Holland’s single most important weapon : a CPE-inspired common sense analysis of the consensus on wage restraint in the Netherlands

(1)

‘Holland’s single most important weapon’

A CPE-inspired Common Sense Analysis of the Consensus on Wage Restraint in the

Netherlands

Tom van de Haar 10379029

09-07-18 De Deken

(2)

2

Abstract

Although wage restraint is often used as an explanans for the Dutch ‘miracle’, it is rarely viewed as an explanandum: that which needs to be explained. This thesis investigates how wage restraint became the ‘go-to’ solution during economic downturns in the Netherlands. Despite the occasional reluctance of trade unions to accept lower wage increases, the consensus of the social partners on wage restraint and its apparent benefits throughout the last few decades has been remarkable. To understand why trade unions have subscribed to a logic that not necessarily seems beneficial for its members it is necessary to adopt a CPE-inspired common sense framework. With the help of this framework, this thesis argues that the consensus of the government and the social partners on wage restraint has been underpinned and shaped by a competitiveness discourse, which over time has been incorporated into common sense. This competitiveness discourse has been propagating an unequivocal positive link between wage restraint and employment. The incorporation of the logic of the competitiveness discourse into the common sense of a variety of actors opened up space for a synthesis of the different perspectives, and subsequently fostered a broad-based consensus on wage restraint.

(3)

3

Introduction

In 1995 the Netherlands Bureau for Economic Policy Analysis (CPB) claimed that wage restraint has been “Holland’s single most important weapon in international competition (as cited in Hemerijck & Visser, 1997: 26)". The quote perfectly captures the broad-based consensus on the benefits of wage restraint that has dominated the Dutch wage bargaining process in the last 40 years. Wage restraint is, as an economic strategy, strongly embedded in the culture as well as the formal institutions of the Netherlands.

The emphasis on wage restraint has had clear consequences for labour’s share of income. Following the majority of capitalist economies, the Netherlands experienced two interrelated phenomena from the late 1970s onwards: a downward trend in labour’s share of income, and the emergence of a persistent gap between real wage growth and labour productivity growth. Although once regarded as a stylised fact (see Kaldor, 1957; or Keynes, 1939), the alleged stability of the proportion of national income received by labour and capital has in the last few decades been seriously challenged by a vast amount of research (see i.a. OECD, 2012; ILO, 2013). After the post-war period until the 1970s real wages in the Netherlands and similar western capitalist countries typically grew in proportion to labour productivity (van Klaveren, Salverda & Tijdens, 2009: 414). However, after the 1970s labour was no longer able to translate productivity growth into real wage growth (Lavoie & Stockhammer, 2013: 3; Stockhammer, 2013: 1). The extra real income generated as a result of increasing levels of labour productivity has, instead, been largely appropriated by capital, which resulted in big shifts in the national income away from labour and towards profits (Mitchell, 2013: 6-7; Lapavitsas, 2013: 190).

In the Netherlands the real wage-productivity nexus has deliberately been broken, since supressing real wage growth forms an integral part of the Dutch export-led growth strategy. The signing of the famous Wassenaar Agreement in 1982 heralded the (re)introduction of wage restraint, and marked a turning point in the Dutch political economy. The social partners subscribed to the logic that increasing the profitability of companies and improving the international competitiveness of the Dutch economy was the way out of the turbulent economic times of the early 1980s (Visser & Hemerijck, 1997: 13, 16; Hemerijck & Schludi, 2000: 159). In the subsequent decades wage restraint became an indispensable tool for promoting job growth, and was almost continuously pursued by both the government and the social partners. Wage formation in the Netherlands has since 1982 been characterised by voluntary wage restraint in exchange for jobs, resulting in a real wage growth of around 0% between 1982 and 2015 (de Beer & Keune, 2017: 223; see also Storm & Naastepad, 2013: 101).

(5)

5

However, an increasing number of voices argue either that wage restraint has not been the best strategy for the Netherlands (see i.a. Kleinknecht et al., 2006; Vergeer & Kleinknecht, 2007; Storm & Naastepad, 2013) or that in the current economic climate there is no need to restraint the wages (see i.a. Eggelte et al., 2014). Even the president of the Dutch Central Bank Klaas Knot has in the last few years repeatedly argued for stronger wage increases (de Boer, 2017). These pleas have so far not resulted in a significant rise of real wage growth. The consensus of the social partners on wage

restraint and its apparent benefits throughout the last few decades has been remarkable. Wage

restraint enjoys an almost unquestionable legitimacy in the Netherlands. The wage bargaining process has been dominated by concerns surrounding the international competitive position of the Netherlands (see i.a. Salverda, 2013: 371; Visser & Hemerijck, 1997: 26, 109; Becker, 2001: 26), which begs the question where these specific concerns originated from, and how these concerns were able to secure a place at the forefront of the Dutch social dialogue.

Different explanations for the Dutch consensus on the wage restraint strategy have been put forward. The majority of these explanations offer either a depoliticised narrative that frame wage restraint as a long-awaited sensible decision, or stress that trade unions simply lacked the power to oppose to the wage restraint wishes of the employers. However, the first type of explanations cannot account for the shaky theoretical and empirical foundation on which the proclaimed benefits of wage restraint are build; while the second type of explanations fall short in explaining why the strategy of wage restraint has been embraced by the social partners in the Netherlands for the last 40 years but not to a similar extent in the other countries.

To understand why trade unions have subscribed to a logic that not necessarily seems beneficial for its members it is crucial to adopt the Gramscian notion of common sense. This thesis argues that the consensus of the government and the social partners on wage restraint has been underpinned and shaped by a competitiveness discourse, which over time has been incorporated into the common sense of the relevant actors. The hegemonic discourse that presents wage restraint as unambiguously beneficial for the general interest gained legitimacy and became throughout the years widely accepted. The discourse has been an essential instrument in producing and sustaining the necessary support for the wage restraint strategy. The Dutch preoccupation with (unit) labour costs and the profitability of firms has been premised on the pervasive beliefs that the competitiveness discourse propagates.

The thesis is structured as follows. The first chapter examines the debates in the literature. The second chapter presents the research objective and the research questions. Chapter three discusses

(6)

6

the research methods, and explains the guiding framework for the analysis. The fourth chapter presents the analysis. Lastly, chapter five offers a conclusion.

(7)

7

1: Theoretical Framework

A number of economic shocks, such as the end of the Bretton Woods system (1971), the first (1973) and the second (1979) oil crisis, and the subsequent global economic recession of the early 1980s, left many Western capitalist economies in dire straits. For the Netherlands the period was characterised by economic difficulties and societal conflicts. Because of the severity of the problems, the term ‘Dutch disease’ was coined to describe the political and economic stagnation of the Netherlands during the 1970s and early 1980s (Visser & Hemerijck, 1997: 9). The country experienced a high level of unemployment, a big current account deficit, a declining level of competitiveness, falling growth rates, and a rapidly rising public debt (Hemerijck & Schludi, 2000: 158); while on a societal level fierce labour conflicts dominated the labour relations (Scharpf, 2000: 60; van der Velden, 2009: 84, 143).

