Unnecessary radicalism: the limits of economic ideas in political thinking

ABSTRACT

This article analyses late 1960s’ and early 1970s’ policy debate on issues concerning balance of payments in Sweden. Part of this debate was the question of fiscal austerity as a tool to achieve external balance, and if it could be used without risking economic and social unrest. The aim is twofold: first to empirically shine new light on modern Swedish economic policy in a historic context. Second to theoretically explore new ways of interpreting the relationship between political thinking and economic ideas. Special focus within the second aim are the consequences of political thinking on Keynesian economic ideas as a framework of economic understanding at the time. The study is qualitative in its methods and pays attention to limits within the relationship between economic policymaking and economic expertise. The article highlights conflicting perspectives on Keynesian ideas and the heterogeneity of these perspectives among economic experts. A heterogeneity of this kind is also shown to complicate the assumed close relationship between Social Democracy and Keynesianism in a historic context. In essence, the article shows that studying policy debates in close historic detail makes for new conclusions on the development of modern economic ideas and the part political thinking plays in it.

KEYWORDS:

• History of economic thought
• economic policymaking
• balance of payments
• Keynesianism
• social democracy

JEL CODES:

• B2
• B22
• B27

1. Introduction

Austerity policies have been a main feature of the Western political economy between the financial crisis and the Covid crisis and were widely debated by scholars of economic history and economics (Alesina et al., 2019; Blyth, 2015; Eichengreen, 2015; Tooze, 2019).

Austerity has by some scholars been traced as a cause of neoliberal ideas (Blyth, 2015; Mudge, 2018). But neoliberalism has not always influenced every type of austerity policy. Austerity was a feature of economic policymaking in the era of the gold standard (Eichengreen, 2015), it returned during the Bretton Woods era to fight off inflation (Bordo & James, 2019; Helleiner, 1994), and it has been widely used post-Bretton Woods as a policy regime (Haffert, 2019). Austerity can be thought of as a persistent political idea with a shifting theoretical base.

John Maynard Keynes famously said that:

The ideas of economists and political philosophers, both when they are right and when they are wrong are more powerful than is commonly understood. Indeed, the world is ruled by little else. Practical men, who believe themselves to be quite exempt from any intellectual influences, are usually slaves of some defunct economist. (Keynes, 2018).

This statement can be interpreted as an illustration to that persistency. An economic idea can be this persistent because it relies on a political framework that puts limits on the influence of academic theories which do not fit with its political goals.

However, as the case of austerity shows, there is not always a particular ‘defunct economist’ to be found in political thinking. It is not a novelty, but a point that has been made before by the likes of Thorstein Veblen, who has described the use of economic ideas in politics as a ‘habit of thought’. The interaction between the limits that these habits create and economic ideas, rather than the ideas themselves, are more rarely studied in economics and economic history. In this article the political logic of these limits is the main area of study. For example, the persistence of politicians to revert to fiscal austerity in a time of crisis even when economic ideas tell them otherwise is a case of a habit of thought. But how does this habit of thought present itself in relation to other economic ideas when economic policy is debated? That question will be answered by archival evidence from Sweden, showing that even during the height of what can be referred to as the Keynesian era (Blyth, 2002; Skidelsky, 2005), fiscal austerity was a tool that when in doubt seems to have provided some certainty to policymakers. Expansion on the other hand can be seen as the exception to the rule.

The cause of the Swedish uncertainty was perceived troubles with the balance of payments in the late 1960s (Magnusson, 2011; Stahl, 2021). Relatively harsh measures to combat them released an infected debate on fiscal austerity among economic experts. Public support for the goal of full employment also put the political debate on its head when the bourgeois opposition to the Social Democratic government took a stance to the left of said government's economic policy. This political triangulation during the first years of the 1970s resulted in a battle over the heritage of Keynesian ideas in Sweden (Lindvall, 2004; Löwdin, 1998).

At the same time, important trade partners, such as West Germany and the United Kingdom, had embarked upon austere policy programmes (Mudge, 2018; Schenk, 2010; Schmidt & Thatcher, 2013). Pressure on Sweden to choose a path in policymaking was mounting, which at first lead to austerity during the years 1971–1972, then to expansion in 1974–1975. Between 1976 and 1982 four bourgeois governments launched several devaluations and packages of state support to failing industries. Finally, economic policy returned to austerity again from 1980 (and even more so after 1982 when the Social Democrats returned to political office) (Larsson & Söderberg, 2017; Löwdin, 1998).

All the while, balance of payments was at the heart of the political struggles. That is why debate on this issue is chosen as the object for this study. It included economic experts as well as long-term Swedish finance minister Gunnar Sträng. Sträng was a Social Democrat and held the position as head of the Treasury during the years 1955–1976. He contradicts the automatic link between Social Democracy and Keynesianism that is sometimes used to describe post-war economic policymaking in Sweden; he was not a Keynesian at heart and usually thought restrictive policies were fundamental to economic stability (Johansson, 2018).

The period from the late 1960s and throughout the 1970s was a moment of economic turbulence for the global economy, relating to the fall of the Bretton Woods agreement and the first OPEC crisis (Helleiner, 1994; Schön, 2007; Tooze, 2019). As such it proved a challenging time for economic policymaking everywhere, and made technical quick-fixes attractive to policymakers (Best, 2004).

The aim of this article is twofold: first, it is empirical and seeks to present archival evidence of the Swedish debate on balance of payments as part of the Bretton Woods turmoil and a battle on the use of fiscal austerity. Second, it is theoretical and explores new interpretations of how political thinking limits the use of economic ideas. Special focus within the second part is how political thinking affected Keynesian economic ideas in the late 1960s and early 1970s. Keynesian economic ideas are understood as both a dominant theoretical framework in economics and an economic philosophy against financial conservatism (Carter, 2020; Coats, 2005; Robinson, 2015). From the Swedish example, we can learn that Keynesian economic ideas became a tool in a political struggle over the limits of economic policymaking.

This leads to the question: how can economic ideas be understood as limited by political thinking in a historic context of economic turmoil?

1.1. Example of political thinking framing changing economic ideas

An example to illustrate the question is found in a short international outlook. In 1975, the Organization for Economic Co-operation and Development (OECD) assigned a group of experts to present recommendations for economic policymaking to its member states. A report was presented in 1977 and became known as the McCracken report (McCracken & OECD, 1977) due to its editor Paul McCracken, who was an American economist and economic advisor to President Nixon between 1969 and 1971.

As political scientist Vincent Gayon (Gayon, 2017) shows in his study of the McCracken group, the final report was a compromise between differentiating sets of economic ideas. These were as far divided as Keynesianism on the one hand and the Chicago school of market liberalism on the other. A compromise had to be reached that would be politically acceptable in the OECD member states or else the report would have been useless. Experts could push their own agenda, but only as far as political thinking allowed. In this case that entailed a fight on inflation without large increases in unemployment (Gayon, 2017).

Reading (Gayon, 2017) we learn that the conventional history of Keynesian demand management being replaced by neoliberal supply-side theories of deregulated markets, as for example (Schmelzer, 2012) describes, is a simplified version of the development of economic ideas. It obscures the role of politics, and the mutual dependence political thinking and economic ideas share with one another. Economic ideas must continually be adjusted to political priorities as well as the other way around. The separation of economics and politics, as Joan Robinson has described (Robinson, 2015), is perhaps the greatest mistake mainstream economics has made to the understanding of the role of economic ideas in society. There are always ideologies and values to be considered in both political and economic frameworks (which has also been elaborated on by Gunnar Myrdal (Appelqvist, 2000)).

1.2. The finance minister and his priorities

Gunnar Sträng embodied Swedish Social Democratic economic policymaking for 21 years, and was one of the most popular politicians in Sweden when he resigned due to the Social Democrats losing the 1976 election after 44 consecutive years in office (Johansson, 2018). Consequently, during his long rule of the Treasury there was not much scope for drastic changes in the use of economic ideas.

As a Social Democrat Sträng had his political upbringing in the trade union branch of the Swedish Labour Movement during the 1920s and 1930s (Åsbrink, 2019; Johansson, 2018). Sträng was neither a scholar nor an ideologue. Such features had been the characteristics of his predecessor Ernst Wigforss, who became one of the most important thinkers of Swedish Social Democracy and was a figure-head for the ‘New Economic Policy’ as the Swedish version of the Keynesian breakthrough was known (Wigforss, 1938, 1960). According to Sträng, Wigforss always preferred fiscal policy over monetary policy, and saw the state as responsible for the aggregate level of investment (Werin, 1993), which rhymes well with Keynesian thinking (Skidelsky, 2005). In this respect, the Wigforssian ideas can be seen as the theoretical core of economic policymaking by the Social Democratic Party. A core that lasted until the early 1970s when it began to wither due to international economic turbulence (Helleiner, 1994), and new institutional norms in Swedish government of individual utility maximisation (Ryner, 2004). Changes which eventually led to a more market-based approach to public investment (Svensson, 2002).

