How do ideas, through discourse, influence the adoption of workfare activation policies in different welfare states?

(1)

Master’s Thesis Public

Administration

Research question: How do ideas, through discourse, influence the

adoption of workfare activation policies in different welfare

states?

Name: Maria K. Charles Student number: s0101745

Stream: Economics and Governance Thesis supervisor: Dr. Natascha van der Zwan Study year: 2017/2018

(2)

List of Abbreviations

CDA Christen-Democratisch Appèl (Christian Democratic Appeal) D66 Democraten 66 (Democrats66)

NYPD New Deal for Young People PvdA Partij van de Arbeid (Labor Party)

VVD Volkspartij voor Vrijheid en Democratie (People's Party for Freedom and Democracy)

(5)

Introduction

Welfare states remains a popular subject for study among scholars, with much emphasis on its evolution over the years. This master’s thesis deals with the role of ideas, measured through discourse, and the role it plays in one aspect of welfare state development.

Macroeconomic forces pressure politicians to implement social policies (expanding or retrenching the welfare state) which may contrast their institutional past. In the 1990s, governments struggled to cope with the consequences of the aging population and early retirements (Walters, 1997), the increased demand from welfare recipients for social protection (Allen, Cook, Mitchell, & Watts, 2007), changes in the labor market with the decrease in demand for production based workers and increase in demand of knowledge/service-based workers (Iversen & Wren, 1998), and high unemployment as a result of the economic crisis of the 1980s (Cook, 2008). The former protective welfare state where social benefits were provided as a supplement to individual’s income was no longer sustainable. So a new way of thinking about the welfare state evolved. A new strategy called the social investment strategy, under the Third Way thinking, was introduced in many European countries (Van Kersbergen & Hemerijck, 2012). Social benefits were used as an incentive to get individual to work (Cox, 1998). This was the dawn of the activation era. Under this new way of thinking, benefit recipients were encouraged and empowered to find jobs, on the one hand, and obliged to keep these jobs, on the other (Barbier & Ludwig-Mayerhofer, 2004). Over the years, there has been a shift to adopt stricter approaches as it regards active labor market policies (Vis, 2007). This stricter approach is referred to as workfare. Simply put, this is social benefits in exchange for (testable) job search efforts (Dingeldey, 2007).

Welfare regime types came to the forefront in academic literature particularly because of the work of Gøsta Esping-Andersen, who typified three forms of welfare states: the Liberal, Social-Democratic and Corporatist (Esping-Andersen, 1990). In keeping with this work, Barbier & Ludwig-Mayerhofer (2004) also identified welfare states according to their level of activation: the liberal and universal activation regimes. These various regime types are characterized by distinct ideologies on welfare which shape social policies (Arts & Gelissen, 2002; Esping-Andersen, 1990). Ideally, one would expect different responses from countries of different welfare regime types, as it is believed that different sets of ideas cause decision makers to act differently in the same circumstance (Hay, 2002). However, we see governments across Europe implementing workfare activation policies (Vis, 2007). This is puzzling because I

(6)

reason that the workfare approach is a better fit within the liberal activation type because of their focus on re-commodification (Mascini, Soentken, & Van der Veen, 2012). What’s more, a country characterized as having a universalistic activation welfare regime, should theoretically embrace decommodification and therefore not favor the workfare approach.

If the ideology or way of thinking about the welfare state shapes social reform, then before a welfare policy reform happens there should also be a change in the thinking (Blyth, 1997; Hay, 2002). According to Vivien Schmidt (2002), a leading proponent of discursive institutionalism, ideas (which according to her are measured through discourse) has a superior propensity to provide a richer account of changes in the welfare state when put together with other forms of institutionalism. If this is to be believed, then we could look at the discourse surrounding a social reform to describe how ideas can influence a policy. According to Schmidt (2002), ideas, through discourse, can be measured by looking at the coordinative, i.e.: discourse among political actors, and communicative discourse, i.e.: discourse to the public, which is essentially whatever politicians communicate to fellow political actors and to the general public in order to create and justify their ideas. This is essentially framing, or in other words choosing to portray something in a particular so as to create classifications in the minds of people (Stone, 2012). Politicians frame their ideas in such a way so that it is normatively and cognitively acceptable to their peer and the public (Schmidt, 2002). So in essence, I assumed that by looking at the framing of a social policy, I would be able to show the strategies used to gain acceptance for a policy.

