Explaining Employer Support for Welfare State Development in The Netherlands
Dennie Oude Nijhuis
TSEG 16 (1): -52
In recent years, business-centered explanations of welfare state development have challenged conventional perspectives on the welfare state. This new scholarship argues that employer and other business groups have acted as major proponents of welfare state expansion during crucial moments in history. This article investigates the claims of this new scholarship through an analysis of the attitudes of the main employer associations towards the introduction and expansion of social insurance programs in the Netherlands. The article finds no evidence for the claim that these associations supported the introduction and expansion of social insurance programs because they expected to derive economic benefits from the development of these programs. It shows that instances of active employer support for social insurance expansion generally came about as a strategic response aimed at preventing the coming about of more costly alternatives.
Ever since the introduction of the first modern social insurance program in the Netherlands in the beginning of the twentieth century, employer groups have played a central role in the development of these programs. The nature of their involvement seems to pose a puzzle for conventional perspectives on the welfare state, which mostly depart from the assumption of business hostility towards the development of social insurance and related social programs.1 While various studies on the development of the Dutch welfare state have observed instances of organized employer opposition to the introduction and expansion of social programs, few have taken the position that Dutch employer groups’ views of social policy development have always been innately adversarial. In fact, many studies have instead suggested that the main employer associations in the Netherlands ‘on the whole, supported [the process of] welfare state expansion’ or important parts thereof, in particular during the postwar period.2
In recent years, the assumption of business hostility towards welfare state development has also been called into question in the broader literature on the welfare state. According to a large and influential scholarship that originated in the social sciences, but has also attracted much attention from historians in recent years, employer and other business groups have been much more supportive of the introduction and expansion of social insurance and related social programs than traditional studies on welfare state development have realized. This scholarship has supported its claim by arguing that these programs do not only impose costs and labour market rigidities on businesses, but also provide direct and tangible benefits to firms that have been sufficiently important to warrant active business support for their introduction and expansion.3 Other studies have downplayed the importance of these benefits, and argued that the costs of social policies for employers are generally so severe that employer groups are unlikely to have displayed an active interest in their development, even when these policies also provided some advantages to their members.4
While various historically oriented studies have empirically investigated the role of business in welfare state development in recent years, scholarly disagreement on the extent of business support for this development continues to persist. An important reason for this persistence is that business representatives may support the introduction of particular welfare initiatives, and even come forward with expansive welfare initiatives of their own, not because they have a genuine interest in the development of particular social programs, but to gain access to negotiations and prevent the coming about of more generous or otherwise costly alternatives. Moreover, and as recently noted by for instance Paster, under such circumstances business representatives also have a strong incentive to strategically misrepresent their preferences and appear supportive of expansive welfare reform.5 As a result, scholars may be led to believe that support is genuine, while it is in fact strategic. This problem is all the more serious as scholars do not always have access to the internal communications and deliberations of business groups.
Contrary to those of their counterparts in many other countries, the internal communications of the most powerful representatives of the Dutch business community, the main employer associations, are open to scholarly investigation. Moreover, as a result of their early involvement in the development of social insurance programs, these associations have produced a wealth of sources on their involvement in welfare state development.6 The existence of these archives provides a unique opportunity to investigate the claims of the new business scholarship. To date, they have not been used for this purpose. In fact, despite the many references to the involvement of business in the literature on the Dutch welfare state, in-depth analyses of employer attitudes towards social insurance development in the Netherlands are scarce, in particular for the postwar period.
An important purpose of this article is consequently to investigate to what extent the claims of this scholarship can explain instances of organized employer support for social insurance expansion in the Netherlands. As a result of the immensely fragmented nature of Dutch employer organization during most of the twentieth century, it is not possible to investigate the attitudes of all of the peak employer associations that played a role in social insurance development in the period investigated in this article. The analysis will consequently focus on the largest and most influential peak employer associations. These were, respectively, the Association of Dutch Employers (Vereeniging van Nederlandsche Werkgevers, henceforth vnw), which was formed in 1901, the Central Social Employers’ Federation (Centraal Sociaal Werkgeversverbond, henceforth cswv), which resulted from a merger between the vnw’s social department and various other employer associations in 1945, and its 1967 successor, the Federation of Dutch Industries (Verbond Nederlandse Ondernemingen, henceforth vno).7
As a result of their organizational strength, relatively centralized nature, and close involvement in corporatist policy-making, these associations are most likely to have appreciated the possible benefits that social polices provided to employers.8 For practical reasons, the analysis thus pays less attention to the confessional employer associations, which were more numerous and fragmented during much of the twentieth century. This omission should not affect the rigor of the analysis, however, as the new business-oriented scholarship does not suggest that liberal and confessional employer associations might differ in their appreciation of the benefits that social policies provide to firms. The focus on social insurance programs, which have always formed the core of the welfare state, is also in accordance with the deliberations of the new business scholarship.