To combat the economic deterioration the Dutch reformed their welfare state, and reintroduced their post-war strategy of wage restraint (Visser & Hemerijck, 1997: 9). This strategy included a chain of reasoning that started with wage restraint and ended with job growth. Wage restraint was therefore regarded as the correct medicine to cure the ‘Dutch disease’ of high and sclerotic unemployment, which was over 11% in 1982 (Storm & Naastepad, 2013: 101).

The ‘sick man’ image of the Netherlands changed profoundly in the second half of the 1990s. While the majority of western economies experienced stagnation, the Dutch economy was one of the few that was able to perform comparatively well (Becker, 2001: 19; see also Engelen, 2016: 116). The relative success of the Netherlands had not gone unnoticed. In the end of the 1990s international media praised the Dutch corporatist model for its ability to function as a ‘jobs machine’ (Visser & Hemerijck, 1997: 9; Marx, 2007: 383-384).

The transformation from the Dutch disease of the 1970s to the Dutch miracle in the 1990s, and the international acclaim the Netherlands received for its socio-economic turnaround, grasped the attention of many scholars in the late 1990s and early 2000s. These scholars have begun to offer important theoretical and empirical evidence, insights, and explanations with regard to the importance and consequences of, and the consensus on, the wage restraint formula. This chapter will examine these debates in the literature and develop a framework of analysis that shapes the empirical arguments of the thesis. This literature review is composed of three parts. The first part focuses on the reasoning behind the Dutch strategy to restrain the wages. The second part is concerned with the macroeconomic and social consequences of the wage restraint formula for the

(8)

8

Netherlands. While the third part critically explores different arguments for the Dutch corporatist consensus on wage restraint.

1.1: The wage restraint-employment nexus

Governments have different non-cooperative strategies at their disposal to improve the competitiveness of their economy. To diminish the relative production costs, governments could for instance opt for ‘competitive devaluation’ (or ‘external devaluation’), which entails measures that will cause the exchange rate of their currency to fall relatively to the currency rates of other countries (Dufresne, 2013: 118). Another option is ‘competitive disinflation’ (or ‘internal devaluation’); measures that try to suppress inflation or wages. The idea is, in the case of suppressing wages, to ensure that the nominal wage growth is lower than the productivity growth and preferably lower than the nominal wage growth of the neighbouring countries. This will, as the logic goes, lead to a decrease in the relative unit labour costs, and an increase in the competitiveness of the economy. The increased competitiveness will improve the current account position of the country, and reduce the level of unemployment (Blanchard et al., 1993: 13; Clift, 2003: 182; Blanchard, 2007: 2, 7).

The export-led growth strategy of the Netherlands incorporates the idea of ‘competitive disinflation’. Suppressing wage growth forms a crucial element of the Dutch strategy. The envisioned effects of wage restraint include competitiveness as well as job creation. Wage restraint will, following the underlying logic, foster job creation via three different channels. Firstly, by ensuring that the relative costs of nation’s firms are relatively lower compared to their foreign rivals, the nation’s firms have a comparative advantage, which will lead to higher output and therefore employment. Secondly, wage restraint should induce firms to shift from capital- to labour intensive production. The demand for labour will as a result increase, and the employment level will rise. Thirdly, if the nominal wage growth is lower than the productivity growth, the profitability of the nation’s firms will rise. The more capital friendly climate of the country will ensure that the capital remains within borders, and will be invested in labour intensive production (Naastepad & Kleinknecht, 2004: 138). The policy of wage restraint will in the short run reduce the domestic purchasing power, but is offset by the increased levels of export, and the employment generating effects that comes with it. The Dutch strategy is therefore characterised by a wage restraint-employment nexus; inspired by neo-classical economic theory, a positive causal link between wage restraint, competitiveness, and employment growth is envisioned.

(9)

9

The Dutch strategy to pursuit competitiveness and employment growth via wage restraint contains substantial flaws. The competitive corporatist strategy of the Netherlands to restrain the wages could yield advantages if the domestic wage demands are relatively lower compared to their competitors. This kind of ‘beggar-thy-neighbour’ policy is not particularly useful if a great number of countries employ the same strategy. It could even turn into a collective ‘beggar-thyself-and-thy-neighbour’ situation, with minor competitiveness improvements and major domestic effects (see Lapavitsas et al., 2010). The purchasing power and the domestic demand will drop, while employment in the countries will not increase, since the competitive gains cancel each other out (Blecker, 1989: 407).

There are, however, more fundamental shortcomings. Restraining the wages, or even lowering them, does not necessarily improve the competitiveness of a country. The unit labour costs depend on two factors: the total labour costs and labour productivity. However, labour productivity is not constant, and could possibly be negatively affected by wage restraints. Higher wage growth could potentially have the opposite effect on productivity growth (see i.a. Vergeer & Kleinknecht, 2007; Storm & Naastepad, 2011). Higher wages could, for instance, heighten the pressure for firms to introduce labour-saving techniques (de Beer, 2001: 325-326). Furthermore, they could also serve as what Streeck (1991, as cited in Streeck, 2005: 271) describes as a ‘benevolent constraint’: a situation in which an employer feels more obliged to invest in skills and high-quality products. Because of the intricate interrelationship between labour costs and productivity it seems that a clear outcome is not a given.

Furthermore, the logic behind the Dutch wage restraint strategy reduces international competitiveness, as it is often done, to the growth of relative unit labour costs (Fagerberg, 1988: 355). Despite the lack of an agreed upon definition (Boltho, 1996: 1), the popular measure of relative unit labour costs is too simplistic for two reasons. Firstly, wages are not the only cost component. Import and export respond more strongly to changes in the gross output price, of which unit labour costs only constitute a relatively small part (Storm & Naastepad, 2015: 14). Secondly, and more importantly, export market shares are less dependent on relative costs than often is argued (see ‘Kaldors paradox’ in Kaldor, 1978; Carlin, Glyn, Van Reenen, 2001). Competitiveness seems to consist of several aspects, one of which being the price. One of the main factors determining the international competitiveness of an economy is non-price technological competitiveness; the relative unit labour costs are therefore less important for competitiveness than is often stressed (Faberberg, 1988; Storm & Naastepad, 2016).

(10)

10

So far the point has not been to deny that there may be a positive relationship between wage restraint, competitiveness, and employment growth. Instead, the aim was to show that the underlying logic of The Dutch wage restraint strategy to foster employment growth is far from robust. Although the relationship between the different elements is less straightforward than often is assumed or presented, it is possible that the strategy has had a positive impact. The belief that the wage restraint strategy has been a success or ‘Holland’s single most important weapon’ enjoys widespread popularity among the Dutch society. The focus of the next part will therefore be on the economic and social consequences of the wage restraint strategy for the Netherlands.

