The third way: How different is the Swedish model?

Swedish exceptionalism stems from the country's unique experience of 19th and 20th century economic change.

Adobe Stock

Sweden was forced to abandon its position as a local great power in northern Europe as early as the eighteenth century. Although a result of necessity, this proved largely advantageous in the long run and was the basic condition for the two centuries of peace that Sweden has subsequently enjoyed. While it is difficult to draw definite conclusions, it is clear that the absence of geopolitical obligations and the burden of an empire have created unusual scope for economic growth and prosperity. Not least, Sweden has been spared the financial burdens of war. However, the renunciation of great-power status also created a dualism that has been a characteristic of the Swedish mental climate, especially during the last century. On the one hand there has been a strong feeling that it is necessary to adapt to external factors – this applies in both the political and the economic fields. On the other hand there is probably no other country that has so frequently proclaimed itself a ‘special case’, or as representing a particular ‘model’. The latter naturally applies above all in the field of labour market and welfare policy. Almost until the end of the twentieth century, many politicians were still speaking of Sweden as the ‘third way’ or quite simply as ‘the country with a difference’, even though there was no longer any justification for this. Such attitudes were especially prevalent during the period between 1960 and 1990. In the 1960’s the slogan was ‘politics is everything’. During the 1970’s the structure of the welfare state was debated as if old economic truths about the need for resources and economic growth were no longer valid. Distribution became everything and no consideration was given to the preconditions that made distribution possible.

Most political players, social commentators and partners in the labour market seemed almost without exception to agree that what applied to everyone else did not apply to the Swedes. The severe crisis of the 1990’s undoubtedly made us more humble. However, this insight was slow in coming, and there are still many who do not wish to recognise our limitations in this respect.

Before  the  late  1950’s,  however,  few  spoke  seriously  about  a  specific ‘Swedish model’. It is true that Marquis Childs’s Sweden – The Middle Way was published as early as 1936, but the concept did not come into its own until much later, during the 1960’s and 1970’s. Like so many other constructs of this kind, it has been provided with a historic past, in this case one going back to the 1930’s. However, it is unlikely that at that time many people were aware of the Saltsjöbaden Agreement of 1938, as a ‘formative historic step’ – the term frequently used by political scientists today, with more than a tinge of anachronism. Contemporaries were probably impressed more by the fact that the agreement seemed to prevent an industrial conflict at a time when Europe was threatened by extremism of different kinds. Even in hindsight, it is no simple matter to draw a line from 1938 to the type of labour market and welfare model which became the norm in Sweden twenty or thirty years later. Conditions for this were undoubtedly different. The historical construction of a ‘Swedish model’ with its roots in the 1930’s is consequently a good example of the process which Eric Hobsbawm and Terence Ranger called ‘the invention of tradition’.

Gigantic mental leaps between adaptation and self-assertion have thus been a Swedish characteristic during the twentieth century. The former approach is perhaps simpler to explain than the latter. Awareness of a small country’s vulnerability has naturally stimulated skill in adapting. This applies particularly to a country like Sweden, which has long been integrated in an external world market, and been exposed to its unbending demands. In this sense, the term ‘globalisation’ is not particularly appropriate to Sweden’s situation, at least if limited to the economic processes of the last decades. Sweden has actually been ‘globalised’ for a long time. As a proportion of GNP, Sweden has had a large export-import sector by international standards since the nineteenth century. Most of the crises and macroeconomic shocks that affected our country in the last century resulted primarily from external causes.