Sträng on the other hand adhered to the logic of the trade unions and the institutional constraints of the Bretton Woods system. According to biographer Anders L. Johansson, Sträng was firmly rooted in the labour market policies of the Swedish Confederation of Trade Unions (Landsorganisationen) (Johansson, 2018). Its research department had a strong position of power vis-à-vis the Treasury during the 1950s and 1960s (Pontusson, 1992).

Wigforss’ interpretation of an active government and the possibility of using fiscal policy to counteract instability in markets (Wigforss, 1938, 1960) could be known as progressive Keynesianism and was closer to the original Keynesian theories (Skidelsky, 2005). Secondly, there was the Neoclassical synthesis, which became a go-to theory for many Swedish macroeconomists. Its approach to government and markets were of a more technical nature, and it would therefore not be amiss to call it technical Keynesianism (Best, 2004; Giraud, 2014). Finally, Sträng's own branch of policy can be known as a labour-market based regime, or, with Peter A. Hall's words, a type of ‘activist manpower policy’ (Hall, 2022).

The Confederation of Trade Unions’ policies had the aim of maintaining centralised wage bargaining as a mean of achieving both full employment and price stability. Strict discipline was required of the individual trade union to contribute to the larger aims of common prosperity of the working class, as well as the larger society by keeping the price of Swedish exports competitive. This was achieved by holding wages for the same type of job equal no matter the individual firm's abilities to pay. Profits would be kept in check and less productive firms pushed out of markets (Pontusson, 1992). The policy became known as the ‘solidaristic wage policy’ (Schön, 2014).

In the late 1960s this order began to crack as internal dissatisfaction with working conditions and wage levels led to strikes first among miners, and later among public employees (Johansson & Magnusson, 2020). Some economic experts and individual trade unions were worried that long-run aggregate investment would shrink if continual priority was not given to public investment by economic policymakers (Pontusson, 1992). Failure to prioritise public investment could increase unemployment to unacceptable levels (Pontusson, 1992; Schön, 2014). However, the radical wave was not to Gunnar Sträng's liking as it risked undermining the effectiveness of the labour market policies (Andersson, 2003; Johansson, 2018).

The other leg was the Treasury's wish to keep Sweden as a prosperous member of the Bretton Woods agreement (Ahlström & Carlson, 2006). Sweden was a loyal partner to the International Monetary Fund and kept its economic policy in line with the goal of not devaluating the currency (Ahlström & Carlson, 2010). For Sträng and his governor of the central bank (Riksbanken), Per Åsbrink, it was a matter of principle and national pride not to devaluate the Swedish krona (Larsson & Söderberg, 2017). This meant that between 1949 and 1976 the krona was not devalued once (Werin, 1993). It also meant that fiscal policy, and occasionally also monetary policy, had to be kept strict as not to bring the exchange rate out of order. Besides promoting free trade, the fixed exchange rate was thought to help trade unions keep wage demands in check. It also meant that balance of payments became a measurement of the strength of the overall economy (Johansson, 2018; Werin, 1993).

The role of government in the Swedish corporatist system was mainly to create economic policies that fit the centralised bargaining model, while still trusting that the labour market organisations adhered to the main policy goals (Johansson & Magnusson, 2020; Swenson, 2002). In the late 1960s, these goals were price stability (and a stable exchange rate), full employment, and a high level of capacity utility and growth (Johansson & Magnusson, 2020; Schön, 2014).

1.3. The experts and their affiliations

To form a bridge from Gunnar Sträng to the economic experts who appear in this article it falls naturally to begin with the trade union economists. In the 1950s the two most prominent of their kind, Gösta Rehn and Rudolf Meidner, had concocted a model of supply-side, productivity inducing policies, which became known as the Rehn-Meidner model (Johansson & Magnusson, 2020). As political scientist Jonas Pontusson has described, the model was a middle way between Keynesianism and supply-side oriented policies. It relied heavily on the success of large export-oriented firms and paid little attention to domestic demand for goods from domestic firms. This increasingly led to a conflict between export-led growth and internal demand (Pontusson, 1992). In this article, the trade union economists are represented by Clas-Erik Odhner and Gösta Dahlström, who both worked for the Swedish Confederations of Trade Unions.

The Rehn-Meidner model was at times controversial to academic economists. Lennart Erixon has shown how prominent Keynesian economist Erik Lundberg continuously criticised it for being too heavily relied on squeezed private profits and restrictive tax policies to work. Lundberg being influenced by Friedrich von Hayek's Road to Serfdom, and a participant in the Swedish debate on planning in the late 1940s and early 1950s, saw dangers with the Social Democratic model of increased state influence. But eventually in the late 1960s, he came around to see the merits of the model and its contribution to increased productivity levels of the Swedish economy (Erixon, 2021).

Lundberg was of the second generation of Keynesian economists in Sweden. He was a leading figure in debates on economic policy, and served as chair for the Nobel Prize Committee responsible for the Nobel Prize in economics during the 1970s (Erixon, 2021). Before him there had been the generation of economists like Gunnar Myrdal and Bertil Ohlin, who made up the original Stockholm School and even predated Keynes’ General Theory in proposing deficit spending and state interventions to elevate unemployment and improve aggregate investment. It was a group with progressive views and played a significant role in shaping economic policymaking in Sweden from the 1930s and forwards (Erixon, 2021; Jonung, 2006). As economist Bo Sandelin has put it: economics in Sweden, as a small country, was policy-intensive, but it also made economic policymaking economist-intensive (Coats, 2005, pp. 49–50). Electoral success for the Social Democratic Party in Sweden, and the Liberal Party (Folkpartiet) as the leading opposition meant that these economists were given a political base to use for their ideas (Berggren, 2010; Blyth, 2002; Mudge, 2018).

They were also in many ways corner stones in the build-up of the Scandinavian welfare state during the post-war years. The model saw the transformation of academic economists from public intellectuals to political actors. They became planners and experts essential to the work of government, private business and trade unions (Lundin & Stenlås, 2015; Mudge, 2018; Wikman, 2019).

The late 1930s and 1940s also saw the rise of privately funded research institutes such as the Industrial Institute for Social and Economic Research (Industriens utredningsinstitut), as well as private think-tanks such as the Centre for Business and Policy Studies (Studieförbundet Näringsliv och Samhälle) (Henrekson, 2009; Ullenhag, 1998).

A founding member for both institutions was Schumpeterian economist Erik Dahmén (Erixon, 2021), who kept a close link to the highest ranks of Swedish private business during his entire career. Another economist who kept to the side of private business within the Industrial Institute was Gunnar Eliasson (Eliasson & Lindegren, 2020). The mentioning of two ‘sides’ comes from the fact that the Industrial Institute also produced economists like Lars Wohlin and Carl Johan Åberg, who worked as secretaries of state to the Treasury. Wohlin to the centre-right government from 1976 to 1979 and Åberg to Social Democratic governments from 1967 to 1976 (Henrekson, 2009; Sverenius, 2018; Ullenhag, 1998).

The note about Wohlin and Åberg is of special importance, as the Industrial Institute in fact had a long-standing co-operative link to the Swedish Treasury. Its experts made forecasts to the important special branch of official government reports called the long-term planning reports (Långtidsutredningar). These forecasts and reports in turn created a foundation for policy recommendations made by the Treasury's own group of experts within the Economic Planning Committee (Ekonomiska planeringsrådet). The cooperation lasted until the middle of the 1970s when the Treasury and the Industrial Institute parted ways. Government ministries increasingly recruited their own experts and the Industrial Institute wanted to keep a scientific profile rather than be involved in political communication (Coats, 2005; Henrekson, 2009).

Aside from the business-affiliated institutions, there were also publicly funded research institutes, such as the Institute for International Economics (IIES) and The National Institute of Economic Research (Konjunkturinstitutet). The latter had Erik Lundberg as its first chair from 1937 to 1955 (Erixon, 2021). IIES on the other hand was founded in 1962 by Gunnar Myrdal (Lindbeck, 2014). It produced such economists as Assar Lindbeck (who also participated as an expert in the McCracken group, see (Gayon, 2017)), and Sven Grassman who became a controversial voice against austerity in Sweden (Sverenius, 2000).

On the private side, the Stockholm School of Economics was home to Erik Dahmén, as well as economist Ragnar Bentzel, who in the 1960s joined the Industrial Institute and later Uppsala University (Ysander & Bentzel, 1993). They also briefly had the company of Lars Matthiessen, a Danish economist who served in the Danish Treasury's Economic Council (Matthiessen, 2011; Stahl, 2021).

Finally, one economist held positions with all the above-mentioned institutions except the trade unions. His name was Göran Ohlin, nephew to the previously mentioned Bertil Ohlin. In the 1980s he even became Deputy General Secretary to the United Nations (Coats, 2005).