1.2: The consequences of wage restraint

Despite the traps and pitfalls of wage restraint, the Netherlands adopted the strategy to overcome the economic problems of the 1970s and early 1980s. The Dutch response and its apparent success attracted much attention, and quickly became the corporatist revelation of the mid-1990s. In a relatively short time span the Dutch reform became known to be a model for effective labour market adjustment. Despite the fact that there is no direct and unambiguous positive relationship between wage restraint, competitiveness, and employment growth, it has been one of the primary concerns in the Dutch wage bargaining process. The signing of the Wassenaar Agreement in 1982 is retrospectively often regarded as the return of the wage restraint strategy. Ever since the agreement, Dutch unions have persistently placed jobs before income (Hemerijck & Schludi, 200: 162). The social and economic consequences of this strategy are widely debated. After the international appraisal of the late 1990s positive studies dominated the scientific debate, though later on a number of scholars started to dispute the importance and benefits of wage restraint. The Wassenaar Agreement resulted, according to i.a. Hemerijck & Visser (1997), Visser (1998), Hartog (1999), Levy (1999), Nickell & van Ours (2000), and Scharpf (2000), in a relative success. The strategy to restrain the wages clearly succeeded; the unit wage costs in 1996 were at the same level as in 1981. Germany experienced over the same time period an increase of 40%, while within the EU it rose on average 15% (Hartog, 1999, 464-465). Labour’s share of income fell as well; from circa 65% in the early 1980s to 56% during the period 1984-2000 (Storm & Naastepad, 2013: 101; see also Scharpf, 2000: 61) Furthermore, the Dutch unemployment rate fell from almost 15% in the early 1980s to 3.8% in the mid-1990s, the lowest of all the EU countries. In addition, the labour force participation experienced a spectacular surge from 52% in 1983 to 66% in 1996 (Levy, 1999: 258). Hemerijck & Visser (1997: 28, 42, 181) note that the rate of job growth was higher, and the fall in the rate of unemployment was steeper compared to other European countries. The wage restraint

(11)

11

strategy restored the profitability of businesses, led to increasing levels of export, and kept a growing number of people on the payroll. Although they concede that labour productivity growth has been low, and that the majority of the newly created jobs are part-time, they see wage restraint as the foundation and as a success story.

Though there have also been critics who claim that the relationship between wage restraint and the Dutch labour market success is less straightforward than often is assumed. Becker (2005: 1085-1089) asserts that wage restraint was not the foundation on which the Dutch ‘employment miracle’ of the 1980s and 1990s was built; the new jobs that emerged were not the result of ‘wage-restraint-induced export’, but were caused by wage dispersion and the rise of (female) part-time work. Furthermore, the new jobs emerged in the service sector, primarily in branches that do not compete on an international level. Salverda (1999: 11, 14) notes that the Dutch export performance after 1982 was mediocre, while consumption and import was strikingly low. The current account surpluses the Netherlands experienced in this period can therefore not be the result of rising net exports, but could be explained by falling imports. In addition, Kleinknecht & Naastepad (2005: 1200-1201) argue that wage restraint resulted in higher job growth, but was at the expense of productivity growth and innovation. However, in the long run this makes the Dutch economy more vulnerable. While Storm & Naastepad (2013: 102) argue that wage restraint had a negative impact on aggregate demand. The modest economic growth performance of the Dutch economy was not due to wage restraint, but was the result of an increasing reliance on export demand, and a domestic consumption demand fuelled by debts and (housing) wealth increases.

Despite the theoretical and empirical shaky foundation of the wage restraint-employment nexus, the social partners and the government time and time again highlighted the importance of wage restraint and its apparent benefits. Furthermore, it is not only possible to question the dominant positive narrative of the economic impact of the Dutch strategy, but also the social costs it incurred. Delsen (2002: 81) and Marx (2007: 393, 395) point out that the distribution of the consequences of the wage restraint strategy has not been even across the society. People at the lower end of the income distribution immediately felt the income effects of the strategy; while the new jobs that emerged ended up at the middle income households. The two-earner middle income households that emerged as a result experienced an increase in their standard of living. The lower income households fell behind, and even faced the risk of experiencing declining living standards.

(12)

12

1.3: The explanations for the consensus on wage restraint

As we have seen, the Dutch wage restraint export-led growth strategy has serious theoretical shortcomings, while the economic and social consequences of this strategy are widely debated. Both the theoretical and empirical evidence do not unambiguously show positive results for the Netherlands; which begs the question why trade unions subscribed to the wage restraint strategy in the first place. Different explanations for the initial and continued consensus on wage restraint as the solution to overcome economic problems have been put forward.

Several explanations underline the importance of the high level of unemployment that the Netherlands faced during the early 1980s. For instance, De Beer (1997: 328) concludes that the consensus is located in the severe recession that the Netherlands experienced during this time. As a result of the recession a lot of permanent full-time jobs of trade union members were for the very first time at risk. This pushed trade unions to accept wage restraint in exchange for i.a. shorter working hours. A similar explanation is put forward by Green & Pedersen (2001: 146; for another example see Hassel, 2006: 86), who argue that the Wassenaar Agreement is the result of the government’s ability to launch a credible threat by showing its willingness to impose solutions to resolve the dire economic circumstances that neither side preferred. These explanations, however, are only aimed at the consensus of the 1980s, and are difficult to reconcile with the continued reliance on wage restraint in the following decades.

One recurrent theme in the literature is the ‘polder’ explanation, which stresses the importance of the Dutch consensual corporatist tradition. The foundation of the so called Dutch success story is the shift from competitive to cooperative social relations. Both parties realised the importance of wage restraint for the viability of the Dutch economy, which then was exchanged for new jobs in the Wassenaar Agreement. These explanations offer a depoliticised narrative that frame the trade unions acceptance of wage restraint as a long-awaited sensible decision. The most famous example of a ‘polder’ explanation is the often cited work A Dutch Miracle by Visser & Hemerijck (1997). The book was published at the peak of the international appraisal for the Dutch model, and quickly became the dominant storyline. Visser & Hemerijck (1997: 13) put the learning capacity of trade unions at the centre of their explanation. Under pressure of unemployment, the trade unions learned that a higher profitability for capital was required to combat the high levels of unemployment and to restore economic growth. The severe economic crises the Netherlands had experienced changed the mentality of the trade unions, and resulted in the acceptance of the new demands of the world economy (Visser & Hemerijck, 1997:13; Hemerijck & Schludi, 2000: 138). Hemerijck & Visser (2000: 237, 252) note that with high levels of unemployment in the beginning of

(13)

13

the 1980s trade unions were in no position to demand real wage growth, and could not afford a wage conflict. Instead, Dutch unions decided from 1982 onwards to place jobs before income. The new insights trade unions gained as a result of the learning process, led to the ‘revitalisation’ of the Dutch consensual corporatist model in the subsequent decades (Hemerijck, Unger & Visser, 1999: 262).

Although implicitly mentioned, the aforementioned explanations do not necessarily stress the importance of power with regard to long-term implementation of wage restraint in the Netherlands. The analysis of Hemerijck & Visser (1997) relies heavily on the idea that trade unions had to learn to accept the correct solution to the economic malaise of the early 1980s, which was wage restraint. Even if we accept that, in hindsight, the solution was correct; to what extent can we ascribe the social consensus on wage restraint to a genuine learning process? According to Hemerijck & Visser (1997: 58) a learning process ensues when the involved actors adjust their policy practices and preferences based on their experiences with previous policy failures. However, they also note that “those who lose power must learn (p. 78)”, which stresses the importance of shifting power relations. A learning process therefore de facto entails a learning curve favouring the dominant actor. One could argue that the trade unions simply did not have the necessary power to oppose to the wishes of the employer’s organisations. As a result of the high levels of unemployment and the declining membership numbers, Dutch labour relations changed in favour of the employer’s organisations. However, these trends did not only occur in the Netherlands, but have been a striking development in the majority of western capitalist countries (Becker, 2001: 470). Obviously power matters, but power alone is not able to explain why the strategy of wage restraint was embraced by the social partners in the Netherlands, but not to a similar extent in the other countries. Additional factors must therefore be at play.

