The Swedish economy and wage formation
Introduction
First I should like to express my thanks for the invitation to discuss Sweden's economy and wage formation. The topic is important. Economic activity has steadily improved in recent years but the persistently high unemployment is evidence that the labour market does not function properly.
Total production has recovered the drop at the beginning of the decade and its level is now above the high in 1990. GDP growth in recent years has also been stronger than the rates to which we were accustomed in the 1970s and '80s. Moreover, the increase in industrial production is the largest since the early 1960s.
In spite of this, the number of employed persons is at its lowest level since 1974. We now have an employment gap. In other words, Sweden is experiencing something that is characteristic of many other countries in Western Europe, namely jobless growth.
Since price formation is the basic market mechanism, it follows, as I see it, that weak employment and permanently high unemployment stem from wage formation. Whatever the causes of the high unemployment, important factors for turning the trend are how the labour market functions and the atmosphere between the parties represented there. The part played by the Swedish Employers' Confederation in this domain is therefore important.
As Governor of the Riksbank I have several reasons for discussing the labour market and wage formation.
* Firstly, unduly high wage increases can lead to a higher path for inflation - combating inflation is a central bank's primary function.
* Secondly, there is a need to explain what monetary policy can and cannot do by way of contributing to growth and employment. Most people today consider that permanently low unemployment cannot be achieved by accepting high inflation. Demand policy must be adapted to the economy's potential long-term output. If structural unemployment has risen - and in Sweden's case there are many indications that it has - measures must be aimed at bringing it down. That is not a matter for monetary policy.
* Thirdly, a favourable trend for jobs and unemployment helps to strengthen the long-term credibility of a low inflation regime. Conversely, a labour market that functions poorly can influence the expectations of savers and investors about the low inflation regime's sustainability and thereby the stability of bond rates and the exchange rate. Another way of putting this is to say that macroeconomic stability is supported by microeconomic flexibility.
Wage formation in focus
For a long time now, the labour market, like the economy in general, has been changing rapidly. Manual industrial work has been replaced by machines and there has been growing competition from countries with a considerably lower wage level. In Sweden this has been countered by expanding both public and private service industries. Fewer jobs in manufacturing have been accompanied by more in public and private services. For many decades, however, there has been no net increase in private sector jobs. The expansion of employment has been confined to the public sector, where one million new jobs were created between 1960 and 1990. Now that the political system has been obliged, for various reasons, to arrest public sector growth, total employment has fallen.
The process of de-industrialisation in Western industrialised countries is, generally speaking, a sign of economic success and rising standards of living. In this process, rapid productivity growth means that employment in manufacturing remains unchanged or even falls. The process has been at work for a long time in the "old" industrialised countries and recently seems to have spread to newly industrialised countries.
In that jobs in manufacturing account for a diminishing share of total employment, the development of industrial productivity will be less and less significant for the economy as a whole. Overall productivity growth and thereby the development of real wages will increasingly depend on the situation in service industries, where average productivity growth is lower.
There are countries where the decline of employment in manufacturing may pose problems if service industries do not expand and recruit those who lose their jobs elsewhere. The pace of de-industrialisation and the extent to which service jobs rise are dependent in part on how wage formation functions. The altered conditions in the labour market therefore place demands on wage formation so that the elimination of existing jobs can be offset by the creation of new employment. A good development of real wages may be desirable but the average path should not jeopardise the process whereby new jobs arise as old jobs disappear. The development of wages must also allow for productivity differences between industries and sectors.
In the United States the average real wage has risen in the past quarter-century by a total of 15 per cent, as against as much as 60 per cent in Europe. The picture of employment is the reverse: a 60 per cent increase in the United States as against 15 per cent in Europe. That tells us something about the importance of wage formation for jobs.
A particular problem in the European labour market - and to date at least this includes Sweden - is that increased unemployment does not seem to instil wage restraint. One conceivable explanation may be that, in practice, those who are out of work have a very limited say in wage negotiations. Instead of being restrained, with consequently lower labour costs and the possibility of hiring additional labour, wage increases remain high and so does unemployment.