It is worth speculating on the origins of the Swedish tendency to self-assertion. Perhaps the explanation can be sought in the long-term trauma caused by the loss of true great power status. As compensation, Sweden would instead become a great power in the field of welfare – and when this road apparently closed during the 1990’s – a great power in the field of morals. Or perhaps the tendency is due to self-righteousness, simply because the Swedes have retained a high degree of homogeneity and because, despite globalisation and openness to the outside world, there has long been a fairly small influx of people from overseas. The renunciation of great-power ambitions has in turn, of course, been an important precondition for this single-track culture and high degree of homogeneity. On the whole, the view of Sweden as a small state greatly dependent on other countries seems the closest to reality in the long-term. This of course does not preclude Sweden’s own room for manoeuvre. However, sometimes an excess of self-assertiveness can be the worst enemy of one’s own freedom of choice. By overestimating one’s ability one can in fact undermine one’s freedom of action, for example by pursuing a policy that radically impairs one’s economic conditions. Examples of this can be found in our most recent history. The bridging policy of the 1970’s and 1980’s – to which I shall return – not only contributed to the severe crisis at the beginning of the 1990’s but also, in a more long-term perspective, hampered our prospects of pursuing a more independent economic policy. Since the early nineteenth century, the West has been through three industrial revolutions, which have also radically affected Sweden’s development. The first took place in the nineteenth century and primarily meant the introduction of the factory system and the introduction of new machines driven by water and steam. This revolution reached different sectors of industry at different rates and with differing force. A tendency in international research in recent years has been both to put the date of this revolution later – Great Britain at the beginning of the nineteenth century was still dominated by pre-industrial production forms – and to play down the force of the transition. In many places, the transition was evolutionary rather than revolutionary, and in certain countries – for example Italy and France – many industrial sectors continued to be dominated by small-scale industry, cottage industry or domestic industry. In Sweden, the full-scale breakthrough of new production methods did not come until the beginning of the twentieth century.

The second industrial revolution had its breakthrough first in the USA, at roughly the same time, i.e. around the year 1900. It did not spread to other countries until after the First World War, while its peak was even later (partly due to the disruption in economic communications brought about by the Second World War) during what came to be known as the ‘golden years’ from the end of the war to the mid-1960’s. The most important characteristic of the second industrial revolution was mass production aimed at achieving diminishing unit costs and advantages of scale. One important precondition for the breakthrough was a series of technological innovations, for example motors for cars and other purposes, but above all the increasingly widespread use of electricity, both for production and consumption purposes. The second industrial revolution was also based on the development of a mass market for relatively homogeneous consumer articles. The rapid growth in the use of cars, household machines and other so-called durables is often used to symbolise the second industrial revolution. Mass production on a large scale meant new ways of organising work, with extensive division of labour, characterised principally by the conveyor belt. With increased mechanisation, factory work became increasingly subjected to routine and, between the wars, scientific work management, sometimes referred to as Taylorism, began to have an impact.

Generally speaking, the second industrial revolution meant homogenisation of work and working conditions. In its wake, industrial production methods reached their peak. In most industrialised countries, the industrial sector achieved a proportion of GNP of around 40 per cent, as did the proportion of workforce employed in industry. The golden age of the second industrial revolution was in the 1950’s and early 1960’s – an era characterised by a unique rate of growth in most countries, not least Sweden.

Towards the end of the 1960’s, the golden age was replaced by a longterm crisis in the second industrial revolution. The third industrial revolution was around the corner. The traditional industrial sector in the USA and Western Europe shrank rapidly. The immediate origins of this industrial crisis lay in external shocks such as the oil crises of 1974 and 1978 and the end of the dollar’s role as a guarantor of stable monetary value with the abandonment of the Bretton Woods system. However, more long-term factors were also significant.

This was to do first with the globalisation of the world economy, and industrial growth in many areas outside the ‘old’ industrial core area. A massive increase in the world trade in commodities, services and capital went hand in hand with the reduction of excise duties and the deregulation of financial markets, which in turn created further conditions for increased international competition. This powerful competitive pressure, both in industrial production and the provision of services was something that affected everyone: it strictly limited the political room for manoeuvre of each nation state. Transnational organisations such as the EU also played a role here in regulating this fast-growing international market economy, at the same time as they further speeded up internationalisation by homogenising institutional barriers and moving towards the introduction of a common currency.

These changes derived secondly from a revolutionary innovation: information and communication technology (ICT). In the 1990’s this new technology led to a dramatic increase in productivity and the emergence of new services and products that have revolutionised our lives. Old industry was replaced by new production forms and an increasing proportion of the economy was based on services. The consequences for people’s working lives have been far-reaching. Increased demand for skills, the phasing-out of simple industrial jobs and a larger ‘service sector’, in combination with a new business philosophy which entailed among other things ‘outsourcing’, have brought further demands for flexibility in the labour market and in working life. Higher demands are being made for the adaptation of the workforce and the updating of skills. Today’s trend toward shorter periods of employment and more project-oriented work for the more qualified employee, together with short term supply employment for the less qualified workforce and self-employment of various types, will probably continue to increase. A further effect of the reduced demand for ‘old’ industrial jobs and of increased flexibility is an increased salary differentiation in most old industrialised countries. As regards Sweden, data going back to the mid-1980’s indicate that the previous strong tendency to even out the pay structure has been reversed. Increased demands for individual skills, a high degree of flexibility, social skills and so on are probably among the main reasons for this.

