from leading to a weaker economic trend in Sweden and thereby a lower path for inflation, the Riksbank reduced the repo rate twice, in February and March, bringing it down to 2.90 per cent.

The situation stabilised during the late spring and summer. The relatively relaxed monetary policy that had been implemented for a number of years – both in Sweden and in many other parts of the world economy – played an important part in the improvement that occurred during 1999 and contributed to more positive economic forecasts.

It was then important for the Riksbank to take timely action so as not to have to implement a more pronounced monetary tightening and economic retardation to cope with rising inflationary pressure at a later stage. A process was therefore initiated in November that amounts to successively taking our foot off the monetary ‘accelerator’. As the after-effects of the Asian crisis and the financial market turbulence in the autumn of 1998 subsided, a strong economic stimulus from monetary policy was no longer required. In November 1999 and February this year the repo rate was therefore increased by a total of 0.85 percentage points.

The analytical and monetary policy challenges that confronted the Riksbank during 1999 were not minor. The strong growth of demand in recent years has been accompanied by paths for underlying and trend inflation that have been weaker than earlier relationships seemed to suggest. A number of reasons have been put forward as to why these relationships may have changed.

- One reason may be that when inflation has been brought down to a low level, price and wage setting behaviour is different. Households as well as firms in Sweden count on inflation remaining low.

- Another reason may be increased competition, for instance due to deregulations, globalisation and IT commerce.

- A third reason may be that our assessments of resource utilisation are faulty and the actual level is lower than we have assumed to date. Economic growth can then be higher for a time, without generating increased price pressure.

- A fourth reason may be a higher productivity trend and an ability to produce more goods and services without generating increased prices; in other words, a higher potential growth rate. This could lead to higher growth being feasible for longer without inflation taking off.

There are arguments in favour of each of these explanations but the empirical support is not clear-cut.

The issue of potential growth and higher productivity has recently attracted particular attention in the light of economic developments in the United States, for example. Productivity growth in Sweden has improved but we do not yet have any notable evidence that potential growth has risen as it has in the American economy.

An expected increase in trend productivity can also affect components of demand. The prospect of higher corporate profits leads to higher share prices for listed companies. Besides prompting companies to step up investment and improving the supply of venture capital, the appreciation of shareholdings may make households feel more wealthy and stimulate consumer demand. The strong upward stock market trend in Stockholm in recent years may be a sign that players have incorporated expectations that economic growth in Sweden will be appreciably higher than earlier.

The relative paths of supply and demand in the Swedish economy and the development of the inflation propensity are obviously central issues in the assessment of inflation. If supply and demand develop at the same rate, no inflationary pressure will build up. This is what seems to have happened for a time in the United States. But there is, of course, a risk of one of them predominating and that would alter the conclusions for monetary policy.

The questions about trend productivity and other matters that may have altered the relationship between growth and inflation are one of several elements of uncertainty in the Riksbank’s forecasting work. When constructing assessments and material for decisions it is therefore important to check what is happening and be alert to any signs of changes that should be taken into account.

The current assessment

The Inflation Report we are publishing today presents a picture of the Swedish economy that remains bright, with good growth and moderate, though rising inflation.

External economic prospects have gone on improving. It is above all GDP growth in the U.S. economy that continues to be remarkably positive but it also looks as though growth in the euro area and the United Kingdom could be higher than assumed earlier. International consumer prices, however, are not expected to move up to the same extent. Less relaxed monetary policies among a number of central banks and growing pressure from international competition are judged to hold external prices back. But partly because the krona is not expected to appreciate quite as rapidly as we counted on earlier, import prices to producers are now assumed to be somewhat higher. On account of strong competition, however, we do not foresee a full pass-through to consumer prices.

Domestic demand is also expected to be stronger than we foresaw in the December Report, notwithstanding the subsequent repo rate increase of 0.5 percentage points. Rising asset prices and increased household wealth are also contributing to the growth of private consumption. Moreover, the strong economic activity and good stock market trend should contribute to a favourable development of investment.

The Riksbank foresees GDP growth rates of 4.0 per cent this year, 3.5 per cent in 2001 and 2.6 per cent in 2002. These are high rates compared with the 1970s and ’80s. By itself, the upward revision of GDP growth implies that unutilised resources in the economy are brought into production more quickly.

The strong economic activity points to a higher rate of wage increases in the coming years. At the same time, it is possible that changes in the labour market will hold wage increases down. A growing proportion of wage increases is being allocated locally, for example. Moreover, inflation expectations among labour market organisations are still in line with the target. This means that, as we have seen in recent years, a real wage increase can be secured without such a large increase in nominal wages. Historically high real wage increases in recent years should likewise contribute to a moderate development of wages.

Against this background, we judge that the level of nominal wages can move up at an annual rate of just over 4 per cent. This is marginally less than foreseen in the December Report. In the longer run, however, wages cannot go on rising at this rate because they would then exceed the room that now looks like being provided by the probable trend for productivity growth. One illustration of this is that in the coming one to two years, underlying domestic inflation will be somewhat above 2 per cent, though this is countered by downward effects from import prices.

In our main scenario we judge that overall inflation prospects are in line with the target one to two years ahead but that UND1X inflation will be marginally above the target at the end of the period. With an unchanged repo rate, UND1X inflation is expected to be 1.6 per cent one year ahead and 2.1 per cent after two years.

The transitory factors that the Riksbank should disregard in the formation of monetary policy are judged to be small in the coming one to two years.

Moreover, the Riksbank considers that the risk spectrum for inflation is balanced. In other words, the risks of inflation being higher and lower, respectively, than in the main scenario are equal.

On the one hand, inflation could be lower than in our main scenario if there is a marked correction to American stock markets or if deregulations and trade liberalisation have greater effects than we have allowed for. On the other hand, inflation could be higher if oil prices do not fall at the rate we assume or if the growth of demand, instead of slackening at the end of the forecast period as we foresee, remains strong.

The strong economic activity and gradually growing inflationary pressure point to a future need for a further repo rate increase. The timing and size of the increase are considered in the light of, for example, new information and its significance for the Riksbank’s overall inflation assessment. Our current assessment is that in the greater part of the coming one to two years, inflation is expected to be below 2 per cent. This speaks in favour of leaving the repo rate unchanged for now.