Bäckström: Sweden's economy and monetary policy
Thank you for the invitation to your meeting this evening. In my talk today I shall be discussing the construction of monetary policy and developments in the Swedish economy.
The Riksbank safeguards the value of money
The Riksbank manufactures and distributes money. We also have the task of maintaining its value - we safeguard the value of money. If the Riksbank were to issue too much money - in practice, set the repo rate below the level that is conducive to macroeconomic balance - it would risk generating inflation. Conversely, issuing too little money - setting the repo rate too high - could lead to deflation. Inflation and deflation are both undesirable; they tend to result in economic instability.
Monetary policy is thus broadly a matter of issuing money in appropriate amounts. The Swedish word lagom ("just right") describes this aptly. The repo rate must be set continuously so that both inflation and deflation are avoided and price stability is achieved. The Riksbank's 2 per cent inflation target, with a tolerance of 1 percentage point up and down, is fairly close to what is usually regarded as price stability.
The Riksbank is sometimes criticised for concentrating unduly on the price stability target. Some say we should also consider growth. Our answer is that we do - this is part and parcel of the way in which monetary policy is conducted by targeting inflation.
Once a low-inflation regime has been established and people have adapted to a future with 2 per cent inflation, the rate of inflation is contingent on the relationship between demand and production capacity. A risk of demand becoming too strong, so that the economy overheats, prompts an increase in the repo rate. Conversely, an economic slowdown with the risk of a recession calls for a repo rate cut.
In this way, a monetary policy aimed at an inflation target ultimately tends to smooth fluctuations in economic activity. It accordingly contributes to more sustained growth in line with the economy's long-term trend. In other words, the inflation target helps to ensure that the economy expands as rapidly as is possible in relation to such factors as new technology, the level of investment, how well the labour market functions, people's education and other matters that determine growth in the longer run. In addition, low inflation as such makes the economy function better, thereby promoting long-term growth.
Expectations also influence inflation
The Swedish economy is now in its fifth year of growth since the profound crisis in the early 1990s. Since the upturn in the summer of 1992, annual GDP growth has averaged over 2.5 per cent. This has been combined with an annual inflation rate of about 2 per cent. Here we have a marked contrast with the situation in the 1970s and '80s, when growth averaged around 2 per cent and inflation about 8 per cent.
In one respect, however, the situation is still dismal, namely as regards the labour market and employment. If an economy is growing well but employment remains weak and unemployment high, there are good reasons for supposing that the labour market is affected by structural problems. The problems concern how the labour market works in general and, more specifically, how wage formation functions. But there is nothing monetary policy can do about this. An expansionary monetary stance that lasts too long simply leads to increased inflation. Experience has taught us that sooner or later an inflation economy generates imbalances that lead to higher unemployment. So a lasting reduction of unemployment cannot be achieved by accepting higher inflation.
Although growth in recent years has been satisfactory, the generation of a broader upswing in the economy has proved difficult. Initially, about three-quarters of GDP's increase came from manufacturing, which represents only about 20 per cent of Sweden's economy. This meant that bottlenecks occurred fairly quickly, accompanied by various types of inflationary impulses. It goes without saying that our relatively small industrial sector cannot be expected to sustain growth on its own.
The problem with the dual economy - a flourishing export-related industrial sector and a rather weak recovery in sectors with a more domestic orientation - stemmed in part from the persistence of high inflation expectations. People simply did not believe that in the longer run the Riksbank would be able to keep inflation in line with the target. This was evident from, for instance, high bond rates, a weak exchange rate, rapid wage increases and so on.
Monetary policy therefore had to concentrate on demonstrating that the price stability target was to be taken seriously. The repo rate was increased a number of times between the summers of 1994 and 1995. This was accompanied by a far-reaching consolidation of government finances. The alternative to this line in economic policy would have been successively rising inflation; it would not have generated higher long-term growth.
The lesson from this period is that inflation is not determined solely by resource utilisation, that is, whether the level of activity is high or low. People's expectations of future inflation also play a part.
During the spring of 1996 households, firms, savers and investors began seriously to believe that even the future level of inflation would be around 2 per cent. The repo rate could then be cut by degrees to 4.10 per cent, the lowest level for the Riksbank's instrumental rate since the early 1960s. Lastingly low inflation is the best guarantee of low interest rates.
Low interest rates support an upswing
Most assessments currently indicate that the Swedish economy is moving into a period of high activity, with strong domestic demand as well as strong export growth.
The expansionary monetary conditions are beginning to exert their full effect on the development of demand. The level of the Riksbank's instrumental rate is comparatively low, which has contributed to low market rates. The exchange rate, moreover, is weaker than we believe it should be in the long run. Its effect is therefore also stimulatory. The combined effect of short-term interest rates and the exchange rate is accordingly tending to support the economic upswing.
A continuation of this direction in monetary policy throughout the economic cycle would be incompatible with price stability. Expansionary monetary conditions cause the growth of demand to accelerate so that it exceeds the economy's long-term growth rate. In such a process, unutilised capacity is activated initially but the time comes when the pressure for resources becomes so great that it generates inflationary tendencies.
