THE BUSINESS OF WELFARE

Abstract

A common narrative in welfare state research is that Sweden exemplifies a specific model of welfare, ‘the Swedish model’, or ‘the Social democratic welfare regime’. From this perspective the emerging welfare state left little room for private initiatives – the stage was set for the development of an encompassing welfare state in the 1950s. In this article I argue that this, virtually hegemonic, perspective has hindered an analysis of how private insurance co-existed and thrived within the emerging Swedish welfare state. As an alternative approach to ‘modelling’ – the concept of welfare-formation is developed to analyse mutually sustaining practices of welfare. I show how the insurance business and its protagonists influenced the settings of public pension schemes in a way that underpinned their own interests. A close cooperation with the state apparatus was fundamental for creating a trustworthy insurance market and legitimizing the business claim of fulfilling a social mission. The business adaptability in the shifting landscape of social policy also influenced perceptions of security and welfare in general. Commercial ideals became an essential dimension of the welfare state. By exploring this marginalized history – the business of welfare – the study deepens our understanding of modern welfare societies.

Keywords:

• life insurance
• welfare state
• social insurance
• social policy

Introduction

The concept of a ‘Swedish model’ or a ‘Social democratic welfare regime’ has been of paramount importance for the contemporary understanding of welfare policy in Sweden. In this vein of research social policy in Sweden is characterized by universal and income-related welfare schemes with far-reaching redistributive effects.¹ One cannot overestimate the imprint of Gøsta Esping-Andersen on this concept. His comparative analysis of ‘welfare capitalism’ has become a classic with an almost paradigmatic position in welfare state research.²

According to Esping-Andersen the goal of the social democrats was to avoid a split where social provision was divided between the market and the state. The market was purposely side-lined with the direct intention of creating a cross-class solidarity in favour of the welfare state. Social democrats could achieve these goals, Esping-Andersen asserts, because the market for welfare was underdeveloped in Sweden.³

The concept of a Social democratic welfare regime, ‘free’ from commercial interests, has turned out to be exceptionally powerful. The overall concept is still reproduced in welfare state research and is taken for granted in standard history.⁴ The notion of a distinct Swedish model of welfare is also used in research on ‘liberal welfare regimes’ as a self-evident fact and convenient antithesis.⁵

Researchers do tend to disagree about how to perceive the welfare state in general: Some describe Swedish social policy as a ‘success story’ where citizens were pulled out of poverty by social rights and welfare systems. Others view it as a paternalistic state apparatus where practices of social engineering threaten individual freedom and hamper the economy in general.⁶ There is also disagreement about the roots of the Swedish welfare model: Some focuses on class struggle and Social democracy, while others highlight the power of particular social groups like the agrarian middle classes or the interests of employers. Still others seek the answer in a specific Swedish mentality and substitute a class-based explanation with the powers of a long-standing tradition of co-operation.⁷ All of these approaches have a common denominator – in welfare policy development it is the ‘unique’ relationship between state and citizen that is to be explained. Hence, the ‘market’ is rarely part of the analysis.

Consequently, when Peter Baldwin states that social policy early on was determined by the interests of the agrarian middle classes, the focus is still on explaining the unique features associated with the Nordic welfare states. His study also mainly supports the role of farmers as a ‘veto group’.⁸ And even though Peter Swenson focuses on capitalist interests he does not examine whether employers were actually involved in the shaping of policy proposals or if they had any elaborate ideas about welfare policy in general. Thus, his study mainly supports the role of employers as consenters to ‘fait accompli’.⁹ In the same manner the ‘mentality approach’ to the Swedish welfare state is based on the assumption that social security was organized by the state. According to Berggren and Trägårdh the first major welfare reform, the 1913 public pension scheme, was shaped by a mentality that sought independence through the state and this spirit eventually led to the implementation of the Swedish model of welfare – ‘private insurance companies’ simply had no part in the evolving welfare state.¹⁰

It has been shown in previous research that ‘insurance experts’ dominated the public social insurance investigations for decades around the turn of the 20th century.¹¹ Obviously, influence is not analogous to the size of the insurance industry. But even though these experts are perceived as instrumental, they are either left out of the picture when social policy discourse is analysed or their affiliation with the insurance industry is never scrutinized.¹² In his classic account of social politics in Britain and Sweden Hugh Heclo points out, that public pension schemes ‘clearly did not originate as a substitute for voluntary retirement provisions’. Still, Heclo´s analysis of the Swedish case is underpinned by the assumption that the market for insurances was underdeveloped and that insurers did not influence early welfare policy in Sweden.¹³ The general agreement in welfare state research is that the growth of the Swedish insurance industry was limited by a ‘crowding out’ effect from statutory insurance schemes.¹⁴

The argument for a distinct welfare regime by pointing to a supposedly underdeveloped market is, however, not entirely convincing.¹⁵ A case in point is the business of Industrial life insurance introduced in the year 1899 in Sweden. Specifically aimed at manual labourers this insurance promised pension savings and compensation for surviving relatives and was an immediate success. Within 10 years 300 000 insurances had been sold. By the mid 1900s the four companies that completely controlled this market were responsible for almost three million insurances in a total population of 7 million people.¹⁶

The successful dissemination of life insurance among the working class demanded intense work by a vast number of agents – the so-called agency system.¹⁷ The number of life insurance agents also escalated rapidly in Sweden, peaking around 1930 when about 80 000 agents were employed to sell insurances.¹⁸ Another way of measuring the importance of life insurance is to compare it with expenditures in the public pension system. Even though Sweden is renowned for its encompassing pension systems it was not until the late 1930s that public benefits surpassed payments from insurance policies.¹⁹

A reason for why ‘markets’ have been excluded from historical explanations is that comparative welfare research has tended to concentrate on the period 1950–1980. Within the confines of this chronological timeframe, focus has been on explaining why private pension schemes only grew marginally in Sweden. This overlooks, however, the crucial fact that this sector at the outset of the period, in 1950, actually was more extensive in Sweden than in ‘liberal welfare regimes’ such as the US and Switzerland.²⁰

The common appraisal of Sweden as an unequivocal representative of a Nordic model or Social democratic welfare regime is in fact based on a static view of Swedish welfare policy from a teleological perspective. Hence, my objection is not about the model itself, but that the concept has become self-explanatory and thereby sustaining a vantage point that tends to neglect market-based social security. It has underpinned a state-oriented analysis of social policy and the result is, that one important part of the puzzle is still missing.

This article will show that the insurance industry in fact played a pivotal role in the emerging Swedish welfare state. And it will also explain how the insurance business – by promoting mutually sustaining practices of public and private insurance – could thrive within the emerging social democratic welfare regime. The public pension scheme implemented in 1913 was a critical juncture in Swedish social policy. Ideas developed in the commercial sector were important for the outline of the scheme, and the insurance industry continued to influence pension politics leading up to a completely reformed pension scheme in 1946.

