How does a coordinated market economy evolve? Effects of policy learning in the Netherlands in the 1980s

PLEASE SCROLL DOWN FOR ARTICLE

This article was downloaded by: [Touwen, L. J.][Touwen, L. Jeroen]On: 16 October 2008Access details: Access Details: [subscription number 903256121]Publisher RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,37-41 Mortimer Street, London W1T 3JH, UK

Labor HistoryPublication details, including instructions for authors and subscription information:http://www.informaworld.com/smpp/title~content=t713436999

How does a coordinated market economy evolve? Effects of policy learning inthe Netherlands in the 1980sL. Jeroen Touwen a

a Leiden University, The Netherlands

Online Publication Date: 01 November 2008

To cite this Article Touwen, L. Jeroen(2008)'How does a coordinated market economy evolve? Effects of policy learning in theNetherlands in the 1980s',Labor History,49:4,439 — 464

To link to this Article: DOI: 10.1080/00236560802376904

URL: http://dx.doi.org/10.1080/00236560802376904

Full terms and conditions of use: http://www.informaworld.com/terms-and-conditions-of-access.pdf

This article may be used for research, teaching and private study purposes. Any substantial orsystematic reproduction, re-distribution, re-selling, loan or sub-licensing, systematic supply ordistribution in any form to anyone is expressly forbidden.

The publisher does not give any warranty express or implied or make any representation that the contentswill be complete or accurate or up to date. The accuracy of any instructions, formulae and drug dosesshould be independently verified with primary sources. The publisher shall not be liable for any loss,actions, claims, proceedings, demand or costs or damages whatsoever or howsoever caused arising directlyor indirectly in connection with or arising out of the use of this material.

http://www.informaworld.com/smpp/title~content=t713436999

http://dx.doi.org/10.1080/00236560802376904

http://www.informaworld.com/terms-and-conditions-of-access.pdf

Labor HistoryVol. 49, No. 4, November 2008, 439–464

How does a coordinated market economy evolve? Effects of policy

learning in the Netherlands in the 1980s

L. Jeroen Touwen*

Leiden University, The Netherlands

This article studies the changes in economic policy in the Netherlands and theireffects on the coordinated market economy (CME). In order to gain insight intothe development of the CME, the neo-corporatist literature and the concept of‘policy learning’ are called upon. The article takes the analysis of Visser andHemerijck on the changes in the Dutch labor market in the 1980s and 1990s asa starting point for the analysis of the development of the Dutch CME.The analysis includes a systematic review of macroeconomic developments. Thefrequent introduction of market-oriented solutions is related to a paradigmchange in macroeconomic policy. ‘Policy learning’ explains the transition inpriorities of economic policy in the Dutch corporatist economy. After a decadeof increasing polarization in the 1970s, economic urgency increased andstimulated a regime change towards more market-oriented policy, creating amore liberal CME. In any CME, either new forms of coordination will develop,or market-based solutions will be sought. In the Netherlands, in various policyfields coordination was transferred to the market. Labor market reform wasaccompanied by supply-side reform and deregulation. The tension betweenglobal trends in economic ideas and their application in a CME was resolved bypolicy concertation (as institutionalized regular consultation between partiesis often called) and learning, creating a more liberal business environment withinthe CME.

Introduction: paradigm changes in a coordinated market economy

In recent years, the ‘varieties of capitalism’ theory (VOC) has climbed the charts ofcomparative capitalism. The theory gives a firm-based explanation of an essentiallybipolar divergence of advanced capitalist countries into ‘liberal market economies’ (LME)such as the USA, and ‘coordinated market economies’ (CME) such as Germany. Althoughrich in explanatory power, the theory is relatively weak in explaining change within eachideal type, and insufficiently addresses the hybrid forms of mixed economies. In short,according to VOC, CMEs tend to remain coordinated due to institutional complementa-rities and comparative institutional advantages. Likewise LMEs tend to remain liberal.However, the supposed bifurcation is weak in explaining the broad range of differentdevelopments that can be observed in various OECD countries. VOC ‘brings firms backinto the center of analysis’ and claims to provide us with a micro-level analysis, but thisis only partially true: in fact, the ideal types CME and LME are descriptions ofmacroeconomic settings in which firms operate.1

*Email: [email protected]

ISSN 0023–656X print/ISSN 1469–9702 online

� 2008 Taylor & Francis

DOI: 10.1080/00236560802376904

http://www.informaworld.com

Downloaded By: [Touwen, L. J.][Touwen, L. Jeroen] At: 10:13 16 October 2008

The aim of this article is to analyze the process of change in one CME, the Netherlands,by invoking and extending the existing literature on neo-corporatism. I will reviewmacroeconomic development and contrast this with the ideal type of the CME. In VOC

neo-corporatist countries – where concertation between firms, their employers and the statetakes place – are classified as CMEs. (Concertation is the institutionalized and regularconsultation between parties.) The changes in the relations between these groups areexamined in the literature on neo-corporatism. This literature provides insights that can becombined or contrasted with the perspective of institutional complementarities, which is

central in VOC in explaining the behavior of firms. As we will see, corporatist negotiationshave an effect on the CME and produce changes in its coordinating institutions.2

One of the major assets of the VOC is the call for a comparison of states ‘more deeplyrooted in the organization of the private sector’.3 This has been widely picked up. Inanother branch of political science literature, the productive gains for employers forsustaining ‘cross-class alliances’, as for example in certain types of publicly organizedsocial security, are brought to the forefront.4 Together these offshoots form a supply-side

revolution in institutional studies concerning the political economy.In studies of neo-corporatism, there has been an emphasis on class conflict. Hall and

Soskice state concisely: ‘a nation’s capacity for neo-corporatism was generally said todepend on the centralization or concentration of the trade union movement’, and thatthis was ‘generally associated with the capacity of a state to negotiate durable bargainswith employers and the trade union movement regarding wages, working conditions, andsocial or economic policy’.5 By contrast, VOC emphasizes the voluntary positive attitude

of firms towards non-market institutions that results in profitable interaction effects.This seems very appropriate for the Dutch situation, in which, since the 1980s, thecentral government has scaled back its intervention in industrial relations and wagenegotiations. VOC claims that in CMEs ‘institutional complementarities’ or ‘cross-classcoalitions’ develop that inspire employers to reject deregulation and display a morewilling attitude towards issues such as works councils, earnings-related entitlements, or

public social insurance systems than is often surmised. The main reason for firms andworkers to do this is the investment in industry-specific skills.6 Although this mayoverstate the commitment of the employers in co-specific issues (issues whose resultsdepend on cooperation with others), at least it explains a tolerance of employers towardstopics that are often hard won by the labor unions. In this article I argue that the results

of neo-corporatist studies can be used to evaluate the strengths and weaknesses of CMEsand their cross-class interaction effects. I will examine the cooperation between employerorganizations and trade unions in the Netherlands in a survey of the developments in theDutch CME in the 1980s, focusing on macroeconomic policy and its influence onthe coordinating institutions.

The primary study on this period, Visser and Hemerijck’s ‘Dutch miracle’, forms thestarting point of the analysis. Essential in Visser and Hemerijck’s analysis of change in the

Dutch corporatist economy (or CME) is the ability of social partners to learn. Policylearning is a concept that is used to explain that new views slowly spread among variousgroups creating support for policy change. As a result, practices that were liberal incharacter were adopted and transformed the CME into a new type of CME. The title ofthis article refers to the book by Kathleen Thelen, How Institutions Evolve, which shows

that institutions display continuities through reconfiguration even in reaction to majorexternal shocks.7

440 L.J. Touwen


Visser and Hemerijck: ‘learning’ at the core of job creation

Unemployment was the most pressing economic problem in the Netherlands in the early1980s. It formed part of the recession that was triggered by the second oil shock of 1979.Its urgency propelled a learning process that, as Visser and Hemerijck conclude in theirinternationally acclaimed book on policy change in this period, resulted in the revival ofresponsive corporatism (as opposed to immobile corporatism) in labor relations, socialsecurity, and (to a lesser extent) labor market policy in 1982.8

Essential in this turnaround was the ‘Wassenaar Agreement’, a social pact signed in1982 that formed the key symbol of a new approach to wage-setting.9 In this social pact,wage moderation was exchanged for shorter working hours and early retirementopportunities. Thus, employers could restore profits and increase their competitivepower, while the unions received better working conditions and more jobs. The earlyretirement opportunities implied a redistribution of work: an outflow of older employeesfacilitated hiring young (lower-paid) employees. In effect, rather than shorter workinghours, lower wages were essential in stimulating employment growth. Flexibility of short-term labor contracts and part-time work were stimulated, both also resulting in aredistribution of work, as well as lower youth unemployment.10

As a result of the Wassenaar Agreement, wage restraint, which also had been asignificant factor in Dutch economic growth in the 1950s, again became a central featureof the Dutch economy.11 One major difference with the 1950s was that it was not centrallysuperimposed. Another difference was that shorter working hours did not imply a wageincrease, but instead a wage restraint, since payment per hour was not raised tocompensate for fewer hours worked. In 1984–1988, the minimum wage was kept constant,so that as a result its value lowered in real terms. New in the 1980s was the emphasis onpart-time employment and short-term contracts.12

The ‘Wassenaar Agreement’ had a top–down character, containing less than two pagesand aiming in the general direction of improving industrial relations. The agreement itselfdid not mention the word ‘wage restraint’. However, ‘redistribution of work’ wasmentioned three times. It explicitly aimed to solve labor market problems without theintervention of the government (through the combined efforts of employers and unions)and stated that full employment (the old Keynesian ideal!) should be discarded as anunrealistic goal, even when economic recovery took place. Instead, profitability of theprivate sector should be pursued.