But surely, people may say, there is no danger at present of overheating and rising inflation. It is precisely here that a central bank has difficulty in clarifying the situation.
The reason lies in the long time lag before monetary policy measures take effect; between one and two years may pass before their full impact materialises. Today, therefore, the Riksbank has to assess what the economic situation will be towards the end of 1999, for example as regards resource utilisation. Provided a central bank acts in time, it is possible to avoid sharp interest rate adjustments. Such action often has to be taken when inflationary pressure is still low. When most people think everything is still going well and the central bank raises its instrumental rate, it is often accused of tilting at windmills. If our efforts are successful, however, inflation will never pick up. As I mentioned earlier, that is how the Riksbank supports persistently sustainable economic growth.
Work on the coming inflation report
Our inflation reports present a detailed picture of our assessments of economic trends, resource utilisation, inflation expectations and the inflation outlook for the total economy.
In the September inflation report it was estimated that economic growth would be 2 per cent in 1997 and around 3 per cent in 1998, followed by the prospect of continued acceleration to over 3 per cent in 1999. With a current difference of around 2 per cent between total demand and total resources and a potential growth trend of 2 per cent, the output gap would then be likely to close some time during 1999 and in any event towards the end of that year. In September the Riksbank concluded that the gap would not close in the future period of one to two years that is relevant for monetary policy. At the same time it noted that monetary policy is being conducted in the context of rising activity. Risks of inflation have to be assessed not only with reference to the degree of unutilised resources but also in the light of how quickly the output gap is closing. An upward tendency was foreseen in the CPI increase during 1998 and 1999.
In the coming inflation report the relevant period for monetary policy extends somewhat further into the future, to the end of 1999. Given that the earlier economic forecast holds, in the course of this period the gap between total demand and total resources will close. Statements from the Riksbank in the course of the autumn have drawn attention to this. If a central bank is to take timely action, such an analysis accordingly implies that the time is approaching when some reduction of the stimulatory effect will be necessary.
Savers and investors in the market are familiar with this line of reasoning. For them the question is not whether but when the Riksbank will raise its instrumental rate. An expected increase in the repo rate is, in fact, discernible in the structure of market interest rates. The general opinion among financial agents seems to be that an increase in the repo rate is probable some time in the coming winter.
All this serves as a starting point for the work that is in progress on the Riksbank's coming inflation report, which provides a foundation for the discussion by the Governing Board and the ensuing conclusions.
Some important problems
The Riksbank has to analyse whether the demand forecast still holds or whether additional factors warrant a revision. One factor, which so far looks like being of secondary importance, is the stock market turmoil in Southeast Asia. But the possibility of other factors also has to be considered.
Another issue is resource utilisation. Capacity utilisation in manufacturing is high. The National Institute's business tendency surveys shows that the supply of unutilised resources is dwindling. While there are no general labour shortages, signs of this are evident in specific fields. More and more firms are reporting longer delivery times, though upward pressure on prices still seems to be limited. This highlights the issue of economic flexibility - how smoothly will it be possible to transfer resources in the coming years to areas where demand is good?
It is likewise necessary to consider the wage negotiations; together with wage drift, they will determine the overall wage rise. Will there be a complete adjustment to what is needed to fulfil the inflation target or will wage increases tend to depress profit margins and thereby generate upward pressure on prices from costs?
In a balanced economic situation, the rate of wage increases that is consistent with the inflation target and with sustained economic equilibrium is given by the sum of the targeted rate of inflation and the trend growth rate for productivity. Thus it is important to distinguish between productivity's long-term trend and its cyclical component. The somewhat higher productivity growth that is indicated by the national accounts for the first half of this year is no doubt attributable to the economy being in the beginning of a broader expansion rather than to an increase in the trend.
At present, however, the economy is not in equilibrium. Unemployment's high level is evidence of this. The room for wage increases may therefore not be the same as in a balanced situation. Greater wage moderation may be needed in order to promote a balanced increase in employment while maintaining price stability.
Yet another major component in the assessment of future inflation is the development of inflation expectations and thereby of confidence in economic policy.
Europe is now entering a period when the monetary union will be established. Important decisions are to be taken next spring about which countries will participate from the start. The new European Central Bank is to begin its work no later than July 1st. It looks as though the monetary union, in little more than a year from now, will begin to conduct a single monetary policy for the countries in the euro area.
In that most things suggest that Sweden will not be participating from the start, there must be no cause to question the credibility of our monetary policy in the run-up to Stage Three. There must be no doubt whatsoever that Sweden's inflation target is a serious commitment.
Conclusion
Monetary policy acts on inflation with a time lag and therefore has to have its sights on the future.
The Riksbank is now working on the coming inflation report. The period ahead will call for a number of complex assessments. The probabilities of alternative scenarios have to be weighted together to arrive at a comprehensive assessment of future inflation. From what I have said about demand, resource utilisation and other factors, it is clear that some time in the future monetary policy will have to be given a somewhat less expansionary direction.
It is by consistently working to achieve the 2 per cent inflation target that the Riksbank can best contribute to sustainably high and stable growth in the Swedish economy.