1.4. The common link: the Swedish Economics Association

The common connection between finance minister Gunnar Sträng and the economic expertise in this article is the Swedish Economics Association (SEA) (Nationalekonomiska föreningen). The SEA was founded in 1877, not as an academic association, but a forum for politicians and civil servants with mainly liberal opinions to discuss economics-related topics. It grew in status during the twentieth century as did the discipline of economics (Coats, 2005).¹ During this time the membership increasingly shifted from politicians to academic experts. From Ernst Wigforss and forwards, the SEA became an important stage for the incumbent finance minister to give an annual budget presentation to the members of the association.

The unique feature of these debates was their close interaction between politics and economic ideas. They were shaped like an academic seminar where the finance minister presented the budget and was afterwards treated to an opposition from the gathered economic experts. Since the meetings were for members only, it meant that they could express their opinions rather freely to each other. Such characteristics make the SEA interesting to study more closely for details on directions of economic ideas. The minutes from the meetings were published with some delay, and from 1973 they could be found in the association's newly installed journal Economic Debate (Ekonomisk Debatt) for subscribers only.² Nowadays, the material is available on the association's website.

Like (Gayon, 2017), I am treating the SEA as the stage where different sets of economic ideas are confronted by political thinking. As seen in the presentation of the experts there was a spread of affiliations, which could also affect opinions on economic matters.

1.5. Article outline

The rest of the article is outlined as follows: in section two I will describe the methodological and theoretical contributions I intend to make. Section three gives a brief discussion of how previous literature has dealt with political change and political thinking during the 1960s and 1970s in relation to economic expertise. Section four contains the empirical study and section five has conclusions drawn from it. Finally, after the references in section six an Appendix with a list of all the actors and their affiliations can be found.

2. Method and theoretical contributions

The investigation in this article is focused on the policy debate on balance of payments between 1967 and 1976. As mentioned in the introduction, a stable price level, and thereby a stable exchange rate, was a political priority during this period. At the same time, it was a priority that became increasingly contested as the Bretton Woods system crumbled and labour market consensus was put under pressure (Helleiner, 1994; Magnusson, 2011). It was at this point, in the late 1960s, that the question of fiscal austerity, and its compatibility with Keynesian ideas, seems to have taken off as a debated topic among Swedish economic experts and politicians (Johansson, 2018; Löwdin, 1998).

Already, during the recession of the Korean war in the late 1950s, Gunnar Sträng had refused to stimulate general demand against slowing economic growth, due to what he saw as excessive inflation (Johansson, 2018; Löwdin, 1998). It was the 1973 OPEC crisis which made an exception to the rule. Sträng was persuaded by younger party colleagues, such as prime minister Olof Palme, to consider counter measures to falling demand because of the oil price shock, and he launched the so called ‘bridging policy’ (överbryggningspolitiken) to ensure the goal of full employment could be maintained (Johansson, 2018). That move seems to have afterwards become one of the few things remembered by several Swedish economists, who later deemed economic policymaking in the 1970s as a political failure (Henrekson, 1996; Jonung, 1999; Lindbeck, 1998; Ståhl et al., 1993).

The arguments studied in the article were used in the exchanges between finance minister Gunnar Sträng and Swedish economic experts at the budget presentations hosted by the SEA. The method for studying these exchanges is primarily to use history of economic thought to put policymaking in its historic context, as for example economic historian Benny Carlson (Carlson, 2018) did with Swedish economists and the debate on economic planning in the 1930s. Rather than using historical examples to illustrate a theoretical framework, I will present an interpretation of how history of economic thought can be used to widen the study economic policymaking in a historic context.

2.1. Methodological discussion

A few examples of how studies of the relationship between economic ideas and political thinking have been made by economic historians in the Swedish context come primarily from studies in history of economic thought.

First, the study of economic thought can in itself be a source of historicising political thinking. Economic historian Lars Magnusson's study of mercantilism and its school of thought concerns economic ideas as well as political thinking. They are intertwined in such a way that they are difficult to separate, and especially in so early a time when economics was still not a developed discipline. By reading not just academic works, but also more popular circulars such as pamphlets, the development of the political and economic language of mercantilism is recreated (Magnusson, 1994).

Second, if we move to more modern times there are studies such as the already mentioned Benny Carlson's overview of Swedish economists in the debate on economic planning, which began in the 1930s. Carlson lets the economists’ arguments in newspaper articles make the case of how economic ideas and political thinking were separated between an older generation of laissez-faire economists and the younger generation in the Stockholm School (Carlson, 2018). Carlson has also studied Sweden's relationship to the Bretton Woods agreement and the International Monetary Fund together with economic historian Göran Ahlström. These studies also offer a method of studying political thinking with the help of history of economic thought as they use a variety of empirical material such as official documents, and economists’ arguments in books and newspapers (Ahlström & Carlson, 2006, 2010). This makes the development of political thinking and economic ideas even in practical arrangements such as a partnership to the IMF into a broader analysis.

Finally, there are studies that have a more straightforward focus on the development of economics in Sweden and its effects on political thinking. Here we find economist Bo Sandelin as well as economic historians Avner Offer and Gabriel Söderberg. There is also economist Lars Jonung, who has mostly written the early twentieth-century history of Swedish economic thought (Jonung, 2006), and business economist Lars Engwall, who has amongst other things studied the structures of economics in Sweden (Engwall, 1992).

Sandelin has studied both the discipline of economics in Sweden and history of economic thought. Especially the first-mentioned inevitably means some focus on policy change, as Swedish economists have been involved in policymaking during at least a century (Coats, 2005; Sandelin, 2011). Offer and Söderberg have together written the history of the Swedish Riksbank's Prize in Economic Sciences to the Memory of Alfred Nobel, more commonly known as the Nobel Prize in Economics (Offer & Söderberg, 2016). This study is just not a description of the prize and its recipients, but also studies change in political thinking of the Swedish Social Democrats. This change is interpreted as a consequence of the initial thoughts, by economists like Assar Lindbeck and Erik Lundberg, of why the prize was needed; to influence the Social Democrats away from a radical left-turn, towards market liberalism (Offer & Söderberg, 2016).

Political thinking can be studied through the lenses of history of economic thought. By concentrating on areas where politics and economics are in crossroads with each other, there is much to be gained in knowledge of their relationship during history. In this article, this method will be applied in the use of protocols from the SEA. Since economics has long been a part of economic policymaking in Sweden, and politics has been crucial for economists to make themselves into important actors, the Swedish context is fruitful ground for the use history of economic thought as means to study political change.

2.2. Theoretical contribution

Method-wise, history of economic thought has been covered by the previous section. But that still leaves the question of how to define political thinking. Economic ideas are a more straightforward concept as they can usually be put into different schools of thought: Keynesianism, the Neoclassical synthesis, the Austrian School, the Chicago School, Monetarism, and so on.

Political thinking in the other hand should in this article not be confused with ideology. Ideology is a part of political thinking, but the latter should be seen as a broader concept. If we return to the words of Keynes and Veblen from the introduction: ideas of defunct economists or habits of thought, we are getting closer but not entirely to the end goal. In this article, there is a reason to define a habit of thought more in terms of a framework. My reason for this is that I believe political thinking to be a more conscious act than Keynes and Veblen seem to have done. For example, Danish political scientist Rune Møller Stahl defines his views on the political thinking in Gramscian terms in his study of the Danish Economic Council during the 1970s and 1980s (Stahl, 2021). In such a definition political thinking is a language which moves from one school of economic thought to another, depending on how it fits into the preferred type of policymaking.

Language is important, but it also obscures the actors and their affiliations. If Swedish economic policymaking with Sandelin's words has been ‘economist-intensive’ and economics has been ‘policy-intensive’, then the actors cannot be excluded from influence (Coats, 2005). Gayon gives a contrary view to Stahl, were the actors’ ideas and affiliations are what constitutes their political thinking. This can also be linked to sociologist Stephanie Mudge who studies experts influencing social democratic parties in United Kingdom, Germany, Sweden and the American Democrats from the 1960s to the 1990s (Mudge, 2018). However, Mudge puts more emphasis on the expert as an influential actor, than on political thinking steering this expert. Gayon on the other hand recognises that there is a political framework to limit economic ideas.

This framework, how economic ideas are thought to be useful or not in politics, is an interesting concept. That is not comparable to studies in political science of paradigm shifts, which views economic ideas as formative to political thinking in politically turbulent times (Blyth, 2002; Hall, 1993). Like Gayon, I believe there is always a political framework to be considered and to that end economic ideas can be deemed either politically useful or useless. Economic ideas play an important part in making political frameworks seem relevant by ruling out other economic ideas. Unlike Swedish political scientist Johannes Lindvall (who also sees economic ideas as limited in their influence, albeit not during certain phases of paradigmatic change (Lindvall, 2009)) I assume that this limit is always present.

My approach is however another than Gayon's in that I include political thinking directly into the study of economic ideas. It is not just a distant framework to set a limit, but a present judge of the usefulness of these ideas. I believe that politicians playing an active part in defining economic ideas have been lacking in previous studies of economic expertise. To include such active political thinking into history of economic thought can be a way to further develop this area of study.