A spectacular example of this is Sweden's construction sector, where unemployment is a good bit above 20 per cent. Yet the rate of wage increases there is almost on a par with the average for all employees. This illustrates what economists refer to as the theory of labour market insiders and outsiders.
The task of eliminating effects of inefficient wage formation and the ensuing structural unemployment lies outside the scope of monetary policy. But it is very much a matter that can be influenced by the labour market parties - representatives of employers as well as employees. The free negotiations that are a long-standing tradition in Sweden make these parties accountable for wage formation. A problem lies in the circumstance that the parties do not have to carry the full costs when wage formation functions poorly.
The 1995 round of wage settlements must be regarded as yet another failure. Despite a level of unemployment that was very high for Sweden, wage increases were negotiated that pushed up labour costs, necessitating further rationalisation and hampering new recruitment. Employment fell and unemployment rose.
There is certainly a good deal of new thinking about wage formation but the intended norm seems to be to negotiate wage increases at the level for Western Europe, and we are not even there yet. Has it not occurred to anyone that this can also led to unemployment becoming permanently fixed at the level for Western Europe?
From cooperation to collapse
At the time of the Saltsjöbaden agreement in 1938, Sweden was characterised by an organisational capability, pragmatism, cooperation and a firm set of rules. The agreement, which formalised a labour market mood of positive cooperation, probably mirrored the prevailing atmosphere, with both parties clearly aware of economic realities. Sweden's economy and labour market had experienced a very difficult period from the 1920s onwards, with exceptionally high unemployment in the early '30s. It was generally recognised that a key factor for changing the trend would be wage formation.
The results were not long in coming and in the decades after World War II economic development in Sweden was highly favourable. Inflation was low, growth high and the industrial peace was highly beneficial. The unemployment from the '30s was transformed into full employment.
The primary ingredient of the Saltsjöbaden agreement was mutual understanding and respect for the other party. Employers and employees negotiated in an atmosphere in which both parties weighed costs against benefits. It is indicative that Bertil Kugelberg on the employers' side has described the personality and character of his opposite number, August Lindberg, with great respect and admiration.
The mutual understanding was broken in the early 1970s. There were several reasons for this. One was the collapse of the Bretton Woods system for exchange rate cooperation between the industrialised countries. In the post-war period up to then this exchange rate regime had not only completely regulated the frame for Swedish stabilisation policy but also served as a clear benchmark for wage formation and mutual understanding in the labour market.
After the collapse of Bretton Woods it proved difficult to establish an exchange rate regime that was correspondingly stable. Economic policy and wage formation no longer had a clear benchmark. Attempts were made to revive the former atmosphere and sense of responsibility but without success. One of the reasons for this was the lack of an underlying consensus - cooperation for its own sake does not work, it needs a set of shared insights. This was compounded by the indistinct stabilisation policy regime. In the 1970s and '80s there was a succession of devaluation cycles, fiscal policy was expansionary, structural imbalances accumulated and wage formation functioned very poorly. Deregulation of the financial sector in the mid 1980s made the economic problems more and more conspicuous.
Growth in Sweden in the 1970s and '80s weakened in relation to other countries. Even so, the economy tended to overheat and this, together with a continued expansion of the public sector, kept unemployment temporarily - and I mean temporarily - in check.
The problem of poorly functioning wage formation and other economic problems were veiled and too many people were lulled into believing that inflation and over-expansion had rendered Sweden immune to economic laws.
The situation was not sustainable. An inflationary, speculative bubble cannot grow indefinitely; sooner or later it is bound to burst. Moreover, the scope for a further expansion of fiscally financed employment was limited. The problems could no longer be concealed. In the late 1980s the Swedish economy was in urgent need of changes.