In addition, the article critically engages with another popular claim on employer support for welfare state expansion, which is also frequently used to explain the overall supportive stance of Dutch employers for this expansion in the postwar period. According to this claim, employer associations operating in highly centralized labour markets sometimes supported welfare state expansion as part of a ‘political exchange’ under which they ‘compensated’ workers for their support for wage restraint policies.9 As a result of the exceptionally long duration of the guided wage policy, a postwar incomes policy that aimed to improve the competitiveness of Dutch industry by moderating the growth of wages and prices and which lasted from 1946 up to the mid-1960s, many scholars have argued that the rapid growth of the Dutch welfare state in the postwar period was the result of a political exchange under which ‘wage restraint was secured in return for […] extensions in state welfare.’10
The current analysis finds little evidence for the claims of the new business scholarship. It shows that most instances of Dutch employer support for welfare state expansion were strategic in nature and served to limit the costs of this expansion for firms. Moreover, it shows why it makes little sense to attribute the overall supportive stance of Dutch employers for the postwar expansion of the welfare state to a political exchange under which they ‘bought’ union support for wage restraint. The analysis does suggest that the employer associations may have found it easier to concede to worker demands for social insurance expansion when the unions offered to finance this out of the margin for social pay (in Dutch: loonruimte), as this reduced their concerns that these demands would increase labour costs. Yet it emphasizes that this practice only emerged after the guided wage policy broke down. Moreover, it shows that this practice was by no means always sufficient to persuade the employer associations to support social insurance expansion.
The article consists of two parts. The first part outlines the claims of the new business scholarship and shows how they will be investigated. The second, empirical, part of the article is divided into two subparts. The first of these investigates employer attitudes towards the introduction of mandatory social insurance programs for the main labour market risks (respectively disability, old age, sickness and unemployment). The second part investigates employer attitudes towards the expansion of these programs in the postwar period. In this second subpart I will return to the claim that the overall supportive stance of Dutch employers for welfare state expansion in this period came about as a result of attempts to buy labour union support for wage restraint.
Analyzing the role of business in welfare state development
Most scholars would probably agree that there are at least three reasons why businesses may be reluctant to support the introduction and expansion of social insurance and related transfer programs. First, they tend to increase labour costs or taxation levels. Second, they may raise workers’ reservation wages and thus their willingness to work. Finally, and in particular with regards to old age pensions, they may threaten the viability of existing private schemes that are used by businesses for investment purposes.11 As a result, much of the historic and comparative literature has traditionally assumed that the relationship between business and the welfare state is by nature an adversarial one. At the same time, however, this view has frequently been challenged over the years. While some studies have done so after observing instances of business consent for the introduction and expansion of social programs, others have done so based on the belief that at least some businesses may also benefit from the development and existence of these programs.
Academic debate on the welfare preferences of business dates back to at least the 1970s, when Marxist-inspired accounts first noted the possibility of business support for social policy development as part of a broader strategy to placate the unions and thereby stabilize and legitimize capitalism.12 It was also around this time that a very different scholarship on neo-corporatism first came to link welfare state expansion to wage restraint, suggesting that business support for the former could be exchanged for labour union support for the latter.13 The recent resurgence in scholarly attention for the role of business in the development of social programs stems from a very different theoretical concern, however, which is to show that the introduction and expansion of social insurance and other social programs could in many countries count on active support from business groups because these groups recognized the importance of these programs in shoring up competitiveness and ‘improve[ing] the operation of markets.’14
While this new business scholarship is quite diverse, it is nevertheless possible to identify two main strands. The first strand explains the coming about of pro-welfare views by dominant segments of business as strategies developed in the context of constraints imposed on them by the existence of labour unions and private or public welfare programs. Examples of such strategies include support for mandatory membership of social insurance programs by large and unionized firms that already provided private provision to their workers – and which are often presumed to have dominated the main business groups by the early to mid-twentieth century – in an attempt to level the playing field vis-à-vis competitors who did not yet do so.15 It also includes situations where business groups supported welfare state expansion in order to prevent competition over fringe benefits.16 Finally, it includes the possibility that businesses in high-risk industries supported the development of solidaristic welfare policies in an attempt to off-load costs onto competitors or society as a whole.17
The second strand, which is part of a broader scholarship on the Varieties of Capitalism, differs from the first in that it argues that traditional social insurance programs can actually ‘serve a productive function’, in particular by overcoming obstacles towards human capital formation.18 According to this strand, these programs are connected to business demands for human capital formation in the following way. Drawing on key insights from institutional economics, it argues that workers need to be convinced to invest in specific skills that are not easily transferable.19 It adds to this claim that employers consequently have an interest in making sure that workers are protected against the risk that their investment in skills does not pay off. As a result, employers operating in markets that rely on specific skills – which according to this literature includes most employers in coordinated market economies like the Netherlands – are for instance expected to have supported the introduction of earnings-related unemployment benefits that offer high replacement rates for skilled workers, provisions that enable unemployment benefit recipients to turn down job offers outside their previous industry or occupation, and benefits that are offered for a relatively long period. Moreover, as individual employers may renege on their commitments, these benefits have to be public in nature.20
While various studies have empirically investigated the claims of this business scholarship in recent years, these have often focused on particular policies and benefits or a relatively narrow time frame. Moreover, while there has been much attention for the role of business in the development of social insurance programs in larger countries like the United States, few studies have focused on smaller West-European countries, even though employers there are expected to have been most supportive of the introduction and expansion of social insurance programs.21 In addition, and as noted above, many empirical studies on the role of employers in welfare state development are based on a limited use of internal employer sources.