Other studies have focused on the demonstrating the causal link between ideas and politics in general (for an account of different views see Béland & Cox 2011; Schmidt & Radaelli, 2004, welfare state adjustment from an economic perspective (Schmidt, 2002), and the development of activation policies (Weishaupt, 2011). However, I want to look specifically at the framing of ideas, and how this influenced the adoption of workfare policies in different welfare states (Barbier & Ludwig-Mayerhofer, 2004; Esping-Andersen, 1990) to see if and why there is a similarity in policy outcome despite different ideologies. Discursive intuitionalism is fairly new in comparison to other three forms of institutionalism and workfare activation have only started gaining popularity over the past two decades (Mascini, Soentken, & Van der Veen, 2012). With this thesis, I want to add to the growing contribution of work on discursive approaches by identifying and describing how it works in one of the newest aspects of welfare development, i.e.: workfare activation.

The type of research I carried out was descriptive research. I address the puzzle concerning the convergence among European welfare state to embrace more workfare activation policies

(7)

despite these countries having differing welfare ideologies. The goal of the research is not to show causal links between ideas and social policy but to describe how ideas, through discourse, influenced the adoption of workfare activation policies in two different welfare states. I aim to show how ideas were framed to gain acceptance. The variables of interest for this study are ideas, welfare states (dependent variables), and workfare activation policies (dependent variable). The independent variable in this study is ideas, or rather, discourse.

I ask the following research question and sub-questions: How do ideas, through

discourse, influence the adoption of workfare activation policies in different welfare regimes?

 What is the prevailing idea(s)? When can we first observe ideational influence,

and how was this communicated?

 How have these ideas been translated into plans for national policy?  How was the discourse around the time of adoption?

Considering the nature of my study, I elected to carry out a comparative case study design because this is the best design to study few or single cases (Gerring, 2007, Yin, 2009). It afforded me the ability to carry out an in-depth investigation of the case as opposed to a study with larger samples. I also used a crucial case selection approach, which a theory-driven case selection method founded on three pillars, namely the theory used has to have reasonably clear-cut expectations, the cases selected to test the theory should be led by said theory and make the best use of it, and lastly, the researcher has to be clear beforehand about the types of evidence he or she will accept as evidence (Gerring, 2007). The cases I selected are the countries the United Kingdom and the Netherlands. These countries have two different welfare state ideologies, the United Kingdom is considered liberal and the Netherlands is social democratic1 (Arts & Gelissen, 2002), yet both have been frontrunners with regards to activation policies (Verhoeven & Tonkens, 2013). Since the 1990s both countries have also embraced a workfare approach. The units of analysis are reforms implemented during this period. In the United Kingdom it is the New Deal for Young People and in the Netherlands, it is the Wet Inschakeling

Werkzoekenden. I employ a most-likely and a least-likely case selection strategy. The two cases

offered the opportunity to compare two welfare states of which one is, or rather should be, (theoretically) receptive, i.e.: the United Kingdom, and the other not receptive, i.e.: the Netherlands, to workfare activation approaches. The aforementioned test assessment is achieved through a research technique called process tracing. It is a method that uses evidence

1_{According to Barbier and Mayerhofer’s (2004) classification, the United Kingdom is considered a}

liberal activation regime idea type, and the Netherlands is considered a universalistic activation regime ideal type.

(8)

identified from studying the case to make inferences about causal explanations about that case (Bennett & Checkel, 2014, Toshkov, 2016).

Moreover, my data collection strategy was to look at spoken and written texts. To gauge the ideas of political actors found in the coordinative and communicative discourse, I looked at primary data sources such as speeches, political manifestos, legislation, newspaper and television interviews, and debates. For background information about the cases, I looked at secondary data sources such as academic journals, books, and newspaper articles.

