Heikensten: Economic policy and inflation
Thank you for the invitation to attend your meeting and talk about economic policy and inflation.
Today I shall begin with a brief review of the historical background to the fight against inflation in the Swedish economy. It is worth recalling that more than eight years have passed since it was decided in Sweden that combating inflation would have top priority. This policy has been pursued consistently ever since. After a couple of initial years that were troublesome in many ways, the results have been encouraging and Swedish stabilisation policy now looks good even in a European comparison. This will be followed by a discussion of three policy fields—fiscal policy, labour market and wage formation—that are closely linked to monetary policy, as well as some remarks about competitive conditions in Sweden's economy. Successful results in these policy fields enlarges the room to manoeuvre in monetary policy. In conclusion I shall consider some topical monetary policy issues, including our current appraisal of inflation and monetary policy's construction.
1. The fight against inflation and its results
High inflation characterised Sweden's economy for a number of years in the 1970s and '80s, a period that clearly illustrates an inflation economy's negative consequences.
Repeated attempts were made to rectify the economic imbalances by means of devaluations and fiscal stimuli but development remained unstable. The problems with inflation were then accentuated when the financial markets were deregulated. Prices and wages rose steeply and Sweden was caught in wage-price spirals. The occurrence of price bubbles made investment decisions more hazardous and led to weak productivity. A high level of inflation became a permanent feature of our economy. As monetary policy at that time was focused on maintaining a fixed exchange rate, fiscal policy ought to have been tighter.
Bit by bit, there was a growing awareness across party political lines that in the longer run this economic policy strategy was not viable, a perception that had already dawned on most countries in our part of the world in the early 1980s. It was now that low inflation in Sweden came to be seen as a necessary condition for achieving stable and sustained growth and thereby high employment. This was clearly expressed officially for the first time in the Budget Statement in January 1991. Combating inflation was given pride of place on the economic policy agenda. The final policy objectives were the same as before but the perception of the conditions that were needed to attain them had changed.
This realignment of stabilisation policy occurred, however, at a time when the earlier policy's negative consequences had begun to leave their mark. Due to the high level of costs in manufacturing, for example, economic growth in Sweden was weak. In addition, a rising real interest rate coincided with a period when the tax reform made savers and investors more sensitive to interest rates. All this had a marked downward effect on domestic demand.
The drop in output—almost 6 per cent in the course of three years—meant in turn that government finances worsened dramatically. From a surplus at 4.2 per cent of GDP in 1990, the budget balance swung to a deficit of 12.3 per cent in 1993. In this period, moreover, registered unemployment shot up from 1.7 to 8.2 per cent.
In a way, the economic crisis in the early 1990s can be said to have been generated by political decisions. The combination of a stabilisation policy realignment with some structural reforms did play a part in the course of the crisis. But it should also be underscored that such measures were broadly essential. The problem was not so much the measures that were implemented when an inflationary bubble had already developed as the fact that policy had not been realigned earlier.
In recent years, economic developments in Sweden have been favourable. Growth has been stronger than in the high phase in the late 1980s and it has proved possible to combine good growth with low inflation. A couple of years after the realignment of economic policy, inflation had fallen sharply to around 3 per cent. Moreover, the reduction of current inflation has been accompanied by a downward revision of inflation expectations. Today, neither households, firms nor financial market agents perceive any serious threat from inflation in the coming five years.
Government finances have also improved markedly since the effects of the acute crisis in the early 1990s were overcome. After rapid budget consolidation, a surplus of 1.6 per cent is forecast for 1998. Government debt has been reduced since 1995. Still, the restoration of confidence did take a long time. There were strong and persistent doubts about both the determination and the ability of the Government and the Riksdag (Sweden's parliament) to carry out the proposed measures.
Finally it is worth noting that the steps for reorienting economic policy on stability were taken with a consensus that was considerably broader than is sometimes suggested by the political debate. A good foundation for the years ahead has now been laid, although high unemployment is still a serious problem for Sweden's economy. The positive picture is underpinned by institutional reforms to the budget process and the status of the Riksbank.