To test the claims of the business scholarship and investigate the possibility that instances of business support instead resulted from strategic considerations rather than a genuine interest in social policy development, the following section relies on two types of empirical implications. First, it analyzes internal communications of the three selected employer associations to directly establish their policy positions and the arguments they used to justify these positions. These communications include the deliberations of the Association for Central Consultation of Employer Associations in Labour Affairs (Vereniging Centraal Overleg in Arbeidszaken voor Werkgeversbonden, henceforth coa), a platform created in 1920 on the vnw’s initiative to facilitate coordination among the then existing employer associations on all matters relating to social legislation, and which was incorporated into the cswv in 1945. They also include the meetings of the permanent and ad hoc committees on social affairs created by the cswv and vno as well as the governing boards of these associations, which frequently discussed major social insurance initiatives.
Secondly, it pays close attention to the timing and political context under which instances of employer support for the introduction and expansion of social insurance programs took place. When the employer associations displayed a genuine interest in progressive social insurance reform, we would for instance expect them to have played a proactive role in such reform rather than signal consent once a particular social insurance initiative had been placed on the agenda.22 And vice versa, when instances of employer support followed on the emergence of a parliamentary majority or government in favor of progressive reform, it makes sense to conclude that this support was motivated by strategic considerations. To investigate these possibilities, the following section will therefore analyze whether employer’ attitudes changed over time depending on the political constraints they faced.
Employers and the development of the Dutch welfare state
The prewar and postwar histories of Dutch welfare state differ from each other in at least one way that is of obvious importance for the following analysis. Whereas the debate over welfare reform in the prewar period primarily focused on the desirability of mandatory social insurance programs, the postwar debate mostly revolved around the need to increase the generosity of existing programs. The first part of the analysis therefore investigates to what extent Dutch employers displayed an active interest in mandatory social insurance programs in the prewar period. It covers the years ranging from the turn of the century to the outbreak of World War II, when Dutch parliament introduced mandatory social insurance programs for industrial injuries (1901), long-term invalidity and old-age (1913/1919), non-work-related sickness (1913/1929), and unemployment (1939 - never enacted). The second part focuses on employer attitudes towards attempts to expand the generosity of social insurance programs in the first two and a half decades of the postwar period.
Employers and mandatory social insurance membership in the prewar period
For a variety of reasons, which include the country’s relatively late industrialization, the process of social insurance development started slightly later in the Netherlands than in most surrounding countries. When Germany for instance introduced its first mandatory social insurance program with the Health Insurance Act of 1883, parliamentary debate over the merits of mandatory social insurance membership had only just gotten underway in the Netherlands. And it was not until 1894 that a state committee first recommended introducing a mandatory social insurance program to cater for old age pensioners. By that time, newly established national labour union organizations like the General Dutch Workers’ Association (Algemeen Nederlandsch Werkliedenverbond) and Dutch Workers’ Association Patrominium (Nederlandsch Werkliedenverbond Patrominium) had already spoken out in favor of mandatory social insurance membership for a large variety of labour market risks. In consecutive years, these organizations would continue to lobby actively for the introduction of mandatory social insurance programs.23
Employer representatives by contrast mostly remained aloof from discussions over the possible introduction of a mandatory social insurance program in this period. Their silence can largely be attributed to the absence of centrally organized employer associations at the time. There is nevertheless some documented evidence of employer attitudes towards compulsory social insurance development in this period. In the run-up to the introduction of the 1901 Industrial Injuries Act (Ongevallenwet), a liberal-progressive government consulted a large number of local employer associations on their views regarding its introduction. Of those employers who responded, various remarked that they supported the Act because it would ‘eliminate competitive disadvantages’ for those firms that already offered private provision to their workers. In addition, local labour inspectors noted that many large employers supported the introduction of mandatory insurance against work-related labour market risks for this reason.24 Neither of these two sources provides information on the extent to which these views were shared among employers though.