This thesis is divided into four parts. Chapter one is the theoretical framework. This section is intended to provide the reader with an indication of the lens through which I see and, ultimately, address the puzzle expressed in this paper. I begin by sketching a picture of the development of the welfare state and highlighting the change in thinking that came about in as a result of different socio-economic factors. Subsequently, I address how activation policies have become a normal response among welfare states, and conclude by touching on how ideas, through discourse can lend a hand in understanding how ideational change happens, resulting in the implementation of certain social policies. I include the expectations I have for this study. Chapter two contains information about the methodology used in designing and selecting cases for this research. In addition, I explain how I collected my data, and how I analyzed this according to the theoretical framework provided in the Literature Review section of this dissertation. Principally, I want to show how this strategy helped me to obtain the answers to my research question.

In Chapter three I handle the findings from the research carried out and an analysis of the evidence I found. I use a discourse analysis strategy to analyze spoken and written texts. The goal is not only to show similarities and/or difference but also to describe the role of ideas in the adoption of workfare activation policies.

(9)

Chapter 1: Theoretical framework

1.1 The development of the welfare state

Though the concept of welfare is age-old, the term welfare state only started gaining prominence in the 1930s when governments of advanced industrialized societies struggled to stabilize crippled economies as a result of the Great Depression (Greve, 2008). The notion of welfare state centers on the role of the government or state to provide for the well-being of its citizens. According to Greve, in order to adequately measure the welfare state in terms of outcome and effectiveness, governments should consider more factors besides economic aspects, i.e., social expenditures. In providing for the welfare of its inhabitants, states must also consider societal developmental aspects such as health, education, aging, and pensions.

To an extent, this was the view held by Gøsta Esping-Andersen in his seminal work The

Three Worlds of Welfare Capitalism. Esping-Andersen (1990, pp. 26-29) outlines three regime

types: the Liberal welfare state, the Social Democratic welfare state, and the Conservative/ Corporatist welfare state. These three classifications are ideal types in which, in reality, no state fits specifically. Nevertheless, Esping-Andersen contribution with these typologies continues to be helpful in aiding further research, particularly when used to explain the development within and across welfare regimes. Esping-Andersen (1990, pp. 21-26) characterizes the three welfare state regimes according to two essential dimensions: (1) the degree of decommodification and (2) the level of social stratification. The former dimension is the degree to which individuals are able to maintain a living without having to participate in the labor market. In this case, the government provides welfare benefits (e.g.: unemployment benefits and tax credits) that would enable a person to still function within society, i.e., meet their basic needs, without work. The second dimension denotes how the policies implemented stimulates the system of stratification and how this influences social solidarity and the formation of stigmas (e.g.: social policies for poor people automatically categorizes the individuals that make use of this benefit) (Arts & Gelissen, 2002).

The three welfare regimes each have distinct policy approaches, ideologies and structures of welfare and how it is implemented. The Liberal welfare state places much emphasis on the labor market and is characterized by a low level of decommodification (Esping-Andersen, 1990, pp. 26-27). Social benefits are modest and only target the very needy. The Social Democratic welfare state comprises a generous, universalistic and highly distributive

(10)

welfare system (Esping-Andersen, 1990, p. 27). It is considered to have a high level of decommodification and there is a high degree of benefit equality. The Conservative/Corporatist welfare state comprises a moderate level of decommodification but this is dependent on status (Esping-Andersen, 1990, p. 27). Moreover, it emphasizes the role of the family and traditional welfare providers, such as the Catholic Church, to care for the needs of its members. In this case, the government only steps in if the aforementioned welfare providers are unable to fulfill their roles. Social benefits are then based on past contributions.

Over the years, European welfare states have gone through structural changes with periods of expansion and retrenchment. According to Allan and Scruggs (2004), the following distinction can be made: a period of expansion before the 1980s and a period of retrenchment after the 1980s. After the 1980s, the social and political climate had changed because of globalization, deindustrialization, the changes in the traditional family and the role of women in the labor market, combined with modest economic growth over a period of years (Myles & Quadagno, 2002, p. 35). As a result, social policies were shaped by demands for more austere measures, i.e.: retrenchment. Governments worldwide struggled to provide answers to social problems such as rising unemployment, the aging population and the respective problems regarding pensions (Castles et al, 2010).