In due time, after a lengthy bargaining process at various lower levels in the branchesof industry and services, the Wassenaar goals did indeed materialize. The WassenaarAgreement was officially a bi-partite agreement, but it was backed by the government,which promised to refrain from intervening in the wage negotiations (which it had done onand off throughout the 1970s by issuing so-called wage measures). The shadow ofhierarchy influenced the social pact by the government’s threat to superimpose a wagefreeze when disagreement persisted. In an interview, the Prime Minister Ruud Lubbersconceded that Wassenaar originated under the influence of a ‘poker game’ by the Ministerof Social Affairs Jan de Koning, who supposedly told the peak organizations: ‘If you don’tdo it now, we will really interfere!’.13 The absence of the government was appreciated bythe employers, who had since the 1960s increasingly opposed centralist governmentaction. Indeed, the Wassenaar Agreement formed an important step in a trendtowards decentralized wage bargaining that continued throughout the next decades. Thefaltering system of collective bargaining was replaced with a decentralized (yet highly

Labor History 441


coordinated) system. The next key moment occurred in 1993, when a new pact was signed

(ironically called ‘Een nieuwe koers’, ‘a new direction’) that again stimulated wage

moderation, decentralization, and flexible wage negotiations.Wage constraint, a new active labor market policy, and social security reforms were

successful in creating new employment opportunities. Many new jobs were temporary or

part-time. The coordinated solution to unemployment was successful partly because the

world economy recovered its strength. The Dutch economy profited from external growth

factors by successfully implementing labor market reforms (Figure 1).14 Growth was thus

facilitated by institutional reform. Visser and Hemerijck describe a process of third-order

policy learning that they deem so important that they describe job growth as ‘a miracle’.15

In the next section we will take a closer look at ‘policy learning’. As a result of policy

learning, the government and the labor movement discarded many aspects of the older

paradigm of regulationist government intervention: the former by developing new market-

oriented, decentralized labor market policy and the latter by accepting them.

Policy learning changes the CME

How can we connect this trajectory of change to the characteristics of the CME as outlined

in VOC? VOC focuses on the mutual interaction between firms and the interaction

between firms and their workforce. The degree of being ‘liberal’ or ‘coordinated’ is

determined by the way coordination problems are solved in five institutional subspheres.

Hall and Soskice argue that firms in CMEs have different ways of recruiting or

reorganizing their work force, treating knowledge and skills, achieving innovation and

financing their enterprise from firms in LMEs. These ideal types provide an analytical

0.0

2.0

4.0

6.0

8.0

10.0

12.0

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

United Kingdom

United States

Netherlands

Figure 1. Unemployment rates in the Netherlands, the UK and the USA, 1984–2002 (standardized;as a percentage of civilian labor force). Source: OECD.

442 L.J. Touwen


framework that takes into account embedded company behavior. Thus, VOC recordspotential dynamics of economies which in the eyes of orthodox free-market advocates are

doomed to stagnate or go bankrupt.VOC defines five spheres of institutions between which institutional complementarities

occur in coordinated economies: (i) labor relations (consultation of employees on the firmlevel); (ii) vocational training and education of employees; (iii) corporate governance and

acquisition of investment capital; (iv) inter-firm relations; (v) information sharing withemployees (motivation of individual workers by participation, human resource manage-ment). Between these spheres, the strategies of firms influence and reinforce each other,

thereby creating a comparative advantage.16

The mutual reinforcement between institutions located in different spheres of thepolitical economy strengthens the respective CME or LME character. The concept of‘institutional complementarities’ is analogous to the familiar economic term ‘complemen-

tary goods’: for example, bread and butter are complementary goods since an increase indemand for one (for example due to a lower price) also boosts the demand for the other.Analogously, long-term employment relationships are more feasible in states where the

financial system provides capital on terms that are not sensitive to current profitability, byway of long-term bank loans instead of the stock market. Conversely, if financial marketsare keen to transfer resources towards specific promising endeavors by enterprises, they

create short-term demand for labor and encourage fluid labor markets. In the same way,investment in training is more feasible in economies with long-term labor contracts. Also,inter-firm relationships may be less competitive when the system of corporate governance

includes supervision by trustees from other firms.Institutional complementarities explain why firms can successfully operate in

coordinated economies, which seem at first sight to be less efficient than liberal economies.The features of the CME are not detrimental to economic growth because firms

understand how to cope with specific institutions. This, as we shall see later on, contrastswith the neo-liberal policy view, popular since the 1980s, that the best way to reformcoordinated economies (or extensive welfare states) is to liberalize. The characteristics of

the CME create a different, but a no less favorable climate for economic expansion.Upon scrutiny, one may observe that institutional complementarities are dynamic by

nature, since they are the result of a process of choice. For example, when changes takeplace in the CME that create a shift in the balance of power, the degree to which

institutional complementarities occur will change. One could call this institutionalelasticity. In labor relations: if the private sector gains governmental support for reformthat increase flexibility in the labor market, then co-specific commitments such as

information sharing, provision of vocational training, and employee consultation maydecrease. Mechanisms of coordination will not immediately disappear, but policy reformopens the door to institutional reform because new strategic options are created that

decrease the strength of institutional complementarities. When stock market financingreplaces long-term bank loans, shortening the scope of strategic decisions, some firms maycontinue to apply life-long employment because they value its co-specific advantages, but

others will not.In this context, the role of the state is vital. It creates the institutional background that

enhances either market-oriented solutions or ‘coordinated’, information-sharing solutions.In the Netherlands, neo-corporatism was strongly stimulated when in 1950 the government

issued a law called ‘PBO’, demanding regular concertation on central and lower levels,

Labor History 443


between the state, the employers and the unions. This arrangement created the legal basisof wage negotiations and policy concertation, and provided the Dutch unions with a stable

negotiating platform, notwithstanding their relative weakness in membership rates. In thisway, the government helped to create a CME.

As Crouch states, the centrality of the nation state ‘is found in most neo-institutionalist studies, including those on ‘‘social [that is, national] systems ofinnovation and production’’.’17 It also take a central place in the parallel but distinct

literature on ‘national systems of innovation’ of the Regulation Theory.18 In a CME, wecan think of state support for research consortia, fiscal arrangements supportinginnovative firms, or state laws requiring negotiations between firms and employees onvarious levels (central, intermediate of firm-level). Thus, economic policy has an

enormous influence on the type of solutions that are chosen by economic actors (firms,organizations, government institutions). When dealing with coordination problems thatcan be solved by the market or by other types of (non-market) regulation, governmentpolicy pushes economies into the direction of either LME or CME. When we want to

study changes within the CME and LME, it is of vital importance to include an analysisof government policy.

One of the co-authors of the VOC theory, Peter Hall, coined a useful theory on policychange in a much-quoted article on policy learning in the UK.19 In this article heintroduces the phrase ‘third-order learning’ for a paradigm shift in the overarching goals

of policy, thus connecting ‘social learning’ to the concept of a ‘paradigm’. In Hall’s theory,‘first-order learning’ consists of incremental improvements in policy, and ‘second-orderlearning’ entails the development of new policy instruments. When these do not suffice,a breakthrough in policy views may occur, which is termed ‘third-order learning’. ‘Policy

learning’ is used in the analysis of major policy changes, sometimes referring to cross-national policy learning, sometimes to the spread of new insights based on, for example,academic research.

The concept of ‘policy learning’ provides a perspective on economic change: it focuseson the emergence of new ideas and new solutions (whether widely shared or forced upon

society from above) and attempts to distinguish between minor adjustments and majorrenewal. The emergence of a pattern in types of solutions, such as for example theintroduction of the market in a range of areas previously dominated by the government,inspires the use of the term paradigm. A paradigm can be defined as a reigning set of ideas,

beliefs, values and techniques, shared by a specific social school, group or community.20

The use of third-order learning defined as a paradigm shift is inspired by Thomas Kuhn’sfamous essay on scientific paradigms.21 Kuhn’s ideas can be considered commonknowledge among academic scholars but it should be kept in mind that in a social-science

context the term ‘paradigm’ is often applied very intuitively.Visser and Hemerijck apply Hall’s theory on ‘third-order policy learning’ in their

analysis of wage moderation and job redistribution in the neo-corporatist Dutch economy.These labor market reforms were achieved by a process of ‘puzzling and powering’.22 Thisrefers to a complicated process of change which is intrinsically connected to the

concertation economy: because so many groups are involved in the decision-makingprocess, widespread support (possibly as a result of third-order learning) is necessary for amajor policy change – whereas in a non-corporatist LME an electoral mandate can besufficient. Visser and Hemerijck emphasize that learning takes place in an institutional

context in which at least four other factors play a role: the formation of necessary

444 L.J. Touwen


coalitions, the shadow of hierarchy, formal policy rules, and the influence of supportgroups.23

Summarizing, we can state that in order to study change in the CME, we need toreview both the elasticity in institutional complementarities and the direction of economicpolicy. Before pursuing in greater depth how policy learning in the 1980s affected theDutch CME, let us first review briefly the international context of the changing policyparadigm.

The neo-liberal revolution in the 1980s

In most countries, in the early 1980s a transition took place from Keynesian, demand-oriented policy to a predominantly monetarist and supply-side-oriented policy. There is arange of literature in political science studying this transition, often applying the conceptof ‘policy learning’.24 According to Meseguer, ‘the argument that a learning processgenerated the spread of liberal economic policies during the 1980s and 1990s is wellestablished among both scholars and practitioners’.25

One should be careful with the use of the term ‘a departure from Keynesianism’because it is too crude. Hall states that in the first post-war decades, Keynesianism was‘the prism through which policy makers saw the economy as well as their own role withinit’. He observes a third-order policy change when Margaret Thatcher was elected in 1979.26

As Clift and Tomlinson emphasize, not all characteristics of Keynesianism (which is oftenused in a very broad sense) disappeared.27 For example, many policy makers andpoliticians continued to give much weight to government intervention and a large groupdefended the economic benefit of deficit spending. Orthodox anti-cyclical budget policy(which in the Netherlands was never consistently applied) was abandoned in principle, butstimulating the demand side by applying fiscal or monetary policy has continued to bepopular. While creating and protecting employment to attract voters has continued to bepopular with politicians everywhere, and also in economic research, several otherKeynesian ideas continue to be studied, for example rigidities of prices and wages (smallfrictions in price flexibility may result from rational or ‘near-rational’ behavior of firms,but generate important effects at the macroeconomic level).28

Keeping these caveats in mind, in this article I distinguish three aspects of thetransition. First, a neo-liberal policy of market-oriented economic growth gained priorityover more government-regulated growth objectives. Second, anti-inflationary policiesgained priority over the willingness to accept deficit finance (this includes both monetaristand fiscal discipline). Third, central government responsibility for employment by eitherstimulating aggregate demand or by creating public sector jobs was abandoned. In sum,market forces, under proper conditions, were now believed to be more successful in leadingan economy to full employment. This implied a departure from almost forty years ofstrong belief in a specific economic role of the government. It also severely decreased theamount of income redistribution that distinguished the post-war period from earlierdecades.