Welfare-formation

In Esping-Andersen’s analysis of welfare states the interplay between the state and the market is essential. A key point of departure in his comparative study is the assumption that the state created the market for welfare, but also that the market created the welfare state. A mapping of the welfare state therefore needs an analysis of private welfare schemes. Esping-Andersen stresses the fundamental changes in welfare policy that occurred around the turn of the 20th century. Pre-modern forms of welfare were now pushed aside by the new private-public mix of welfare that was emerging. The process is described as ‘the thrift element shifted from the friendly society to the modern insurance company; the employer gratuity pension was gradually transformed into a contractual fringe benefit in collective bargaining; government poor relief became social security’.²¹

However, Esping-Andersen´s analysis is one-sided. To what extent his description is valid for the Swedish case is left open, and the role and influence of the insurance business is not scrutinized. The state is perceived as instrumental for developing both sectors; the state acts and the market reacts. Emphasis is put on the state´s ability to foster or constrain the market, and the market´s influence on the state is limited to an analysis of how the scope of the market acted as a catalyst for public welfare schemes.²²

In a recent book political scientist Kimberly J. Morgan and sociologist Ann Shola Orloff points out, that viewing the state as a uniform and cohesive entity tend to obscure the multiple actors and processes at work within the state. We need to scrutinize the real-world practices of governance – the mix of public and private actors charged with implementing policies. In essence, shifting boundaries reflect political contestation over the state´s meaning, purpose and resources.²³ Hence, to make sense of the welfare state requires that we examine the processes by which state-society boundaries are being defined.

In order to emphasize a historical open-ended process, I choose to speak about welfare-formation rather than welfare regime. By describing social policy as a process I want to avoid the inherent generalizations of comparative typologies and the static and monolithic metaphors of regimes or models.²⁴ The concept of welfare-formation draws attention to a history of competing discourses and institutions, rather than the neat and unified building of a single system.²⁵ From the perspective of welfare-formation, I interpret the emergence and expansion of life insurance as a process shaped by – and shaping – the social policy arena (Figure 1).²⁶

FIGURE 1 The social policy arena.

In the years around 1900 social policy was defined much more broadly than today. Welfare was not equated with public schemes and there were competing interpretations of how to meet the needs of the working class.²⁷ The social policy arena was made up by three interrelated sectors. This study focuses on the interplay between the business and the public sector, or more specifically, between life insurance and social insurance. In turn, these terms refer primarily to insurances related to pension policy.²⁸

Welfare-formation in the social policy arena took place in a grey zone characterized by both competition and co-existence. The boundaries between private and public were not clear-cut in any way. An illustrative example is that the first social democratic prime minister in Sweden was also one of the founders of an Industrial life insurance company in 1914.²⁹ Welfare-formation thus highlights the hybrid form of public-private action involved in the making of the welfare state.³⁰

Life insurance influence on welfare-formation can be discerned into two forms of impact: The direct impact was achieved through propaganda, political influence in parliament and insurers involvement as experts in governmental committee reports. The other form of influence was indirect where actuarial practices as well as societal ideas about insurance in general were mediated from the private to the public sphere.³¹

Life insurance and social policy

The formation and conflation of private and public social policy

The business of life insurance has always been entangled with moral issues and infused with ideas about social improvement.³² Already in the 1870s the practice of profit sharing was established as a result of pressure on the industry to conform to societal ideas about life insurance. Dividends were limited and policyholders were granted a share of the returns. This principle became a key characteristic of life insurance in Sweden.³³ In 1875 the ‘Swedish Insurance Society’ was founded as a forum where leading insurers and high-ranking politicians could meet to promote a ‘sound’ market. A platform for resolving internal disputes and political lobbying had been established. From now on the Insurance Society successfully promoted ideas about life insurance importance for both personal and public finances.³⁴

The more rapid development of life insurance coincided with the growth of the working class and the formative years of the social democratic party during the 1890s. Consequently, the product was now framed in terms of social policy, and a thorough regulation was presented by the Insurance Society as necessary to safeguard and develop life insurance among the working classes. In 1903 the first comprehensive laws on insurance were implemented.³⁵

From the perspective of welfare-formation this regulation was important for three reasons: First, by establishing clear-cut actuarial principles, a disparate and chaotic market turned into a more or less homogeneous industry infused with a ‘social purpose’. Second, on behalf of demands from the industry, the public Insurance inspectorate was established. A government body had now been created to monitor the market and promote the development of commercial insurance in general. This new authority with very close ties to the industry could exert influence over welfare-formation. Third, in the process the company manager Sven Palme emerged as the leading representative of the industry. In his dual role as business leader and politician Sven Palme would play an important part in the emerging industrial Sweden. Palme was a productive writer who was especially interested in the social aspects of life insurance and the relationship between public and private social policy.³⁶ In the hands of actors like Palme the business cultivated the self-perception of a social movement. This perception led to a strong sense of identity and purpose, emphasizing self-discipline and laying the basis for collective mobilization.³⁷

The question at hand is not whether the insurance companies’ motives ‘in fact’ were commercially self-serving or socially altruistic. Case in point is that the ‘social purpose’ provided means to a close cooperation with the state. Through direct impact the conditions on the social policy arena could be monitored; a market for life insurance sustained and developed and competition from the non-profit sector checked.³⁸ The entanglement with social politics was a key for adapting to the changing landscape of social policy. But the industry was also confined by the social purpose that propelled the market – life insurance was ‘caught in a web of social policy’.³⁹

The 1913 pension scheme

It was the development in the public sector that was to be the main issue for commercial insurers to deal with. Welfare historians have sought to explain why Sweden was the first country in the world to introduce universal pension insurance in 1913.⁴⁰ The substantial support in Parliament for this scheme has been interpreted as a typical expression of consensus politics in Swedish social policy. But the implementation of a public pension scheme was actually a political mess. The question of social insurance was first raised in Parliament in 1884 and it would take three public committees and a series of bills over a period of 30 years to reach a solution that finally could be accepted by Parliament.⁴¹

The future of social insurance was actually discussed in the Insurance Society before it was raised in Parliament. The society’s members were advised to form an opinion on social insurance because it was ‘only a matter of time’ before the question should become a political issue. At this point the industry still believed that the best policy was to develop the business of Industrial life insurance in Sweden. Life insurance had to be adapted to the needs and conditions of the working class.⁴²

However, this position within the industry slowly gave way to a positive view on social insurance based on an ideology of complementarity described in the next section. In the beginning of the 1900s the business support for a universal pension scheme was firmly rooted. The question at hand was instead how the scheme would be designed.⁴³ The business leading spokesman on social issues, Sven Palme, was also part of the public committee that outlined the universal pension scheme introduced in 1913 – the industry had insight as well as influence over the process.