There are theories that claim that globalization exerts pressure on governments to conform fiscally, causing instances of expansion or retrenchment. One example is the so-called efficiency hypothesis, which claims that retrenchment is brought about when a government lowers taxes in an effort to attract large companies desiring to be established in countries with low production costs (Garrett & Mitchell, 2001). However, this efficiency hypothesis has been rejected by many studies over the years (Glatzer & Rueschemeyer, 2005; Starke, Obinger, & Castles, 2008). Another is the compensation hypothesis (Walter, 2010), which asserts that governments try to compensate the losers of globalization, i.e.: workers who lose their jobs, for risks associated with international competition through redistribution (Walter, 2010). Opponents of this theory credit deindustrialization as the major reason for the expansion of the welfare state (Iversen & Cusack, 2000).

Besides the pressure of a shifting global economy, partisanship also plays a role in the discussion about welfare state development. Esping-Andersen credits the working classes ability to mobilize, and push support for leftist parties who, in turn, implement more generous social policies, as the reason for welfare expansion (Esping-Andersen, 1990, p. 29). This is the Power Resource Theory and contrasts with an earlier view which reasoned that the welfare state

(11)

expansion came as the result of economic growth and demographic change (Wilensky, 1975 as cited in Myles & Quadagno, 2002).

According to the New Politics of the Welfare State (1996), Paul Pierson claims that welfare expansion theories, such as the Power Resource Theory, do not play a decisive role in the evolution of the welfare state. He argues that welfare state expansion and retrenchment are distinctively different because the latter has a different goal (the retrenchment reforms) and context (the possible political backlash for these reforms, which cause politicians to engage in blame avoidance tactics) which creates a new political dynamic for the welfare state. While it is true that welfare expansion and retrenchment have different political contexts, Pierson does not deny the part that partisanship plays in that right-wing governments are more likely to retrench (Pierson, 1996, pp. 176-177). Examples are the reforms of Margaret Thatcher in England and Ronald Reagan’s in the United States of America.

Moreover, the type of political decision-making, i.e.: social policies and program, can be attributed to the type of welfare state regime it is (i.e.: Liberal, Conservative, or Social Democratic), as this determines the level of generosity and focus on commodification (Vis, 2007). Left wing governments are credited with more generous social spending and, thus, larger welfare states, while right-wing governments are said to be more likely to retrench (Allan & Scruggs, 2004; Klitgaard & Elmelund-Præstekær, 2013). There are a number of scholars that make an argument for the role of partisanship in welfare state development. For example, Hicks and Swank showed that welfare expansion is dampened with rightist party participation in government as opposed to leftist parties (Hicks & Swank, 1984). Allan and Scruggs (2004) analyzed partisanship by looking at political manifestos and programs and found that differences between left and right wing politics stand. Kwon & Pontusson (2010) also provide some evidence for the role of partisanship in the expansion of the welfare state by looking at the interaction between globalization, the power of unions and partisan politics. They argue that partisanship matters during globalization if unions are strong. In keeping with the Power Resource Theory, strong unions put pressure on leftist parties for favorable social security arrangements and/or protection.

However, sometimes left and right wing parties behave differently in times of crisis, engaging in behavior that would not normally fit within their political ideology. Starke, Kaasch, & Van Hooren (2014) maintain that countries with low welfare state expenditure and few economic stabilizing programs such unemployment benefits to balance out fluctuations in the economy tend to be more influenced by partisan politics. Whereas, in countries with larger welfare state expenditure, the direction of change is determined by other things such as