The paradigm change of the 1980s was international in character but generated specific‘coordinated’ responses in many countries. Applying the VOC taxonomy, we observe amilder version of the supply-side revolution in the CMEs, such as Germany and theNetherlands, than in the LMEs, such as the USA, the UK, or New Zealand. The outcomeof the more radical type of supply-side revolution partly inspired the classification of VOC

Labor History 445


since VOC theory is based on data from the 1990s. In CMEs, mechanisms of coordinationand corporatism have not disappeared under the supply-side paradigm and, as mentionedbefore, institutional complementarities explain why coordination may survive in a marketeconomy at all.29 The question is legitimate whether a CME under the forces ofglobalization and neo-liberal domestic policy renewal develops into another type of CME(in which some institutions are replaced by others) or shifts into the direction of an LME.

The next section surveys the political changes in the Netherlands that took place duringthe 1980s and examines the policy results in fiscal discipline and private sector revival.Policy learning in the labor market, as analyzed by Visser and Hemerijck, was the result ofa reinvented responsive corporatism. Does learning also take place in macroeconomicpolicy, and if so, what were the effects on the CME?

Reform in the Dutch concertation economy

In the wake of Margaret Thatcher’s economic reforms in the UK and Ronald Reagan’sintroduction of supply-side economics in the USA, in 1982 the Dutch government adopted‘no nonsense’ politics. Although it raised some protest, there was compromise involved,and even if with hindsight it seems to have been an important change, the Dutch version ofsupply-side revolution was much more moderate than its Atlantic counterparts. ‘Nononsense’ politics introduced more rigid fiscal discipline and structural supply-side reform,replacing the belief in ‘big government’ and its capacities with an increasing belief in the‘free market’. In addition, it set the agenda for major reforms in the social security system.

Market deregulation and fiscal discipline were spearheads of the new policy measuresof the 1980s. The government had to find answers to the globalization of trade and capital,competition from low-wage countries, high oil prices and volatile exchange rates. This shiftin policy was intense in character, because in previous decades the Netherlands haddeveloped a generous and extensive welfare system with an extremely dominant staterole.30 Policy change in the Netherlands was not as revolutionary as in the UK underThatcher, nor as radical as in New Zealand, where the so-called Rogernomics produced acomprehensive transformation. Reform took a less jagged and protracted course than inSweden, where the social-democrat government tried to prolong its regulationist approachduring the 1980s. In the Netherlands, Keynesianism only existed in a moderate form,because there was no strict anti-cyclical fiscal policy, but the government was heldresponsible for supporting aggregate demand by supporting troubled branches of industry(such as textiles and shipbuilding) and by creating public employment opportunitiesduring periods of recession. After a number of transitional years, the government and thesocial partners accepted that a departure from this type of policy was necessary to dealwith persistent problems in the economy. As the OECD reported about the Netherlands:‘The deterioration in its growth and employment performance was indeed worse than thatof most other countries.’31 By the time the OECD wrote this, in 1985, recovery was alreadyon its way.

Regime change and regained consensus

In the period 1973–1982, the Dutch economy experienced stagnating economic growth,steadily increasing unemployment, decreasing investment, an entrepreneurial climate thatwas asphyxiated by high taxation, and a lack of budget discipline. Generous government

446 L.J. Touwen


expenditure and exports of natural gas mitigated the sense of crisis, but the latter also ledto a high exchange rate hindering foreign exports (known as ‘Dutch disease’). Until at least1977, the main belief was that economic problems were cyclical by nature and would

resolve when the business cycle picked up. The turnaround in economic policy in 1982 waspreceded by a four-year transition period during which the Prime Minister Dries van Agtled three cabinets. In his policy agenda Bestek’81 (published in 1978), Van Agt expressedclear political goals, which included restructuring industry and reducing governmentspending. However, these goals were not reached, owing to either weak statesmanship or

lacking support. In the plan, wage constraint should have restored the profitability of theprivate sector, but it did not come about. Social expenses should have been seriously cut,but the time was not right for severe budget cuts. Opposition to welfare state reforms,perhaps most clearly embodied by the Minister of Social Affairs Wil Albeda, effectivelyblocked budget cuts. Albeda, who was a very influential figure in Dutch industrial

relations, instead argued that voluntary wage moderation should be aimed for, with thesupport of the labor unions, avoiding welfare state reform. In 1979, Albeda nearlymanaged to organize a social pact (known as the ‘bijna-akkoord’, the ‘almost-agreement’).The peak workers organization FNV withdrew from the pact at the last moment. It wasdivided on the subject (the union of government employees in particular blocked theagreement) and did not yet accept wage moderation as a solution to the problems of

unemployment and economic stagnation.32

In 1982, the Christian-Democrat/Liberal (CDA/VVD) coalition cabinet of RuudLubbers came into power. This coalition was successful where Van Agt had failed.Although his coalition government was publicly viewed as rigorously pursuing its ‘nononsense’ policy, in fact Lubbers embodied compromise as both an entrepreneur and aformer minister of the left-wing cabinet of Den Uyl (1973–1977). He set out by makingsevere cuts in the government budget and started a process of privatization of state-owned

enterprises such as the state mail and telephone company PTT (privatized in 1989).33

He also prepared a reform of the unemployment law (realized in 1986).Policy change was simultaneously introduced in three fields: government budget, labor

market policy, and supply-side reform. This was called the ‘three-track policy’ foreconomic recovery (driesporenbeleid) and entailed the familiar goals of reducing thegovernment deficit (by restructuring the social sector), restoring profit margins in theprivate sector, and limiting unemployment by job redistribution. This started the regime

change that Van Agt had intended in 1978, and illustrates that the transition took place inseveral major policy fields simultaneously (not only in labor market policy, which iscentral in Visser and Hemerijck’s study).

It is remarkable that the pragmatic neo-liberal course was sustained when twelve yearslater Lubbers was succeeded as prime minister by Wim Kok, who had been the leader ofthe peak unions organization who signed the Wassenaar agreement. As the leader of thesocialist party PvdA, Kok gave a speech in 1995 in which he stated: ‘The old ideology is

not capable of answering the key questions of our times. Shaking off ideological feathers isnot only a problem for a political party like ours, in several regards it is a liberatingexperience as well.’ As De Rooy emphasizes, ideologies develop continuously and cannotbe discarded as a closed entity; however, as an expression of neo-liberal ideas, Kok’s wordswere unambiguously clear.34 Thus, the leader of the social-democrats openly continued the

pragmatic course facilitating further recovery of the supply side and supported the newparadigm. We may acknowledge an element of ‘power shift’ in the policy priorities of

Labor History 447


Lubbers in 1982, but eventually this was accepted as a solution to economic problems, alsoby the social-democrats. The corporatist system of checks and balances was held in place,a context of negotiations that was typical of a CME, but the employers received morebreathing space and obtained a stronger power base.

Priorities of economic policy in the Netherlands

In 1950, the Social-Economic Council (SER), the tripartite advisory council of theDutch government, stated five official government goals of economic policy. These fivegoals symbolized Keynesian thinking: (1) full employment; (2) a healthy balance ofpayments; (3) stability in price levels; (4) economic growth; and (5) an equal (or just)distribution of income. Combining these goals was not typically Dutch: the first four ofthem have been called the Keynesian ‘magic square’, which was rather successfullypursued in the 1950s and 1960 in many countries. In several countries, including theNetherlands, Denmark, Ireland and Austria, the magic square was restored in the 1990sand early 2000s.35 It is remarkable that in the Netherlands, income equality was addedas an official policy goal, and genuinely pursued between roughly 1950 and 1980, evenby right-wing coalitions.

‘Keynesian-type’ objectives such as full employment held a strong position on thepolicy agenda. In the original 1950 SER-report, a healthy balance of payments and stableprices were merely viewed as preconditions and thus subordinate to the other goals. Theexpansion of the welfare state and the institutional structure of concertation (in bodiessuch as the SER) facilitated the redistribution of income and the expansion of government-subsidized employment. Although only the left-wing cabinet of Joop den Uyl (1973–1977)turned redistribution of income into an explicit policy aim, other Dutch cabinets in theperiod 1950–1980 accepted this aim as well. In the mid 1970s, economic growth became apolicy goal with a lower priority: in committees of the SER and in official reports of theCentral Planning Bureau (CPB), the wish was expressed to concentrate more onimmaterial goals such as well-being and a healthy and sustainable environment – this wascalled ‘selective growth’.36 Selective growth can be viewed as ‘second order learning’because it included new policy instruments to deal with the recession, but failed to generatea breakthrough in policy views. This may have been due to the lack of economic urgency.As Pemberton states, ‘whilst networks may facilitate learning, competing networks mayblock the translation of this learning into effective policies’.37 The Netherlands containedmany such networks, among which the unions, and their blocking power had not yet beeneroded by mass unemployment.