Becker (2001: 28) notes that the profitability of Dutch firms was not particularly low compared to other countries, and that the unit labour costs of Dutch firms were below average in the late 1970s and early 1980s. Furthermore, he stresses that there is no clear causal link between wage restraint and job creation. He therefore argues that the learning argument of Hemerijck & Visser is weak; instead of a learning process, the trade unions reluctantly accepted the hegemonic interpretation of the benefits of wage restraint. As a result of the consensual nature of the Dutch corporatism, actors will have to follow ‘the rules of the game’, and try to avoid conflict. In case the social partners are not able to agree, the actor in the weaker position will need to submit to the dominant view, either the one supported by science or the government. For the Dutch case this meant that the trade unions had to submit to the hegemonic interpretation of the general interest; a particular view of the

(14)

14

general interest that was broadly supported by the government (Becker, 2001: 28; Becker, 2005: 1086). Van Apeldoorn (2009: 214) argues that the main reward trade unions, the weaker party, gained from the Wassenaar Agreement was that it remained one of the key players in the socio-economic governance of the Netherlands. However, one could question what the benefits of a seat at the table are when the consensual corporatist exchanges are skewed towards capital. As Becker (2001: 29) notes, the consensualist culture of Dutch corporatism could turn out to be a mechanism of sustaining hegemony.

To explain the Dutch corporatist tradition and the choice for wage restraint a number of scholars stress the importance of the perceived vulnerability of the Dutch economy. Katzenstein (1985: 191) argues that small European countries tend to resort to democratic corporatism as a response to increasing international pressures. The economies of these small countries tend to be open, and vulnerable. The lack of global economic power, and the export-dependent nature of their economy, creates a sense of vulnerability, which in turn provides a motive for policy coordination. To deal with the harsh economic conditions, wage restraint is therefore more often introduced in these types of countries (Delsen, 2002: 10). However, these explanations do tell us exactly why trade unions accepted the narrative on the benefits of the wage restraint strategy for the last few decades. This is especially puzzling when taking into account that the signing of the Wassenaar Agreement was not necessarily in the interest of the employees that trade unions represent (Bruff, 2010: 628).

Bruff (2010; see also Bruff, 2008) adopts the Gramscian notion of ‘common sense’ to explain why Dutch trade unions have been accepting wage restraint for the last 40 years. He (2010: 622) argues that common sense is the basis for how humans make sense of the world. At the centre of social and political conflicts different social groups struggle to “render their conception of the world common sense and thus the basis for how humans in that society think and act (625)”. In the Netherlands the consensus on wage restraint is rooted in the common understanding of the vulnerability of the Dutch economy. Bruff shows, via interviews with representatives of political parties and trade unions, that the relevant actors assume that the economy is vulnerable to external economic factors. It is this shared understanding of the world that binds the interests of the principal actors. As a result of an overlap in the intersubjective versions of common sense of the different actors, wage restraint is regarded as a logical solution, since nobody would want to jeopardise the foundation upon which the Dutch prosperity is build. The consensus on wage restraint is therefore not solely rooted in power or the institutional environment alone. Instead, we also need to consider the role of culture in the creation of consensus. A certain degree of overlap in the different versions of common sense opens up space for a synthesis of the different perspectives, and subsequently a consensus (Bruff, 2008: 158).

(15)

15

The learning argument of Visser & Hemerijck (1997) cannot adequately explain why wage restraint was regarded as the correct solution, and how the underlying competitiveness-employment nexus was created. Instead, by stressing the economic necessity of wage restraint they take its legitimacy more or less for granted. The power shift that occurred in the Dutch industrial relations in the early 1980s (van Apeldoorn, 2009: 214), and the fact that the signing of the Wassenaar Agreement triggered a process that was predominantly favourable to the wishes of capital (Bruff, 2010: 628), shows the relevancy of the interests and power of the principal actors. However, power alone is not sufficiently able to explain the Dutch wage restraint strategy. We therefore also touched upon alternative explanations that stress the role of hegemony and the Dutch corporatist structure, and the perceived vulnerability of the Dutch economy. Using the Gramscian notion of common sense, Bruff (2008; 2010) shows the continued legitimacy of the dominant narrative that asserts that there is an unequivocal positive link between wage restraint and job growth. However, it does not give us any insights into the creation of this dominant narrative, or how it was sustained by all the relevant actors. The social partners and the government have formally embraced the discourse on the benefits of the wage restraint-employment nexus for the last 40 years. However, it remains unclear how a theoretically and empirically shaky discourse became a stable and recurring element in the Dutch wage bargaining process. Because of this lacuna, more research into the emergence and the reproduction of this ‘competitiveness discourse’ is necessary.

(16)

16

2: Research design

2.1: Research objective

Although wage restraint is often used as an explanans for the Dutch ‘miracle’, it is rarely viewed as an explanandum; that which needs to be explained. Despite the occasional reluctance of trade unions to accept lower wage increases, the consensus of the social partners on wage restraint and its apparent benefits throughout the last few decades has been remarkable. The formal basis for this consensus seems to be the mutual understanding that to generate economic growth and foster job creation it is essential for the Dutch export-led economy to be competitive. The dominant narrative asserts that there is a positive causal link between wage restraint, competitiveness, and employment growth. Time and time again it is therefore stressed that rapid nominal wage increases should be avoided, since in order to remain internationally competitive the unit labour costs need to be suppressed. Increasing the competitiveness of the economy without jeopardising the profitability of firms, will, according to this logic, result in higher levels of export and domestic investments, which will enhance the demand for labour. The aforementioned logic is entrenched in the wage formation process of the Netherlands. National competitiveness has become a clear priority, which resulted in a critical shift from productivity bargaining to employment bargaining. Wage-setting was transformed into a function of competitiveness; agreements were aimed at the stimulation or preservation of employment without obstructing the competitiveness of the economy (Leonard, 2001: 37-39).

The theoretical and empirical evidence that supports the competitive-employment nexus has serious shortcomings. Furthermore, it seems that the Dutch wage restraint strategy did not result in the proclaimed miracle; Real GDP growth after 1982 was inferior to the performance in the 1960s and 1970s (Storm & Naastepad, 2013: 102), while the jobs that emerged were low-paid flexible service sector jobs (Becker, 2001: 26; Becker, 2005: 1088). This begs the question why trade unions continuously accepted wage restraint, and adopted the competitiveness discourse that stresses the causal relationship between wage restraint, international competitiveness, and job growth. Because of the firm grip that this specific discourse seems to have, this thesis will investigate how wage restraint became the common sense notion, adopted by both the government and the social partners. To do this, it is necessary to uncover the role of the government and its affiliated organisations, the employer’s organisations, and the trade unions in mobilising and sustaining the view that improving the cost-competitiveness via wage restraint is an economic necessity favouring the national interest. The aim is therefore not to offer an alternative explanation of the proclaimed success story of the Dutch wage restraint strategy. The main subject is the hegemonic character of

(17)

17

the underlying competitiveness discourse that for the last 40 years continuously propagated the unequivocal benefits of wage restraint.