Finally, the third industrial revolution was sustained by changed preferences among consumers. Increasingly fastidious consumers in western industrial countries made demands for more customised consumption. The conformity and uniformity of the mass-consumption society were no longer acceptable. It is not easy to understand the heavy emphasis of modern management teaching on ‘lean production’, ‘just-in-time’, customer control and so on, unless one takes into account this fundamental shift in preferences. This applies also to the consumption of services. Here too we detest uniformity. Much of the criticism in recent years of, for example, the school sector or, perhaps to an even greater degree, the care sector in a country like Sweden with an almost total public monopoly is ultimately based on the fact that we are no longer prepared to stand cap in hand and accept what is offered by experts and politicians. We increasingly demand freedom of choice and personal attention.

Having reviewed the general pattern, let me return to Sweden and take a closer look at how the social structure which we have often perceived as a specifically Swedish phenomenon – with a ‘Swedish model’ – in fact has been strongly influenced and formed by the pattern which we have just discussed. Without doubt, the main characteristics of this development can be found in Sweden too. At the same time, specific characteristics can be identified which have been significant for the tempo and dynamics of the Swedish development process.

Although, with good reason, more importance has been attributed to the role of the home market in the breakthrough of Swedish industrialism at the end of the nineteenth century in recent years’ research, one cannot deny the extremely strong foreign contribution to Swedish development. Sweden’s industrial breakthrough took place in leaps and bounds during the latter half of the nineteenth century, with marked peaks during the 1850’s and 1870’s. Nevertheless, it is difficult to speak of a massive transition to industrialism earlier than the end of the 1890’s and the time around the turn of the twentieth century. Growth was extremely rapid at this time, and the size of the industrial sector was increasing fast. A number of ‘innovation’ industries in the engineering sector, which today have wholly or partly ceased to be Swedish companies – Ericsson, AGA, Alfa Laval, SKF and so on – were founded around this time and began export operations. Before the 1890’s, only a minority of people had in fact been involved in what can be called modern industry. Most made a living in cottage industries and handicraft in rural areas. But then everything speeded up. From 1900 to 1920, the proportion of people employed in agriculture (including related employment) diminished from 55 per cent to 44 per cent. Over the same period, the proportion employed in industry increased from 28 per cent to 35 per cent. This was a revolution whose effects were felt throughout society.

So by the time Sweden began to be industrialised in earnest, the process had reached a degree of maturity elsewhere. Of course, this had consequences, not least the fact that Sweden could take advantage of the demand generated by a number of fast-growing economies in the wider world. There were also certain advantages in beginning large-scale industrialisation somewhat later. It was possible to use the latest technology without being hampered by earlier investments.

After the First World War, Sweden was hit hard by the international post-war depression. The radical restructuring that followed in the 1920’s and ’30’s was also characterised by the transition from the first to the second industrial revolution. Increased liberalisation of trade in the 1920’s accelerated this restructuring, which was characterised by keen cost competition. As a consequence, industry was rapidly mechanised, there was a transition to electrically powered production, strengthening of corporate enterprise, and the phasing-out of old-fashioned and small-scale capacity. In particular the Swedish iron and steel industry, as well as the engineering industry, felt the cold wind of change.

In the 1930’s, this tendency continued, at the same time as the home market grew. Despite the financial crisis of 1931 and the Kreuger crash in 1932, Sweden was not so hard hit by the Depression that generally followed the Wall Street crash in 1929. By 1934, the wheels were turning at full speed again. There were several reasons for this successful turn-around. The theory that the main reason was the new Keynesian crisis policy adopted by finance minister Wigforss and prime minister Per Albin Hansson must unfortunately be discounted. It was too half-hearted and moderately expansionist to achieve this. Instead, the main reason was probably the combined effect of an undervalued krona, which was linked to the pound sterling after the gold standard was abandoned in 1931, and the considerably improved competitive position of Sweden’s export industries. Extensive restructuring and the influx of people from rural districts created productivity gains, at the same time as Sweden had a favourable age pyramid, with many young people coming on to the labour market, enabling an increase in the ‘productive’ proportion of households. Agriculture had been put under considerable pressure by liberalisation and reductions in trade tariffs in the 1920’s. This forced rationalisation, mechanisation and mergers in agriculture, which led to further workforce reductions.

