The economy and monetary policy
Let me begin with a word of thanks for being invited to talk with you about the Swedish economy and monetary policy.
Today I shall be considering three issues. First some words about the construction of monetary policy and why low inflation is a good thing. Then a look at what has happened in the Swedish economy since the crisis culminated in 1993. Finally I shall be saying something about the future outlook for the economy and for monetary policy.
1. Low inflation and economic stability
The objective is low inflation
The Riksbank conducts monetary policy with a target that is clear and unmistakable: the annual rate of change in the consumer price index is to be kept at 2 per cent, with a tolerance interval of *1 percentage point. In order to fulfil the target the Riksbank varies its instrumental rate (the repo rate), which influences the level of economic activity and thereby inflation. Experience has taught us that the time lag before an interest rate adjustment has its full effect on inflation ranges from one to two years. This means that the Riksbank has to operate with a forward-looking perspective - the interest rate must be set in the light of forecast inflation one to two years ahead.
The central function of the Riksbank is to achieve price stability or at least keep the rate of price increases low and stable. This approach to monetary policy is based on the insight that in the longer run, higher price increases cannot generate growth and employment. Growth stems from other factors, such as technological advances, the way the economy functions, demographic trends, etc. Attempts to stimulate economic growth by accepting higher inflation soon tend to boomerang. This is clear both from our own experience in recent decades and from what has happened in other countries. The players in the economy adapt to the higher inflation; prices, wages and interest rates rise; the end result is normally an economic trend that is weaker instead of stronger.
Throughout the western world today, economic policy is less concerned with "stop and go" policies to influence short-run activity and focuses more on creating good and stable long-term conditions for economic growth and employment. In such an approach, the function of monetary policy is to safeguard the value of money. In the past decade support for a policy of this type has also grown markedly in Sweden. This year's Spring Economic Bill states that "The overriding task for monetary policy is price stability". In keeping with this and in order to facilitate the implementation of a credible monetary policy, more and more countries are striving to separate monetary policy decisions from day-to-day policy-making. The latest instance of this comes from Britain, where one of the first things the new government did was relinquish control of interest rate policy to the Bank of England.
The harm done by inflation
High inflation impairs the workings of the economy. General price increases make it more difficult for households and firms to keep tabs on how the prices of certain goods and services are moving in relation to others. It then also becomes more difficult for firms to know which products are worth marketing, which investments will pay best and so on.
A case in point and one that is certainly familiar to some of you is Sweden's property boom in the 1980s. The combination of high inflation and the tax system in those days made speculation in real estate profitable. The property market then collapsed in the early 1990s and occasioned huge economic costs. The overheated real estate market had crowded out other, more productive investments and when the bubble burst, finance companies and banks were hit by loan losses at the same time as the ensuing adjustment entailed massive costs in the form of lost production and unemployment.
There are also grounds for supposing that low, stable inflation moderates cyclical fluctuations in economic activity. This seems to be the case, at least, in the United States, where since 1982 there have been two long expansionary phases separated by a single, relatively brief downturn. Experience in Sweden points to the same conclusions. The first two post-war decades, for example, when inflation was low, were considerably more stable than the 1970s and '80s, when inflation was left to rise and there was a series of devaluations, with negative effects on productivity and growth.
Moreover, the higher the rate of inflation, the more likely it is that the rate will fluctuate, which accentuates the uncertainty about future inflation. Savers and creditors then require higher compensation for risk: besides compensation for the expected rate of inflation, they will demand a premium for the degree of uncertainty about future inflation. This means higher real interest rates, which weaken the propensity to invest.
Furthermore, the more inflation fluctuates and is difficult to forecast, the greater will be its arbitrary and random effects on the distribution of income and wealth. Inflation tends to redistribute from savers and lenders to borrowers, from those with fixed nominal incomes to those who pay such incomes, and so on. Clearly, such arbitrary redistributional effects are hardly in line with a distribution policy based on fairness.
2. Developments since the crisis
Several years of good growth
Since the upturn from the profound crisis in 1991-93, economic growth in Sweden has averaged 2.8 per cent, which is higher than we were accustomed to in the inflationary 1970s and '80s. Moreover, the composition of growth has been favourable, with exports and investment predominating (Diagram 1). It should be underscored that this has been accompanied by the lowest level of inflation for decades.