2.2: Research questions

Wage restraint was in the post-war period until mid-1960s an important element of the low wage strategy of the Netherlands (Visser & Hemerijck, 1997: 92-93). In the early 1980s this strategy was reintroduced, embedded in a corporatist consensus. To fully understand how wage restraint became the ‘go-to’ solution during economic downturns, it is essential to look at the competitiveness discourse underlying the formal consensus of the social partners on employment. Where did it come from, and how is it possible that the discourse has been so stable for the last 40 years, while its claims are rather shaky?

To accomplish the objective of this thesis the following research question has been formulated: How was the discourse on the wage restraint-employment nexus created and sustained in the Netherlands?

The answer to the research question will be composed of three parts. Firstly, we will focus on how the discourse was created, and what role the relevant institutions played in creating the wage restraint-employment narrative. The real wage-productivity nexus has been replaced by a wage restraint-employment nexus; a causal link between wage restraint, competitiveness, and job creation. However, it remains unclear how wage restraint became equated with competitiveness, and why increasing the cost-competitiveness was regarded as the solution to improve the employment levels in the Netherlands. A socio-historic reconstruction that uncovers the creation of the discourse is therefore necessary. The first sub-question will therefore track the creation of the discourse from the 1970s onwards. Secondly, we will examine the utilisation and transformation of the discourse. How was the theoretical discourse translated into practice, and how did the discourse develop over time? Thirdly, the sustentation of the discourse by the social partners and the government will be studied. The focus will be on the different ways the competitiveness discourse was retained and reinforced after the initial consensus of 1982.

(18)

18

To give a proper answer to the aforementioned parts three sub-questions have been formulated: 1. How was the discourse that places wage restraint at the centre of the competiveness

strategy socially constructed?

2. How did the competitiveness discourse evolve over time? 3. How has this competitiveness discourse been sustained?

(19)

19

3: Methodology

3.1: Research methods

To study the creation and sustentation of the competitiveness discourse process tracing will be employed. Process tracing is a qualitative research method that aims to deductively or inductively uncover social causal mechanisms or outcomes in a single-case research design (Beach & Pedersen, 2013: 2). Although it is regarded as a fundamental tool of qualitative analysists, it often is not adequately understood (Collier, 2011: 823). The confusion surrounding the research method is in part because of the different versions that seem to exist. Beach & Pedersen (2013: 9-13, 18) argue that there are three different variants of process-tracing, which all have distinct research purposes. The variant that will be used in this thesis is, what they define as, ‘explaining-outcome process-tracing’. This most commonly used variant is particularly useful when the aim is to account for a puzzling historical outcome. The goal is to build a minimally sufficient explanation of the particular outcome. Compared to the other variants of process-tracing, this variant is neither concerned with theory-testing nor theory-building. The causal mechanisms are therefore not theorised as sufficient causes. Instead, causal mechanisms are understood as relatively simple pathways between two variables; variable x is seen as contributing to producing variable y. However, theories and concepts do play a role as heuristic instruments that have analytical utility. Not the testing or building of theories is the goal, but to show how a particular theory or set of concepts is useful in helping to produce the best possible explanation.

To build a minimal sufficient explanation for the creation and sustentation of the competitiveness discourse this thesis will therefore focus on description and logical inferences. Furthermore, special attention will be paid to the actors that are directly or indirectly involved in the Dutch wage bargaining process, such as the social partners, the government, the Stichting van de Arbeid (Labour Foundation), the SER, the CPB, the WRR, and (scientific) experts. Moreover, there will be a clear focus on critical junctures, such as specific bipartite or tripartite agreements since the 1970s onwards.

Following the distinction of Bonnell (1980) between theoretically informed and conceptually informed historical sociology, this thesis will “rely primarily on concepts to facilitate the selection, organization, and interpretation of empirical material (166)”. The aim of this thesis is therefore to offer a conceptually informed reconstruction of the history of the competitiveness discourse. Instead of theories, a variety of concepts will be used as a heuristic device. The guiding framework for the analysis will be explained in the next part of this chapter.

(20)

20

3.2: Framework for analysis

To interpret the continued reliance on wage restraint the functional ‘polder’ explanations fall short in explaining why trade unions accepted the shaky discourse on the benefits of wage restraint. This continued acceptance is especially puzzling when one takes into account the fact that the consensual corporatist exchanges have been skewed in favour of capital. To explain why trade unions accepted a strategy that is not necessarily beneficial for labour it is necessary to adopt an alternative framework that takes the role of power seriously. However, changing power relations alone is not able to sufficiently explain why the wage restraint strategy was embraced by the key players in the Netherlands and not in the other countries. Bruff’s (2010) use of common sense offers an alternative perspective to understand why trade unions subscribed to the wage restraint strategy. Bruff mainly focusses on demonstrating the continued legitimacy of wage restraint in the Netherlands. This legitimacy is, according to Bruff, anchored in the common sense notion that wage restraint is, as a result of the assumption of economic vulnerability, a necessity. By focussing on the creation, reproduction, and utilisation of the discourse underlying the alleged causal link between wage restraint, competitiveness, and job creation this thesis moves beyond the work of Bruff.

To research how the competitiveness discourse was created, sustained, and became dominant in the Netherlands this thesis will use a framework that incorporates the Gramscian notion of common sense and certain elements of Cultural Political Economy (CPE). Hoare & Nowell-Smith (1971), the translators of the works of Gramsci, note that Gramsci used the term common sense “to mean the uncritical and largely unconscious way of perceiving and understanding the world that has become “common” in any given epoch (322)”. Common sense therefore refers to the incoherent set of assumptions and beliefs that are broadly held by the members of a society (Hoare & Nowell-Smith, 1971: 322). The concept expresses the taken-for-granted, but is “not something rigid and static; rather it changes continuously, enriched by scientific notions and philosophical opinions which have entered into common usage (Hoare & Nowell-Smith, 1971: 326)”.

As an important element in the struggle for hegemony, the notion of common sense helps in explaining why trade unions subscribed to a logic that does not necessarily seem to benefit labour. However, to understand how the competitiveness discourse was created and sustained additional concepts and theoretical insights are necessary. CPE offers the tools necessary to study the mechanisms, processes, and practices through which hegemonies are constituted. CPE is an analytical approach that combines concepts and tools from critical semiotic analysis and from critical political economy (Jessop, 2004: 159). It highlights the importance of the cultural dimension by

(21)

21

emphasizing the foundational nature of meaning and the process of meaning-making in social relations (Jessop, 2010: 337).

The aim of this thesis is, however, not to offer a full CPE-account of the competitiveness discourse underlying the consensus on wage restraint. It is therefore not necessary to explain CPE in full detail. Instead, the goal is to explore the diverse processes and mechanisms involved in the production of this particular hegemonic discourse, how it developed over time, and how it constantly is renegotiated within civil society and government institutions. The actors involved in the ongoing reproduction of the competitiveness discourse may have heterogeneous motives and do not necessarily work together to maintain the discourse. The exact motives of the actors that reproduce the Dutch competitiveness discourse are, however, not a concern that will be addressed in this thesis. For this exploration it suffices to use a limited number of concepts that play a central role in CPE. To track the development of a particular discourse it useful to focus on the emergence of a discourse, the privileging of the particular discourse, and the ongoing realisation of the discourse. The evolutionary mechanisms of variation, selection, and retention will be used to study this process. However, before explaining these mechanisms we will first briefly discuss the concept of discourse. Discourse is a versatile and ambiguous concept that is prone to misunderstandings. In this thesis discourse refers to particular sets of semiotic practices that produce and communicate sense and/or meaning (Jessop & Sum, 2013: 151-153). A discourse could therefore be conceived as a narrative that attributes sense or meaning to particular objects, and envisions certain relationships between these different objects. Because discourses selectively assign meaning to some objects and not to others, it structures and therefore restricts the possible variety of interpretations and solutions. In this thesis we are not concerned with the linguistic features of the discourse, but with its semiotic impact, and its articulation into social practices.