As regards Sweden, the second industrial revolution did not reach its peak until after the war. The groundwork had of course been created through the intensive restructuring between the wars, together with the very high level of capital investments. This led to a considerable buildup of industrial capacity, which began to be utilised in earnest after the Second World War. The industrial sector therefore grew rapidly. Industrial facilities were built and mechanisation occurred at a rapid rate. Small and medium-sized companies expanded especially fast. They absorbed their labour force from an agricultural sector that could no longer support everyone. New industry provided jobs and, in addition, much higher wages than agriculture could offer.

But there was another factor. Sweden could move into the post-war era with its production facilities intact. After the devastation of the war, the rest of the world was in enormous need of raw materials and engineering products. This paved the way for exceptional economic growth in the 1950’s and 1960’s. Growth was higher than ever before. Productivity in industry increased rapidly at the same time, indicating that growth was accompanied by industrial transition and rationalisation. In the 1960’s, industry employed as much as 45 per cent of the total labour force. But problems followed. Growth stagnated in the late 1960’s, and the proportion of industrial employees began to drop relatively. Similarly, traditional industry represented a constantly diminishing share of the GNP.

In the 1970’s, Sweden, through a bridging policy, could still ward off the worst effects of the industrial restructuring crisis which in Western Europe and the USA had a devastating effect on the “old” basic industries. The main aim of the bridging policy was to maintain employment levels, and the method used was firstly a series of devaluations, from the mid-1970’s until 1982. Competitiveness could be maintained through a falling exchange rate for the krona. As an effect of the general structural crisis, growth dropped to half its rate in the 1950’s and 1960’s, while at the same time there was a downturn in productivity. One effect of the down-ward adjustment of the exchange rates was that growth became even slower. The repeated devaluations quite simply meant there was too little pressure to restructure in industry, which slowed necessary productivity improvements. At the same time, the devaluations threatened to create greater imbalance in income levels, which in turn led to vociferous demands for compensation from different groups in the community. Such compensation claims in turn fanned the flames of inflation. This ‘crisis policy’, which was launched in 1982 as the ‘third way policy’ by the social democratic government with finance minister Kjell-Olof Feldt at its helm, and which was based on a major devaluation in the same year, only served to reinforce the process of low growth and high inflation. Towards the end of the 1980’s, Sweden was in the unenviable position of having stagnating growth, rampant inflation, a highly overvalued property and finance market characterised by overgenerous loans from the banking sector (something encouraged by a high marginal tax rate, which led to negative real interest rates), and an extreme shortage of labour in many areas of the economy. Against this background, it was not surprising that the bottom fell out of the economy at the beginning of the 1990’s, with a subsequent financial crisis, unemployment, and an enormous deficit in state finances.

The Swedish bridging policy was an ambitious attempt to avoid the dramatic increase in unemployment that affected large parts of the Western world during the same period. The unemployment crisis was in turn a consequence of the decline of traditional industry in the advanced industrialised countries, and the transition to the third industrial revolution. Taking this into consideration, this policy must be seen as an example of the type of overconfidence mentioned earlier. The prospects of pursuing an economic policy radically different from that in the world around – although few wished to acknowledge it – were minute.

Despite falling growth and considerable structural problems in industry, we still afforded ourselves the luxury, alongside ambitious employment goals, of increasing both private and public consumption. Under the yoke of rising inflation, the partners in the labour market in the 1970’s and 1980’s signed a series of record-breaking pay agreements. At the same time, wage drift was extremely high, and caused envy in those who were excluded, especially public-sector employees. Unions in negotiation with their counterparts succeeded in pushing through different kinds of pay harmonisation clauses, which in turn increased nominal pay levels. But the results were dismal. Due to a high rate of inflation, the pay agreements were largely meaningless. As early as the beginning of the 1990’s Stig Malm, the then leader of LO, the Swedish trade union federation, stated despondently that real pay levels in practice had remained unchanged since the early 1970’s. The only result of the heroic efforts had been total imbalance and an economy in decline. The historian Barbara Tuchman has used the term ‘the march of folly’ to describe the exhibitions of qualified foolishness and overconfidence that seem constantly to afflict humanity. One wonders whether Sweden in the 1970’s and 1980’s could serve as a further example of such folly.