The low inflation as such has no doubt helped to make the economy function better. The conditions for a long, stable upswing have improved. Together with clearer price signals, this makes it easier for households and firms to plan and make decisions. The workings of the economy have also been improved by the tax reform, various deregulations and changes to social security systems. Competitive pressure has grown as a result of EU membership and the globalisation of trade. The result is discernible in a higher rate of capital formation and structural changes in the economy.
In both 1994 and 1995 GDP growth topped 3 per cent. In 1996 the growth of final demand remained fairly high, 2.1 per cent, and the rate accelerated in the second half-year to 2.6 per cent from the corresponding period the year before. Total production, however, did not match the growth of final demand, the reason being that firms chose to trim stocks. The growth of GDP, which represents total output, stopped at 1.1 per cent in 1996.
Stock trimming in the current cyclical phase is not uncommon; it occurred at much the same juncture in the 1970s and '80s. It has to do with the tendency to produce in excess of demand when activity is rising so as to avoid being unable to deliver if demand grows rapidly. In time, firms find they have accumulated surplus stocks and initiate an adjustment. This has the reverse effect - production rises more slowly than final demand.
Under certain circumstances stock adjustments may trigger a more marked economic slowdown. This may happen, for instance, if effects of the adjustments either spread to private consumption or have a negative impact on the labour market and household confidence. The Riksbank was therefore alert to the risk of private consumption being weaker last spring and summer, which would have pointed to a further curb on inflation. In practice, however, private consumption went on rising in the summer and its growth became somewhat stronger in the autumn.
Low inflation and interest rate reductions
The rate of inflation decreased during 1996 and the rates that different economic agents expected in the future also moved down markedly (Diagram 2). This enabled the Riksbank to lower the repo rate from 8.91 to 4.10 per cent. As a result, the initial position for the ongoing growth phase in Sweden's economy is more favourable than for many years.
The drop in the rate of consumer price increases was appreciably accentuated by the Riksbank's repo rate cuts in that these contributed to the reduction of mortgage interest costs. Under normal circumstances the tolerance interval around the price stability target should suffice to absorb transitory effects on the price level. Last year, however, the fall in interest rates was very marked because many factors had a concerted effect that rapidly altered the picture of inflation. The magnitude of that interest rate fall is also largely responsible for the weak price trend to date in 1997.
Different indicators of underlying inflation also show a declining rate of increase, though the picture measured in this way is not quite as dramatic (Diagram 3). Excluding house mortgage interest costs, the current rate of inflation is about 1 percentage point higher than indicated by the CPI.
It seems reasonable to conclude, however, that an evaluation of the Riksbank's policy for inflation must recognise that a part of the changes in the rate of inflation stems from transitory effects occasioned by the Riksbank's own measures.
3. The current assessment
The March inflation report
I should now like to turn to the outlook for inflation in the coming years. In order to provide as open and clear a picture as possible of how the Riksbank analyses the future prospect for inflation and approaches the construction of monetary policy, we present reports on this four times a year. In the latest report, published this March, it was considered that the GDP growth rates for 1997 and 1978 would be 2 and 3 per cent, respectively. It was also foreseen that no appreciable capacity restrictions should arise in these years. With an unchanged monetary stance and given that households and firms continue to count on inflation remaining low, underlying domestic inflation (the CPI change excluding transitory effects on inflation) one to two years ahead was judged to be between 1.5 and 2 per cent.
In that the transitory effects will be dropping out of the 12-month change figures, however, registered inflation will move up by degrees in 1997 and 1998. In March we counted on the annual rate of inflation in 1997 being around 0.5 per cent, which would still be clearly below the inflation target. We also envisaged that in 1998 inflation would be somewhat below 2 per cent.
As we all know, forecasting economic activity and prices is not a simple matter. Since such assessments cannot be very exact, one has to weigh up risks and probabilities. Two secondary alternatives were therefore discussed in the March report, where future inflation might be lower and higher, respectively, than in the main scenario.
In the former alternative, lower inflation might result from a weaker development of consumption. In the latter, on the other hand, higher inflation might stem from a weaker exchange rate that leads to higher growth for exports and GDP, thereby generating stronger inflationary pressure.
Against this background, I want to say something about our current view of tendencies for private consumption and the exchange rate.