2.3. Sources

The policy debate on balance of payments has been selected because it samples so many of the pressing issues of the 1970s’ economic conditions in Sweden. The choice to preserve the fixed exchange rate in Sweden after the disintegration of the Bretton Woods regime in the beginning of the 1970s was a pivotal political choice (Werin, 1993). It kept balance of payments at the top of the economic policy agenda at a time when international economic conditions were rapidly changing.

In the SEA, debate on balance of payments was on its way to become a regular topic at the lectures where finance minister Gunnar Sträng presented his budget proposals. These lectures will be the main object of study. One external example from 1972 will also be used. It is taken from the Treasury's Economic Planning Committee. This source will be object of further scrutiny in my forthcoming doctoral thesis, but the example stands for itself and can be used to complement the debate.

The source material from the SEA is helpful in showing what experts argued for year by year. Changes in arguments are clearly noticeable. History can therefore be detailed in a way that actors’ own narratives are not because of selective memory and the need to tell a story from a certain perspective.

Less helpful is the fact that we cannot follow the exact same actors year by year. Even if several experts are present for the lectures repeatedly, it is arbitrary when this happens. Arguments from the same person can therefore have gaps of a year or two. Conclusions will therefore have to be drawn from an aggregate of actors expressing similar views. Individually though, their opinions can differ somewhat. This has been considered and the individual differences are shown as much as possible.

I also must mention that two actors are not mentioned either in the introduction or included in the Appendix; banker Tore Sellberg and economist Lars Jacobsson. The reason for this exclusion is that they both appeared just once in the material, and they were only included because of their single statements on the balance issue.

One actor who, on the other hand, is present at every lecture is finance minister Gunnar Sträng. This means that there is a risk of giving the views expressed by him a greater space than those of others. Such a risk has also been considered and dealt with by always keeping note of the opposing views to Sträng's arguments.

The SEA is of course not the only area where economic policy was debated by experts. Debates in newspapers were very common even on technical topics such as balance of payments. Several experts also wrote books to express their views on current economic affairs, as well as reports for both government committees and private research institutes. It is not within the scope of this article to study all these sources. They will be left to further research.

2.4. Limitations

Main limitation to the study is the periodisation where the years 1967–1976 was chosen first because of an official government report on the balance of payments statistics that launched in 1967. It can be interpreted as a formalised sign of how important this issue was in economic policymaking at the time. Such a launch also made it a likely topic of discussion in the SEA.

The end year was chosen because of the 1976 general election when the Social Democrats lost for the first time since 1932. Thus, 1976 was the year of Gunnar Sträng's last budget presentation, which limits the scope here to concern Social Democratic economic policymaking. Such a limitation is relevant not least because of assumed historic links between Swedish Social Democracy and Keynesianism (Mudge, 2018). Finally, the investigation should not be seen as an all-encompassing study of economic experts during this period in Swedish economic policy debate, or even on the matter of balance of payments. It should rather be seen as a case of political thinking at odds with economic ideas.

2.5. Presentation of the study

First, the debate will be displayed as a development over two time periods: 1967–1972 and 1973–1976. The first period represents the initial formation of the debate, and the second represents the new situation after the fixed exchange rate system was entirely abolished.

Second, the included arguments are formulated by individuals who appeared repeatedly in the debates. This is to show how the debate developed as different sets of different economic ideas and their limits in relation to the political framework of Gunnar Sträng.

Third, arguments have been selected on the basis that they relate directly to the issue of balance of payments. For reasons of scope, no economic area indirectly related to balance of payments will be dealt with in this paper, i.e. taxes, public investments, wage bargaining, or public sector policy. Actors who do not argue on the matter of external balance have been excluded from the study even if they were present for the budget presentation.

3. Literature on political change and economic policymaking 1960–1980

Last before the empirical study is a brief overview of literature on political change in the studied period. The literature can be separated into three different themes:

First, a structural shift in the industrial economy with the return of deregulated international finance. This process accelerated after the Bretton Woods system of fixed exchange rates was dissolved in 1971 (Altamura, 2017; Bordo & James, 2019; Krippner, 2012; Monnet, 2018; Schenk, 2010).

Second, an institutional shift taking place toward the end of the 1960s within public institutions as the Bretton Woods regime came to an end. At this time technocratic views of market efficiency gained in strength. The ideas were rooted in the technical models of the Neoclassic synthesis rather than more orthodox Keynesian thoughts (Best, 2004; Helleiner, 1994; Ruggie, 1982).

Third, a political shift towards economic policies of fiscal austerity and marketisation of the public sector. This is supposed to coincide with a turn from Keynesian economic theory towards Monetarism and neoliberal ideas of the market (Blyth, 2002; Mudge, 2018; Schmidt & Thatcher, 2013).

These three explanations also apply to Swedish conditions. Structural explanations in Sweden have pointed to the 1970s as a moment of permanent shift in economic structures. It was the end of the post-war era of heavy mechanical industries and public investments (Magnusson, 2011; Schön, 2007).

Institutional explanations have focused on increased pressure on the relationship between capital and labour. The Swedish model of centralised bargaining was decentralised and post-war consensus dissolved (Ahlström & Carlson, 2010; Andersson-Skog & Ottosson, 2018; Korpi, 1985; Larsson & Söderberg, 2017; Svensson, 2002; Thelen, 2014; Werin, 1993).

Explanations of changing political conditions include studies of how the Social Democratic Party shifted from prioritising full employment in the immediate post-war decades, to using austerity measures in the early 1990s as a means to control inflation (Andersson, 2003; Lindvall, 2004; Mudge, 2018; Offer & Söderberg, 2016; Ryner, 2004). Scholars have also investigated how organised business launched an ideological offensive from the late 1970s onwards (Blyth, 2002; Boréus, 1994; Westerberg, 2020).

In this article the institutional and ideational explanations are at the forefront.

3.1. Few restrictions in economic policymaking during the 1970s

The late 1960s and early 1970s have been described as a ‘long decade’ from the 1971 Nixon shock to 1983³, when French president François Mitterrand conclusively abandoned Keynesian stabilisation policies (Tooze, 2014). Economic historian Adam Tooze wrote in a response to sociologist Wolfgang Streeck's analysis of the 1970s as a debt-ridden, and politically chaotic decade (Streeck, 2013) that there is another side to this description (Tooze, 2014). In his view the 1970s can be seen as the zenith of democratic economic policymaking; there were few restrictions from fixed exchange rates or international agreements.

In Sweden, the Streeck view can be found in some economists’ opinions of the 1970s. For example Erik Lundberg, Lars Jonung and Assar Lindbeck (Jonung, 1999; Lindbeck, 1998; Lundberg, 2010) have all taken on the view that economic policy was too experimental and politically incoherent with earlier decades of institutional stability in Sweden.

Adam Tooze's critique is applicable to Sweden in more ways than one. ‘Experimentation’ has had a negative connotation, with Assar Lindbeck even naming a book on what he thought went wrong with the Swedish economy in the 1970s ‘The Swedish Experiment’ (Lindbeck, 1998). Rather than Swedish economic expertise being more radically Keynesian than in other countries, it seems the contrary was true for at least a handful of the most prominent experts.

3.2. Swedish economic policymaking in hindsight

It is not uncommon to suggest that Keynesianism in Sweden ended in the early 1990s when the goal of full employment was abandoned (Lindvall, 2004). In this policy change, economic expertise as well as organised business interests are supposed to have played a significant part in transforming public perceptions of the economy (Blyth, 2002; Lindvall, 2004, 2009; Mudge, 2018). Studies such as these focus more on the goals of policymaking, and the ideas of economic expertise. Politics plays a less active part in the making of economic ideas.

An exception is political scientist Magnus Ryner's analysis of the Swedish Social Democracy and what he describes as its ‘neoliberalization’ in the 1980s. Ryner tries to explain that political thinking had already changed before the neoliberal turn took place. This is not captured in enough detail in previous studies (Ryner, 2004).

Ryner traces ideational shifts between different time periods. That makes the 1950s and middle of the 1960s a period of institutional-technocratic understanding of politics, which is then replaced by a utilitarian understanding from the middle of the 1960s to the 1990s. A shift from collective and institutional solutions to utility maximisation for the individual firm or household (Ryner, 2004, p. 109).

How the shift in paradigmatic understanding of politics occurred sometime in the late 1960s and early 1970s is beyond the scope of Ryner's study (Ryner, 2004, p. 110). The actual economic arguments taking shape in the late 1960s and early 1970s seem to be a blank that requires further research, which is the starting point for the study in this article.

Then economic historian, nowadays professor of history of ideas, Jenny Andersson also makes good points on the late 1960s and early 1970s as a pivotal period for Swedish Social Democracy (Andersson, 2003). Radical ideas in the late 1960s on social justice and increased economic equality were dismissed by finance minister Gunnar Sträng on grounds of economic efficiency. At first, his resistance received criticism from within the Labour movement (Johansson, 2018). But after having lost office in 1976 and two election defeats later, the Social Democrats embraced the efficiency argument rather than the one calling for increased equality. This from the conclusion that economic efficiency was lacking while social justice was not (Andersson, 2003; Löwdin, 1998).