It becomes easier to establish employer attitudes towards mandatory social insurance membership after the introduction of the Industrial Injuries Act, which prompted the formation of the country’s first employer association with a national orientation, the VNW, by various disgruntled representatives of big industry. Until its dissolution by the German authorities in 1941, the VNW would remain the most influential voice of business in the Netherlands.25 While formed as a vehicle to lobby parliament to retract its proposal for the introduction of an industrial injuries insurance program, its representatives emphasized that they did not oppose the scheme’s mandatory nature, and only objected to the government’s proposal to place responsibility for the program’s administration in the hands of state officials, which they deemed too costly. As the VNW’s founders were large employers who had already created relatively generous private funds of their own, there likely is some truth to this assertion. At the same time, however, it should be noted that the VNW’s insistence that it supported mandatory membership served a strategic purpose, which was to strengthen its campaign against state administration. The VNW’s founder, Dirk Willem Stork, was quite clear on this when he argued that ‘we should distance ourselves from those who have joined our movement because they do not want to provide in the consequences of injuries. I believe that these people have done our movement more harm than good.’26
In subsequent years, the VNW even went so far as to bar radical opponents of any form of state intervention from membership.27 Yet this does not mean that it supported mandatory membership for all types of social insurance programs. The association did not voice strong objections to the introduction of a mandatory insurance program for workers who were unable to work because of long-term invalidity and old age when a confessional government first proposed to do so at the beginning of the twentieth century. Of major importance to the association’s relatively accommodative stance towards this initiative was undoubtedly that it did not threaten to increase costs or affect labour supply in a major way, as the program was to be completely financed by worker contributions and the old age retirement age was set at 70 years. While the long-term disabled were entitled to a benefit before this age, they had to go through a thorough medical assessment. Moreover, as the benefit rate depended on the number of yearly contributions, only the older disabled could obtain a benefit rate that was even remotely sufficient to live on.28 The proposed program passed parliament in 1913, and eventually came into operation in a somewhat different form in 1919.29
The VNW did, however, speak out forcefully against the proposed introduction of a mandatory insurance for non-work-related sicknesses by the same confessional government. The association did so even though nearly all of its members already offered private provision to their workers in case of temporary sickness.30 This opposition shows that employers who offered private provision were not nearly as concerned with eliminating domestic competitive disadvantages as recent writings have suggested. In fact, in a series of publications on the matter, the VNW’s general secretary Henry Smissaert referred to the existence of private sickness funds as a reason to oppose the introduction of a mandatory scheme, arguing that the government should first limit itself to facilitating private initiative in this area.31 By lambasting the government’s decision to finance the scheme from employer contributions, to opt for a relatively generous benefit, and to include medical assistance in addition to cash benefits, Smissaert’s publications also show why the VNW took this stance: a mandatory sickness insurance program was likely to be more costly to employers than voluntary private schemes were.32
In the following decade, the VNW would no longer speak out against mandatory membership. As subsequent governments of various political affiliations had by then produced public sickness insurance proposals, showing that a large majority in parliament favored some type of obligatory arrangement, it is safe to explain the VNW’s acceptance of mandatory membership as an adaption to prevailing political circumstances.33 Instead of questioning the need for a mandatory sickness insurance program altogether, the association now sought to limit the adverse consequences of such a program for employers. It mainly did so by attempting to obtain control over the administration of sickness insurance and by lobbying for the exclusion of medical costs from the program. To strengthen its bargaining position, the association had already stated its willingness to accept partial employer financing.34 In a clear bid to obtain labour union support for industry-based implementation, the VNW would later also offer to administer the program in a bipartite manner. The need to assuage organized labour became particularly important after the adoption of universal suffrage in 1919, which greatly strengthened the position of working class parties in parliament.
Within a year after the adoption of universal suffrage, a confessional government also created the country’s first national consultative body, named the High Council of Labour (Hoge Raad van Arbeid). The creation of this body further affected the nature of employer involvement in social insurance development. After all, it not only meant that the VNW and other employer associations would now be formally consulted on the merits of specific social insurance initiatives; they were now also directly involved in drafting outlines of these initiatives. One the one hand, this provided an unprecedented opportunity to influence policy outcomes in a direct manner. At the same time, however, it demanded that employer representatives worked together with union representatives and other members of the Council.35 And as these actors often had views and interests that differed from those of their employer counterparts, the latter’s involvement in corporatist policymaking frequently required them to make concessions.