The pension reform was a compromise between the three major types advanced for providing pensions through state action: (i) a compulsory and contributory insurance scheme like in Germany, (ii) a supplementary tax financed and means-tested assistance system of the Danish type, (iii) a state subsidised optional insurance scheme like the post office scheme in the UK.⁴⁴ In welfare state research the optional third part is almost always ignored and the pension scheme is described as a dual system. But in the on-going process of welfare-formation this part is equally important. In a sense the pension system was a combination of three different views on social policy. The fee-financed compulsory scheme represented social insurance as a social responsibility for each citizen to practice self-help. The supplementary system represented social insurance as a social right to a bearable old age without the social stigmata of poor-relief. The optional insurance represented the values of voluntariness and thrift and insurance as a moral institution.⁴⁵ Besides from the inherent logics the pension scheme was also a critical juncture with three possible development trajectories. Crucial for these trajectories was in turn the relationship that the insurance industry had to these different views on social policy.

The ideology of complementarity

Business representatives, like Sven Palme, were firm in their belief that the state had to intervene; the uncertainty in the actual position of the workers implied it. A voluntary solution was not feasible if pauperism was to be abolished.⁴⁶ But this societal perspective on social insurance rested on a dogma that emphasized the importance of complementary spheres. According to business representatives there were some key aspects that guided a proper balance between private insurance and social insurance.

Business representatives observed that private companies could administer social insurance, but at the same time the state could also pursue private-like insurance schemes. Hence, the public-private dichotomy was not sufficient for establishing proper boundaries. The punch line was that ‘one can be a devoted follower of social insurance as a concept, and yet be a firm opponent to the non-social state insurance’.⁴⁷

According to insurers the key element defining social insurance was the ‘social policy element’ based on the principle of assistance. Exactly this element was lacking in individual (private) insurance that was based on the actuarial principle. From this perspective the most important policy issue to address was how to establish a boundary where a redistributive social insurance could be justified.⁴⁸ Two conditions determined this boundary according to insurers like Palme: The first was that social insurance exclusively intended to protect against the need that arises when the capacity for work has diminished and when work is the main source of income:

if one by social insurance has something else in mind, something more, one will tear away the support for the logical and natural determination of boundaries.⁴⁹

The second condition was that the objective of social insurance was preventive, to render poor relief unwarranted. The idea that social insurance should be universal did not mean that it should cover the needs of the entire population. Or as one insurer put it: The aim is to prevent pauperism, not to administer pensions for everyone.⁵⁰ In accordance with this line of argument the pension benefits should be set to an appropriate ‘subsistence level’, determined by the cost of ‘bare necessities of life’. Even if this meant a deteriorated standard of living, Palme believed it to be of secondary importance compared to the advantage of a subsistence level eligible to everyone.⁵¹

More important, from the viewpoint of our contemporary concept of moral hazard, business representatives could present these boundaries as self-evident. In the same manner as an insurance policy reduced incentives in individuals to act responsibly, statutory social insurance could endanger the moral fabric of society. A precondition for social policy was an elevated work ethic based on practices that considered and battled ‘indifference and concessions regarding individual responsibility’.⁵² Already before the introduction of a universal pension scheme insurers highlighted and were well aware of the interdependence between social insurance and ‘the work approach’ in unemployment policies.

According to business representatives social insurance had to be based on the principle of assistance, hence the business opted for a state subsidised system based on progressive taxation.⁵³ The important question of financing has always informed the politics of pensions. Already in 1890 the question was debated at length in the Insurance Society: should social insurance be financed through contributions in an insurance system, or should it be financed through taxes in a redistributive system?

If an insurance model was used there were two possibilities; an actuarial system, like in the commercial sector, or a pay-as-you-go system, like in the non-profit sector. Worst-case scenario was, according to business representatives, an actuarial model. A fully funded public pension scheme would mean a massive accumulation of capital in the hands of a government body, which in turn would increase the state´s influence over the entire economy. In the long run this was also expected to have a negative effect on returns on capital and thereby threaten the development of life insurance in general.

A pay-as-you-go system was also considered to have drawbacks. The lack of robustness compared to the funded system made it prone to political manoeuvrings. There were no guaranties against short-sighted politics, and the solidarity it created between generations was only artificial. But the main problem with any insurance model was that it only meant that the costs of poor-relief was transferred from the state to the individual, i.e. from the well-to-do to the people with lesser means. The Insurance Society reached the conclusion that social insurance had to be financed by taxes if pauperism was to be abolished.⁵⁴

At an early stage business representatives began to describe social insurance as a ‘social right’.⁵⁵ In 1912 the concept of a tax financed ‘citizen-provision’ had been put forward in the Insurance Society. The idea was that pensions should be funded by the state without demands for means-testing or contributions. The stigmatized poor-relief should be replaced by a public pension system based on ‘the individual right’ to a bearable old-age.⁵⁶ A year before the implementation of a statutory pension scheme business representatives were in fact already discussing a national allowance system as the most preferable option.

There were other reasons too for the industry’s advocacy on behalf of social security. From a strategic perspective it was better to be seen as a proponent of pro-active social policy, as political agitation directed against such reforms would interfere with the message of social responsibility. Even more important was the belief that social insurance could be used to promote life insurance sales. Actually, this was one prime objective behind the first draft for a universal public pension scheme in 1889.⁵⁷ The universal character of pension policy in Sweden was not a socialist principle per se. On the contrary it was founded on the liberal principles of thrift and voluntariness.

Thus, already in the 1880s insurers were in fact promoting social insurance based on the expectations that it could facilitate the expansion of the commercial sector. This assumption was also confirmed when the social insurance scheme introduced by Bismarck in 1889 turned out to be beneficial for life insurance sales in Germany.⁵⁸

In other words, social insurance was considered a legitimate undertaking by the industry because it could be used to foster thrift and individual savings. This notion of positive effects was based on two arguments: First, social insurance promoted business by working as a forceful propaganda machine for the idea of life insurance. Secondly, social insurance created a more solid economic basis for the poor, which in turn also promoted the interest in supplementary private provision.⁵⁹ Or in the words of Palme; social insurance kept within its proper framework would:

be of great value to the nation and not at all unwelcome to the insurance business. It is a well-known fact that a healthy social insurance prepares the way for private insurance and makes individuals, who otherwise would not be thinking of insurance, available to the offers of insurers.⁶⁰

In the eyes of insurers commercial and public insurance did not represent different societal aspirations that were in conflict, but as ‘two parts of a whole that were mutually sustaining’.⁶¹