(12)

supranational organizations and public debt. Conversely, Raess & Pontusson (2014) reject this view; for them, it is the other way around. They say that if a welfare state is larger, and thus has more economic stabilizers, then partisanship matters more. They argue that smaller welfare states, in the absence of safety nets in a crisis, may expose ruling parties to electoral backlash. As a result, parties may act differently to gain or save votes. According to Rueda's (2005) governments do not always support welfare expansion policies when it is politically unfavorable (Rueda, 2005, p. 71). Here, Pierson’s central argument about the new politics of the welfare state becomes evident, namely that ‘contemporary politics of the welfare state is

the politics of blame avoidance’ (Pierson, 1996, p. 179). In any case, as Pierson (1996) points

out, a purely historical analysis may be insufficient to explain why political parties act in an adverse way, given the new goal and context under which they operate. A social policy’s ‘unfolding over time’ (Pierson, 1996, p. 178) requires an analytical framework that can explain change. However, I will return to this point below.

In sum, we see that over the years the welfare state has evolved. Whether a government chooses to expand or decrease the size or scope of the welfare state as a result of macroeconomic forces normally depends on the political ideology it holds. Although given much attention in scholarly articles, the abovementioned responses to social problems are not the only answers to the socio-economic pressures described above. Van Kersbergen, Vis, & Hemerijck, 2014 believe that besides expansion and retrenchment, governments can respond to social and economic crises by embracing a social investment strategy. This social investment strategy presented another way of thinking about the welfare state.

1.2 A shift in thinking about the welfare state

After the Second World War, welfare states2 were fairly protective, in the sense that it was generally accepted that the government’s role was to provide for the well-being of its citizens (Castles et al, 2010). By the 1980s the idealistic hopes and dreams of the post-war era – a time when a number of governments had expanded their welfare programs – were over. Burdened with inflation, foreign policy pressures, and high unemployment rates as a result of the 1970s economic crisis, governments responded by establishing public employment services and providing financial assistance for the vulnerable in a type of Keynesian intervention3_(Cook,

2_{Refers to advanced industrial countries within the OECD. Southern European countries have developed relatively}

little in terms of activation as compared to its western counterparts (Bonoli, 2013).

(13)

2008). However, they soon needed to develop new ideas and strategies to address their social (e.g.: inequality) and economic (e.g.: unemployment) problems. Firstly, there was concern about the aging population and early retirements, which put stress on government resources, as pension and healthcare benefits form a large part of social expenditure (Walters, 1997). Secondly, governments could not maintain the expansive and protective welfare state because of the increased demand from welfare recipients for social protection (Allen, Cook, Mitchell, & Watts, 2007). Thirdly, deindustrialization, globalization and rise in technology changed the dynamics of the work environment, where the demand for production based workers, common to the industrial age, decreased and the call for knowledge/service-based employment increased (Iversen & Wren, 1998). Fourthly, as a result of the former point, new strategies were necessary to address the mismatches between skills

Consequently, European governments changed the emphasis from providing full employment4 to embracing policies that were created to improve the employability of vulnerable individuals without work. This change in thinking from a passive welfare state to an active welfare state can be termed a ‘paradigm shift’ (Eichhorst & Konle-Seidl, 2008; Hall, 1993). Traditionally, social security benefits were seen as a replacement or supplement for the missed income (a sort of safety net) for when a worker lost his/her livelihood because of unemployment, injury or sickness (Cox, 1998; Walters, 1997). According to Van Kersbergen & Hemerijck (2012), this use of social expenditures to help develop the capacities of the unemployed is referred to as the social investment strategy. Under this social investment strategy, the social benefits system was no longer seen as a financial burden for governments, but as an incentive to improve production and economic growth (Cox, 1998). Social policies are then considered as an investment in human capital (for example: education and vocational training, active labor market policies, childcare and benefits for working mothers). By investing in human capital on the short term, governments hope to increase labor participation in the long run (Van Kersbergen & Hemerijck, 2012). In other words, it can be said that the goal of the social investment strategy is to balance social and economic goals (Morel, Palier, & Palme, 2012, p. 354) by making individuals dependent on the labor market again because the investment helps them back to work and lowers the dependence on the state.