As a result, the proposals of the High Council should not be taken as indicative of the first-order preferences of employers. Take for instance the VNW’s support for a new sickness insurance proposal in 1921, which followed on a formal government request for advice by a confessional government a year earlier. In addition to accepting bipartite administration, the employer delegation to the committee eventually agreed to set the benefit rate at 80 per cent of the previous wage, set the maximum duration of the benefit at 26 weeks, and finance the program completely from employer contributions. As this proposal was more generous than the one proposed by the government (which had, for instance, opted for a benefit rate of only 70 per cent), it is tempting to conclude that the employer associations apparently had little difficulty in accepting generous outcomes. Yet such a conclusion would neglect that the Council’s proposal reflected a delicate compromise that came about after much deliberation. In exchange for the above-mentioned concessions, the employer representatives to the Council (among other things) managed to exclude medical costs from the program, which was to be exclusively financed by workers in a separate program.
This exclusion was viewed as an important achievement, and the leader of the employer delegation of the committee responsible for drafting the Council’s sickness insurance proposal, Folkert Posthuma, consequently managed to persuade the executive board of the VNW to support the proposal by arguing that it was cheaper than the government’s plan.36 Not all VNW members agreed with this decision though. During a special meeting on the matter, some advocated that their association should withdraw its support for the High Council’s proposal, arguing that it was too generous and costly. A majority of its members, however, accepted that a rejection of the Council’s proposal might do little to persuade parliament from adopting the features that they disliked while it would undermine the VNW’s credibility.37 Mainly because parliament remained divided over the matter of the scheme’s implementation, the Council’s proposal did not immediately prompt legislative actively. While a first bill on this had already passed parliament in 1913, the sickness insurance program only came into operation in 1929.
In sum, the VNW’s supportive stance towards the introduction of the sickness insurance program in the 1920s can clearly be attributed to the need to adapt to political circumstances. Similar considerations can explain its support for the introduction of an unemployment insurance program one decade later – which dealt with the last major labour market risk for which a mandatory workers’ insurance program was yet to be created. In previous years, the VNW and other employer associations had always been particularly hostile towards the introduction of a mandatory unemployment insurance program. Not only did they regard unemployment as a working class risk for which they bore no responsibility; they also feared that any initiative in this area would undermine workers’ willingness to work.38 As a result, they took a rather principled stance at first. However, as the state began to provide subsidies to union-run unemployment funds during World War I, a practice that the employer associations greatly regretted, they gradually changed their stance.
From the perspective of the employer associations, state subsidized unemployment funds were even more problematic than a mandatory unemployment insurance program, as the former not only provided a strong recruitment tool for the unions, but also gave them full discretion over the administration of unemployment benefits. During the 1920s, they therefore regularly met up in the COA to discuss how to respond to the practice of state funding of union-run unemployment funds. One of the options discussed at the time was to support the introduction of a mandatory unemployment insurance program that would be administered by sectoral employer and union representatives on a bipartite basis. While some employer representatives now supported the introduction of an unemployment insurance government as a preferable alternative to the practice of state subsidized union-run funds, a majority continued to oppose a solution that would force employers to contribute to benefits for unemployed workers on a mandatory basis.39
Of major importance to the rejection was also that the unions strongly resisted sharing control over their unemployment funds with the employer associations. This changed as the economic crisis of the 1930s gradually made it more important for the unions to obtain employer contributions for their funds though. As a majority of employer representatives had by the mid-1930s come to view these contributions as the lesser evil, this paved the way for an agreement between the two sides of industry. After lengthy discussions, they agreed to the introduction of a mandatory unemployment insurance program to which employers would contribute as well and that would be administrated in a bipartite manner. As a result of the outbreak of the World War II, the agreement was never implemented though.40
Explaining employer support for postwar welfare expansion
By the early postwar period, the Netherlands thus had mandatory social insurance programs in place for all major labour market risks with the exception of unemployment. As a result, discussions over social insurance reform would from then on primarily focus on the need to increase the accessibility, duration, and level of existing social insurance benefits. As noted by a host of scholars over the years, the main employer associations, through their membership of a new advisory council, the Social-Economic Council (Sociaal-Economische Raad), actively contributed to the process of social insurance expansion in this period.41 Their cooperative stance should not be taken to mean that they were also strong proponents of this expansion, however. As we will see, their priority remained to limit the adverse consequences of progressive welfare reform to employers. As such, they generally sought to limit the scope of reform and emphasized the necessity for worker (co-)financing of social insurance initiatives. The newly established CSWV motivated the latter by pointing out that ‘sums, which have to be raised for different social insurance programs, are economically just as much part of total labour costs as those sums, which are paid to provide for living expenses […] social costs are part of total wage costs and should therefore be treated as such.’42
The CSWV’s statement clearly shows why it makes little sense to argue that social insurance expansion in this period, let alone employer support for this expansion, rested on a ‘political exchange’ that served to compensate workers for wage restraint under the aforementioned guided wage policy. The obvious problem with this notion is that any attempt to compensate workers for wage restraint would have undermined the very purpose of the guided wage policy, which was to moderate labour costs. As a result, rather than providing the employer associations with an incentive to offer welfare state expansion in a bid to prolong labour acquiescence for wage restraint, it only reconfirmed their commitment to limiting the scope of this expansion. Take for instance the introduction of the first major postwar welfare initiative, the 1949 Unemployment Insurance Act. According to various scholars, this Act came about as ‘quid pro quo for wage restraint’ and was designed to ‘compensate workers for the lower wage levels’ that resulted from the guided wage policy.43 This claim is not only at odds with historical studies, which have shown that the government enacted the Act because it viewed the introduction of a mandatory unemployment insurance program as a matter of unfinished business;44 it is also at odds with the behavior of the employer associations.