2. Three fields of importance for monetary policy
Now I should like to turn to some issues or policy fields that are of major practical importance for monetary policy. They are: fiscal policy, the labour market, wage formation and competitive conditions in the Swedish economy. It is generally the case that the more price stability is supported by other policy fields, the greater the contribution monetary policy can make to other economic policy goals.
Good fiscal foundation but storing up for lean years is advisable
Fiscal policy is relevant for monetary policy in a number of ways. Fiscal decisions affect the room to manoeuvre in monetary policy not only via their direct impact on demand—and hence on inflation—but also by how they influence economic policy's credibility. A fiscal policy that is perceived as being untenable in the longer run can raise doubts about the overall policy commitment to low inflation and stability. To this should be added fiscal policy's influence on the macro economy. The ways in which taxes are levied and public revenue is spent influence the behaviour of households and firms, which may thereby—generally in the longer run—alter the economy's inflation propensity. Today, however, I shall not be developing this aspect.
In the first respect—fiscal policy's impact on economic activity—the picture is clear-cut: in recent years the fiscal stance has been very tight. Of course, this has contributed to the low rate of inflation and in this sense it has assisted monetary policy. With the current fiscal proposals for the years ahead, however, the stance will not be as tight as before. But to some extent the growth of household income is still being restrained by the overall change in taxation and public spending.
As regards credibility, there has been a marked improvement. Last year the Government laid down a long-term objective for government finances which goes further than the average of the corresponding commitments in Europe. Over the business cycle the budget balance is to average a surplus of 2 per cent of GDP. The Riksdag has approved this objective.
The Riksbank finds this objective reasonable. A budget surplus over the business cycle can contribute to a speedier reduction of government debt, which is currently high. This will create greater freedom of action for future economic policy, which is particularly important in that Sweden's public finances are no doubt still relatively sensitive to the cyclical economic situation. But the future has further challenges in store for fiscal policy.
An ageing population implies new and growing requirements, for example. Another matter that may alter the conditions for fiscal policy is the increased mobility of tax bases, which in some respects may make it difficult to maintain present levels of taxation.
Better wage settlements but what about the next round?
As things now stand, fiscal policy will continue to provide good room to manoeuvre in monetary policy in the coming years. So what about the contributions from the labour market and wage formation?
Of the aggregate value-added in the Swedish economy, wages make up almost 70 per cent. Wage trends are therefore crucial for the path of inflation. For many years, the wage rise in Sweden has been above the rate in most other countries in Europe. But as we all know, instead of this giving Swedish employees a better real wage, the outcome has been the reverse. In the 1990s the rate of wage increases in Sweden has been curbed but the negotiated increases in the 1995 round were above the European average.
As I mentioned earlier, economic policy in Sweden has been focused on low inflation since 1991. A consistent implementation of this policy, with an inflation target, clarifies the rules of the game for wage formation. High wage increases are liable to boost inflation, which necessitates higher interest rates.
The inflationary consequences of the wage increases can be modified if productivity growth is good, as has been the case in recent years. But at least a part of this productivity growth is achieved at the expense of jobs; this is one aspect of the process whereby higher labour costs generate stronger rationalisation and maybe also an increased capital input. In this way, too, and not just because the higher interest rates directly affect demand and employment, a rapid wage rise is liable to endanger jobs.
Against this background, the outcome of the latest wage negotiations is better than in earlier years. Agreements for the next three years have been concluded at levels around 3 per cent a year. It has been said that this takes wage drift into account but that is probably not entirely the case. Instead one should no doubt count on wage drift averaging around 1 per cent. The average annual wage outcome in the period 1998–2000 would then be 3.5 to 4 per cent. This does still represent some downward shift in the rate of wage increases, which in turn assists the Riksbank's task of ensuring low inflation. All else equal, we can allow ourselves a somewhat lower level of interest rates and a somewhat stronger growth of demand.