The notion of a Swedish model of welfare presupposes a ‘crowding out’ effect, that statutory insurance limited the growth of the private market. In a recent article Andersson and Eriksson investigate this basic assumption. By examining cost-of-living surveys and firm data from 1884–1914, they show that compulsory social insurance led to a substantial rise in lapses during 1914. Hence, they conclude that there was a significant reduction in the demand for life insurances after the public pension scheme was implemented in 1914. They also point out that this result is contrary to the view held by insurers at the time, i.e. that social insurance would stimulate the market for life insurances.⁶²

However, from a business perspective temporality is essential. In the short-term a ‘crowding out’ effect was anticipated by the industry as people in general could not know what to expect from a statutory pension scheme. The ever-expanding sales forces were put to work and life insurance was expected ‘to ride the wave of social insurance’.⁶³ As a matter of fact, the market for life insurances exploded after the introduction of a statutory pension scheme. The sharpest rise in policyholders during the period 1870–1946 took place between 1915 and 1920.⁶⁴ Another aspect of the development has to do with ‘signalling effects’ on other levels than mere sales. The rise of private occupational pensions was directly related to the statutory scheme and two new Industrial life insurance companies were also founded with the explicit intent to supplement the public scheme.⁶⁵

A public investigation of saving habits in the 1920s pointed out that the Swedish population had become much more disposed to saving in life insurances after the war. There was no evidence to support the view of a ‘crowding out’ effect from the statutory pension. The inquiry was carried out by the chief actuary from one of the major insurance companies who rounded off with the following statement: as of old it is also common knowledge among perceptive business representatives that ‘the different branches of the rich tree of insurance’ do not compete with each other but promote each other’s well-being.⁶⁶ Thus, a one-sided concept of competition or ‘crowding out’ effects does not grasp the dynamics in the private-public mix of welfare. In the long-term the ‘market’ for social security was not perceived as a zero-sum game. Social insurance was expected to expand the scope of the market by strengthening the material base as well as the mental predisposition for self-help.

To summarize: The business support of social insurance was a mixture of social pathos, strategic reasons and profit reasons. It was also posited in the idea of a natural boundary conditioned by specific criteria. The core concept was that social insurance was equivalent with preventive poor relief. By carving out a limited policy area suitable for public initiatives, space was simultaneously created that was ‘naturally’ suited for private initiatives. The prized principle of self-help could thus live on and be accentuated despite the introduction of compulsory pension schemes.

Economizing in social policy: state insurance vs social assistance

The 1913 public pension scheme consisted of three interrelated parts: a contributory part, a means-tested part and an optional insurance. These parts were largely within the boundaries envisioned by the insurance industry and key figures like Sven Palme.⁶⁷ The first threat against the ‘natural’ containment of social security was the development of the optional insurance. This ‘third leg’ in the pension scheme had been important to appease critique of the scheme’s meagre benefits and lack of employer contributions. The pension plan’s advocates trusted that this part would be used by both employers and recipients, thereby endorsing its development.⁶⁸ There was in other words a built-in complementary solution in the general scheme.

The optional insurance was administered by the newly founded National Pension Board. A noteworthy fact is that the first director, Adolf af Jochnick, was the one who introduced Industrial life insurance in Sweden. Under his direction the optional pension scheme was turning into a private-like company, yet managed, subsidised and owned by the state.

When the state began running insurance schemes under the same principles as the private companies, social insurance was according to insurers turning into ‘state insurance’. This was seen as the big threat, and it was this front that the industry constantly policed. The optional insurance scheme eventually met with fierce resistance from the insurance lobby in the 1920s. The gist of the insurers critique was that the state was guilty of unfair competition on the insurance market.⁶⁹

The timing of the coordinated counter-offensive paralleled the development of occupational pensions in Sweden. The insurance industry did not venture into this market until the mid-1920s when a joint company was founded.⁷⁰ Thus, when the new sector for occupational pensions took off, the optional ‘state insurance’ scheme became a serious threat to commercial activities. The counter offensive turned out successfully and changed regulations from the late 1920s seriously hampered the National Pension Board’s aspirations. The optional insurance was confined to play a negligible role in the overall development of the public pension scheme.⁷¹

During the 1920s it also became clear that the pre-emptive role of the overall pension scheme was failing – a significant portion of retired workers still had to rely on poor relief.⁷² Financing was in the midst of the disputes that surrounded the development of the public pension scheme.⁷³ The insurance industry was involved, paving the way for a complete remake of the pension scheme.

The main issue was the balance between individual contributions and tax revenues, or between the insurance principle and the assistance principle. This conflict not only engaged party politics but also the state apparatus was divided. On one side were conservatives and liberals who – from a state insurance view on social provision – wanted to shift the balance in favour of a fully funded system based on individual contributions. This side was also supported by representatives of public poor-relief as well as the National Pension Board that administered the public pension scheme. On the other side were the Social democrats who wanted to increase tax subsidies and redistributive effects. The public Insurance Inspectorate and the insurance industry supported this social assistance view on social provision.⁷⁴

In this way the politics of pensions was split in two camps, though both fractions built upon ideas and concepts that evolved at an earlier stage within the commercial sector. In a sense these two views on welfare-formation allocated the state to different roles. Proponents of state insurance assumed that public poor relief would take care of the needs of the present-day generation, while the National Pension Board took care of the needs of the coming generations. Proponents of social assistance instead assumed that the public pension scheme was to replace poor relief, while the insurance industry was to address all the needs above basic security.⁷⁵

These views on social policy were in other words fuelled by self-interest. A public pension scheme based on the principle of assistance would more or less mean the end of public poor-relief. A far-reaching contributory public scheme would on the other hand mean, as was the case of the optional insurance discussed above, a pre-emption of the role of the private industry. The scope of the market as well as the profitability in life insurance in general would be endangered.

The conservative fraction made a push in the mid 1920s to develop the contributory funded part of the pension scheme. A right wing dominated public commission led to a bill that failed in parliament. One reason for this is that it was resisted all the way by the insurance industry and the Insurance Inspectorate. They even countered with a joint proposal for a reform that would build upon the pay-as-you-go system instead.⁷⁶

In the 1930s and 1940s the pension system was completely reformed. In 1935 the funded parts of the system were changed to a pay-as-you-go system and the actuarial principle was replaced by ‘a reasonable trade-off’ principle. In 1946, finally, the Parliament decided to break away completely from the insurance principles borrowed from the commercial sector, and converted the public pension system into a tax financed flat-rate ‘People´s Pension’.⁷⁷ All these changes meant a relocation of the pension system from insurance to assistance, which was completely in line with the agenda of the insurance industry.