The growing resistance from economists within the Swedish Social Democratic Party to the radical social movements as Andersson has described (Andersson, 2003) began in the 1970s and continued during the 1980s. If radical social policies were in line with Keynesian fiscal policy was however a debated issue, where for example Erik Lundberg withheld that they were not. But neither was any resemblance to the fiscal austerity of the gold-standard era (Lundberg, 2010).

Keynesianism as a balancing act between orthodox economics and radical leftism is a theme which follows us into the empirical section of the article. Let us move on to see the growing tension between a theoretical framework of Keynesianism that pointed to the necessity of active fiscal policy, and a political framework of centralised bargaining and fixed exchange rates which pointed to restrictive measures.

4. Balance of payments at the centre of Swedish economic policy debate

4.1. Introduction: deflation as a political strategy

In 1964 the Swedish Economics’ Association (SEA) organised a lecture on the problems facing the British economy. It was titled ‘Can Britain escape from her balance of payments difficulties?’ and speaker was British economist C. T. Saunders from the National Institute of Economic and Social Research.⁴

The United Kingdom had faced several balance-of-payments crises since the 1950s and had dealt with them by putting internal demand on hold. According to Saunders, the measures usually consisted of raising taxes, cutting public investment programmes, and keeping a strict monetary policy. In contrast to the theoretical assumptions, he argued, these measures had reduced imports but not increased exports proportionately.

British governments have become increasingly adept in one new technology – that of correcting dangerous balance of payments situations by the instrument of deflation. There is much to interest the economist and politician in the relative effectiveness of the respective instruments – taxation, manipulation of public expenditure, and monetary measures. But at the moment I want to make one point only. This is that deflation corrects the balance of payments by reducing imports, not by increasing exports. It is not always defended in this way. It is more often suggested that the purpose of depressing home demand is to “make room” for exports. In fact, the conclusion to be drawn from British experience is that the effect of deflation on exports is small. Room has been made – but the exports have not come in to fill it (Saunders, SEA, 1964, p. 106).

At the core, deflationary policies seemed to shrink economic growth in the domestic economy a lot more than it could stabilise imports and make exports more competitive. Rather the connection between these factors seemed to be reversed; a country needed growth to allow exports to increase in the first place. Then, to further export-led growth, a devaluation of the currency could help. The connection between growth, exports and currency would not work the other way around. The British example had shown this to be true. Deflationary policies had suppressed growth in the domestic economy, creating crisis after crisis (SEA, 1964).

One of the prominent economists present at the lecture, Professor Erik Lundberg, mentioned that Sweden had a similar asymmetric relationship between imports and exports but had hitherto managed to escape the difficulties that had put such pressure on the United Kingdom. One of the reasons seemed to be a more competitive exporting sector.

In Sweden we are aware of the possibility of the same problem arising and we discuss it, but so far we have had little experience of balance of payments crises and we wonder why. The Swedish economy seems rather geared to a leading rise of exports; we get our booms mainly just because of rising exports (Lundberg, SEA, 1964, p. 129).

Another prominent guest, Professor Gunnar Myrdal argued that even if a country in great difficulty with its external balance could devalue its currency, it would not mean that you could escape the fixed currency system altogether. It was too important for business. Competitive devaluations were never a good idea, as shown by the great drive of 1949 when most industrialised countries had used such a measure.⁵ The result was inflation and exchange rate instability.

If it was true in 1964 that Sweden had managed to stay away from most of the unpleasantness coming from issues with the balance of payments, the next few years would see a change of tune. That is why the British example was interesting in a Swedish context. It was a bad sign. In the late 1960s, debate on the balance of payments grew rapidly as a topic among Swedish economists. Finance minister Gunnar Sträng made it his priority to keep Sweden out of trouble. But could deflation really work as a political strategy?

4.2. 1967–1973: overheating and ‘The lost years’

Since 1965 the imbalances in Swedish external affairs had begun pressing economic policymaking to a shift in priorities. In the Social Democratic government's view, at a minimum, external balance needed to be restored in 1970, but preferably earlier. This ambition was presented by finance minister Gunnar Sträng in 1967 (SEA, 1967, p. 2).

It would be a matter for fiscal and monetary policy to be kept strict, and credit to be further regulated. However, Sträng promised that the harsh measures would not affect the most vital areas of society. These included expansion of industry and housing, education and health care. But the balance of payments issue was seen as greater than any other in the economic policy area, affecting fiscal policy in its core function of regulating total demand (SEA, 1967, pp. 2–12).

Finally, I would like to stress that this year's budget proposal is characterized by a problem greater than others, namely our unsatisfactory situation with the balance of payments. Strict fiscal policy is needed at all times with the aim of erasing the deficit [in the balance of payments] (Sträng, SEA 1967, p. 12).

Just a few years after Saunder's presentation to the SEA, the Swedish government seemed to take a first step in following the British path. But so far, the consequences of such a move were hard to predict.

At the same lecture, Professor Göran Ohlin, pointed out that a great deal of uncertainty of the future had been built into the economic system. Almost all industrialised countries experienced similar worries over balance of payments and currency instability as Sweden did. But how great the problem truly was had not become altogether clear (SEA, 1967, p. 13).

This year the speed [of the economic activity] might not be as high as it used to, but instead we are heading into a curve, and nobody knows what hides at its end. The question is if we should hit the brakes, or if it is time to give more gas, if we should turn more to the right or to the left. There is uncertainty of the future, and it should be considered regarding our economic policy (Ohlin, SEA, 1967, p. 13).

Ohlin argued that the situation felt almost like the era of the gold standard in the 1920s and 1930s had returned. Even more so since devaluation of the Swedish krona had been ruled out as a policy measure. International cooperation put a great amount of pressure on economic policymaking to achieve external balance. In a worst-case scenario, economic activity in its entirety might have to be put on hold. As the 1920s had showed, Ohlin reminded the auditorium, going to such extremes could create a very volatile economic situation (SEA, 1967, p. 13). Dealing with the balance problem with permanent austerity was an option that had to be avoided.

Moreover, the statistics on the balance of payments could not provide the certainty that was required. There were worries that the numbers in the statistics on foreign exchange reserves did not hold sufficient quality to give a true resemblance of the balance problem. Gösta Dahlström, economic expert at the Swedish Confederation of Trade Unions, asked if maybe the crisis had more to do with statistics than with the economy itself (SEA, 1967, p. 22). Perhaps the government had been caught up in a mirage of a problem that was related to incorrect information.

Professor Erik Lundberg also criticised measures that combined stabilisation policies and structural change. These two areas were not the same in his mind. If industrial investments did not reach a satisfying level, it meant that credit policy and monetary policy were too strict, not that more state regulation was required. This was not the first time general fiscal and monetary policies would be preferred by leading economists rather than selective measures. The issue was that the balance problem made the Treasury prefer selection before general stabilisation. According to Lundberg this made economic policymaking too interventionist at the same time as it was too macroeconomically restrictive. A worst-case scenario to him (SEA, 1967, p. 19).

Why did the government prefer selective measures to general fiscal and monetary policy? One explanation, aside from the balance problem, was the Rehn-Meidner model. It held the specifics that it was supposed to promote employment in competitive sectors, mostly in the exporting industries. When unemployment increased the Social Democratic recipe was to expand employment in the public sector rather than stimulating total demand, which was thought to raise high levels of inflation.

But strict fiscal and monetary policies meant that it would be hard to keep demand in the domestic private sector alive. The result in time, if taken to the limit, would be a dual economy dominated solely by the exporting industry and the public sector.⁶ Economists arguing against selective measures preferred economic policy to be more neutral, not promoting single sectors but rather regulating demand by fiscal and monetary policies on an aggregate level. Thereby creating more room for private demand.

Lundberg interpreted Keynesian theory as demand management alone here. That interpretation did not take any inherent instabilities of markets into account, which selective measures could handle by regulation. Another interpretation of Keynesianism, apart from demand management and state intervention, was the argument of uncertainty brought forward by Göran Ohlin. If economic actors were uncertain of the future, a permanent shift to austerity would not be wise as it would lead to even more uncertainty of the stability of the economy.⁷ This interpretation was more progressive than the technical one as it viewed demand management not just as a quick fix, but as a way of reshaping beliefs in the future strength of the economy.

In 1968, Göran Ohlin criticised Sträng's fiscal austerity calling it ‘conservative’. Ohlin returned to his thoughts of the gold standard and mentioned that back then there had been several examples of countries falling into economic depression while attempting to maintain external balance. Economic downturns should not be interpreted as ‘stability’. After the Second World War there had been an agreement among economists that internal demand was not to be sacrificed for the sake of external balance. This agreement seemed to have vanished now that exchange rate stability was valued above all else. Such a choice could mean that a country would have to live below its economic potential (SEA, 1968, pp. 19–21).