Speaking to you, it seems natural to take up two more aspects of this issue:
- The outcome of the 1998 wage round does not appear to be sufficiently low to make a major contribution to employment in Sweden. But our wage level is still rising faster, if anything, than the European average. This may prove to be manageable as regards inflation and Sweden's competitive position, thanks to good productivity growth, but what will it then entail for employment?
- Neither is it easy to feel really confident about the next wage round. What was really responsible for the better outcome this year compared with 1995? The greater credibility of the inflation target can be cited as one explanation but is it the whole story? And if not, what were the other arguments on this occasion and will they still hold good in two to three years time? These are important questions that need to be analysed and answered, in the first place by the parties in the labour market.
Better competition and price formation but there is more to do
A third issue, which presumably has had practical consequences for monetary policy and the fight against inflation in recent years, although it is seldom discussed in these contexts, is the competitive situation in the Swedish economy.
In work at the Riksbank on monitoring prices in the Swedish economy, we could not fail to notice some positive changes in competitive conditions and pricing in recent years, as well as some more negative patterns that seem to be persisting.
Let me take the positive aspects first. One is the favourable price trend for imported goods in recent years. This is particularly notable in that for much of this period the exchange rate has been comparatively weak, which by itself should push import prices up. A favourable price trend has also been observed for food products, for example; this seems to be a consequence, at least in part, of Sweden's membership of the European Union. Today, moreover, there are many indications that the combination of European monetary union and increased IT-based trade, for instance, will make competition even stronger.
There is a general difference between the development of administered prices and market prices for goods and services. The administered prices have risen faster. This may be a result of increased fee-setting in the public sector in recent years, both in connection with structural moves away from tax financing and to fill existing “holes” in public finances. Then there is the classic explanation that it is not so easy to raise productivity growth in some parts of the services sector. This no doubt applies in particular to a number of the services that are mainly provided under public auspices, such as medical care and care of the aged.
I have raised the questions of competitive conditions and price formation today mainly so as to help initiate a debate on these matters. For monetary policy it is self-evidently the case that increased competition, by keeping costs and prices down, can lead to better inflation prospects. This in turn can improve the conditions for monetary policy.
In recent years there have been indications of tendencies in this direction, in Sweden as well as globally. At the same time, it is difficult to tell just how much of the reduction of inflation can be attributed specifically to changes in competition and price formation.
Neither is it clear for how long such a process can continue and whether the same potential exists for keeping prices down over a series of years. These are matters to which we intend to return.
3. Some monetary policy issues
Sweden has had good experience, as I demonstrated earlier, of efforts to establish credibility by means of a direct inflation target. Inflation is currently low and the level is expected to remain low in the coming years. The conditions for continuing with monetary policy's focus on stability are being further improved by the far-reaching delegation of monetary policy that is involved in the Riksbank's enhanced status from the turn of this year. Against this background, I should like to comment on some of monetary policy's topical issues.
The exchange rate
The part played by the krona's exchange rate is a matter for debate that surfaces from time to time. A recent instance is, for example, the run up to European monetary union and the attention this drew to Sweden's commitments as a member of the European Union. Some observers believe that the Riksbank may have an (unacknowledged) exchange rate target. That is a misconception. The Riksbank does not target the krona's exchange rate.
In a flexible exchange rate regime with a price stability target, the exchange rate is admittedly a major factor, but only one of many factors that impinge on inflation. A permanent appreciation or depreciation of the krona affects inflationary pressure via import prices, for example. A given exchange rate trend may well support monetary policy's drive for price stability; when this is not the case, however, the Riksbank has to intervene and adjust the repo rate.
At the same time, the exchange rate is the price of an asset that is current throughout the economy; the value of the krona is highly important in many economic decisions, not least for export companies. This makes a stable exchange rate fundamentally desirable.
It is then easier for all economic agents to make decisions. All else equal, this should lead to higher growth.
To this basic endeavour to keep the price of the krona stable should be added our European undertakings. As a member of the European Union we are committed to treating our exchange rate policy as a matter of common interest. Opinions in the Union presumably differ from time to time about the practical application of this commitment but the basic requirement is that policy focuses consistently on stability.