Which discourse prevails not only depends on the persuasiveness of the argumentations, or whether it has a scientifically supported foundation, but also on the extra-semiotic (material) features of social practices, such as power or institutional logics. Most discourses lack the support of actors with the appropriate economic, administrative or legislative resources to create consensus on a particular interpretation and solution. A carefully constructed and powerful narrative without the backing of important actors that help to implement the narrative will be less effective compared to more arbitrary accounts that do have powerful backing (Jessop, 2013: 9). Power therefore clearly matters. It is in periods of crisis that non-dominant discourses are able to gain footing. Crises are generally characterised by profound disorientation. The shock crises tend to give to prevailing economic wisdom and dominant policy paradigms creates space for alternative economic perspectives (Jessop,

(22)

22

2010: 341). This gives room for the (re)politicisation of the sedimented discourses and practices (Jessop, 2013: 5). Faced with feelings of urgency, the profound disorienting effects of a political and/or economic crisis initially leads to a proliferation of a variety of responses by different social forces. A limited number of discourses are selected for interpreting the crisis and used as legitimation for actions. However, which discourse is privileged is not reducible to its semiotic features. Although it is possible that different plausible discourses emerge, the narrators are not equally effective in conveying their message or gathering support (Jessop, 2004: 167). Different extra-semiotic factors also play a role. In short, the selection of a discourse depends in part on the structure, the agents involved, and technology. Structures favour certain interests or strategies. In the case of the wage restraint strategy, Becker’s explanation of the ‘rules of the game’ of Dutch corporatism could be regarded as a structural factor. Furthermore, the capacities of the agents involved and their abilities also play an important role in the selection of a particular discourse. Lastly, technologies, regarded as knowledge, expertise and apparatuses, construct authority, and have the potential to convey legitimacy and guide action (Jessop & Sum, 2013: 175, 216-222).

If a discourse is able to successfully obtain a privileged position it is not only able to reduce the ways in which a crisis can be interpreted, but also the number of appropriate solutions to overcome the crisis. It therefore offers an imagined recovery that forms the basis for economic strategies and policies. After the selection, some discourses are retained via sedimentation of the discourse and the consolidation of the discourse into social practices. The selected discourses are translated into policy, routines, and new social arrangements (Jessop, 2013: 4, 7). It is the retention of the discourse that helps form the basis of the Gramscian notion of common sense. The institutionalisation of a discourse has a naturalising and path-shaping effect. The incorporation of the discourses into routines, the continued expression of the discourse in economic strategies, and the depoliticising effect of specific technologies, such as statistics and indexes, reinforce the position of the discourse, and foster the translation of the logic of the hegemonic discourse into common sense (Jessop & Sum, 2013: 282-283; Jessop, 2010: 346).

However, as argued, common sense is not something rigid and static. If the incorporation into common sense is successful, the actors share an intersubjective version of common sense; there is an overlap in the assumptions and beliefs of the actors. However, even intersubjective versions of common sense are fragmentary, incoherent and inconsequential (Gramsci, 1971: 419). The common sense notions an actor holds can therefore be contradictory in nature. It is as a result possible that the perspective of an actor is informed by both a hegemonic logic and forms of critical insights (Sum & Jessop, 2013: 222; see also Bruff, 2008: 48-49). Because of these internal contradictions within

(23)

23

specific versions of common sense, it is possible that actors who share an overlap in common sense emphasize to a varying degree the policy and institutional changes sought (Bruff, 2010: 626).

A Full CPE-account would focus on the dialectic of path-dependency and path-shaping that emerges as a result of interaction between semiotic and extra-semiotic (material) processes. However, this is beyond the scope of this thesis. Furthermore, some aspects are visible in the arguments of Becker (structure) or widely regarded as having an impact (changing power relations). Instead, the thesis will use the outlined evolutionary mechanisms, and pay close attention to the role of technology in naturalising the competitiveness discourse.

3.3: Data

A variety of data will be used to examine the creation and sustentation of the competitiveness discourse. This data will primarily be qualitative in nature, and comprises i.a. annual reports, research reports, magazines, letters, speeches, minutes and statements of the cabinet, governmental organisations (SER, CPB, WRR), the VNO-NCW and its precursors (the Dutch employer’s federation), the FNV and its precursors (Federation of Dutch Trade Unions), the CNV (National Federation of Christian Trade Unions in the Netherlands) and the Stichting van de Arbeid. In addition, interviews in newspapers with representatives of the government or with the members of the different organisations will also be examined. Moreover, a variety of news articles by the Dutch scientific community will also be consulted.

Some of the reports and statements of the relevant organisations are accessible online, while the physical historical records of a number of other organisations can be consulted via the restricted archives of the International Institute of Social History:

• A digital archive that consists of historical newspapers, books, and magazines:

https://www.delpher.nl

• The digital archive of the Scientific Council for Government Policy (WRR):

https://www.wrr.nl/archief

•_{The digital archive of the Social-Economic Council (SER): https://www.ser.nl/nl/publicaties/}

•_{The archive of CNV: https://search.socialhistory.org/Record/ARCH00296}

•_{The archive of FNV: https://search.socialhistory.org/Record/ARCH00419}

•_{The archive of the Stichting van de Arbeid:}

(24)

24

4: Analysis

Discourses on competition or competitiveness have a long history, and have been linked through radically different political and economic interpretations (Sum, 2010: 48). These interpretations are liable to change; one only needs to compare the mercantilist strategies that dominated Europe from

the 16th_{to 18}th_{century to the Schumpeterian notions of innovation and competition to realise that}

throughout history there has been a lot of disagreement on the relevant units of competition. The diversity of interpretations has also led to a variety of proposed competitive strategies, and the determinants of their success. In the post-war period the indicators that were regarded as relevant for economic competitiveness include, among others, unit labour costs and the real effective exchange rates. Actors may try to improve the competitiveness of a country with the help of these indicators either because of their perceived impact on the overall competitiveness of firms within a country or because of their envisioned effects on alternative economic objectives, such as employment or inflation (Sum & Jessop, 2013: 266-267).