In the 1980s, however, priority policy goals became economic growth and budgetcontrol. Full employment as a government responsibility and income redistribution as anexplicit policy goal were pushed to the background.38 The fact that these changes in manypolicy areas were accepted and supported by the labor organizations, who had to give upsome of their 1970s ambitions, is essential. The new regime managed to reset the agenda,but simultaneously sought corporatist consensus. Several solutions seemed traditionalrather than revolutionary, such as aiming towards wage constraint, a traditional tool thathad been successfully applied in the 1950s and was regarded as the Dutch answer torecession. However, the neo-classical emphasis on the market mechanism wasrevolutionary to many who believed firmly in Keynesian government regulation andshared a statist (or etatist) belief in the welfare state.

448 L.J. Touwen


A continuous fight against price inflation

The rise of monetarism in the 1980s did not constitute a major policy change in the

Netherlands. The explicit aim of restricting price inflation had been pursued since at least

1973 when, in reaction to inflationary pressure, the exchange rate was strengthened and

interest rates went up. Ever since the guided wage policy had been slowly abandoned

during the 1960s, a constant threat of an upward wage–price spiral had been present.

Director of the Dutch Central Bank (1967–1982) and former Prime Minister (1966–1967)

Jelle Zijlstra called Dutch economic policy under his authority ‘moderate monetarism’,

thus expressing the large weight given to monetary stability.39 There were several threats to

the stability of the Dutch guilder. After the wage restriction policy of the 1950s gave way

to more liberal wage formation in the 1960s, wages and prices spiraled upward, until in the

early 1970s the economic crisis brought down excessive wage claims. In the course of the

1970s a new inflationary pressure arose: the Dutch welfare state had linked social

provisions payments to the general wage and price level, which was facilitated by income

from natural gas exports.Between 1969 and 1973, inflation in the Netherlands was slightly higher than

inflation in surrounding countries (except in the UK). However, the growth rate of price

inflation hardly increased at this time, while in many surrounding countries it still did.

As Figure 2 shows, the consumer price index in the Netherlands after 1976 was

consistently below that of the USA and the average for the Euro-area (the 12 countries

that would introduce the Euro in 1999–2002). While Sweden applied drastic currency

devaluations during the period 1982–1991 in order to boost exports, the Netherlands

refrained from doing so, fearing inflation. The guilder stayed close to the German mark,

improving competitiveness with Germany (where wage moderation did not take place).

This may have slowed down recovery on the short run, but improved the economic

structure on the longer run.40

0.0

2.0

4.0

6.0

8.0

10.0

12.0

1976–1982 1983–1989 1990–1996

Netherlands

Euro-area

United States

Figure 2. Average consumer price index, Netherlands, Euro-area and the USA, 1976–96. Source:CPB, Macro Economic Outlook 2006.

Labor History 449


During the period 1973–1979, average annual inflation was 7.2%, while it was 15.6%in the UK and 16.1% in Italy. In 1979–1985, average inflation was 8.5% in the UK and4.4% in the Netherlands. Finally, in 1985–1994 average annual inflation in theNetherlands was only 1.7%, which was lower than in all surrounding countries. Astrong exchange rate helped to resist importing inflation from abroad. Also, increasingunemployment during the 1970s effectively dealt with the important inflationary cause ofthe 1960s, full employment, which had pushed wages upward.

Thus, theoretical views on monetarism did not fundamentally change monetary policyin the Netherlands. This was not part of the new priorities agreed upon, simply becausefighting inflation was already a high priority. In the late 1960s and early 1970s, priceinflation received a lot of attention in policy discussions. In the course of the 1970s, it lostthis priority position to the problem of unemployment.

Revitalizing the private sector

While monetarism was not new in the Netherlands, supply-side policy was. During the late1970s, the Dutch private sector suffered from high tax pressure. As I already mentioned,the policy goals of ‘selective growth’ at this time prolonged the expansion of welfareprovisions, aiming at a more equal income distribution, and increasing environmentalprotection. In hindsight, the new priorities of 1982 seem to be indicative of the theoreticalframework that replaced Keynesian ideals, often referred to as neo-liberal, monetarist, orneo-classical, and oriented towards the supply side. These new priorities surfaced among alarge group as the best way to deal with the problems, as a shared economic understanding.41

In literature on economic development, a distinction is made between macroeconomicpolicies aimed at economic stability, and liberalization policies aimed at structural reformand growth.42 While the common understanding nowadays is that increased welfarespending and the resulting tax pressure suffocated private investment in the Netherlands,there was initially no consensus on the necessity of cutting welfare provisions – although thecall for reformwas already expressed. However, many believed that the private sector wouldrecover automatically when world trade got back on its feet. In the meantime, investmentdwindled. Average TFP growth 1970–1985 in the Netherlands was 0.89% annually. WithNorway, UK and Sweden, it formed the slowest group of 14 OECD Countries (0.74, 0.61and 0.67% respectively).43 After 1981, investment picked up (Figure 3). The combinationof new industrial policy and wage restraint clearly boosted investment. Investmentsstarted to increase and private sector expenses on research and development increased.44

During the Golden Years of economic growth (1950–1970) a mechanism of expandingsocial spending had been set in motion that could not easily be stopped, despite warningsfrom many directions. During the 1970s, gas revenues and left-wing government policyreduced budget discipline. The expansion of the welfare state during the 1960s and 1970shad led, as in many other countries, to an increasing government budget, an increasing taxburden and, as a result, a declining profitability in the private sector. The ‘profit squeeze’that was harming Dutch companies during the 1970s was both the result of sluggisheconomic performance and increased tax pressure (which financed increasing amounts ofsocial spending).45 Thus, the high labor share of income put pressure on investment rates.

In part, the private sector was itself to blame for this situation. In a concertationeconomy employers play an active role in organizing welfare arrangements. Earlywarnings had been cast, even though the sudden downturn in the growth trend in the early

450 L.J. Touwen


1970s came rather unexpectedly. In 1976, a group of nine managers of large Dutch firmssent an urgent letter to the government, published as an open letter in the major Dutchnewspapers. They addressed problems such as economic stagnation, ballooning socialexpenditure, and strong opposition to industrial restructuring (unions and governmentblocked mass dismissals, and poorly performing industries such as the shipbuildingindustry were heavily subsidized). Their focus was explicitly on the structure of theeconomy, although they expressed confidence that the recession in the business cycle wouldquickly come to an end. This ‘Letter from the Nine’ argued that, since profits werecontinuously eroding, the financial vulnerability of many companies had increased, andinnovation and industrial growth were seriously endangered.46 The employers warned thatthe government budget deficit would boost interest rates, creating an additional burden forthe private sector. Maybe, they suggested, the social sector would have to shrink to preventworse things from happening. This early supply-side statement was met with muchindignation. It would take five more years before these proposals would gain ground.In 1976, politicians, unions, and public opinion were not ready for this message.

In September 1977, the Scientific Council for Government Policy (WRR, a governmentthink tank that has the task to supply scientific information on behalf of long term policy)published a report called Maken wij er werk van? (‘Shall we work on it?’).47 In this report,redistribution of jobs and the introduction of shorter working hours were offered assolutions to restore the low labor market participation. The report states that investmentshould be stimulated by the government and this should not be linked to labor marketpolicy, as was done previously. This is an obvious departure from previous Keynesiandemand-oriented policy: the number of new jobs involved was not a decisive factor

–15.0

–10.0

–5.0

0.0

5.0

10.0

15.0

20.0

1979 1981 1983 1985 1987 1989

Netherlands

Euro-area

USA

Figure 3. Gross fixed investment (private non-residential investments) after the second oil shock(annual growth rates). Source: CPB, Macro Economic Outlook 2006. Appendix A4, B4.

Labor History 451


any more. The report argues that economic policy urgently needed to be renewed, since theproportion of unemployed (or inactive) population was expected to increase substantiallyin the 1980s. The new idea was to make investments unrelated to the resulting increase inemployment (aimed towards profits, not towards Keynesian demand management). Thelarge role of the government, however, was not yet discarded: when these investmentsgenerated profits, jobs should be created by the government in public services. As one ofthe leading newspapers correctly observed, it was not clear how funds could be transferredfrom the private sector to the public sector without again harming the private sector byhigh taxation. The fact that economic growth was given a higher priority than fullemployment was called ‘shocking’ by a major newspaper. However, it was conceded thatwage differentiation was desirable to provide incentives for vocational training.48

This initiative of formulating new policy instruments, an example of second-orderlearning on the verge of third-order breakthrough, was only hesitantly accepted. Therepresentatives of the unions reacted by stating that the government should only stimulateinvestment if this created new jobs. Wim Kok, at that time chairman of the largest peakunion organization FNV, stated that employees should be consulted in decisions relatingto stimulating investment, and that investment in healthy (instead of troubled) firmsshould be preferred only if this also created new employment opportunities. In otherwords, he did not object to a new policy agenda as long as the unions could participate inthe decision-making process and employment remained a secondary goal (his commitmentto a constructive solution led him five years later to be one of the architects of theWassenaar Agreement). Many others union leaders were afraid that their position wouldbe weakened if it was openly stated that lucrative industries were allowed to merely focuson profits, without being compelled to create jobs.49

80.0

85.0

90.0

95.0

100.0

105.0

110.0

115.0

120.0

125.0

130.0

1975

1977

1979

1981

1983

1985

1987

1989

1991

1993

1995

1997

1999

2001

Netherlands

France

USA

UK

W-Germany

Figure 4. Changes in the labor share of income in processing industry (index 1991¼100). Source:CPB, Central Economic Plan 2003. Appendix B3. Note: Index figures show differences indevelopment relative to 1991¼100, and do not show absolute differences between countries.

452 L.J. Touwen


During the Keynesian era, the responsibility of the government for supportingemployment had resulted in generous support for weak or unprofitable branches ofindustry, in order to sustain them until the business cycle made an upward turn. In theNetherlands, particularly in the shipbuilding and textiles sectors, companies were backedby government funding during the period 1973–1977. These sectors were characterized bylabor-intensive production and suffered from international competition. After 1979 theattitude towards business support changed. In 1979 the Innovatienota stressed the need forinvestment in research and development, and the report Plaats en Toekomst van deNederlandse Industrie (‘Place and Future of Dutch Industry’) by the WRR marks a turningpoint by identifying the weaknesses of the Dutch economic structure.