More important, the government apparatus was now expressing the vision of complementary spheres developed in the commercial sector. Thus, it had become an important matter for the state to fashion the public pension system as a foundation for the extension of private pension provision.⁷⁸

In direct correspondence to the development in the public sector the Industrial life insurance companies set up a joint institution in 1933 for the development of occupational pensions among the working classes – The Workers Pension Society (Arbetare Pensions Kassan, APK). Chairman of the board in APK was the social democrat Bernhard Eriksson, who also was a member of the public commission preceding the pension reform in the mid 1930s.⁷⁹ Together with the managers from the Industrial life insurance companies Eriksson was now to realize the public-private welfare mix he himself had been paving the way for.

Eriksson later on became the chairman of the public commission behind the 1946 flat-rate Peoples´ Pension. This pension reform explicitly related to a societal perspective on social security. Proponents of an encompassing basic security system did not deny that abolishing both individual contributions and means-testing would drastically increase state financing. However, they were convinced that the socio-economic impact, its incentives on wage-work and personal savings, in the long-term would be of greater value. The explicit purpose, according to the commission report, was to design a system that would work as a proper foundation for private provision:

a platform which could be supplemented by the individual citizen according to ability, when it comes to old age security, and which could be supplemented by the employer with occupational pensions according to desired level for the employees.⁸⁰

The 1946 pension reform was a clear expression of social security as a phenomenon with three sides: (i) public basic security; (ii) occupational pensions; (iii) voluntary and individual pension savings. This vision of welfare-formation had been anticipated by the insurance industry, and the parallel is also striking with the three-legged stool metaphor that was promoted by US insurers in the late 1940s.⁸¹ This specific way of visualizing welfare-formation, originating in the private sector, has nowadays become an almost unquestionable concept that structures pension policy worldwide.

In welfare state research the pension reform of 1946 has been described as the first example of a universal citizen-based pension model, and Gustav Möller, the Minister of Social Affairs, has been championed as a defender of basic security or universal allowance.⁸² Within the commercial sector this pension model was of course old news, as described in the previous section, and leading business representatives had been propagating that means-testing had to be replaced by basic-security if the growth of occupational pensions was to be advanced.⁸³

The People´s Pension was not only a victory for the principle of a citizen-based flat-rate pension, it was also a success for the vision of complementary spheres.⁸⁴ The changes in the public pension system between 1913 and 1946 had followed in phases that more or less corresponded with ideals emanating from the insurance industry. However, to maintain that the Minister of Social Affairs acted as an agent for the insurance industry would be to put the argument upside down. Above all Möller wanted to abolish poverty among the elderly once and for all, and this aim coincided with the interests of insurers during this period.⁸⁵

Swedish pension policy in perspective

The public pension scheme introduced in Sweden in 1913 included an optional insurance part. But the burgeoning business of supplementary private provision was not keen on the idea of a competing solution in the hands of the government. In the mid-1930s the draft for Social Security in the US also included a proposal for the sale of voluntary government annuities. Consequently, after fierce opposition from the insurance industry this proposal was removed.⁸⁶ Both the Swedish and the US cases are examples of how the state apparatus co-opted actuarial principles developed in the commercial sector, but also how the insurance industry lobbied to control this indirect impact.

The 1946 public pension scheme in Sweden was clearly part of a comprehensive take on social policy, where private supplements were expected to thrive in the social democratic welfare state. This broader take on welfare is clearly visible in the support of key officials, the regulation of life insurance, as well as in how Industrial life insurers interpreted their mission in welfare policy.⁸⁷ This pension paradigm, where public basic security is intended to promote private security, has in welfare state research been equated with the ‘Beveridge plan’.⁸⁸

The public pension schemes in UK and Sweden were reformed during the same years and the recipe for dealing with poverty was the same; basic security. But this apparent similarity is ostensible from a financial perspective. The Beveridge plan was primarily a contributory scheme, while the Swedish pension plan was primarily a universal allowance scheme.⁸⁹ The end-result is seemingly the same, but the effects on redistribution and responsibility is radically different. Beveridge wanted to promote individual responsibility and was not keen on redistributing wealth between classes. The Swedish Peoples´ Pension, on the other hand, was by its very nature redistributive and individual responsibility was addressed with moral incentives rather than pecuniary.⁹⁰

Moreover, from a business perspective the Beveridge plan was in fact more of a foe than a friend to life insurance. One important aspect of the plan, which is often overlooked in welfare state research, is that the Beveridge report was based on a severe and thorough criticism of the commercial sector and the inefficiency of its agency system. Beveridge explicitly wanted to promote voluntary savings among the working-class but, and in here lies the main difference compared to the Swedish case, he wanted the state to administer voluntary insurance for blue collar workers. Phase two of the implementation of his plan was to nationalize Industrial life insurance altogether. On the other hand, he had a positive view on friendly societies and wanted the state to form a partnership with the non-profit sector.⁹¹ The proposals for basic security advocated by Beveridge in the UK and Minister Möller in Sweden were based on a comprehensive view on social policy. But the part business played in the overall schemes were radically different and put them in different camps.

The Beveridge plan actually advocated a comprehensive ‘state insurance scheme’ usually associated with a social democratic welfare regime. And pension policy in Sweden – generally the key determinant in typologies of welfare states – was moving in a different direction than towards the so-called Swedish model of welfare. In compliance with the vision of complementary spheres, the markets for life insurance and occupational pensions were to be reaffirmed and expanded, not reined in. The insurance industry as well as high-ranking officials also believed that this pension paradigm would last for a considerable future – but they were wrong.

Conclusion

The established narrative of the Swedish welfare state builds on the axiomatic assumption that commercial insurance companies in no important way influenced the development of the Swedish welfare state. Institutional path dependency, the argument goes, enhanced by a Social Democratic ideology or a specifically Swedish mentality led to the introduction of a Swedish model of welfare in the 1950s.

In contrast to existing research, I have shown that the insurance industry in fact was actively involved in the shaping of the emerging Swedish welfare state. The key to success rested on the ability of commercial insurers to govern the private-public interplay of social policy. Firstly, already in the 1870s life insurers began to act as a uniform and cohesive entity with a common goal. The ‘collectivized’ market was an apparent advantage when facing a much more diversified and incoherent state apparatus. The insurance industry could assert direct impact by acting as an influential extra-parliamentary actor. Secondly, the blurred boundaries between the market and the state also opened up a space for action where insurers could defend their position on the ‘social policy arena’ through the state. As public figures and experts on various committees, representatives of the industry could advocate a process of welfare-formation that left ample space for their own business interests.