Improving the competitiveness of the Dutch economy to achieve alternative economic objectives has been the primary reason for the Dutch wage restraint strategy. The Netherlands has a long tradition in wage restraint. In order to avoid jeopardising the systematic rebuilding of economic life in the post-war period, the Dutch implemented a centrally directed wage policy. The policy initially served two purposes; it was implemented because of both economic and income-related motives. However, gradually the economic motives became more dominant at the expense of the wage distribution concerns. The centrally directed wage policy was transformed into an important instrument for controlling the business cycle to maintain price stability and, more importantly, improve the competitiveness of the Dutch economy in order to foster economic activity and employment (Fase, 1980: 259; van Zanden, 1998: 464). Restraining the wages was however not without a struggle. Because full employment was virtually achieved in the 1950s, the wage restraint strategy lost part of its legitimacy for the trade unions. Restraining the wages seemed solely intended to reduce wage costs, and lost therefore its appeal to labour (Fase, 1980: 259). This eventually lead to a wage explosion in 1963 and 1964, and heralded the end of the centrally directed wage policy; wage formation became from 1963 onwards characterised by free wage bargaining (De Kam, van Drimmelen, van Hulst, 1995: 11, 23-24). Though, from time to time there was still government interference in the wage bargaining process; even after the formal introduction of the system of free wage formation in the 1970s (van Hulst, 1984, 256-257).

The signing of the famous Wassenaar Agreement in 1982 is widely regarded as the reintroduction of the wage restraint strategy. However, the implementation of the post-1982 wage restraint strategy

(25)

25

fundamentally differs from its post-war precursor. The post-1982 wage restraint strategy has been embedded in a corporatist consensus without the formal interference of the government; while in the post-war period wage restraint was actively pursued and imposed by the Dutch government. The lack of formal government interference in the wage bargaining process in the early 1980s and the subsequent decades shows the changing role of the government. The responsibility of wage bargaining has since been the prime concern of the social partners, without the government having an initiating or controlling role in the process (van der Meer, Visser, Wilthagen & van der Heijden, 2003: 62-63). Furthermore, the post-war implementation of wage restraint was meant to help achieve a variety of goals; the post-1982 wage restraint strategy was primarily meant to have a positive impact on employment. Moreover, both were situated in different periods of crisis. Certain elements of the reasoning, such as the relationship between wage restraint and employment, that the competitiveness discourse incorporates, can also be traced back to the post-war wage restraint strategy. However, it seems that during the post-war period a clear causal link between the two variables was never propagated. Notwithstanding that path-dependent influences are important; the competitiveness discourse that underpins the consensus on the wage restraint strategy seems to be primarily situated in the crisis of the post-war mode of growth: Keynesianism. The competitiveness discourse therefore builds upon, but cannot be reduced to the post-war wage restraint logic.

Using the outlined framework for analysis, this chapter will first focus on the Keynesian crisis of the 1970s, and the variety of responses that it triggered. Secondly, we will trace how the competitiveness discourse was able to get selected for interpreting and solving the crisis via wage restraint. Thirdly, we will examine the use and the evolution of the competitiveness discourse. Fourthly, we will explore the mechanisms that rendered the wage restraint strategy so stable throughout the last few decades: the consolidation of the discourse into practices, the repetition of the discourse, and the incorporation of the discourse into the common sense. The last part is concerned with the role of technology in the selection and the retention of the discourse.

(26)

26

4.1: The Keynesian crisis and the variety of responses

The 1970s were characterised by the declining effectiveness of Keynesian economic strategies (Jessop, 2013: 8). The dire economic situation many countries were in led to a crisis and the subsequent displacement of Keynesianism. The most important targets during the post-war period, such as full employment, growth, and price stability, could no longer be delivered via the nation state (Sum & Jessop, 2013: 269). Furthermore, the high levels of unemployment and the rapidly rising public deficit undermined the government’s capacity to maintain the welfare state in a sustainable manner. To overcome the economic crisis of the 1970s a reordering of the Dutch priorities and a redesign of the economic strategies was therefore necessary. The previous economic configuration inspired by the Keynesian school had failed, which gave room to alternative interpretations, strategies, and imagined recoveries.

The social partners and the government reached a consensus on the most salient problems of the Dutch economy rather quickly. Besides the outdated economic structure of the Dutch economy, both the level of the inflation and the level of unemployment were regarded as too high. However, to what extent these issues needed to be reversed, and what the best way would be to achieve this reversal, was still a topic of fierce debate. The discussion in the Netherlands was therefore not necessarily one dominated by rival theoretical interpretations. Instead, it was more concerned with the search for the appropriate policies to resolve the crisis, and how to allocate the costs of the heavy measures that needed to be taken. The problems of the economy in the Netherlands were therefore framed early on.

In the early 1970s one of the most pressing concerns in the Netherlands was the wage-price spiral. Already at, what in hindsight was, the beginning of the crisis there was an agreement on the necessity to contain inflation. In November of 1970 Sieb Bakkenist, the then current chairman of VNO, the largest employer’s organisation of the Netherlands, already warned for the possible negative consequences of a wage-price spiral (VNO, 1970: 13-14). In a joined letter addressed to the trade unions, employer’s organisations VNO and NCW argued that a reduction of the wage increases was necessary to reduce the inflation, to minimise the threats to the competitive position of the Dutch economy, and to ensure the continuity of the level of employment (VNO & NCW, 1971: 1-2). The trade unions regarded themselves of having the difficult job of fighting inflation while at the same time ensuring that the employees did not have to foot the entire bill (Overlegorgaan NVV-NKV-CNV, 1972: 2). They were therefore willing to weaken real wage growth below the increase in productivity to fight i.a. inflation. However, one of the demands was that the lower wage growth had to be translated into lower prices, and was not meant to increase the profitability of firms

(27)

27

(Overlegorgaan NVV-NKV-CNV. 1971: 1-2; 1972: 2). In 1972 the trade unions and the employer’s organisations were able to come to an agreement in the Stichting van de Arbeid (1972: 1, 7). In the Central Agreement for 1973, both parties stressed the necessity of containing the inflationary development. The real wage growth would be limited to 3.5%. Price adjustments on the domestic market would be limited to what is necessary to maintain the average profit per unit. Despite the agreement, it would not lead to the recovery of the Dutch economy.

Although the problem of inflation was regarded as important, the debate became soon to be dominated by concerns surrounding the level of unemployment in the Netherlands. However, inflation and unemployment were by all parties not regarded as separate issues. As we have seen, the VNO and NCW saw a reduction of the wage increases as a way to fight both inflation and job destruction. The trade unions argued in 1972 that there was a clear link between the level of inflation and employment. In addition, they also stressed that the level of employment strongly fluctuates with the level of investments (Overlegorgaan NVV-NKV-CNV, 1972: 8). Jaap Boersma, the Minister of Social Affairs at the time, stated that the main goals of the cabinet in 1974 were to reduce the inflation and the unemployment, between which a significant interaction exists (Ministerie van Sociale Zaken, 1973: 1). While in 1976, the CNV regarded controlling the inflation as one of the most important prerequisites for reducing the level of unemployment (CNV, 1976 C: 1). Because of the rising levels of unemployment, the Dutch trade unions refocused their attention in the mid-1970s from inflation to unemployment, and saw combating the inflation as an instrument to diminish the level of unemployment (NCW: vakbeweging laat inflatie gaan, 1973: 1).

With rapidly rising levels of unemployment during the 1970s, the imagined recovery plans the different parties proposed were predominantly aimed at restoring or improving the level of employment. Concerns surrounding the inflation and ways to overcome this problem were often inherent or incorporated into the proposed solutions. Throughout the 1970s and early 1980s a variety of such measures were put forward. These measures were, however, not always mutually exclusive, and could sometimes be regarded as complimentary. The variety of the solutions proposed by the trade unions and the government will briefly be discussed. The solution of wage restraint which the VNO put forward will be discussed in the next chapter. Here it suffices to note that, besides wage restraint, VNO regarded tax reductions for both capital and labour, and reducing public expenditure also as ways to promote investments, employment, and economic growth.