As had been the case in the late 1930s, the CSWV did not speak out against the introduction of a mandatory unemployment insurance program when the government first outlined its plans to do so in the late 1940s. Given the government’s firm commitment to its introduction, it simply would not have made sense to do so. Yet the association continued to express worries about the consequences of such a program for labour costs and supply.45 As a result, it sought to limit the program’s generosity and scope by insisting on a low maximum benefit duration of only thirteen weeks, the inclusion of a rather low wage limit to membership, and strict eligibility criteria – including a much more limited definition of the ‘suitable job’-criterion than the one proposed by labour union representatives. In addition, it insisted on premium differentiation as ‘individual firms [will] only be motivated to keep costs down when contribution levels reflect actual business risks.’46 Finally, and in a move that most clearly illustrates why it makes little sense to argue that the Act’s introduction aimed to compensate workers for wage restraint, it demanded that workers contributed to the new insurance program as well. When the labour unions steadfastly refused to accept worker co-financing, the CSWV and the other employer associations successfully lobbied parliament to delay the program’s introduction with several years.47 When the program was finally implemented in 1952, CSWV representatives would gleefully note that they had been quite successful in limiting its generosity.48
Following the introduction of the 1949 Unemployment Insurance Act, the CSWV and the other main employer associations resisted all union attempts in the newly created Social-Economic Council to coax them into supporting increases in the generosity of the new unemployment insurance program. Their ability to do so was facilitated by the long duration of the guided wage policy.49 It was only after this policy collapsed during the early 1960s that the employer associations were willing to consider an increase in the program’s generosity. In exchange, they demanded a more strict definition of the suitable job-criterion – one of the few features of the 1949 Act on which the unions’ view had prevailed. During the mid-1960s, and on the Social-Economic Council’s recommendation, parliament consequently raised the maximum duration of the benefit by five weeks and increased the level of the benefit by a minimal amount.50 Out of sheer frustration with the employers’ intransigent stance, parliament at the same time introduced a tax-financed unemployment insurance provision that effectively extended the maximum duration of the benefit by two more years.51
In subsequent years, the employer associations’ preoccupation with work incentives continued to pre-empt employer consent for increases in the generosity of unemployment protection. The only exception to this obstructionist stance can be found during the mid-1970s, when employer representatives to the Social-Economic Council proposed to increase the duration of the unemployment provision benefit for older unemployed workers and release them from the requirement to look for work. As illustrated elsewhere, however, this initiative was clearly strategic in nature, as it served to deflect union calls for an increase in the duration of the benefit for all workers and an extension of unemployment insurance protection to the self-employed.52 Moreover, soon after the 1976 increase in the maximum benefit duration of the unemployment insurance provision for older workers, the employer associations began to develop various proposals to reduce the generosity of the social insurance system. Many of these would focus specifically on the unemployment insurance system.53
The goal of minimizing the costs of social insurance expansion also guided employer behavior in other welfare domains. In the area of old age pension development, these ‘costs’ included thwarting the expansion of private pension industry. To protect this industry, the associations took a position on pension reform in the early postwar period that was more favorable for pensioners than the government’s stance in one important respect: when the latter put forward a proposal for a new old age pension scheme in 1949, it aimed to include a means-test, which meant that a certain per centage of personal income was to be deducted from the benefit.54 As this income largely originated from private pension funds, such a move would have been disastrous for private pension industry. As a result, the employer associations vehemently opposed the government’s proposal and even went so far as to support a more generous union alternative.55 While they disliked the redistributive nature of the unions’ proposal, which combined earnings-related contributions with flat-rate benefits, they eventually accepted it because of the low level of the benefit. When some representatives of the confessional employer associations expressed objections to the union scheme’s centralized nature in a private meeting with representatives of the CSWV, the latter emphasized that continued resistance to this feature would ‘jeopardize the materialization of consensus in industry,’ which would in turn make it more difficult to prevent the government from going ahead with its means-tested proposal.56 After the employer associations threw their weight behind the unions’ proposal, the government immediately dropped its own proposal, thus paving the way for the introduction of the 1956 General Old Age Act (Algemene Ouderdomswet).