The indirect impact of life insurance, where the state deliberately imitated commercial practices, is discernible in the contributory and fee-funded structure of the 1913 compulsory pension scheme as well as the actuarial structure of the publicly administered optional insurance. The industry acted to rein in this kind of ‘state insurance’ allying itself with those who wanted to increase the emphasis on providing tax-financed social assistance. The commercial sector´s normative take on social policy was affirmed and legitimized in the late 1940s with the introduction of a flat-rate Peoples´ Pension and new laws on commercial insurance. The People´s Pension was in fact the logical outcome of the vision of complementary spheres that the insurance industry had advocated for a period of almost 50 years. Social insurance was equated with tax-financed basic security and everything above was to be supplied by the commercial sector.

The framework of welfare-formation emphasizes competition between different welfare solutions rather than focusing on public institutions and class-based politics. The article points to the relevance of how a specific institution – the insurance industry – influenced social policy. In a way the perspective of the ‘Social democratic welfare regime’ has been inverted.

However, this does not mean that the class-based ‘social interpretation’ of the welfare state should be replaced with a ‘business interpretation’. Quite the opposite. Locating mutual interests is crucial for the analysis of how strong incentives are forged as well as for the explanation of certain developments. But this also makes it hard to determine to what degree the Swedish welfare state is a product of business interests or of, for example, social democratic social policy. A more detailed analysis of the public-private interplay could probably estimate the impact of the insurance industry more accurately, but the basic dilemma would still be unresolved.

The success of life insurance rested on the close relationship between commercial interests and state interventions. The state on one hand assisted commercial actors as they refined their product and created a market infused with a ‘social purpose’. The insurance industry, on the other hand, provided a consistent ideology for the organization of the social policy arena. The vision originating among commercial actors became intimately associated with the welfare state. This process of welfare-formation should not be seen as the result of a compromise between diverging interests. Rather: it constitutes the very essence of the welfare state.

Disclosure statement

No potential conflict of interest was reported by the author.

Additional information

Funding

This work was supported by the Jan Wallanders och Tom Hedelius Stiftelse samt Tore Browaldhs Stiftelse [grant number W16-0475].

Notes on contributors

Alf Sjöblom

Alf Sjöblom holds a PhD in history from Stockholm university, Sweden. He currently works in three different research areas: welfare state research, business history and marketing. In his thesis ‘The Business of Welfare’ (2016) he shows how the life insurance industry in Sweden influenced public pension schemes in a way that underpinned their own interests and also how business influenced perceptions of security and welfare in general. [email: [email protected]]

Notes

1. ‘The Swedish model’ can also refer to labour market relations in Sweden. But it is in the sense of welfare policy that the concept has been mostly recognized internationally, see Hilson, The Nordic Model. The present article is based on one of three parts of my dissertation, Sjöblom, Trygghet som handelsvara.

2. Cf. Arts & Gelissen, ‘Models of the Welfare State’; Emmenegger, Kvist, Marx & Petersen, ‘Three Worlds of Welfare Capitalism’; Wincott, ‘Social Foundations of Welfare (State) Regimes’.

3. Esping-Andersen, Three Worlds of Welfare Capitalism, 26–29, 31–32, 45–46. Cf. Korpi & Palme, ‘Welfare State Institutions’, 661–687; Åmark, Hundra år av välfärdspolitik, 15–18, 23–25. For a critique of the link between solidaristic welfare policy and the Social Democrats, see Baldwin, Politics of Social Solidarity, 61–63; Swenson, Capitalists Against Markets, 7–10.

4. For examples, see Hort, Welfare State in Sweden; Lundberg & Åmark, ‘Social Rights and Social Security’; Swärd, Edebalk & Wadensjö, Vägar till välfärd; Åmark, Hundra år av välfärdspolitik. In standard history and historical surveys this image dominates completely, see Hedenborg & Kvarnström, Det svenska samhället 1720–2010; Morell & Hedenborg, Sverige – en social och ekonomisk historia; Nilsson, Hundra år av vensk politik; Stråth, Sveriges historia 1830–1920.

5. For example, see Leimgruber, Solidarity Without the State?; Rein & Wadensjö, Enterprise and the Welfare State.

6. For a discussion on this topic, see Lundberg & Tydén, ‘Stat och individ’, 17–28.

7. Baldwin, Politics of Social Solidarity; Swenson, Capitalists Against Markets; Berggren & Trägårdh, Är svensken människa; Trägårdh ‘Civila samhällets välfärdstjänster’.

8. Baldwin, Politics of Social Solidarity, 86–94, 289, 292. Farmers shaped social policy by blocking early proposals, but they did not instigate social reforms.

9. Swenson, Capitalists Against Markets, 247–248, 261–267.

10. Berggren & Trägårdh, Är svensken människa, 10–11, 52–54.

11. Andersson & Eriksson, ‘Compulsory public pension’, 250; Englund, Arbetarförsäkringsfrågan; Heclo, Modern Social Politics, 301–303, 313, 319; Hort, Welfare State in Sweden, 85–86.

12. For examples, see Heclo, Modern Social Politics; Hort, Welfare State in Sweden.

13. Heclo, Modern Social Politics, 182, 217, 227 (quote), 228, 299. The British case, on the other hand, highlights the importance of insurers and the friendly society movement for the development of social policy, ibid., 160, 163, 173–175, 208, 297–299. Cf. Swenson, Capitalists Against Markets, 208, 212–213, 230.

14. For a discussion of the impact of the 1913 statutory pension scheme on insurance sales, see Andersson & Eriksson, ‘Compulsory public pension’.

15. Sweden – along with Denmark – could boast the most insurance companies (per capita) in the world around the turn of the 20th century, Bergander, Försäkringsväsendet i Sverige 1814–1914, 337; Hägg, Institutional Analysis of Insurance Regulation, 270. In the early 1900s savings in life insurance surpassed savings in banks among the working classes, and in the 1930s the total expenditure on life insurance was ahead of bank savings, Bäcklund & Lilja, ‘Variation och förnyelse’, 643–645; SOU 1949:25, 16.

16. Sjöblom, Trygghet som handelsvara.

17. In order to establish a lasting relationship with its customers, the companies sought to gain access to the home environment of the workers. Agents created demand for the product and maintained it through their regular visits to collect payments, see Sjöblom, Trygghet som handelsvara. C.f. McFall, Devising Consumption; Zelizer, Morals and Markets.

18. This vast number of agents equals one agent per 80 citizens, or one agent per 22 households, see Sjöblom, Trygghet som handelsvara, 158.

19. Elmér, Folkpensioneringen i Sverige, 284–285; SOU 1949:25, 15. If we look at contributions the importance of life insurance is even more pronounced. Still in the mid 1950s the total amount of premiums equalled two thirds of the taxes that funded the public pension system.