Policy renewal in the wake of ‘Wagner’

The Place and Future report proposed an economic policy directed towards specificindustrial sectors and paid much attention to technological innovation. It advocated aprocess of re-industrialization, inspired by the industrial policy of the 1950s. Thus, it was amix of old and new: maintaining a central role for the government but aiming at structuralreform of the private sector.

In answer, the government set up an advisory committee, the ‘CommissieWagner’ (named after its chairman G. Wagner), which in 1981 issued its report A NewIndustrial Elan.50 Even more than the earlier WRR-report, the Wagner committee hadswitched to a modern approach and advocated industrial self-support. The governmentshould refrain from active intervention and stimulate decentralization. High wages and ahigh exchange rate of the guilder were mentioned as causes for the lack of competitivepower on the international markets. In fact, it did two things: it advocated a new industrialpolicy, and also represented a broad change in thinking. An important characteristic of theWagner Committee was that it included representatives of labor unions.51

In Figure 4, the development of the labor share of income of five countries iscompared, in time series using 1991 as a standard year (the graph shows developmentsrelative to 1991). A rough indicator of the profitability of the private sector, the laborshare of income displays a turning point in 1981, after which year the trend turnsdownwards and a larger share of the gains becomes available for investment. The laborshare of income in the Netherlands decreased impressively after 1981. Although a rathercrude measure, it is highly significant in showing the share of labor cost in themacroeconomy, irrespective of whether it is high wages, high income tax or high socialsecurity fees that make labor expensive to employers. The impressive decrease of the laborshare may be viewed as the combined effect of wage restraint and lower taxes in theNetherlands in the 1980s. It also reflects the increased breathing space for capital gainingsuperiority over labor as a result of the supply-side revolution.

Visser and Hemerijck, who do not study industrial policy but focus on corporatistpolicy renewal in the labor market (in the three domains of labor relations, welfare state andlabor market policy), pay little attention to the Wagner Report, although its implicationswere manifold. They do acknowledge its scope and vision in a brief paragraph but do notsystematically trace the effects of the report on negotiations between labor and capital.Visser and Hemerijck state that in 1982, the signing of the Wassenaar Agreement signifiedthe end of ‘immobile corporatism’ of negotiations without agreement. Afterwards, amore ideal form of ‘responsive corporatism’ emerged. The change, according to their

Labor History 453


hypothesis, was the result both of government authority dictating the new paradigm, and ofa change in the power balance (the weaker party must learn how to cope!).52

Acknowledging that the Wassenaar Agreement signified a turning point in laborrelations, I would like to suggest that the Wagner Report was a strong symbol of thechanging attitude towards the supply-side – a profitable and more dynamic private sector.Its meaning was threefold. First, its ideas conformed with the views of the center-rightgovernments since 1978, which only gained ground in 1982. Second, it explicitly paidattention to a shift of power towards the private sector. Third, since the labor unions wererepresented in the Wagner Committee, it shows that the unions learnt to adapt to the newparadigm, which formed the basis of voluntary wage restraint.

The Wagner Report supplied the ideological basis for replacing governmentintervention with the market mechanism, thus supplying the ‘learning materials’ fromwhich the policy learning in the specific domains of the labor market (a more flexible labormarket, lower social entitlements, a smaller tax burden, revitalizing the private sectorinstead of creating employment in the public sector) were learnt. Therefore, it signifiedthird-order learning in the CME. When the economic crisis was at its nadir and, soonafterwards, when the economy started to recover, the corporatist solutions ‘in the wake ofWassenaar’ could gain ground, but these could only take place after the first major stephad been taken. In contrast with Visser and Hemerijck, I want to emphasize that theWagner Report provided the turning point and signified the paradigmatic change. TheWassenaar Agreement and its aftermath were rather an elaboration of the changed viewon the economy, which was now typified by terms such as self-activation, more emphasison the ability to manage for oneself, and a retreat of government.

Labor relations are necessarily subordinate to the economic situation. However, whilethe Wassenaar Agreement was signed at times of economic urgency, its top–downrealization, which according to Visser and Hemerijck is typical of a modern economy, tookplace in the years after 1982, when wage restraint was sustained even though the economypicked up. Economists state that at the core of the recovery in the late 1980s wereexogenous factors, such as the recovery of the world economy (which was itself largely dueto ongoing trade liberalization) and progressing European integration (e.g. the SingleEuropean Act of 1986). It has been demonstrated that in the Netherlands economic policywas effective in facilitating the profiting from these opportunities.53

It is my contention that the successful economic policy renewal, which was proclaimedin the Wagner Report, effectuated a turnaround which formed the basis of the improvinglabor market. By introducing deregulation and promoting flexibility, this transformed theold-style CME in the Netherlands into a more liberal type of CME. In short, ‘Wagner’facilitated ‘Wassenaar’.

Controlling the government budget and reforming the welfare state

The policy aim of controlling government expenditure and restricting the budget deficit wasnot new, but the degree to which it was deemed necessary was definitely subject to change.In the 1960s, the Minister of Finance Jelle Zijlstra developed a ‘structural budget norm’,stipulating that the budget should only expand at the same rate as the growth trend in theeconomy (disregarding short-term fluctuations). Zijlstra hoped to achieve an anti-cyclicaleffect, or rather, to prevent pro-cyclical spending. On and off, in later years concern wasexpressed about the increasing deficit in the government budget. In a relatively small

454 L.J. Touwen


country with a strong concern for cautious monetarist policy and an open economy, policymakers frequently discussed where to draw the line when the deficit again increased. Theassistance of income from natural gas exports mitigated the urgency of this problem.

In 1975, the Minister of Finance Wim Duisenberg (later known to be the first president

of the European Central Bank) introduced the ‘one per cent operation’, aiming for astricter budget policy, since economic stagnation had disrupted the previously blissfuleconomic outlook. This policy initiative attempted to restrict the increase of thegovernment deficit, while waiting for the upturn of the business cycle, by limiting theincrease in the share of social spending in the national product to 1% annually. Discussioncontinued as to whether this was adequate, particularly since the ‘spending departments’

did not manage to reduce their expenses at all. The aim was merely to limit the increase ofthe share of social spending in the national product. In 1978, the government programBestek’81 had sound ambitions in the field of budget discipline but failed to reducegovernment expenditure. Of course, social security entitlements made it difficult to reducesocial spending during an economic recession. Social payments were linked to the general

wage level, which in turn was linked to the inflation rate (the Lubbers I cabinet unlinkedthem). When the second oil crisis in 1979 could not be accommodated, the resultingrecession helped to change the climate. Eventually it was recognized that the governmentbudget had to be seriously reduced. In 1982–1989, the Minister of Finance Onno Rudingmade severe budget cuts, particularly by reducing social benefits and salaries of civil

servants. The unions refrained from organizing strikes to protest against these reductionsbecause of the high levels of unemployment. In 1986 unemployment benefits were reducedand in 1988 a reduction of income taxes (the so-called ‘Oort’ tax reform) was feasible. WimKok, the Minister of Finance in the period 1989–1994, continued the austerity policy,going in for severe budget cuts, which were necessary due to the recession of the early1990s.54 When in the second half of the 1980s the share of government spending in GDP

fell (see Figure 5), this was in part the result of economic recovery. Only in the course ofthe 1990s did the share of total government expenditure in the national product movebelow that in France and West Germany.

In 1984 total government expenses peaked at more than 62% of GDP, but this yearalso signified a turning point in Dutch government finances. Although this appears ratherpromising, the annual deficit did not disappear. This persistent deficit in the 1980s and1990 can be explained by new types of expenditure accompanied by tax reduction. The

deficit was only seriously reduced in 1991, when the policy towards European monetaryunification as well as a cyclical boom period coincided (Table 1). In the 1990s, the annualgovernment deficit remained under the European average.

The learning period, which resulted in aiming at a smaller government share ofthe national income, took roughly 10 years (1972–1982), but it took another 10 yearsto take effect. Pressure groups, unions, left-wing parties, advisory councils, andspending departments experimented with various ways to cope with recession and

eventually accepted that budget cuts were necessary, if not to reduce the deficit, thenat least to prevent further increase. Within the Keynesian framework, the cyclicaldownturn was expected to be temporary and the government deficit was expected tobuild the bridge towards recovery. It was only when the broad insight disseminatedthat there were structural problems in the economy, and that large government

expenses were part of the problem, that measures to limit further rise of the deficitbecame feasible.

Labor History 455


To summarize, concern about the size of the government expenses and the budgetdeficit was not new at all, but it was given new priority in the early 1980s. Measures had theeffect of limiting excessive growth of government expenditure, rather than reducing totaloutlays. This reflects the consociational character of the political economy: the governmentmade efforts to reduce specific outlays, but pressure groups of various sorts claimedexpenses in other areas.55 Eventually budget discipline triumphed. With a shrinking role forthe government budget, redistribution is limited. The government is no longer willing toreduce unemployment by offering public sector jobs, and the private sector gets moreleeway. The outcome of this was economic growth and new employment opportunities, butdid it not erode the existent mechanisms of coordination by changing the balance of power?

Did policy learning change the CME?

Policy renewal and learning for a large part consisted of setting new priorities in theeconomic agenda, and acceptance of these priorities by the social partners. This is part ofthe learning process because in the tradition of neo-corporatism the ‘conservativecommunity’ (as Kuhn terms it) had to agree to the new paradigm. In the years following

20

25

30

35

40

45

50

55

60

65

1960 1969 1972 1975 1978 1981 1984 1987 1990 1993 1996 1999

Netherlands

France

W Germany

UnitedKingdom

United States

Figure 5. General government total outlays (as a percentage of nominal GDP). Source: OECD.

Table 1. Average government balances for several countries, 1985–99 (surplus (þ) or deficit (�) as apercentage of nominal GDP).

Netherlands France United Kingdom European Union United States

1985/89 �4.5 �2.5 �1.2 �4.0 �4.31990/94 �3.7 �4.0 �5.1 �5.3 �4.81995/99 �1.4 �3.4 �2.2 �3.0 �1.0

Source: OECD.