In the government's budget proposal, the following line is repeated: “a firm determination to defend the Swedish exchange rate. Improvements in the balance of payments has to be maintained by a strict and conscious policy, that imparts confidence with the rest of the world.” I find that to be the most horrible of clichés. Defending the external value of the Swedish currency is not the same as defending king and country (Ohlin, SEA, 1968, p. 20).

Ohlin was backed by academic economist Lars Matthiessen, who agreed that too narrow a perspective on public finance was a counterproductive way to assess the health of the economy (SEA, 1968, p. 26). They seemed to both have the more progressive Keynesian view in mind. Permanent austerity was in their views a threat to economic stability.

The matter of balance-of-payments statistics and if they could show whether there was an actual balance problem came up again at the 1969 lecture. This time it was pointed out by banker Tore Sellberg that the numbers seemed almost arbitrary (SEA, 1969, p. 23). A group of economists had been appointed by Sträng to investigate⁸ how this could be corrected and present the findings in an official government report (SEA, 1969, p. 27). But uncertainty on the statistic realities of Swedish external affairs was still apparent.

However, Sträng did not wait for the results before he continued to stress the weakness of external balance as the most pressing issue for the Treasury. In 1970 he argued to the SEA that economic policy had to be dedicated to creating a trade surplus and reducing internal consumption and imports (SEA, 1970, p. 5).

It was implied, somewhat in jest, by Erik Lundberg that since Sträng had received so much criticism from the radical left flank in his own party for his harsh fiscal policies he would not dare continue now. The claim seems to have been so absurd that it did not require a response from Sträng. Lundberg continued to argue that it was an unhappy circumstance that fiscal policy did nothing to compensate for the strict credit and monetary policies (SEA, 1970, pp. 21–22).

Successful structural changes in the economy had the downside that companies producing for domestic markets had been outsourced, creating more imports as a side effect of a stronger exporting sector. Lundberg argued that domestic firms needed internal demand to not be outcompeted by the exporting industry in terms of productivity and profits. This required another, less restrictive, policy mix than the government proposed.

The downside to the structural changes creating this favourable exporting competitiveness is an increased openness to imports. It says, in the section on business cycles in the budget proposal, that a negative effect of structural rationalization is elimination of several less competitive companies, producing for the demand of the domestic market. What we have won in exports we have through this policy lost in imports (Lundberg, SEA, 1970, p. 19).

Economist Lars Jacobsson took the argument a step further, implying that a decrease in demand in the Western world in 1970 compared to 1969 also made further increases in exports less likely (SEA, 1970, pp. 30–31). It seemed that creating external balance was not as straightforward as just increasing exports, not if domestic firms could not meet internal demand.

In Sträng's 1971 budget presentation, the situation had brightened slightly from the government's perspective. The ratio between imports and exports was predicted to shift in favour of exports. However, favourable forecasts did not mean that it was time to relax; achieving external balance was still a priority for the government (SEA, 1971, p. 5).

Sträng received both support and criticism from Göran Ohlin and Professor Ragnar Bentzel. They agreed that the balance problem was a pressing economic issue, though they both believed the Swedish economy to be built on a strong foundation. Bentzel wanted Sträng to assure private business that it could feel confident of profitable private ownership in the future (SEA, 1971, p. 20). Ohlin on the other hand had more qualms when it came to the duality of economic development. Just as Lundberg had done in 1970, Ohlin warned that domestic production was developing too slowly, and he was also critical of the expansion of the public sector instead (SEA, 1971, pp. 15–18).

The analysis of the balance of payments, with which the finance minister has agreed, says that expansion of industrial capacity during the 1960s has been too slow in comparison to our importing needs (Ohlin, SEA, 1971, p. 16).

[…]

What I have hitherto said is related to the fact that public spending has risen too fast during the last five years. Not as an abstract measure, but in our real-term possibilities of financing them (Ohlin, SEA, 1971, p. 18).

If we assume that both Lundberg and Ohlin were Keynesian economists in this context (after all, they had both argued against permanent austerity and strengthening domestic production to meet internal demand), then Sträng's economic model must have appeared rather ambiguous. He operated from a political thinking where the exporting industry and the public sector were the pillars of the economy. The proposed Keynesian arguments of easing the strains on domestic demand did not seem to fit into that model.

Meanwhile, uncertainty still reigned over the balance-of-payment statistics. As trade union economist Clas-Erik Odhner pointed out, it was still possible that the debate was being held on incorrect terms. Despite this, he was supportive of a strict budget proposal for the sake of that year's round of centralised bargaining, which was about to begin (SEA, 1971, pp. 22–23).

At last, criticism came from economist Gunnar Eliasson. The calculus on the ratio of imports and exports made for a too optimistic view on the balance problem. If the government really wanted to achieve a higher level of industrial investment it needed to check internal demand further. It was not however easy to do so when numbers pointed to an international slump in demand. Alternatively, the government would be forced to implement deflationary policies. The British example was discouraging in that respect. Yet, either exports had to increase, or deflationary reforms would be necessary according to Eliasson (SEA, 1971, pp. 25–26).

1971 was also the first year recorded since the immediate post-war years where there had been a period without growth. It was a time of crisis for the fixed exchange rate system now that the American president Nixon had abandoned the gold parity. The only light in the darkness was that Swedish exports had grown enough to create a trade surplus. However, according to Sträng the unemployment level⁹ should not be compared with the situation in the 1920s and 1930s. Social safety nets made sure that the unemployed maintained a certain purchasing power. Despite the worsened economic situation in 1971–1972, there was no need for stimulation of general demand (SEA, 1972, pp. 2–4, 6, 8, 10).

Gunnar Eliasson did not think the sense of crisis was justly founded either. The imbalances within the fixed exchange rate system had been troubling the world economy for years. It was more a case of economic policymakers underestimating the apparent instability (SEA, 1972, p. 17). In short, there was nothing to be surprised about when it came to the economic downturn and no need to shift political priorities from achieving external balance.

Both Sträng and Eliasson were met by a different set of arguments from Lars Matthiessen, who had re-joined the debate. He pointed out that in 1969 the tendency to overheating had been met with credit and monetary austerity, while 1971 had seen fiscal austerity. Total demand had been lowered enough to create the improved external balance. During a year of economic downturn, it was a measure selected from outside the Keynesian toolbox (SEA, 1972, p. 28).

Matthiessen argued that of course, it could be that statistics and forecasts had been incorrect since 1965 and that the effect was less efficient economic policymaking. But more important in his view was the hesitancy to use general fiscal policy instead of selective measures. Sträng should not ‘dismiss Keynes’ and reject total demand during downturns. General measures could prevent demand-related issues so that selective measure did not need to be used as a last resort when problems had already appeared. Selective measures were merely bandages on an already broken limb. The safety net needed to be rigged before a sudden fall occurred (SEA, 1972, pp. 28–29).

What Matthiessen argued for stood in some contrast to the other economists. Apart from Göran Ohlin's references to the gold standard, and the economic depressions that had been created by the constant priority of external balance to internal demand, few had argued that Sträng was anti-Keynesian.

Either they believed in the balancing problem and Sträng's analysis or they wanted correct statistics to draw conclusions on. Matthiessen's implications were that numbers or not, external imbalances or not, fiscal policy had not been used to regulate total demand for quite some time.

Instead, selective measures were used as an emergency solution when it was already too late in the business cycle. It was not what the Keynesian countercyclical recipe called for. This makes for two extremes in the debate: Gunnar Eliasson who openly argued that deflationary policies might be necessary, and Lars Matthiessen who criticised the government for being pro-cyclical in its economic policy choices. In 1971–1972 the government had taken the easy way of creating external balance by letting the international downturn deflate internal demand.

This section ends with the 1973 lecture. Gunnar Sträng now concluded that the period 1971–1972 could be called ‘The lost years.’ No country was an island now that the international economy trended towards more integration. Imbalances could mean that a government had to use drastic measures to restore balance. Restrictive economic policy had in his view been successfully counteracted by selective labour market and industrial support measures (SEA, 1973, pp. 55–58).

For now, the balance problem seemed to have disappeared. Professor Erik Dahmén and Professor Assar Lindbeck both argued that the government had turned towards procyclical economic policy. Lindbeck referred to it as ‘English procyclical policy’, drawing a direct parallel with the British situation. Dahmén argued that the turn to selective measures instead of general demand policies had meant that unemployment had to rise unnecessarily before it was reduced again. They were backed by Göran Ohlin who argued that in its entirety the public sector did not have any deficits, and fiscal austerity now would thereby mean that total demand was depressed (SEA, 1973, pp. 62–64, 66, 69).

What we have here are two contrasting views and the question how these could co-exist. The Treasury's perspective was obviously that Sweden lived dangerously near, or even over, its economic capacity. The opposing view was that the Treasury did not use its tools to regulate total demand but inefficiently relied on selective measures to counteract the business cycle.