20. As an example and for statistics, see Esping-Andersen, Three Worlds of Welfare Capitalism, 83–84.

21. Esping-Andersen, Three Worlds of Welfare Capitalism, 79–80, 93, 103–104 (quote, 93).

22. Esping-Andersen, Three Worlds of Welfare Capitalism, 88, 95, 100–102.

23. Morgan & Orloff, Many Hands of the State, 8–10, 13, 17–18.

24. An inherent problem in modelling is that it demands a typology that has to weight different policy areas. As a matter of fact, pension policy is deemed so important that Denmark and Sweden sometimes represent different types of regimes, see Åmark, Hundra år av välfärdspolitik, 272. For different types of welfare typologies, see Arts & Gelissen ‘Models of the Welfare State’, table 39:1. For a more general discussion on the use of models in welfare state research, see Kettunen & Petersen, ‘Rethinking Welfare State Models’.

25. Compare with Hilson´s discussion on how ideas about ‘national exceptionalism’ leads to teleological assumptions about national predestination, Hilson, Britain and Sweden 1890–1920, 20–22, 246. See also Swenson’s critique of the power resource theory, Swenson, Capitalists Against Markets, 3–11, 293–300.

26. The social policy arena refers to welfare policy on the harmful effects on work capacity from invalidity, old age and premature death. Other areas like health care are not part of the analysis. On the importance of pension policy to study the growth of welfare states, see Esping-Andersen, Three Worlds of Welfare Capitalism, 79–80.

27. Edling, Det fosterländska hemmet, 19, 364–365.

28. For example pension insurance, endowment insurance, annuities, tontine insurance, burial insurance and so forth.

29. Sjöblom, Trygghet som handelsvara, 54–55, 59–60. The board of Industrial life insurance companies was generally made up by prominent people with political power.

30. Cf. Morgan & Orloff, Many Hands of the State, 9, 14, 16.

31. Cf. Hacker, Divided Welfare State, 105.

32. In the US the idea of life insurance encountered religious resistance and consequently had to be framed in religious terms, see McFall & Dodsworth, ‘Fabricating the Market’; Zelizer, Morals and Markets.

33. Bergander, Försäkringsväsendet i Sverige 1814–1914, 309, 312–313, 317; Hägg, Institutional Analysis of Insurance Regulation, 170–171.

34. Hägg, Institutional Analysis of Insurance Regulation, 175–176, 206, 210; Lönnborg, Internationalisering av svenska försäkringsbolag, 78–79.

35. In Sweden it was not the insurance principle per se that encountered severe criticism but its commercial foundations, Hägg, Institutional Analysis of Insurance Regulation, 238–239; Sjöblom, Trygghet som handelsvara, 217–222.

36. Sjöblom, Trygghet som handelsvara. Sven Palme was also the grandfather of the social democratic Prime Minister Olof Palme. About Sven Palme, see Englund, Skandiamän och andra försäkringsmän 1855–1970, 110–119. Cf. Henrik Berggren, En biografi över Olof Palme, 20–51, 68–69, 188; Kjell Östberg, Olof Palme 1927–1969, 19–26, 33, 39–40, 55–56.

37. Sjöblom, Trygghet som handelsvara, 86–89.

38. If the insurance industry has been overlooked in the history of the Swedish welfare state, this is also true for friendly societies. This sector was in fact thriving, especially among the working classes, around the turn of the 20th century. By 1912 a thorough regulation of the non-profit sector had been put in place. In the long term the non-profit sector for life insurance more or less vanished from the social policy arena, Sjöblom, Trygghet som handelsvara, 254–258, 280–281.

39. The ‘social purpose’ is especially salient in the collective efforts to market the idea of life insurance and to govern the regulation of the industry. But the social framing also restricted business opportunities in conflict with the perception of a social purpose, Sjöblom, Trygghet som handelsvara.

40. The pension scheme actually allowed for exemption in special cases, see Berge, ‘Pensions-separatismen’, 14–20. Thus, it was the obligation to be properly insured that was universal.

41. Englund, Arbetarförsäkringsfrågan.

42. ‘hvilka åtgärder äro att förorda för befordran af försäkringsväsendets utveckling i vårt land?’, FFT (1881), 16, 18–19; ’Om engelska bolaget Prudential’s industrial branch och om utsigterna för arbetareförsäkring att vinna insteg hos den svenske arbetaren’, FFT (1881), 29–31.

43. Sjöblom, Trygghet som handelsvara, 283–292.

44. Ålderdomsförsäkringskommittén, 22–33, 47. Cf. Gosden, Self-help, 266–269. The post office scheme was introduced by Gladstone in 1864 as an alternative to Industrial life insurance. It was an apparent failure.

45. The complex scheme is an example of the kind of ‘haggle’ that Baldwin see as decisive for social reforms, Baldwin, The Politics of Social Solidarity, 289, 293. None of the actors involved were in a position to fully realize their own interest.

46. Palme, Staten och försäkringsväsendet, 65–72.

47. ‘Om gränserna för privat- och socialförsäkring.’ FFT (1912), 160–161 (quote, 161).

48. ‘Diskussion om kommitterades förslag’ FFT (1906), 172, 174, 181–182; ‘Är det Raabska förslaget en typisk försäkring?’, FFT (1906), 205, 213–214; ‘Om gränserna för privat- och socialförsäkring.’ FFT (1912), 159. Cf. Lehtonen & Liukko ‘The Forms and Limits of Insurance Solidarity’, 42.

49. Palme, ‘Socialförsäkringen och privatförsäkringen’, NFT (1922), 7.

50. ‘Diskussion om kommitterades förslag’ FFT (1906), 172–173.

51. Palme, ‘Socialförsäkringen och privatförsäkringen’, NFT (1922), 5, 9. Cf. ’Diskussion av Lindstedts föredrag’, FFT (1893), 124–125; ‘Diskussion om kommitterades förslag’ FFT (1906), 172.

52. ‘Arbetareförsäkringen’, Gjallarhornet (1898), nr 2; ‘Är det Raabska förslaget en typisk försäkring?’, FFT (1906), 207; ‘Om gränserna för privat- och socialförsäkring.’ FFT (1912), 165–166; Palme, ‘Socialförsäkringen och privatförsäkringen’, NFT (1922), 8.

53. ‘Diskussion om kommitterades förslag’ FFT (1906); Palme, ‘Socialförsäkringen och privatförsäkringen’, NFT (1922).

54. ‘Om den sociala försäkringen.’, FFT (1890), 12–15.

55. ‘Ett försvar för det Raab’ska förslaget.’, Gjallarhornet (1906), nr 22.

56. ‘Social invaliditets- och ålderdomsförsäkring i utlandet.’, FFT (1912), 162–163.