456 L.J. Touwen


1982, a renewed consensus developed among both unions and employers’ organizations.This ‘corporatist’ solution in itself was not new in the Dutch setting. The result of ‘learning’was more or less unanimously accepting specific new priorities in order to solve persistent

unemployment and to accept elements of supply-side economics. Some commentators, forexample Delsen, are pessimistic about the survival of concertation and corporatism afterthe neo-liberal wave of the last decades.56 By contrast, VOC argues that negotiatinginstitutions are firmly embedded and form a platform through which institutional renewalis modified before it is implemented.57 According to VOC a recurrent encouragement of the

market as a coordinating mechanism does not necessarily transform the CME into a LME.If, for example, trust in networking, stable industrial relations and commitment ofpersonnel are deeply embedded characteristics of the CME, it can be argued thatderegulation can take place within this setting. If the comparative institutional advantagesare substantial, employers will still prefer the coordinated solution, but is the institutional

setting in a CME a longue-duree parameter that survives short-term fluctuations?Upon scrutiny, this view of a persistent culture of coordination and trust is not entirely

convincing. Their presence may be robust, but their influence on economic decisionmaking will fluctuate. In the Netherlands during the 1980s the private sector was givenmore leeway, tax pressure was reduced, the labor market was made more flexible, andwages were constrained. As a result, Dutch economic growth returned in the wake of therecovery of the international economy. Increasing public approval of profitable

entrepreneurship (not thought of in the 1970s), government deregulation, and a shift inthe balance of power all contributed to offer firms other strategic options than thosetypical of a CME. Firms could still choose to provide long-term employment, to leave lawsrestricting dismissal intact, to keep close ties with the works council (just to mentionseveral VOC institutional subspheres), but when offered the choice, they could as well rentshort-term personnel, pressure the government to lift dismissal restrictions, and ignore the

works council. In other words, at times, firms could prefer to maximize profits by ignoringslow and painful negotiations. In my opinion, in the process of shedding Olsoniansclerosis, renewal in the Dutch economy weakened the institutional complementarities thatconstitute the stereotypical CME.

In her survey of Dutch business in the twentieth century, Sluyterman regards the Dutcheconomy as a CME with features of a LME, which is in fact nominally anathema to VOC,which bases its explanation on the strength of the two opposite ideal types.58 Schmidt, in

her thorough study of types of European capitalism, argues that within categories changesdue to globalization and European integration do occur, but they leave countries withintheir original category (managed capitalism for the Netherlands, Germany, and Sweden;state capitalism for France and Italy; and market capitalism for the US and the UK). In thecontext of France, Schmidt applies the term ‘moderate neo-liberal paradigm’, indicating

that socialist governments have attempted to introduce painless reforms.59 Responsivecorporatism in the Netherlands could not avert neo-liberal reform but it did mitigate itsfocus on the market, which could inspire us to adopt the term ‘moderate neo-liberalparadigm’ for the Dutch CME as well.

In reaction to Visser and Hemerijck, Becker and Schwartz argue that the term ‘miracle’is not appropriate since it is too country-specific and includes too much attention to theassessment of policy choices (‘pluck’). They argue it makes no sense to examine whether

these were ‘ineffective or effective in setting things right’, since they might have beenirrelevant in the view of the larger global economic environment. Instead, according to

Labor History 457


Becker and Schwartz, a larger part of the explanation of success lies in external factors(‘luck’) and in embedded practices conditioned by accidental or incidental qualities oforganizations involved (‘stuck’).60

This emphasis on continuity is useful, but underestimates the amount of change. Itunderstates the fact that within the existing context, policy learning facilitated profitingfrom external change. Applying the concept of ‘learning’ emphasizes that ‘plucking’ is notan isolated process but stems from experiences elsewhere. Lessons learnt in economic policyaround 1980 included that the private sector should be able to expand in profitable activitieswhich do not immediately create new jobs, and that decentralization and deregulationcreated a more beneficial operating environment for private enterprise and (previously)state-owned companies. These ‘supply-side lessons’ were implemented within the frame-work of the concertation economy. Trial and error of course included initiatives that laterturned out to be irrelevant or unsuccessful. The major message of the word ‘miracle’ is thatspecific lessons were learned that provided solutions for problems urgent at the time.

Although the mechanism conveying new policies was typical of the corporatisteconomy or CME, the result of the changes in the 1980s was a more liberal economy.In several subspheres, firms outweighed possible gains by institutional complementaritiesagainst direct gains from a market-oriented way of solving coordination problems.Introduction of more flexibility in the labor market facilitated the recruitment of short-term workers instead of re-training tenured personnel. Works councils lost some of theirinfluence because their interference could hinder competitive efficiency of the firm.Although the service sector was more sheltered from international competition thanmanufacturing, the business climate changed and became more competitive. As a result,the CME in the 1990s was more market-oriented than the CME in the 1970s, even thoughthe basic institutions of coordination and corporatism survived.

A critical observer may remark that ‘learning’ is a euphemism for being forced to acceptthe message of the winners – reducing policy change to a shift in power. Likewise, Visserand Hemerijck state that policy learning in a corporatist state is both ordered from aboveby the government, and made possible by a shift in the balance of power that forces theweaker party to accept new views.61 Firms and government together in their negotiationscan convince the unions to be more flexible and to accept the withdrawal of regulations.Within the institutional setting of organized concertation a shared economic understandingwas necessary. The new paradigm was accepted by various groups partly because they wereweaker, and partly because they accepted its inevitability. At the moment that the regimechange took place in the Netherlands, consensus between employers and labor unions wasrestored, after more than a decade of increasing antagonism. However, it was aconsensus in which the unions partly accepted defeat. The social pact that resulted inwage constraint also comprised accepting the abolition of the Keynesian belief in help fromthe state in creating jobs. This increased the elasticity of institutional complementarities: inthe new CME, the institutional subspheres were less closely linked. Employers were tolesser extent forced or tempted to play within the rules of the CME.

Conclusion

Crouch criticizes VOC theory because ‘typologies of this kind are fixed over time; they makeno provision for changes in characteristics’.62 This article has attempted to connect third-order learning to varieties of capitalism by studying the process of change in the CME.

458 L.J. Touwen


How did the paradigm change of the 1980s affect the Dutch CME? This article assumed asa starting point that the Dutch economy in the 1970s was closer to the ideal type CME as

decribed by Hall & Soskice. In VOC, which is based on the situation in the 1990s, Germanyis closest to an ideal type CME, and the Netherlands is slightly more liberal in character.

Coordination requires information exchange and regular consultation (this evenincludes far-reaching qualities such as trust and consensus). For a major change in

economic policy in a corporatist economy, a broad third-order learning process is required(best described as a paradigm change) because many groups have to support the changes.According to Visser and Hemerijck, this happened in the Netherlands when ‘immobile

corporatism’ turned into ‘responsive corporatism’ in 1982.Three main conclusions are drawn. First, I argue that successful policy renewal in

the context of neo-corporatism affected the institutional foundations of the economy.Reform did not only take place in the labor market, although this has received the most

attention in the literature. By defining the entrepreneurial climate, the paradigm changeincreased the strategic options available to firms in the CME (in effect, the measuresinduced the choice for liberal, non-coordinated options). During the 1980s, a process of

institutional third-order learning can be observed in various policy fields simultaneously.The ‘paradigm shift’ was the result of policy learning among politicians, governmentofficials and union-leaders, and led to a realignment of groups within the concertation

economy. The outcome weakened the level of coordination in the CME because theinstitutional elasticity increased: decisions in one institutional subsphere had smallereffects on decisions in co-specific institutions.

In economic theory, complementarities in demand for goods are illustrated using the

example of ‘bread’ and ‘butter’; an increase in demand for bread will raise the demand forbutter. In this analogy, when margarine is introduced, the demand for butter will be lessclosely linked to bread: customers choose whether they buy either butter or margarine with

their bread. In the Dutch labor market, policy reform resulted in more flexibility. Becauseof these reforms (especially introduction of short-term flex-jobs), the ‘freedom to hire andfire’ of top management of firms (typical for LME) increased. When the top management

can pick from various options (either the ‘coordinated’ solution including co-specificbenefits, or the ‘liberal’ solution), this reminds us of the introduction of margarine next tobutter: the institutional complementarities weaken.

At the time of their introduction, economic urgency and a change in paradigm favored

market-oriented solutions, and joined forces in overruling the search for alternatives basedon possible institutional complementarities. The neo-corporatist approach aptly concen-trates on the changing power relations and the reduced influence of the unions. Visser and

Hemerijck observe that successful Dutch policy renewal was the result of renewed,responsive, neo-corporatist concertation. In turn, this led to a change of the institutionalbase to a more liberal type: a movement of the CME to include more liberal

characteristics. It implied that productive gains on the market sometimes supersededgains from institutional complementarities (which would have reinforced ‘coordinated’solutions). Market-based solutions were now more frequently applied in the CME,

weakening coordination. When capital has more power, its range of options will increase.As a second conclusion, the firm-centered perspective provided by VOC inspires us to

pay more attention to the effects of macroeconomic development and macroeconomicpolicy in the institutional analysis of the ‘Dutch miracle’. The circumstances in which firms

operate are not only defined by the business cycle, but also by the economic policy

Labor History 459


paradigm and the institutional setting (CME of LME). Owing to economic urgencyvarious corporatist groups learned to accept wage moderation and deregulation.International competitive pressure formed part of this pressure to alter the macroeconomicparadigm. Visser and Hemerijck focus on the triplet labor relations, labor market policy,and welfare state and do not extend their argument to institutional effects ofmacroeconomic policy. Focusing on the private sector provides us with an understandingof policy learning in the 1980s: it was the need to breathe air into the private sector thatencouraged the paradigm change and laid at the core of the ‘job miracle’. In order to createthe ‘Dutch miracle’ proclaimed by Visser and Hemerijck, a belief in the need toreinvigorate the private sector was necessary. Thus, the supply-side revolutionaccompanied (and even preceded) policy learning in the labor market. The turningpoint was the Wagner Report rather than the Wassenaar Agreement.