The Riksdag has left the door open for Sweden's participation in the monetary union at a later date and stated that in connection with such a move it can be relevant to join the European exchange rate mechanism (ERM). In this perspective, too, it is an advantage if the krona is more or less stable.
Monetary policy in the euro area is to have the primary objective of price stability. We do not yet know just how the European Central Bank (ECB) will formulate its target but it is unlikely to differ from the norm that has applied in Germany since the mid 1980s. This means that inflation is not to exceed a medium-term level of 2 per cent. In practice, therefore, the differences in this policy field between the ECB and Sweden will probably not be so great as to have any real consequences for the exchange rate.
With policy focused on price stability both in Sweden and in the monetary union, there should be good possibilities of a stable exchange rate between the krona and the euro. This is supported by the path of the krona relative to the German mark in recent years.
As measured, for example, by the long-term interest rate differential with Germany, confidence in Swedish economic policy has been high since 1996. In this period there have been only two occasions when the krona has fluctuated more than 2 per cent against the German mark.
Managing transitory effects
The subject of the Riksbank's attitude to the inflation target is also raised from time to time in the media. It has been claimed that the Riksbank shows less concern about inflation moving outside the tolerance interval's lower limit than it does about the upper limit being exceeded.
That is not the case. A couple of years is too short a period on which to base arguments that deviations from the targeted rate are a sign that policy has not been symmetric. Monetary policy work involves a lot of uncertainty, which is no doubt inevitable when the interest rate has to be set on the basis of forecasts that extend several years into the future. At present, moreover, assessments are particularly difficult so soon after the change-over to a low-inflation regime, with all that this may involve in the form of altered behaviour by households and firms. Another complication is that in recent years, inflation throughout the world has been lower than expected.
Another aspect of this issue is the Riksbank's choice of a target variable. Every conceivable indicator of inflation has its shortcomings and these have to be weighed against the advantages when choosing monetary policy's target variable. The consumer price index (CPI) has the advantages of being well-known and established among economic agents and people in general as a measure of inflation. It also covers a large proportion of household consumption. These were central arguments when the CPI was chosen as the target variable in the beginning of 1993.
A disadvantage with the CPI, however, is that it combines information about inflation's underlying trend with transitory price movements. In the period in which the inflation target has been in force, transitory factors have had a marked effect on the CPI on a number of occasions. Examples of such factors are tax changes (most recently the increases to and subsequent cuts in property and tobacco taxes) and fluctuations in interest rates (which affect the CPI in the form of house mortgage expenditure). During 1996 and 1997 in particular the CPI has been relatively markedly below not just the targeted rate but even the tolerance interval. This has faced the Riksbank with educational problems.
Transitory effects on inflation ought to be a secondary consideration in monetary policy and be absorbed inside the tolerance interval of 1–3 per cent. That is one of the interval's functions. And provided it is a question of temporary price movements around a stable rate of inflation, there is no reason to counter such movements with repo rate adjustments. On the contrary, that might have a destabilising effect in the real economy. Instead, monetary policy's central concern is the direction of inflation's long-term trend. It is this trend which can be affected more directly by total demand and hence by monetary policy.
Against this background, the Riksbank has contacted the CPI Enquiry, set up by the Government, to obtain assistance in specifying an indicator of underlying inflation that is likely to be widely accepted and can be published regularly by Statistics Sweden. In the future, such an indicator might be able to have a more direct function in monetary policy, either by replacing the CPI as the target variable or by serving as an operational target or aid in the continuous work of constructing policy. The Riksbank has also asked Statistics Sweden to publish, for the time being, one or more indicators of underlying inflation together with the CPI. Perhaps I should emphasise that in no way do these initiatives alter monetary policy's fundamental commitment to price stability.
The current situation for monetary policy
The question of transitory effects leads me on to current monetary policy. Transitory effects are playing a major part in the CPI's present development. As a result of tax changes, lower interest rates and a lower oil price, since December the Riksbank has had to revise the 1998 CPI prediction downwards by about 1.2 percentage points. In 1999, too, transitory effects mean that CPI inflation will be lower than we envisaged earlier.
