Sweden's Economy and Monetary Policy
Thank you for inviting me to New York to discuss the Swedish economy and Monetary policy.
Economic developments in recent years
In the past few years Swedish economic policy has achieved tangible results.
o By an improvement in credibility, important progress has been made towards greater economic stability. Favourable economic growth in Sweden has been combined with a rate of inflation that is the lowest for several decades.
o There is broad political agreement in Sweden on the consolidation of government finance. This is manifested in the so called convergence programme that aims to eliminate the budget deficit not later than 1998. The budget reinforcements which the Government and the Riksdag (Sweden's parliament) have undertaken in recent years are extensive. The six-monthly monitoring arrangement provides a means of ensuring that government finance is developing as planned. Moreover, in the Economic Bill this spring the Government states that the long-term target for the general Government budget is a surplus of 2 per cent of GDP as a cyclical average. The gradual adjustment to this goal is also specified.
The general government financial balance has been strengthened from -12.3 per cent of GDP in 1993 to -2.5 per cent in 1996. This is a substantial improvement. The Government calculates that in 1997 the deficit will be 2.1 per cent and that the target of eliminating it in 1998 will be achieved.
o With the improved confidence in budget and monetary policy, together with the tighter interest rate policy from the second half of 1994 up to the end of 1995, it has been possible to check the earlier inflationary tendencies and bring inflation expectations down appreciably. Sweden is now among the countries in Europe with the lowest inflation rate regardless of which measure one uses. This, in turn, paved the way for a gradual lowering of our main instrumental rate - the repo rate - in 1996.
o What used to be an almost chronic deficit on current account has now been turned into a substantial surplus. This is reducing Sweden's external debt, which also contributes to greater financial stability.
o Many structural measures have been implemented since the late 1980s. There are signs that, together with economic policy's focus on stability, this has reduced the economy's inflation propensity and raised potential output.
Against this background, the direction of Sweden's economic policy contrasts sharply with the unstable, inflationary path we experienced in the 1970s and '80s and which contributed to the recession in the early '90s, which was the worst for over sixty years.
A new direction in economic policy
The 1970s and '80s illustrate how a period of inflation creates an atmosphere of uncertainty and instability, with a speculative climate and "short-termism" in economic decisions. Slowly but surely, production's long-term conditions deteriorate.
It was partly in the light of such experiences that Swedish economic policy was realigned at the turn of the 1980s. Since then, the focus on price stability has gained broad support in our society, including the political parties.
However, in view of the severe economic problems in the early 1990s, with inflation expectations based on our historical record and a massive budget deficit, it took time for the new direction of economic policy to acquire credibility. This has been evident, for instance, in the periods of interest and exchange rate unrest. Financial market trends particularly in the longer run can be regarded as an indicator of the prevailing degree of confidence in the determination and ability to conduct an economic policy that is consistent and oriented for stability.
With the free movement of capital and the globalisation of financial markets, economic policy decision-makers need to be highly disciplined. In the light of our experience since the policy realignment at the turn of the '80s, this is a fact that is widely recognised today in Sweden. Turning the clock back would not be feasible.
Experience from other countries, where the policy realignment had already been made in the early 1980s, indicates that the creation of confidence in a low inflation regime takes time and a lot of hard work. But it also shows that after a while, such a process does yield its reward in the form of low interest rates, a stable currency and ultimately a favourable economic development.
The price stability target
Following the move to a flexible exchange rate, monetary policy in Sweden is directed at a quantified inflation target. In this it resembles the situation for monetary policy in Australia, Canada, New Zealand and the United Kingdom, where inflation targets are also used.
In Sweden the target is formulated as being to limit the increase in the consumer price index to 2 per cent. This target is to be assessed, not in terms of isolated twelve-month change figures but in a somewhat longer, annual, perspective. As monetary policy obviously cannot control price developments exactly, the target is accompanied by a tolerance interval of ±1 percentage point. This tolerance interval also serves to absorb temporary deviations from trend inflation. Such deviations may arise, for instance, from political decisions to alter indirect taxes and subsidies, changes in interest rates as well as from international price shocks.
For March 1997 the consumer price index showed a twelve-month fall of 0.4 per cent. However, this figure was to a large extent influenced by last year's sharp repo rate cuts. When home owners renew mortgage loans at lower interest rates we get a transitory downward effect on inflation.