Third, the economic performance of the Dutch economy in the 1980s and 1990s showsthat a CME can introduce market-based solutions to various types of economic problems,without necessarily damaging economic results (as is suggested by Hall and Gingerich) orundermining the corporatist structure. The mixed outcome is not necessarily negative ascompared to the ideal types of a CME and LME. Within the Dutch CME, the newcompromise between labor and capital constituted more room for liberal solutions andresulted in weaker institutional complementarities.63 All European CME’s struggling withmacroeconomic reform such as introducing more fiscal discipline and deregulating labormarkets will face these complicated policy choices. In the new CME, institutionalsubspheres are less closely linked, weakening the coordination mechanism. The varioussocial actors will closely watch policy renewal and attempt to prevent further changetoward either extreme.

Acknowledgement

I would like to thank the reviewers of this journal for their useful comments on an earlier version ofthis article.

Notes on contributor

L. Jeroen Touwen works as a Lecturer in Economic and Social History at Leiden University. He haspreviously published a book on the late-colonial development in Indonesia, Extremes in theArchipelago: Trade and Economic Development in the Outer Islands of Indonesia, 1900–1942 (Leiden:KITLV Press, 2001) and various articles on comparative capitalism.

Notes

1. Hall and Soskice, Varieties of Capitalism; Hall and Gingerich, ‘Varieties of Capitalism’; Crouch,‘Models of Capitalism’; Korpi, ‘Power Resources’; Goodin, ‘Choose your Capitalism’; Schmidt,The Futures, 111.

2. According to VOC, comparative institutional advantages result in a proclivity for radicalinnovation in LME and for incremental innovation in CME. Innovation will not be treated inthis paper. On static aspects of VOC, see Crouch, ‘Models of Capitalism, 445; Hall and Thelen,‘Institutional Change’; Touwen, ‘Varieties of Capitalism’.

3. Hall and Soskice, Varieties of Capitalism, 4.4. Swenson, Capitalists against Markets.5. Hall and Soskice, Varieties of Capitalism, 3.6. This ‘is likely to entail less class conflict’. Hall and Soskice, Varieties of Capitalism, 58.

460 L.J. Touwen


7. Thelen, How Institutions Evolve. Elsewhere, Streeck and Thelen elaborate on different types ofinstitutional change: see Streeck and Thelen, Beyond Continuity.

8. Hall, ‘Policy Paradigms’; Visser and Hemerijck, ‘A Dutch Miracle’, in particular 73–74, 183–4;Becker and Schwartz, Employment ‘Miracles’; see also Keman, ‘Explaining Miracles’; on Dutchemployment 1974–1984 in a comparative perspective, see Therborn, Why Some People.

9. The Wassenaar Agreement was named after the home town of the chairman of the peakemployers organization, C. van Veen, where the pact was signed. It can be found at http://www.stvda.nl/_uploads/akkoord_van_wassenaar.pdf [October 2007].

10. Employers were against collective arrangements for shorter working hours (centrally issued bythe government), which was an aim of the labor movement. The employers won this debate.After 12 months the unions changed this goal to part-time jobs and more flexibility in workinghours. Visser and Hemerijck, ‘A Dutch Miracle’, 102–3.

11. Windmuller and De Galan, Arbeidsverhoudingen in Nederland.12. Ibid., 96–113; Van Zanden, The Economic History, 88.13. Van Bottenburg, ‘Aan den Arbeid’, 193.14. Note that unemployment rates disregard large numbers of workers in early pensions or

inability schemes. Thus, ‘real’ unemployment is often viewed to be higher, but a downwardtrend is still observed. Standardized OECD unemployment data has been adjusted toensure comparability over time and to conform to the guidelines of the International LaborOffice.

15. Visser and Hemerijck, ‘A Dutch Miracle’, 17, 58–60.16. Hall and Soskice, Varieties of Capitalism, 7.17. Crouch, ‘Models of Capitalism’, 450.18. See for example Amable, ‘Institutional Complementarity’.19. Hall, ‘Policy Paradigms’.20. Social science paradigms have an element of fashion (or even cyclical movement) in them

whereas scientific paradigms propel science irreversibly forward. Scholars distinguish between‘radical’ and ‘conservative’ interpretations of paradigms, in which the conservative definitionattempts to separate cognitive and social components from science. See Pinch, ‘Kuhn’.

21. Kuhn, The Structure of Scientific Revolutions.22. Visser and Hemerijck, ‘A Dutch Miracle’, 62.23. Ibid., 63–72; Van den Toren, ‘Meerstromenland ’.24. See for example Hall, ‘Policy Paradigms’; Hay, ‘The ‘‘Crisis’’ of Keynesianism’; Campbell and

Pedersen, ‘Introduction: The Rise of Neoliberalism’. Policy learning concerns more than thepolitical preferences of the government. The term ‘regime change’ is used for a persisting changein direction.

25. Mesguer, ‘Policy Learning’, 69.26. Hall, ‘Policy Paradigms’, 279, 283–4; Visser and Hemerijck, ‘A Dutch Miracle’, 58–61.27. Clift and Tomlinson, ‘Credible Keynesianism?’.28. Beaud and Dostaler, Economic Thought, 134–7.29. Touwen, ‘Varieties of Capitalism’, 103.30. Van Kersbergen, Social Capitalism, 128–36.31. OECD, OECD Economic Surveys 1984/1985, 7.32. Bruggeman and Camijn, Ondernemers Verbonden, 265; Toirkens, Schijn en Werkelijkheid, 70–9;

Albeda, Ik en de Verzorgingsstaat, 201–7; Visser and Hemerijck, ‘A Dutch Miracle’, 97.33. Davids and Van Zanden, ‘A Reluctant State’.34. De Rooy, Republiek der Rivaliteiten, 278–9.35. Becker and Schwartz, Employment ‘Miracles’, 11–12.36. Centraal Planbureau, De Nederlandse Economie, 90–1; SER, Advies.37. Pemberton, ‘Learning, Governance’, 500.38. One can obtain a clear picture of changing policy ideas by studying the policy directives of the

Ministry of Economic Affairs (Rutten, Zeven Kabinetten Wijzer).39. Zijlstra, Gematigd Monetarisme.40. Van Zanden, The Economic History, 175; Henrekson, ‘Economic Growth’, 255.41. Compston argues that three variables are necessary to make lasting three-party agreements

between employers, unions and government: a sense of urgency in the perceived problems,a degree of shared economic understanding among the participants, and a perceived

Labor History 461


implementation capacity of the participants. In the second variable, a degree of learning orparadigmatic change may be at hand. Compston, ‘Beyond Corporatism’.

42. Rodrik, ‘Understanding Economic’, 11.43. Henrekson, ‘Economic Growth’, 245; Van Ark, ‘Characteristics’, 299.44. Van Ark et al., ‘Characteristics’, 309; Bruggeman and Camijn, Ondernemers Verbonden, 271.45. For a survey of Dutch business history in the twentieth century, see Sluyterman, Dutch

Enterprise.46. Van den Brink, Zoeken naar een Heilstaat, 375.47. WRR, Maken Wij Er Werk Van.48. NRC Handelsblad, 8 September 1977.49. NRC Handelsblad, 1 October 1977.50. Published in Van Dellen, Een Nieuw Elan.51. Bruggeman and Camijn, Ondernemers Verbonden, 269–71.52. Visser and Hemerijck, ‘A Dutch Miracle’, 73, 99.53. Van Ark et al., ‘Characteristics’, 311–2.54. Collective expenditure, which refers to government outlays and social security transfers, peaked

in 1982–1983 at over 66% of GDP but its share was reduced to 54% of GDP in 1995.Van Zanden, The Economic History, 69–70; Postma, Honderd jaar Miljoenennota, 74–75.

55. Toirkens, Schijn en Werkelijkheid; Olson, The Rise and Decline of Nations.56. Cf. Delsen, Exit Poldermodel.57. Hall and Soskice, Varieties of Capitalism, 5, 9–12; Hall and Soskice, ‘Varieties of Capitalism’.58. Sluyterman, Kerende Kansen, 323–4 (see under Sluyterman, Dutch Enterprise).59. Schmidt, The Futures, 112–113, 282.60. Becker and Schwartz, Employment ‘Miracles’, 14.61. Visser and Hemerijck, ‘A Dutch Miracle’, 77–79 182.62. Crouch, ‘Models of Capitalism’, 443–4.63. Cf. Goodin, ‘Choose your Capitalism’, 206; Hall and Gingerich, ‘Varieties of Capitalism’;

Thelen, ‘Varieties of Labor Politics’, 74, 78.

References

Albeda, Wil. Ik en de Verzorgingsstaat. Herinneringen van Wil Albeda. (I and the welfare state.

Memoirs of Wil Albeda). Amsterdam: Boom, 2004.Amable, Bruno. ‘Institutional Complementarity and Diversity of Social Systems of Innovation and

Production’. Review of International Political Economy 7, no. 4 (2000): 645–87.Beaud, Michel and Gilles Dostaler. Economic Thought since Keynes. A History and Dictionary of

Major Economists. London: Routledge, 1995.Becker, Uwe and Herman Schwartz, eds. Employment ‘Miracles’: A Critical Comparison of the

Dutch, Scandinavian, Swiss, Australian and Irish Cases versus Germany and the US.