The idea of the social investment strategy contrasts Esping-Andersen’s view of decommodification. Esping-Andersen’s decommodification falls into line with the traditional view of the welfare state described at the beginning of this section, which supports the welfare

4_{Full employment strategies included all actions to provide direct employment to the unemployed, particularly}

(14)

state as safety net perspective. The idea of decommodification allows an individual to maintain a living without having to participate in the labor market because, essentially, the benefits received “replaces” the income from work (Esping-Andersen, 1990, p. 37). Under the social investment strategy, benefits are seen as an incentive for the unemployed to enter or return to the workforce (Cox, 1998). This change in thinking describes the shift from passive to active. Nevertheless, the focus in both the former and the latter welfare society is worker-related. However, where social rights are afforded to non-workers (i.e.: most vulnerable in society) as an integrative mechanism in the welfare society allowing them to still function without a livelihood, the active society integrates through the labor market (Walters, 1997) and promotes the labor participation for all (Bonoli, 2013).

The shift from a passive welfare state to an active welfare state resolves the tradeoff between equality and efficiency (Morel et al, (2009, p.10) as cited in Van Kersbergen & Hemerijck, 2012, p. 475). Governments balance providing social protection, on the one hand, and managing welfare state costs on the other. An example of this is the implementation of ‘make work pay’ or workfare measures in the United Kingdom. While under this scheme benefit recipients receive tax credits, at the same time, the amount and duration of unemployment benefits are reduced (Daguerre & Etherington, 2014). Understandably, embracing active social policy strategies remains attractive for governments for three reasons. Firstly, in the short run, activation policies discourage dependence on social insurance while increasing the workforce participation (Cox, 1998). Secondly, increasing the possibilities for employment could mean more income tax revenues in the long run, which broadens the government’s resource capabilities for social spending (Eichhorst, Kaufmann & Konle-Seidl, 2008). Thirdly, because activation policies are less expensive than traditional forms of welfare expenditure, and appear more sympathetic than the retrenchment policies imposed by right-wing politics, it offers left-right-wing governments the opportunity for ‘affordable credit claiming’ (Bonoli, 2013, p. 8).

In short, after the Second World War, European welfare states underwent a paradigm shift. This change in thinking ushered in an era of active social policies that focused on the development of individuals with the view to improving and encouraging workforce participation and discourage dependence on the welfare state.

(15)

1.3 The activation of the welfare state: from welfare to workfare The OECD’s definition of activation as cited in Martin (2014) is as follows:

To bring more people into the effective labour force, to counteract the potentially negative effects of unemployment and related benefits on work incentives by enforcing their conditionality on active job search and participation in measures to improve employability, and to manage employment services and other labour market measures so that they effectively promote and assist the return to work. (p. 4).

There are two categories of social policies that European governments use to address unemployment: (1) passive labor market policies (PLMPs) and (2) active labor market policies (ALMPs) (Barr, 2012; Hemerijck & Vandenbroucke, 2012). The former encompasses the payment of unemployment benefits such as social assistance and unemployment insurance, and the latter is meant to stimulate people without work in order to increase their employability. For example, training programs for young people or special work placement programs for disabled people. The concept activation has evolved from merely referring to cutbacks in passive labor market policies, to now refer to ‘a dimension of the reform of social protection’ (Barbier & Ludwig-Mayerhofer, 2004, p. 423) It includes entire systems consisting of programs and services that are aimed at increasing human capacity (Van Berkel & Borghi, 2008). Granting that activation policies are not an entirely new concept5, the real distinction between the old and the new approach, i.e.: welfare and workfare, is its punitive and compulsory quality (Barbier & Ludwig-Mayerhofer, 2004). This is where welfare recipients are subject to stricter job search conditions in that they are required to make every effort to (re)enter the workforce, and failure to maximize job opportunities can result in the reduction of or cancellation benefits (Van Berkel & Borghi, 2008, p. 332).