The Act’s introduction also established the principle that improvements to the social insurance system were to be financed out of the margin for pay increases – a development that has been described at length elsewhere.57 Contrary to with the unemployment insurance program, the new old age pension was to be completely financed by worker contributions. While the unions did demand partial wage compensation for these contributions, they did so solely based on the argument that the guided wage policy had kept workers’ wages artificially down for many years.58 When the guided wage policy collapsed in subsequent years, the unions would no longer demand wage compensation for increased contributions. Their willingness to accept full worker financing by no means persuaded the employer associations to accept increases in the generosity of the program though. On the contrary, following the Act’s introduction, the employer associations resisted all union attempts to increase the level of the benefit. According to the CSWV this was because any ‘further extension of the AOW [the public benefit] means that a greater part of pension provision for employees is financed by public provision, which leaves less room for additional supplements provided by industry.’59 To prevent such an increase from coming about they first defended the view that the public pension did not have to provide a subsistence level of income, and then resisted the introduction of a wage-indexation clause under which the level of the benefit would automatically increase in line with average wage increases.60 They also objected to all incremental increases in the level of the benefit, including parliament’s 1965 decision to raise the level of the benefit from 50 to 70 per cent of the minimum wage for married couples and to 70 per cent of a full benefit for single pensioners.61
Then, four years later, the employer associations radically changed their stance by putting forward a proposal to raise the level of the benefit to that of the ‘social minimum’ for married couples, and introduce mandatory membership of occupational pension schemes – something that they had strongly resisted in previous years. Both the timing of this proposal and internal notes on the matter suggest that they purely did so for strategic reasons. Just before the employer associations came forward with their initiative, the unions had proposed to introduce a Swedish-style second pension tier on top of the existing pension scheme. While some employer representatives viewed this proposal as nothing more than an attempt to lure them into making concessions, a majority of them saw it as sufficiently threatening to come up with a counterproposal.62 When various representatives of one of the confessional employer organizations once more raised principled objections, which now focused on mandatory membership of occupational pension schemes, its own leadership responded by pointing out that they ‘lack[ed] a stance that leads to a solution in the short term that is acceptable to the unions. By rigidly sticking to this viewpoint, the unions will only feel it necessary to strive for a completely centralized system.’63 In the end, all of the main employer associations supported the proposal. Once they did so, parliament immediately adopted it.
Increases in the generosity of public protection against sickness and long-term disability did not threaten to crowd out investment-generating private programs in a major way; nor were they viewed as affecting work incentives in a similarly strong way as unemployment protection did. As a result, the employer associations displayed a greater willingness to accept increases in the generosity of the sickness and invalidity insurance programs than they had done with the unemployment and old age insurance programs during the first decades of the postwar period. Their willingness to support these increases became clear quite early in the postwar period, when the employer associations gave their consent to the 1948 report of a tripartite committee that called for a massive increase in the generosity and scope of both programs: in addition to proposing to increase the duration of the sickness benefit from six months to two years and extending it to disabled workers as well, the report proposed to increase the level of the invalidity benefit and improving access to it by reducing the period under which workers need to have contributed in order to obtain a benefit.64
Despite this early consensus on a rather far-reaching set of improvements to both programs, discussions on wholesale reform of the sickness and invalidity insurance programs did not get underway until the second half of the 1950s.65 As these discussions turned out the be much more complex and lengthy than expected, parliament decided to introduce a temporary invalidity program in 1962, before it merged the industrial injuries program and invalidity programs into a single disability insurance program through the Act on Disability Insurance (Wet op Arbeidsongeschiktheidsverzekering) in 1967. In the same year, it also introduced a new sickness insurance program, which was to cater for sicknesses and disabilities that lasted up to a year. The outcome, however, was exceptionally generous as the two programs entitled all workers to a benefit that equaled to 80 per cent of the previous wage (or a per centage of that in case of partial disability), regardless of cause of injury, from the moment they started working for as long as their disability or sickness lasted.