The Dutch trade unions had throughout the 1970s and early 1980s different plans to prevent additional job destruction and to ensure job creation. In mid-1970s the FNV proposed a capital gains distribution; a plan which combined the trade union’s goal of levelling out the income disparities and

(28)

28

to stop further job destruction. In short, the plan was to transfer a part of the profits of the highly profitable companies to the poorly performing businesses. The plan prevented that some people would benefit from extremely favourable conditions in their respective industry, while at the same time other people were being fired (Pen, 1974: 23). 1978 was declared as the year of the ‘arbeidsplaatsenovereenkomst’ (APO; jobs agreement) by The FNV (1978 A: 6; 1978 B: 11; 1978 C: 1-3). These APO’s were very broad agreements with employers on the development of employment in the Netherlands, both in terms of the quantity and the quality of the jobs. The primary goal was to maintain existing jobs, and where possible create new employment opportunities.

In addition, the FNV as well as the CNV regarded a reduction of working time also as a way to create new jobs. A prerequisite to make such a policy a success was that the reduction of working time had to be translated into new jobs for the unemployed population. Reducing the amount of hours worked was already somewhat incorporated into the Central Agreement of 1973 (Stichting van de Arbeid, 1974: 3). Furthermore, it was in 1975 also put on the agenda of the Stichting van de Arbeid by the employees, who wanted discuss whether, and if so, how a working time reduction could help resolve the structural unemployment problems (Stichting van de Arbeid, 1975: 7). A shorter work week was traditionally a social goal that trade unions pursued. However, because of the dire economic situation in the Netherlands it was also regarded as a weapon against the high levels of unemployment. Wim Kok, the chairman of the FNV at the time, stated that a working time reduction was a way to distribute the limited amount of jobs more fairly across the society (FNV, 1976: 24). Redistribution of work via working time reduction was regarded as a valuable strategy for the FNV. Already in 1978 the FNV showed its willingness to discuss whether employees should contribute to help financing the plan; stating that it is unlikely that a reduction in working time of any significance was possible without a meaningful contribution from the employees themselves (FNV, 1978 C: 5). At the end of the 1970s and in the early 1980s the FNV proposed a three-pronged policy approach: a targeted strengthening of the market sector via investments, a selective expansion of jobs in the collective sector, and working time reduction (FNV, 1978 D: 9-10; 1981 A: 3). The redistribution of work via working time reduction was, however, the main negotiating target of the FNV. In October of 1982 it was regarded as a vital part of the solution to the unemployment problem; the FNV was therefore willing to exchange it for wage restraint (FNV, 1982 B: 5-7, 23).

In the early 1970s the CNV mainly referred to the Keynesian instruments as the way to combat the high level of unemployment (Hazenbosch, 2009: 763). From the mid-1970s onwards the Christian trade union seems to take a different approach. The CNV proposed for 1977 to exchange a part of the wage increase for early retirement schemes, provided that the unemployed would end up with the vacant jobs that would result from early retirement (CNV 1976 B: 4-5). Furthermore, in the early

(29)

29

1980s the CNV repeatedly argued for wage investment schemes, which were arrangements whereby a part of the wages would be used for the financing of investments within the company (CNV, 1981: 8; 1982: 4). Moreover, the CNV also suggested an employment fund, which initially had to invest in projects that could strengthen the international competitiveness of the Dutch economy. The difference with the wage investment scheme is that in such a scheme the raised capital would stay within the firm, and would serve the goal of strengthening the firm itself; while the employment fund would have had a national character (CNV, 1981: 6). Lastly, in 1982 the CNV proposed to its members the idea to exchange the automatic indexation of the wages for the next few years for the guarantee that the tax burden would not increase and additional employment policies would be implemented (Hazenbosch, 1982 A: 13).

Throughout the 1970s and the early 1980s the government proposed and implemented a variety of plans to turn the economic tide and to overcome the unemployment problems. These plans varied widely between the different cabinets. The den Uyl cabinet (1974-1977) actively intervened, and initially relied on Keynesian instruments, such as tax reductions and the targeted stimulation of spending to improve the Dutch employment level (see for instance den Uyl, 1974: 12). In 1976, however, the government, inspired by new economic forecasts made by the CPB, slowly moved away from the Keynesian instruments, and decided to try a different approach (van Zanden, 2005: 166). Ruud Lubbers, Minister of Economic Affairs at the time, published in June of 1976 his report ‘nota inzake de selectieve groei’ in which it was stressed that to solve the unemployment crisis, the structure of the economy had to improve (Lubbers, 1976). This could be achieved by stimulating investments, restraining the total labour costs, and shrinking the collective sector. Following the report, the government shifted its focus to the supply-side of the economy (Hazenbosch, 2009: 373). In the same year the den Uyl cabinet tried to encourage entrepreneurs to invest in business assets via a new instrument: the ‘WIR’ (Wet op de Investeringsrekening). The ‘WIR’ provided financial incentives for companies to invest, which did lead to additional investments. However in the first years after its implementation of the law in 1978 the impact had been limited, according to, among others, the CPB (Hazenbosch, 1982 B: 6).

The changing attitude is even more visible in the strategies put forward by the various cabinets led by Dries van Agt (1977-1982). The van Agt cabinets also actively intervened, but were primarily concerned with wage restraint and budget cuts. This already started in 1978, when the first van Agt cabinet presented the report ‘Bestek ‘81’, in which it announced a new series of changes with the aim of combating the high level of unemployment by improving the structure of the economy. The main features of the proposed policy solutions were i.a. wage restraint, budget cuts, and stabilising the collective spending (Stichting van de Arbeid, 1978; see also FNV, 1981 C: 4). In reality, however,

(30)

30

the proposed solutions were hardly implemented. In the early 1980s the government increasingly emphasized the need of wage restraint (see for example Stichting van de Arbeid, 1980: 1-2), and was, for example, willing to exchange wage restraint for tax relief (VNO, 1981: 8). However, because no agreements on wage restraint were reached in the late 1970s and early 1980s, the van Agt cabinets imposed a (partial) wage freeze in 1979, 1980 and 1981.

The Economic crisis opened up space for alternative modes of thinking, which led to a variety of responses. However, over time the proposed solutions were increasingly modelled after the competitiveness discourse. Especially in the plans of the government there is shift visible from Keynesian demand economics to supply-side economics. As we will see in the next chapter, the discourse and its underlying reasoning became more and more noticeable in the government strategies. The proposed solutions of the CNV were also increasingly modelled after the competitiveness discourse. This trend is to a lesser extent visible in the plans of the FNV. But even the FNV became more receptive to wage restraint over time. The Dutch trade unions were increasingly willing to restrain the wages when certain specific conditions were met. Over time the wage bargaining process increasingly became an instrument for political exchange; wage restraint in exchange for the broader socio-economic wishes of the trade unions.

Holland’s single most important weapon : a CPE-inspired common sense analysis of the consensus on wage restraint in the Netherlands