Thus far, however, the altered picture of inflation need not have any sizeable consequences for monetary policy. It is now too late to influence this year's path for inflation more than marginally. Moreover, some of the transitory effects entail a risk of higher inflation, if anything, during 1999; tax cuts and lower interest rates leave households with more money to spend and the same applies to cheaper oil.
But underlying inflation also shows signs of being weaker than expected even though economic activity seems to be relatively strong and in line with the paths we have counted on for some time, with a further increase driven by domestic demand and supported by strong activity in Europe.
The better picture of prices comes from a variety of factors. The rate of wage increases is likely to be somewhat more advantageous than we counted on earlier, even though we have consistently been cautiously optimistic in this respect. Of greater importance is productivity, where good growth in recent years has helped to brighten prospects.
Somewhat lower import prices together with a slightly stronger exchange rate also contribute.
Work on our next Inflation Report is nearing completion. There are several indications that in the two years horizon that is relevant for monetary policy, CPI inflation may be somewhat below our target. The picture of underlying inflation is much the same. Neither are these positive inflation prospects modified by the various risks that lie ahead. In a perspective that extends to the middle of 2000 the element of uncertainty is obviously high; if anything, however, the main risk seems to be that international activity will be marginally weaker and inflation accordingly somewhat lower.
This is a pleasant situation to be in. The Swedish economy appears to be growing at a rate that is reasonable and broadly expected. At the same time, inflation is lower than expected. The prospect that the Riksbank faced last autumn, of inflation two years hence being half a percentage point above the target, no longer seems probable.
For monetary policy, the outlook has accordingly changed. The present inflation prospects clearly imply that the need to raise interest rates has disappeared for now. A strict interpretation of the Riksbank's rule that monetary policy is to focus primarily on the second of the coming two years could even indicate some downward adjustment of the instrumental rate. What the Riksbank now has reason to consider is whether a minor downward interest rate adjustment would lead to good conditions for a stable economic development in the future.
4. Conclusion
Stabilisation policy in Sweden was realigned in the early 1990s and focused on low inflation. The background was our own and other countries' experience in the 1970s and '80s. Repeated attempts to stimulate demand, with a view to boosting employment, and accept the associated increase in inflation had failed to produce the intended results. On the contrary, economic activity had become more unstable and growth in the longer term had, if anything, weakened.
In recent years the results of the policy realignment have been good in many respects. Economic growth in Sweden has been higher than in the latter part of the 1980s and it has proved possible to combine this with low inflation. Government finances have also undergone a rapid improvement. Moreover, revised rules for both fiscal and monetary policy have provided better conditions for future stability. This means that a good foundation has also been established for future economic development.
The 1998 round of settlements agreements represents a step in the right direction in the sense that, compared with the previous round, the agreed rate of wage increases in coming years looks like being easier to combine with low inflation. Unfortunately, however, there are still reasons for some concern about the future. Do the wage settlements create really good conditions for increased employment and has a foundation been laid for further responsible wage settlements in the next negotiating round in two to three years time?
Some positive things in the Swedish economy have also happened in recent years as regards competitive conditions and price formation. With the ongoing process of internationalisation, European monetary union and advances in the IT field, there is likely to be a further intensification of competition and downward pressure on prices, to the benefit of Swedish consumers. But there is cause to consider whether more can be done, which if there is would also assist monetary policy.
Finally I considered aspects of monetary policy and the current situation in this respect. In the past six months inflation prospects have improved. Like many central banks in central Europe, the Riksbank has accordingly moved from a situation that pointed to the likelihood of a relatively quick succession of continued interest rate increases to one where the need of higher interest rates in the near future seems to have disappeared.
With the bright inflation prospects and a strict interpretation of the Riksbank's rule for monetary policy decisions, today there may even be grounds for considering a minor downward repo rate adjustment. What the Riksbank now has reason to consider is whether such an adjustment would lead to good conditions for stable economic.