The much-discussed ‘Swedish model’ has undoubtedly several special characteristics. On the one hand, it is deeply rooted in the second industrial revolution. A correct rendering is perhaps that this ‘model’ was not at all unique to Sweden, but that in Sweden it came to take on an unusually pure form. As we have already indicated, the term ‘Swedish model’ was often used in disparate ways. The foundations were laid perhaps during the 1930’s, through the Saltsjöbaden Agreement, but the model does not appear in its finished form until the 1950’s. One of its cornerstones was the organised cooperation between partners in the labour market, which came to influence not only labour market policy, but also social policy and economic policy in general. During this time, labour conflicts were rare. The Swedish model was characterised not only by far-reaching corporatism, but also by an unusually high degree of centralisation of central union and employers’ organisations. This centralism was perhaps most noticeable in centralised pay negotiations between the parties – which were the norm throughout the period from the mid-1950’s to the mid-1980’s. However, market forces played an important role too, as did the fact that there was full employment. Such a situation almost always creates a tendency to reduce pay differences.

Without doubt, the industrial concept of the second industrial revolution and the dominance of homogenous industrial work, were among the structural conditions that made the ‘Swedish model’ possible. The Saltsjöbaden policy that began with the signing of the general agreement in 1938 must be seen above all as a way of handling problems intimately associated with the industrial society created by the second industrial revolution. The general agreement ushered in a form of negotiation within which issues that concerned collective bargaining could be dealt with. These included the right to strike and other issues which, since the turn of the twentieth century, had been matters of contention between LO on the one hand, and SAF (the employers’ federation), on the other. Saltsjöbaden was followed by a number of further agreements. For example, there was the health and safety agreement in 1941, an apprentice agreement in 1944, a corporate council agreement in 1946 and the last of the ‘Salt sjöbaden Agreements’, the 1948 work study agreement. Through the central labour market committee, with representatives from SAF and LO, a body was created to determine whether or not a conflict was a risk to society.

Other countries came to introduce similar forms of conflict resolution, but perhaps not as consistently. The biggest difference is perhaps that they abandoned these forms much earlier than the Swedes: Great Britain in the 1970’s, and the Netherlands in the 1980’s. The Swedish model is thus comparable with the Nordic models, while there are also many similarities with countries such as the Netherlands, Great Britain and West Germany. There too, relations between the parties are characterised by organised cooperation. There too, one can see a tendency towards corporatism. There too, negotiations and relations between the parties were highly centralised. In these countries too, a homogenous industrial work sector created the very conditions for the high degree of centralisation. Finally, pay differentials diminished there too, as a consequence of a centralised pay policy and full employment.

How intimately the second industrial revolution was a precondition for many of the characteristics we associate with the ‘Swedish model’ is apparent when we look back at the events of the 1990’s. For a long time, as we have seen, relations between the partners, especially in Sweden, but also in the majority of other mature industrialised countries, were characterised by far-reaching centralisation. In recent decades, this model has been abandoned. In many areas, this has meant a dramatic weakening of the central partners’ position. In Sweden, unlike many other countries, it has not led to a reduction in union membership or fewer collective agreements. But in its old sense, the Swedish model or Saltsjöbaden policy ceased to function through the persistent refusal since 1992 of SAF and the employers to act as negotiating parties in a more organised way. There has also been a shift in power among the trade unions from central level to association level. New technology, new patterns of organisation and an increasingly service-based and flexible work sector in many negotiating areas also create pressure to strengthen the local union level. There is general discussion among trade unions of the transition to a more client-based work method.

Sweden is a small country, whose economic fate and ventures are governed largely by external forces. The macroeconomic shocks that hit us in the 1920’s, 1930’s and 1990’s were principally external in origin. Many people certainly realised the vulnerability of Sweden’s position. However, particularly in the wake of the ‘golden years’ in the 1950’s and 1960’s, when growth was very high, the foundations were laid among politicians, parties and the public for unbounded over-confidence. This came to characterise above all the 1970’s and 1980’s. Unfortunately, there were few who paid attention to the fact that the economic basis for such self-esteem was almost non-existent. Through the bridging policy, we tried to escape the effects of restructuring and a third industrial revolution, but the result was instead that the crisis in the 1990’s was considerably worsened. One might think that this has cured us permanently of the ailment of over-confidence. No such luck. We can be sure of this at least: memories are short, and our ability to learn from history is strictly limited.

This essay originally appeared under the title Adaptation or Over-confidence? in The Swedish Success Story: Perspectives from the Engelsberg Seminar, Axess, 1999.


Lars Magnusson