The situation resulted in a profound crisis in the early 1990s. Output fell for three years in a row, by a total of around 5 per cent; registered unemployment rose to almost 9 per cent and total unemployment to around 13 per cent, figures that were totally alien to modern Sweden. Government finance deteriorated sharply, partly due to the massive cost of unemployment's rapid rise. The budget deficit peaked at over 12 per cent of GDP, an almost unprecedented figure in the industrialised world. In the midst of this crisis, moreover, there was a period when virtually the entire banking system threatened to collapse.
A new direction in economic policy
In the late 1980s it thus became evident that Swedish economy policy would have to be realigned. The new direction entailed structural changes and a return to the earlier stabilisation policy regime with low inflation. On account of the extensive economic problems, however, in the short run it proved difficult to make the new line of economic policy credible. Progress was dogged by hesitation and doubts among economic agents. We can recall the spectacular currency unrest in the autumn of 1992, the dramatic rise in bond rates during the spring and summer of 1994 and even the failure of the 1995 round of wage settlements.
But the Government, the Riksdag (Sweden's parliament) and the Riksbank persevered and displayed their determination by working to consolidate government finance and keeping interest rate policy relatively tight in pursuit of price stability. In time this invested the new economic policy regime with growing credibility. Since the autumn of 1995 we have experienced a marked fall in interest rates, a more stable exchange rate and signs of new thinking among parties in the labour market.
The Riksbank's inflation target is clearly supported by the political system and society in general. Together with the work for achieving sounder government finance, this represents a return to the stabilisation policy regime that had long been a Swedish tradition. Here I am thinking not only of the times that produced the Saltsjöbaden agreement and the post-war period up to the early 1970's but also of conditions in the many decades that laid the foundation for our present prosperity. As early as 1830 the Riksdag decided to base our currency on a silver standard; this was replaced by a gold standard in 1863, followed, after World War II, by the Bretton Woods system. So apart from the two world wars and the depression in the 1930s, for more than century the krona was tied to gold. The low rate of inflation in this period, 2 per cent on average, provided a stable basis for the growing prosperity.
Five parties in the Riksdag have agreed on new legislation for the Riksbank. These proposals will further enhance the credibility of the new stabilisation policy regime. The agreement mirrors a political will to resume what once proved to be a successful Swedish tradition that led to low inflation, a more efficient economy and thereby an environment for favourable growth and employment.
What is often called an autonomous, independent Riksbank does not mean that monetary policy is withdrawn from democratic control. On the contrary, the proposed legislation means that the construction of monetary policy is explicitly delegated, while its objective is decided democratically by the Riksbank's principal - Sweden's parliament. It is important to realise that such a delegation of monetary policy to an independent central bank is a way of promoting confidence that the policy aim of government is actually implemented.
New spirit of cooperation in the labour market?
A resumption of Sweden's earlier tradition in stabilisation policy raises the question of the future construction of cooperation between the labour market organisations. When the stabilisation policy regime broke down in the early 1970s the prevailing mood of cooperation was abandoned. Now that a similar stabilisation policy regime is being reinstated, to what extent will relations between labour market organisations be marked by consensus? The right of collective bargaining presumably entails more responsibility for the consequences of wage agreements. Does a shared opinion exist about the fundamental relationships that operate in an efficient market economy?
As central bank governor I am clearly not in a position to voice detailed comments on how any such mood of cooperation should be developed. But I can express the hope that the same type of respect and esteem which marked the parties' relationships after the Saltsjöbaden agreement, as well as the same spirit, can be created in which both parties weigh costs against benefits.
One of the few countries in Europe where employment has risen and unemployment fallen in the past decade is the Netherlands. An important ingredient of the successful labour market trend there has been the evolution of a consensus. In wage formation in the Netherlands, priority is clearly accorded to employment. This has also yielded results, not just in the form of high and sustained economic growth but also in a marked increase in employment and lower unemployment.
Monetary policy's contribution to better wage formation
The Riksbank will be contributing to an improvement in wage formation by working for a stable rate of inflation at the level of the inflation target, which can be described as a good European level. This renders economic factors more predictable and decision-making less uncertain.