4.3. 1974–1976: Keynesianism revisited

In 1972 a council of economic experts belonging to the Treasury called The Economic Planning Committee (EPC) held a meeting on what the new exchange system without gold parities would mean.¹⁰ A memorandum presented to the EPC stated that the British Chancellor of the Exchequer, Anthony Barber, had made a speech in which he said that ‘it is neither necessary nor desirable to distort domestic economies to an unacceptable extent to maintain unrealistic exchange rates.’ (EPC, 1972, p. 35; 14).

Was this statement applicable to Swedish conditions also? The answer provided in the memorandum was hesitant. There was an apparent conflict between managing internal demand to satisfactory levels and at the same time to keep imbalances from too expansionary policies in check.¹¹ Deregulated capital movements were also said to have made nations more vulnerable to imbalances. A cited report to the International Monetary Fund stated that governments in member state countries felt more comfortable with surpluses than deficits. It seemed that for now, Swedish economic policy was suggested to be a middle way between prioritising internal demand and strengthening external balance (EPC, 1972, p. 35).

Was this not exactly the kind of thinking which Gunnar Sträng had expressed publicly to the SEA since 1967? It was perhaps not desirable to deflate internal demand, but contrary to the view expressed by Barber, it was deemed necessary. Sträng was backed in that regard by Gunnar Eliasson¹², Clas-Erik Odhner, Erik Lundberg, Ragnar Bentzel and Göran Ohlin.

At the same time, Ohlin was worried by the tendencies to sacrifice internal demand for the sake of external balance. As we have seen, he was not a pro-austerity economist on this issue like Eliasson, but rather ambivalent in his conclusions. The gold standard-esque manoeuvres were criticised by both Lars Matthiessen and himself. They were also in on the same argument against selective measures as Erik Dahmén, Erik Lundberg and Assar Lindbeck. Especially Matthiessen was a strong proponent of the importance of adhering to Keynesian theories.

The Treasury was at the same time questioned by trade union economists Gösta Dahlström and Clas-Erik Odhner. They thought the lack of verified statistics on the balance problem worrying. To base economic policy on uncertain ground was not an option if centralised bargaining and the Rehn-Meidner model were to function properly. Making decisions on scarce knowledge and without consensus from economic experts was a risky strategy.

In 1974 the finance minister argued that Sweden did well despite the 1973 oil shock affecting economic conditions. Those concerned with too strong a balance position could relax, the reserves would be needed now. But the large increase in the price of oil could lead to volatility in foreign exchange reserves and deteriorate the balance of payments. It would be wise for both government and households to keep reserves for the future (SEA, 1974, pp. 52–54, 57). The message was clear, Sträng thought Ohlin and Matthiessen were worried over nothing. Their radicalism in previous debates had been unnecessary. He would use the surplus to counteract the oil price shock, but in no way should this mean that it was time for long-term fiscal stimulation.

Ohlin responded that demand had to be increased if there was free capacity in the economy. It would be foolish, on top of the oil shock, to introduce home-made deflationary policies to make the country poorer (SEA, 1974, pp. 59–60). Matthiessen also mentioned that in rejecting stimulating internal demand, the finance minister rejected half of what GDP growth consisted of. Too much stimulation could be counteracted later, but welfare losses from unemployment would never be recovered (SEA, 1974, pp. 61–62).

1975 saw a minor shift in the argument. Sträng still held the opinion that his economic policies were in line with the overarching goals of full employment, industrial competitiveness, increased standard of living, and controlled levels of inflation. He also said that it was through increased exports Sweden would pay for more expensive imports. Increased public borrowing would first and foremost be for industrial needs. But the existing deficit in the state budget would be used to support private demand. Should unemployment rise further the government would be ready with more resources (SEA, 1975, pp. 143–146).

Were Ohlin and Matthiessen satisfied now? No. They both pointed to the problem of trusting other countries to lift Sweden out of economic downturns. Unemployment no longer meant that nations began stimulating internal demand. Even if the fixed exchange system had been abandoned, the new European cooperation (‘the currency snake’) did not mean more room for internal demand to be the focus of fiscal policy (SEA, 1975, pp. 153–154). Ohlin congratulated Sträng on finally having reduced some of the surpluses in the public finances, even though he seemed so reluctant to receive the compliments (SEA, 1975, p. 151).

At the same time, Sträng received support from Lars Wohlin, an economist at the Industrial Institute. He thought it was wise to allow fiscal policy more flexibility than he argued Ohlin and Matthiessen wanted (SEA, 1975, p. 155).

However, Wohlin also worried that in the name of reducing imports, every economic downturn had now become an opportunity to steer resources from private demand to the public sector. In the long-term such a move would mean too little for industry to do, and the industrial expansion that Sträng wanted would not be realised (SEA, 1975, pp. 155–157). In using such an argument Wohlin criticised the Social Democratic recipe for tackling unemployment. Austerity might be needed but it should also be directed towards the public sector so that private industry could grow. It was a zero-sum game rather than a total demand problem.

Even if the Treasury decided to stimulate demand during 1974 and 1975, it did so quite reluctantly, and it would only continue until the rest of the world restored its demand for goods. When this moment would arrive was less certain. In 1976 Sträng began speaking of the balance problem again. 1980 was now the goal of resumed surplus in the external balance. This would mean that not only private demand, but public demand too would have to be kept at bay. Except the introduction of a fifth general week of paid vacation and a promised increase in pensions, every other social reform had to be put on hold. Export-led growth and industrial expansion was the road forward in Swedish economic policy (SEA, 1976, pp. 158–159).

Assar Lindbeck, who had been critical of Sträng in 1973 for being too cautious, was now in agreement with the finance minister. Cost-inflation was a greater problem than total demand; inflation was a more alarming threat to full employment than lack of demand (SEA, 1976, pp. 159–160).

Göran Ohlin and Lars Matthiessen's Keynesian argumentation had been joined by academic economist Sven Grassman, who had also been one of the experts employed to investigate the balance-of-payment statistics. In Grassman's view it was not so much that Sweden had difficulties in expanding demand internally as it was impossible to increase it externally. The perspective that growth could only come from exports simply did not rhyme with statistics according to Grassman's calculations. Swedes did not owe the rest of the world more than what it had in financial assets. Rather what had happened during the last few years was a redistribution of foreign reserves and liquidity from corporations to the banking sector. A deficit in the former meant a surplus in the latter. The notion that the external balance was off its mark meant that a reduction in activity in the economy as a counteraction was all for nothing (SEA, 1976, pp. 162–164).

Grassman's line of thinking was supported by Clas-Erik Odhner but rejected by Sträng who sided with Lindbeck (SEA, 1976, pp. 164–165, 170–172). During the budget presentations at the SEA, he had sided more with those economists who worried about the balance problem, than he had with those arguing for Keynesian demand management. Keynesian advice seems to have been presented to deaf ears.

5. Conclusions

Is it fair to say that political thinking limits the influence of economic ideas and their validity in economic policymaking? Yes, so it would seem. Finance minister Gunnar Sträng is repeatedly antagonistic to the Keynesian economic experts in this study. In line with that his most common argument is that of hard times making it necessary for Sweden to rely on exports and restricted internal demand to cope. The 1974–1975 Keynesian moment is just an exception to the rule. His thinking on economic policymaking seems not to have shifted in any way since the 1950s (Löwdin, 1998). New economic conditions were not valid to him in that sense, but a priority for institutional stability was.

If the aim of economic policy making was to achieve a position as a trade-surplus country, then Keynesian expertise could be deemed unsound in its advice. On that ground a finance minister would certainly reject too radical attempts to lower internal costs of adjustment, just as we saw Sträng do repeatedly. It is likely he saw the expanded welfare state as enough to soften the blow. How a continued expansion would be possible to combine with more deflationary policies was less clear.

The problem was also, as Odhner and Dahlström would argue for, that it was not completely beyond doubt that the balance of payments was such a pressing issue. That raises a question of whether Sträng rejected Keynesianism on grounds of him not being sympathetic to the ideas of that school of thought, or if it was indeed because of what he saw as issues in the structure of the Swedish economy. Being the practical politician that he was it is more likely the latter, but of course you also only see ideas you do not like as square pegs in round holes. They simply do not fit the measurements. It is mentioned by trade union economists that the centralised bargaining requires strict fiscal policies, which would have been an aim of Sträng's to achieve.

Interestingly also, there is no clear link between Social Democracy and Keynesian ideas in this context. The most articulate experts – Ohlin, Matthiessen and Grassman – are to my knowledge not Social Democrats here (Grassman became one in 1980). Those who are – Myrdal, Lindbeck, Odhner, Dahlström – differ in their views on Keynesian measures.