The employer federations do not seem to have had major problems with the generosity of the two programs. When the chairman of the cswv social insurance committee in the late 1950s for instance asked its committee members’ opinions on the union demand to set the benefit at 80 per cent of the previous wage, he found that they did not object to this, with some specifically arguing that a lower benefit level was simply ‘insufficient’ for a fully-disabled workers.66 They also agreed with the principle of awarding benefits to the partially disabled.67 In fact, only two union demands were firmly rejected by them. The first was to introduce a minimum benefit rate into the sickness insurance that equaled to the minimum wage. The newly established vno rejected this because it feared that low paid workers would then be much more likely to call in sick.68 Second, the vno resisted pleas to opt for contribution levels that were uniform among industries – just as its predecessor had done with the unemployment insurance program. When parliament sided with the labour unions on this the vno was incensed and, together with its confessional counterparts, continued to insist on rectifying what it viewed as a crucial mistake in subsequent years.69
It would, however, take various decades before the parliament was willing to lend a wiling ear to this complaint. By that time, the unforeseen increase in the number of disability benefit recipients had already forced various governments to reduce the generosity of the sickness and disability insurance programs in incremental steps. Driven by concerns over the financial consequences of this, the employer associations quickly became the strongest proponents of these cutbacks.70
Since the turn of the century, the role of business has returned to the forefront of academic analysis on the welfare state. This resurgence of scholarly interest has been fuelled by an intense debate over the nature and extent of business support for welfare state expansion. An important reason for the continuation of this debate is that business groups – like most political actors – are prone to display strategic behavior. As such, actions and statements that may at first sight seem indicative of a supportive stance towards expansive welfare reform may really be designed to limit its costs. This article suggests that the aim of limiting its costs also explains most instances of business support for social insurance development in the Netherlands. The article focused on the most powerful representatives of Dutch business, the main employer associations. It showed that the attitudes of these associations towards the introduction and expansion of the main social insurance programs were mainly shaped by concerns that these programs would undermine work incentives, raise labour costs, and replace private pension schemes. Most of the time, this meant that they sought to limit the scope of expansive reform. The only instances in which the employer associations took a pro-active role in proposing expansive welfare reform can be explained through strategic behavior that served to prevent outcomes that were even more costly to their members.
The article thus found little evidence for the claims of the new business scholarship. On the contrary, many of the findings directly contradict these claims. The vnw’s opposition to the introduction of a mandatory sickness insurance scheme for instance shows that employers who offered private provision were not nearly as concerned with eliminating domestic competitive disadvantages as recent writings have suggested. The strong resistance of the employer associations to the introduction and expansion of unemployment insurance protection is in turn clearly at odds with the claim that employer attitudes towards social insurance development are primarily shaped by their interest in facilitating human capital development. And although there certainly must have been employers who had an interest in supporting premium levels that were uniform among industries in order to off-load costs on competitors, the main employer associations consistently emphasized the need for premium differentiation in order to reduce the risk of moral hazard. The latter points to a problem that has already been raised in other studies: especially in societies where employers are nationally organized and highly centralized, sectoral interests in leveling the playing field or off-loading costs may not necessarily result in employer support for social policy expansion as there will be many firms who stand to lose from this as well.
Finally, the article showed why it makes little sense to argue that employers in the Netherlands frequently or even occasionally supported social insurance expansion in an attempt to buy labour union support for wage restraint. The main problem with this claim, as the cswv itself pointed out, is that social insurance contributions are an integral part of total labour costs, which means that any attempt to compensate workers for wage restraint by expanding their social rights undermines the very purpose of a wage restraint policy, which is to limit the growth of labour costs. It is therefore not surprising that the rapid growth of the Dutch welfare state only took place after the guided wage policy collapsed in the 1960s. While this collapse and the practice of financing social insurance expansion out of the margin for pay increases, as we have seen, did not necessarily prompt greater employer support for welfare state expansion, it certainly may have facilitated the practice of welfare state expansion in other ways. How this worked exactly falls out of the scope of this article, however, and presents an avenue for future research.
About the author:
Dennie Oude Nijhuis (Ph.D., 2009) is senior researcher at the International Institute of Social History (IISH) and assistant professor at the Institute for History. He specializes in the comparative political economy of labour markets and welfare states and the political economy of European integration. He is the author of Labor Divided in the Postwar European Welfare State (Cambridge 2013) and Religion, Class and the Postwar Development of the Dutch Welfare State (Amsterdam 2018). His current main research project involves a comparative analysis of twentieth-century labour market development in Belgium, Germany, the Netherlands, Sweden, the United Kingdom, and the United States.