It is important to distinguish between CPI changes and underlying inflation. Transitory effects, for example changes in interest rates or in indirect taxes and subsidies, may cause the general price level to jump up and down. In the conduct of monetary policy it is therefore important to keep an eye on underlying inflation; the transitory effects are of secondary significance as long as they do not affect inflation expectations and thereby trend inflation.
Given the time lag that applies to monetary policy, a central bank that targets inflation has to monitor the various factors which determine inflation one to two years ahead.
In a simple analysis of inflation's determinants one can start from the familiar augmented Phillips curve. The rate of inflation is determined by two factors: the relationship between supply and demand in the economy - the output gap - and inflation's expected rate.
Note, however, that while the output gap is a useful starting point for an inflation analysis, there is no simple way of measuring it. This makes it necessary to go into more detail and study a variety of indicators in order to analyse any tensions between demand and supply. Still, the output gap is a suitable starting point for my present purpose.
The simple model tells us that when the output gap is closed, inflation is constant and equal to the expected rate. This means that, with normal capacity utilisation and given a low level of inflation expectations, actual GDP can follow potential output. In such a situation some form of external shock is needed to shift inflation upwards or downwards. This can mean either that production falls below or rises above the level for normal capacity utilisation or that inflation expectations move up or down. The latter case is ultimately a matter of confidence in economic policy's consistency and its future reactions.
The ability of monetary policy to influence inflation has to do with the impact of the instrumental rate on the path of GDP and thereby on the degree of capacity utilisation, together with its effect on inflation expectations. In practice a period of restricted demand has frequently been necessary for the central bank to show that expectations of high inflation are unfounded. Besides monetary policy, expectations are affected by other factors. One of these is the budget. Unsustainably large budget deficits are liable to undermine confidence in long-term price stability.
Our inflation forecast serves as a foundation for decisions about monetary conditions. If monetary policy is directed at stabilising inflation att 2 per cent, it will over time help to stabilise demand so that this is in line with the economy's available capacity.
In order to clarify our line of reasoning and our inflation assessments, the Riksbank publishes an inflation report four times a year. A range of indicators is reviewed in these reports, accompanied by a discussion of risks and opportunities in the field of inflation.
The present situation
In the latest inflation report, published this March, GDP growth in 1997 and 1998 was expected to be around 2 and 3 per cent, respectively. It was also argued that sizeable capacity restrictions should not arise in the course of the period. Given that, and inflation expectations being low, underlying domestic inflation (measured with the CPI excluding transitory effects on inflation) in the coming two years was put at between 1.5 and 2 per cent.
As we all know, to forecast inflation is a complicated task. In view of the shortcomings in such estimates, it is more a matter of assessing the balance of risks. Our latest inflation report therefore includes a discussion of two alternatives in which future inflation might be either lower or higher than in the main scenario.
One alternative is that inflation could be lower than in the main scenario, mainly if private consumption were to be weak. The other envisages the opposite - higher inflation - on account of a weaker exchange rate that stimulates a stronger growth of exports and GDP, leading to increased inflationary pressure.
What are the economic tendencies that have appeared during the last months?
During winter there were signs of a slowdown in manufacturing. The December survey from the National Institute of Economic Research showed an unchanged inflow of orders and expectations of subdued production. The order statistics from Statistics Sweden, which fluctuate fairly widely from month to month, gave the impression of relatively weak orders in the fourth quarter of 1996. In December and January, however, the order inflow has picked up and more recent survey data likewise paint a brighter picture of activity in manufacturing. The improvement mainly applies to production for export but even the domestic market is somewhat optimistic about the future. Against this background, exports and industrial production could develop somewhat better than expected earlier. The changes are not very large but to date various indicators, including the continued undervaluation of the krona, do point in the same direction.
In the case of private consumption, the indicators to date are not uniform. On the one hand, in the first three months of 1997 the number of new car registrations for private individuals has risen in annual terms by more than 60 per cent, albeit from a low level. Moreover, the survey of household purchasing plans by Statistics Sweden shows that since the autumn, household's economic expectations have become more optimistic. The development of monetary aggregates such as M0 is also an indication of rising private consumption. On the other hand, the lower than expected savings ratio in 1996 may increase uncertainty regarding possible increases of private consumption in 1997 and 1998. Another indicator of a slower increase in private consumption could be the weak labour market.
In summary it can be said that, on one hand, exports seem to be doing somewhat better than expected, though the differences are not all that great. On the other hand, the signals about private consumption point either to agreement with the main scenario or to a somewhat weaker course.
The conclusion in the latest inflation report was that monetary policy is well balanced. That