Keynesian thought is in itself a set of theories that can be used for cherry-picking among experts. It also highlights the heterogeneity of the experts involved in this debate. From Matthiessen and Grassman who represent a more progressive Keynesianism of anti-austerity measures, over to Ohlin's, Lindbeck's and Lundberg's criticism of the Treasury's strict economic policies but more from a will to increasing private investment and decrease government regulation. Trade union experts Odhner and Dahlström are somewhat difficult to place as Keynesians, while it is obvious that they did not see the balance of payments as such a pressing matter as others, their economic model still required fiscal restraint.

Then there are the non-Keynesians in Bentzel, Wohlin, Dahmén and Eliasson. They both criticise the Treasury's strictness and propose that austerity will soon become necessary, while also praising Sträng's determination to avoid expansionary fiscal policies. It is not a coherent argument in that respect. Instead, the common thread seems to have been their wish to diminish interventionist policies rather than advocating austerity.

In conclusion, the sets of ideas represented here are as follows: first, a technical sort of Keynesianism focused on demand management and less government intervention. Second, there is a progressive kind of Keynesianism which puts more emphasis on anti-austerity policies and uncertainty. Third, there is the trade union economic model which has strict fiscal policy as a requirement for the centralised bargaining model but does not advocate austerity. Fourth, there is an anti-Keynesian set of ideas which argues against selective government intervention and promotes austerity when required by the exporting sector.

These four sets cannot all be within the limits of Sträng's political thinking, and returning to (Gayon, 2017) this means that only parts of these economic ideas fit into the economic policymaking of the studied period. It might be an obvious thought, but the implication of it is that the late 1960s and early 1970s saw a rise of a debate on fiscal austerity that was beyond the political thinking of the time. Some ideas simply had to be ruled out of the political context.

Sträng's arguments show that it was not plausible to think that Keynesianism – or any other school of thought that challenged status quo for that matter – was politically useful until economic conditions became extreme, and he was not going to let the extreme become reality. It implies that he thought economic policymaking should limit the use of economic ideas that challenged the current regime. His priority was institutional stability.

The four sets also show that the affiliations of the experts can largley be seen as important to their opinions on economic ideas and the government's policies, but are not always straight-forward. Finally, the link between Swedish Social Democracy and Keynesianism in political practice seems to have been weaker than has been argued in previous studies (Blyth, 2002; Mudge, 2018). What we can learn from this study is to always pay attention to the political conditions of the historic context when studying economic ideas. Without these conditions it is easy to forget the active link between politics and economics.

Acknowledgements

First I am indebted to associate professor Nikolas Glover and professor Jan Ottosson at the Department of Economic History, Uppsala university, for their support as well as helpful comments on this article during the whole process from idea to finished manuscript. I am also very grateful to the editors of the Scandinavian Economic History Review in presenting me with comments on my first manuscript in Gothenburg in October 2021. Finally, I am most grateful to the two anonymous reviewers who gave me very helpful comments on my second manuscript, and to the final reviewer who gave me comments on my third manuscript. Any errors remaining are of my own doing.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 For more insight into the history of SEA, see articles on its 50th and 100th jubilees (Nationalekonomiska föreningen 1877–1927, https://www.nationalekonomi.se/filer/nef-pdf/jubileumsvolymer/nf50-minnesskrift.pdf; Nationalekonomiska föreningen 1877–1977, https://www.nationalekonomi.se/filer/nef-pdf/jubileumsvolymer/NF100_0_forord.pdf. Articles in Swedish).

2 Nationalekonomiska föreningen 1877–1927, https://www.nationalekonomi.se/filer/nef–pdf/jubileumsvolymer/nf50; Nationalekonomiska föreningen 1877–1977, https://www.nationalekonomi.se/filer/nef–pdf/jubileumsvolymer/NF100_0_forord.pdf.

3 This is similar to economic historian Lennart Schön's periodization of a ‘new economy’ after OPEC I in 1973, see (Schön, 2007).

4 The entire introduction to the investigation builds on ‘Can Britain escape from her balance of payments difficulties?’, Lecture from the Swedish Economics Association (SEA), 1964-12-07, pp. 105–135.

5 Myrdal had support in his view from an official government report on economic crisis management during the Second World War. It stated that a country should not consider competitive devaluations. It should exclusively be a measure of last resort in deep economic crises (SOU, 1952, p. 49).

6 In the 1980s, the Social Democrats moved this notion from the public sector, embracing ‘industry’ as the sole engine of growth in the economy. The program used for guiding the economic policy after the election victory in 1982 had similar elements to what Sträng presented in 1967: expansion from exports, strict fiscal policy to keep internal demand in check, and capital moved from less productive sectors to productive industry. ‘Rehn-Meidner with supply-side policies’. Excerpt can be read in Swedish finance minister Kjell-Olof Feldt's memoirs; (Feldt, 1991, p. 25, 159). Feldt worked close to Sträng in the Treasury during the first period of this investigation, from 1964 to 1970. What scholars pointing to a great paradigm shift in economic policy in the 1980s probably have had their eyes on are the supply-side policies, such as credit deregulation, tax reductions, and public sector austerity. ‘Supply-side Keynesianism’ has been used to describe this type of social democratic austerity, which had its real breakthrough in the 1990s. See for example (Bremer & McDaniel, 2020). As this paper shows, general demand-side Keynesianism was abandoned as an idea in Social Democratic economic policymaking in the 1970s, relying only on public sector for internal demand.

7 See (Keynes, 2018) for Keynes’ original arguments. Lack of demand that creates unemployment, inherent instability of markets and uncertainty on the future are three of his main arguments on why economic policy is a crucial task for a government.

8 The Balance of Payments report (SOU, 1971, p. 31) was presented in 1971.

9 Unemployment level was doubled from 1.5% to 3% (OECD Economic Outlook, 2017).

10 As mentioned in the introduction, the Planning Committee and its secretariat had close affiliation with the Industrial Institute for Social and Economic Research (Industriens utredningsinstitut) until 1975. The Institute was particularly involved with the long-term planning reports, which was made every five years to predict trends in the economy. Gunnar Sträng is said to have been particularly affected by the one from 1970 where the balance problem was presented as the main economic difficulty. This interesting argument is presented in Carl-Johan Åberg's chapter in (Henrekson, 2009).

11 The Sterling was transitioned from its pegged position to a float in 1972. One of the arguments from the British government and Bank of England was that continuous devaluations would have been too costly for public finances and would have created further balancing problems. It was also thought that both political leaders in the U.S. and academic economists warming to the idea of a more flexible exchange rate system made the transition less frightening than when the pound was devalued in 1967. The years between 1967 and 1972 had changed some of the post-war assumptions on the exchange rate system in the United States and United Kingdom. This was not the case in continental Europe where a fixed exchange rate system was seen as vital for economic cooperation within the European Economic Community. See (Schenk, 2010, pp. 315–340) for further details.

12 Eliasson also wrote a book called Diagnosis on the 1970s (Diagnos på 70-talet) (Eliasson, 1971) on the matter where he expanded his thoughts on the balance problem.

Appendix

Table A1. Actors mentioned in the empirical study.

Gunnar Sträng	Finance minister 1955–1976; Social Democrat.
Gunnar Myrdal	Professor of economics at Stockholm University (founder of Institute of International Economy, IIES); Minister for trade; Social Democrat.
Göran Ohlin	Professor of economics at Uppsala University (positions also at Stanford, Yale and IIES); several international assignments, amongst other deputy general secretary to the UN.
Lars Matthiessen	Professor of economics at University of South Denmark (positions at Stockholm School of Economics and The National Institute of Economic Research, Konjunkturinstitutet); member of Danish Economic Council.
Erik Lundberg	Professor of economics at Stockholm University; first head of The National Institute of Economic Research (Konjunkturinstitutet) from 1944.
Erik Dahmén	Professor of economics at Stockholm School of Economics; scientific advisor to Stockholms Enskilda Bank; founder of the Center for Business and Policy Studies (Studieförbundet Samhälleoch Näringsliv).
Assar Lindbeck	Professor of economics at and chair of IIES; Social Democrat until the 1980s; economic advisor to Prime minister Olof Palme until 1976; founder of Swedish Economics Association's publication Ekonomisk Debatt.
Gunnar Eliasson	PhD in economics; professor of industrial economy at Royal Institute of Technology; researcher and head of The Industrial Institute for Social and Economic Research (Industriensutredningsinstitut).
Clas-Erik Odhner	Head of research at Swedish blue-collar trade union Landsorganisationen (LO).
GöstaDahlström	Resercher at the Swedish blue-collar trade union Landsorganisationen (LO).
Ragnar Bentzel	Professor of economics at Stockholm School of Economics and Uppsala University; head of The Industrial Institute for Social and Economic Research (Industriensutredningsinstitut).
Sven Grassman	PhD in economics; deputy chair of IIES; Social Democrat from the 1980s.
Lars Wohlin	PhD in economics; researcher at The Industrial Institute for Social and Economic Research (Industriensutredningsinstitut); secretary of state to Minister of finance Gösta Bohman; governor of the Bank of Sweden (Riksbanken); member of the Liberal party (Folkpartiet).