Private consumption
In 1996 household disposable income fell, unexpectedly, by 0.1 per cent despite a strong increase in the wage bill. The National Institute of Economic Research and many other observers had expected a 2 per cent increase in disposable income. The growth of consumption was therefore financed entirely from a reduction of the saving ratio to 5.4 per cent. This has altered the starting point for assessing private consumption's future path.
In recent years household income has been greatly affected by measures for consolidating government finance, mainly increased taxes and decreased transfers. But for the impact of public systems, household income in each of the years 1994-98 would have been 3 to 4 per cent higher (Diagram 4). As it is, the level is likely to be virtually unchanged. The withdrawal of household income probably peaked in 1996 but even in the two following years the fiscal stance measured in this way is tight.
A number of structural reforms (cuts in transfer systems, the tax reform, etc.) have probably elicited a permanent increase in the household saving ratio but the future tendency is difficult to foresee. One approach is to compare with countries where the economic framework for households resembles that in Sweden. This suggests that both the household net saving ratio and the ratio of total wealth to disposable income are relatively low in Sweden. It then seems unlikely that in a long-term perspective the saving ratio will fall back to the historically low levels in the 1980s (Diagram 5).
Leading indicators point to a continuation of private consumption's upward tendency in the first quarter of 1997. House prices have shown signs of moving in recent months. A growing money supply and lending to the household sector likewise point to a continued increase in consumption. Retail turnover, above all for consumer durables, strengthened in February and March, as is also evident in the March business tendency survey from the National Institute. In the first four months of this year the number of cars newly registered in the household sector rose in annual terms by around 60 per cent. Households have become more optimistic about their own economy; in April a majority envisaged an improvement 12 months ahead.
There is no cause at present to modify the main assessment in the inflation report - that consumption will show a stable but by no means dramatic rise in the coming years. But an appreciably weaker development of consumption, which was a prominent component of one of the alternative scenarios, seems to have become somewhat less probable.
Exchange rate
The Riksbank does not target the exchange rate. The monetary policy target is price stability. But the exchange rate does feature as one important factor in the Riksbank's assessment of inflation in the coming two years.
The exchange rate conditions demand in the economy, particularly in the export sector, and also affects inflation directly in the form of import prices. The latter, more immediate effect is, however, transitory and of secondary interest provided it does not influence inflation expectations and thereby price and wage setting behaviour.
An assessment of future inflation therefore requires a forecast or judgement of the exchange rate in the coming two years, yet this is a variable that is notoriously difficult to predict and has fluctuated widely for Sweden since 1992. A natural starting point for this is the current exchange rate together with an estimate of what a reasonable long-term level might be in relation to competitiveness, financial saving, etc.
The past six months provides an example of how the argument might go. Since October 1996 the krona's effective exchange rate has weakened by about 7 per cent. Most of this change stems from the appreciation of sterling and the US dollar against currencies in continental Europe (Diagram 6). Cyclical factors have been at work here; the United States and Britain are leading an upward phase and there have been expectations about future monetary policy that tended to strengthen both their currencies.
But the krona has also fallen, particularly during the winter and spring, against most Central European currencies, including the German mark. At times this has had to do with unrest in Europe over the EMU process. Many market observers, however, also point to some uncertainty about the durability of Swedish economic policy.
In the longer run there are good grounds for a stronger krona. Productivity in the Swedish economy has risen rapidly in recent years. The current account shows a large surplus and the expansion of export production should ultimately spread to the domestic sector. The outlook for inflation is favourable and government finance shows a considerable improvement. Still, there is reason to take the weakening of the krona seriously. Policy must continue to focus on consolidating government finance; budgetary improvements in connection with strong economic activity cannot be used for easing the fiscal stance.
Under these circumstances, neither are there grounds as regards the exchange rate for appreciably modifying the March assessment. To date in 1997, however, the krona has been somewhat weaker than envisaged and this, together with the available indicators, points to marginally higher export growth.
-----------------
All in all, today there is no reason to revise the overall picture of the economy and inflation that was presented in the March inflation report. The tendencies to date seem to broadly confirm the assessments there.
Various forward indicators suggest that the Swedish economy is again entering a period of increased growth. At the same time, most things point to price increases remaining low this year and next, which is the period we can assess at present.
The repo rate was lowered most recently last December. Since then we have frequently declared that the monetary stance is well balanced. This still applies.