57. The goal was to foster thrift by compulsory practice, see Berge, ‘Socialpolitik och normgivning’, 77; ‘Diskussion av Lindstedts föredrag’, FFT (1893), 129–131. Baldwin exaggerates farmers influence when he states that a universal and tax-financed pension scheme was ‘their solution’, Baldwin, Politics of Social Solidarity, 93–94, 289, 292.

58. ’Den frivilliga försäkringens ställning till den obligatoriska statsförsäkringen’, FFT (1912), 275; ‘Folkförsäkringen i Tyskland år 1909‘, Gjallarhornet (1910), nr 10. Cf. Andersson & Eriksson, ‘Compulsory Public Pension’, 251–252.

59. ‘Diskussion om kommitterades förslag’ FFT, (1906), 168, 171–173; ’Den frivilliga försäkringens ställning till den obligatoriska statsförsäkringen’, FFT (1912); ‘Försäkringsföreningen om det Raabska folkpensioneringsförslaget.’, Gjallarhornet (1906), nr 21; ‘Ett försvar för det Raab’ska förslaget.’, Gjallarhornet (1906), nr 22; .

60. Palme, ‘Socialförsäkringen och privatförsäkringen’, NFT (1922), 7.

61. ‘Om gränserna för privat- och socialförsäkring.’ FFT (1912), 160.

62. Andersson & Eriksson, ‘Compulsory Public Pension’, 261.

63. For examples of how the agents were instructed to ‘make use’ of social insurance, see ‘Vilket inflytande kan den nu antagna statspensioneringen anses hava på den privata folkförsäkringen i Sverige?’, De Förenade Cirkulär-Meddelande (1913), nr 13; ‘Socialförsäkring och privatförsäkring’ (1913), nr 19; ‘Om folkförsäkring och allmän pensionering’ (1915), nr 6–7, 10, 13.

64. For statistics, see SOU 1946:34, 109.

65. Sjöblom, Trygghet som handelsvara, 58–60, 300–301; Wetterberg, Medelklassens guldägg, 37–38.

66. SOU 1932: 36, 117–169 (quote, 152).

67. Overall, Palme was satisfied with the scheme as long as contributions, benefits and the funds were kept at a minimum, ‘Ålderdoms- och Invaliditetskommitténs betänkande. Föredag af ordföranden direktör Sven Palme.’, FFT (1912), s. 206–225.

68. Sjöblom, Trygghet som handelsvara, 291–292.

69. Palme, Staten och försäkringsväsendet, 71; ‘Socialförsäkringen och privatförsäkringen’, NFT (1922), 8; Pensionsstyrelsens utvidgningsplaner.

70. In the mid-1920s the public Insurance Inspectorate also began lobbying to dismantle the contributory part of the pension scheme in favour of occupational pensions managed by the insurance industry, Berge, ‘Pensions-separatismen’, 24–27, 35.

71. Sjöblom, Trygghet som handelsvara, 300–303. Cf. Elmér, Folkpensioneringen i Sverige, 238–242, 278, 282.

72. The rise of the social democratic party was indisputably an important driving force behind the Parliaments ambition to put an end to the era of poor relief. The class-based politics of social policy has been extensively covered in previous research, for examples see Baldwin, Politics of Social Solidarity; Heclo, Modern Social Politics; Elmér, Folkpensioneringen i Sverige; Åmark, Hundra år av välfärdspolitik.

73. Dissatisfaction with the contributory approach fueled Social Democratic critique and subsequent reforms of the scheme, see Heclo, Modern Social Politics, 189, 191–192, 219, 226. Cf. Baldwin, The Politics of Social Solidarity, 65, 83, 93, 290.

74. For the ‘social assistance’ view, see Proposition 1935:217, 38–39, 95–99, 112–114; SOU 1934:18, 125–128. For the ‘state insurance’ view, see Proposition 1935:217, 94–95, 108–111; SOU 1925:8, 127; SOU 1934:18, 321–322.

75. Cf. Proposition 1935:217, 112–113.

76. SOU 1925:8; Elmér, Folkpensioneringen i Sverige, 56–59; Heclo, Modern Social Politics, 216.

77. Sjöblom, Trygghet som handelsvara, 307–309.

78. Proposition 1935:217, 98, 128–131; 136–137; SOU 1934:19, 22. Cf. Hacker, Divided Welfare State, 72.

79. Frivillig Pensionsförsäkring: Arbetare Pensions Kassan. Cf. Harrysson, Arbetsgivare och pensioner, 54–55.

80. SOU 1945:46, 135–140 (quote, 135).

81. Leimgruber, ‘The Three-Pillar Doctrine’.

82. Edebalk, ‘Möllermodellen’, 36–37; ‘William Beveridge och en storstilad plan’, 193; Åmark, Hundra år av välfärdspolitik, 113.

83. Rörande Thules ansökan om koncession å kollektiv tjänstepensionsförsäkring, 122.

84. ‘Solidaristic features’ could obviously also be part of a comprehensive business concept. Cf. Baldwin, The Politics of Social Solidarity, 51–52, 116, 136, 141.

85. Heclo, Modern Social Politics, 229–230; Åmark, Hundra år av välfärdspolitik, 70–71. In 1943 17% of pensioners, despite the supplementary pensions, still had to rely on public assistance.

86. Hacker, Divided Welfare State, 100. Cf. Swenson, Capitalists Against Markets, 213.

87. On the private-public relationship, see Höjer, ‘Medborgares sociala trygghet’, 147; SOU 1946:34, 14, 155–156; SOU 1949:25, 33–34. On Industrial life insurance, see Utredning om försäkringsverksamheten, 9–10.

88. For example Korpi & Palme, ‘Welfare State Institutions’. In 1942 senior civil servant William Beveridge published his report ‘Social Insurance and Allied Services’. This report has been appreciated as the blueprint for the British welfare state implemented in the late 1940s.

89. SOU 1945:46, 57–60, 124; Åmark, Hundra år av välfärdspolitik, 92–93.

90. Baldwin, The Politics of Social Solidarity, 101, 121, 124; Sjöblom, Trygghet som handelsvara, 327–329.

91. Beveridge, Social Insurance and Allied Services, 72–76 (part II §181–192), 275; Voluntary Action, 53–58, 79–81, 291–304; Harris, William Beveridge: A Biography, 379–380, 408–409, 413–415; Sleeman, The Welfare State, 42. The Beveridge plan also met with fierce resistance from the insurance industry. In the end neither the goal to abolish poor-relief or his plan for state administered voluntary insurance were realized, see Baldwin, The Politics of Social Solidarity, 133, 210; McFall, Devising Consumption, 53–57; O´Malley, ‘Imagining Insurance’, 108. For a discussion on the Beveridge report and the insurance industry, see Morrah, History of Industrial Life Assurance, 143–170.