Amsterdam: Amsterdam University Press, 2005.Bruggeman, Jan and Aart Camijn. Ondernemers Verbonden. 100 jaar Centrale

Ondernemingsorganisatie in Nederland (Entrepreneurs United. 100 years of Peak Employers

Organsiations in the Netherlands). Wormer: Inmerc, 1999.Campbell, John L. and Ove K. Pedersen. ‘Introduction: The Rise of Neoliberalism and Institutional

Analysis’. In The Rise of Neoliberalism and Institutional Analysis, eds. J.L. Campbell and

O.K. Pedersen, 1–23. Princeton: Princeton University Press, 2001.Centraal Planbureau. De Nederlandse Economie in 1980 (The Dutch Economy in 1980). The Hague,

Staatsuitgeverij, 1976.Clift, B. and J. Tomlinson, ‘Credible Keynesianism? New Labour Macroeconomic Policy and the

Political Economy of Coarse Tuning’.British Journal of Political Science 37, no. 1 (2007): 47–69.Compston, Hugh. ‘Beyond Corporatism: A Configurational Theory of Policy Concertation’.

European Journal of Political Research 42, no. 6 (2003): 787–809.

462 L.J. Touwen


Crouch, Colin. ‘Models of Capitalism’. New Politcial Economy 10, no. 4 (2005): 439–56.Davids, M. and J.L. van Zanden. ‘A Reluctant State and Its Enterprises. State-owned Enterprises in

the Netherlands in the ‘‘Long’’ Twentieth Century’. In The Rise and Fall of State-owned

Enterprise in the Western World, ed. Pier Angelo Toninelli, 253–72. Cambridge: CambridgeUniversity Press, 2000.

De Rooy, Piet. Republiek van Rivaliteiten. Nederland sinds 1813 (Republic of Rivalries.

The Netherlands since 1813). Amsterdam: Mets en Schild, 2001.Delsen, Lei. Exit Poldermodel? Sociaal-economische Ontwikkelingen in Nederland (Exit Poldermodel?

Social-economic developments in the Netherlands). Assen: Van Gorcum, 2000.

Goodin, Robert E. ‘Choose your Capitalism’. Comparative European Politics 1 (2003): 203–13.Hall, P.A. ‘Policy Paradigms, Social Learning, and the State. The Case of Economic Policy Making

in Britain’. Comparative Politics 25 (1993): 275–96.

Hall, Peter A. and Daniel W. Gingerich. ‘Varieties of Capitalism and InstitutionalComplementarities in the Macroeconomy: an Empirical Analysis’. In Discussion Papers,Max Planck Institut fur Gesellschaftsforchung, Cologne, 2004. Available at http://www.

mpifg.de/pu/mpifg_dp/dp04-5.pdf (last accessed 31 August 2008).——— and David Soskice, eds. Varieties of Capitalism. The Institutional Foundations of Comparative

Advantage. Oxford: Oxford University Press, 2001.

———. ‘Varieties of Capitalism and Institutional Change: A Response to Three Critics’.Comparative European Politics 1 (2003): 241–50.

——— and Kathleen Thelen. ‘Institutional Change in Varieties of Capitalism’, paper prepared for

the Europeanists Conference, Chicago, March 2006.Hay, Colin. ‘The ‘‘Crisis’’ of Keynesianism and the Rise of Neo-liberalism in Britain: An Ideational

Institutionalist Approach’. In The Rise of Neoliberalism and Institutional Analysis,

eds. J.L. Campbell and O.K. Pedersen, 93–218. Princeton: Princeton University Press, 2001.Henrekson, Magnus, Lars Jonung, and Joakim Stymne. ‘Economic Growth and the Swedish

Model’. In Economic Growth in Europe since 1945, eds. Nicholas Crafts and Gianni Toniolo,

240–56. Cambridge: Cambridge University Press, 1996.Keman, Hans. ‘Explaining Miracles: Third Ways and Work and Welfare’. West European Politics

26, no. 2 (2003): 115–35.

Korpi, Walter. ‘Power Resources and Employer-centered Approaches in Explanations of WelfareStates and Varieties of Capitalism. Protagonists, Consenters and Antagonists’. World Politics58 (2006): 167–206.

Kuhn, Thomas S. The Structure of Scientific Revolutions. Chicago: The University of Chicago Press,1970.

Mesguer, Covadonga. ‘Policy Learning, Policy Diffusion, and the Making of a New Order’. Annals

of the American Academy of Political and Social Sciences 598 (2005): 67–82.OECD. OECD Economic Surveys 1984/1985, Netherlands. Paris: OECD, 1985.Olson, Mancur. The Rise and Decline of Nations. New Haven, CT: Yale University Press, 1982.

Pemberton, Hugh. ‘Learning, Governance and Economic Policy’. British Journal of Politics andInternational Relations 5, no. 4 (2003): 500–24.

Pinch, T.J. ‘Kuhn – The Conservative and Radical Interpretations: Are Some Mertonians

‘‘Kuhnians’’ and some Kuhnians ‘‘Mertonians’’?’ Social Studies of Science 27, no. 3 (1997):465–82.

Postma, J.K.T. Honderd Jaar Miljoenennota (A Hundred Years of Miljoenennota). Den Haag,

SDU, 2006.Rodrik, Dani. ‘Understanding Economic Policy Reform’. Journal of Economic Literature 34 (1996):

9–41.

Rutten, F.W. Zeven Kabinetten Wijzer. De Nieuwe Zakelijkheid bij het Economische Beleid(Seven Cabinets Wiser. New Realism in Economic Policy). Groningen: Wolters-Nordhoff,1993.

Schmidt, Vivien A. The Futures of European Capitalism. Oxford: Oxford University Press, 2002.

Labor History 463


SER. Advies inzake het Sociaal-economisch Structuurbeleid in de Periode 1975–1980, CommissieOntwikkeling Nationale Economie. The Hague, Staatsuitgeverij, 1976.

Sluyterman, Keetie E. Dutch Enterprise in the Twentieth Century: Business Strategies in a Small OpenEconomy. London: Routledge, 2005 (published in Dutch as: Kerende kansen. Het Nederlandse

bedrijfsleven in de twintigste eeuw. Amsterdam: Boom, 2003).Streeck, Wolfgang, and Kathleen Thelen, eds. Beyond Continuity. Institutional Change in Advanced

Political Economies. Oxford: Oxford University Press, 2005.

Swenson, Peter A. Capitalists against Markets. The Making of Labor Markets and Welfare States inthe United States and Sweden. Oxford: Oxford University Press, 2002.

Thelen, Kathleen. ‘Varieties of Labor Politics in the Developed Democracies’. In Varieties of

Capitalism. The Institutional Foundations of Comparative Advantage, eds. P. Hall andD. Soskice, 71–103. Oxford: Oxford University Press, 2001.

———. How Institutions Evolve. The Political Economy of Skills in Germany, Britain, the United

States, and Japan. New York: Cambridge University Press, 2004.Therborn, Goran. Why Some People Are More Unemployed Than Others [The Strange Paradox of

Growth and Unemployment]. London: Verso, 1986.Toirkens, J. Schijn en Werkelijkheid in het Bezuinigingsbeleid, 1975–1986. Een Onderzoek naar de

Besluitvorming over het Bezuinigen in de Ministerraad en het Gedrag van Individuele Ministers(Appearance and reality in austerity policy). Deventer: Kluwer, 1988.

Touwen, L.J. ‘Varieties of Capitalism en de Nederlandse Economie in de Periode 1950–2000’

(Varieties of Capitalism and the Dutch Economy in the Period 1950–2000)’. Tijdschrift voorSociale en Economische Geschiedenis 3, no. 1 (2006): 73–104.

Van Ark, Bart, Jakob de Haan, and Herman J. de Jong. ‘Characteristics of Economic Growth in the

Netherlands during the Postwar Period’. In Economic Growth in Europe since 1945,eds. Nicholas Crafts and Gianni Toniolo, 290–328. Cambridge: Cambridge UniversityPress, 1997.

Van Bottenburg, M. ‘Aan den Arbeid! In de Wandelgangen van de Stichting van de Arbeid, 1945–1995

(Get to Work. Inside the Foundation of Labor, 1945–1995). Amsterdam: Bakker, 1995.Van Dellen, H. ed. Een Nieuw Elan. De Marktsector in de Jaren Tachtig (A New Zest. The Market

Sector in the 1980s). Deventer: Kluwer, 1984.

Van den Brink, J.R.M. Zoeken naar een ‘Heilstaat’. Opbouw, Neergang en Perspectief van deNederlandse Welvaartsstaat (In search of ‘Utopia’. Construction, Decline and Perspective ofthe Dutch Welfare State). Amsterdam: Elsevier, 1984.

Van den Toren, J.P. ‘Meerstromenland in de Polder. De Invloed van het CNV op de Vormgevingvan de Naoorlogse Verzorgingsstaat’. Cahier over de Geschiedenis van de Christelijk-SocialeBeweging 3 (2001): 117–52.

Van Kersbergen, Kees. Social Capitalism. A Study of Christian Democracy and the Welfare State.London: Routledge, 1995.

Van Zanden, J.L. The Economic History of the Netherlands, 1914–1995: A Small Open Economy inthe ‘Long’ Twentieth Century. London: Routledge, 1998.

Visser, Jelle and Anton Hemerijck. ‘Een Nederlands Mirakel’. Beleidsleren in de Verzorgingsstaat.Amsterdam: Amsterdam University Press, 1997 (published in English as ‘A Dutch miracle’.Job Growth, Welfare Reform and Corporatism in the Netherlands. Amsterdam: Amsterdam

University Press, 1998).Windmuller, J.P. and C. de Galan. Arbeidsverhoudingen in Nederland (Labor Relations in the

Netherlands). Utrecht: Spectrum/Aula, 1979.

WRR. Maken wij er Werk van? Verkenningen omtrent de Verhouding tussen Actieve en Niet-actieven.Rapporten aan de Regering Nr. 13. (Shall we work on it? Explorations on the Relationshipbetween Active and Non-Active Persons). The Hague: Staatsuitgeverij, 1977.

Zijlstra, J. Gematigd Monetarisme: 14 Jaarverslagen van De Nederlandsche Bank N.V. 1967–1980

(Modified Monetarism). Leiden: Stenfert Kroese, 1985.

464 L.J. Touwen


How does a coordinated market economy evolve? Effects of policy learning in the Netherlands in the 1980s

Documents