Activation policies can be categorized according to program types and features. Kluve (2010) classified active labor market policies into four main program types: (1) labor market

training, i.e.: in-the-classroom, on-the-job training and work experience; (2) private sector incentive programs (for instance: wage subsidy and self-employment grants); (3) direct employment programs, i.e.: direct employment programs in the public sector); and, (4) services and sanctions, i.e.: all measures to encourage and enhance job search efficiency. The goal of

the first type is to maximizing human capacity by increasing skills. The second ALMP type is

5_{Scandinavian countries such as Denmark and Norway have focused on ‘active’ social policy such as educational}

training for many years (Barbier & Ludwig-Mayerhofer, 2004). However, these differ from the ‘active’ social policy referred to as workfare because their purpose was not to solve social issues or empowering disadvantaged people.

(16)

meant to encourage employers to hire new workers (usually individuals that have been unemployed for a longer period of time), to maintain jobs that would normally be outsourced, or to encourage individuals to become self-employed. Thirdly, direct employment programs are directed at the very underprivileged. The objective is to keep these individuals in contact with the job market and to avoid them falling out completely. The last ALMP type is focused on improving the efficiency of matching the job and the individual. Pascual & Magnusson (2007) distinguish three features of activation policies: (1) an individualized approach (the activation policy targets the individual’s behaviour and motivation); (2) emphasis on employment (individuals are encouraged to be self-sufficient); and (3) contractualization (the joint effort of the state and the individual with regards to recommodification, i.e., independence on social assistance by rejoining the workforce).

The way in which activation policies are implemented is also dependent on the institutional dynamics and socio-economic environment of a country (Clasen and Clegg, 2006). Here, again partisanship comes into play. In the 1990s, Western Europe was characterized by predominantly left-wing government rule6, which could present a reason for the relatively rapid spread of the social investment approach among the European Union community. The EU encouraged member states to embrace policies in order to combat high unemployment rates of the 1990s that would foster social and economic coherence (Martin, 2014). The European Commission introduced the European Employment Strategy (EES) under the Open Method of Coordination (OMC) in 1997 (Eichhorst, Kaufmann, & Konle-Seidl, 2008). The EU established the OMC to help coordinate the different national policies held by member states, and the EES was developed to help European member states reform national employment and social welfare systems (Heidenreich, 2009). Besides the EES, the OECD was also influential in (re)shaping the political climate with regards to employment in the EU when they commissioned the OECD Jobs Study (and later the OECD Jobs Strategy). The aim of the aforementioned study was to help OECD member countries to combat unemployment. The focus of the study emphasized work-related policies and activation agendas typical of the social investment agenda (Van Gerven & Beckers, 2009). It was this type of Third Way7 thinking that ushered in the EES and

6_{Poul Nyrup Rasmussen in Denmark; Wim Kok in the Netherlands; Tony Blair in the United Kingdom; and}

Gerhard Schröder in Germany (Bonoli, 2013).

7_{According to Van Kersbergen, and Hemerijck (2011) the Third Way refers to social policies that are}

characteristically different from conventional social democratic or neo-liberal policies. “The ‘third way'that is emerging is more than anything about the creation of employment and high rates of labor market participation. It is about turning welfare into work. European social democracy is quite flexible about the means to achieve this goal. Thus, ‘third-way ' policy pragmatism involves both a positive view of the ability of the market to provide certain outcomes and a strong emphasis on the active ‘social investment state'. At the level of practical policies, the ‘third way' involves such elements as a robust macroeconomic policy, responsive wage bargaining, combining

(17)

the OECD Jobs Study & Strategy, which encouraged European countries to embrace activation (Clasen & Clegg, 2006; Green-Pedersen, Van Kersbergen, & Hemerijck, 2011; Van Vliet & Koster, 2011).

Like Esping-Andersen, welfare states are also characterized according to activation regime ideology. Barbier and Ludwig-Maerhoffer (2004) distinguish between the liberal and universalistic activation ideal-types regimes8. Firstly, the liberal activation type (for example the United Kingdom) is characterized by modest income protection (for example individuals are encouraged to accept any job on the market as is) and modest social benefits; it can be considered to have a punitive individualized approach. Secondly, the universalistic activation type (for example Denmark) is considered to have generous universalistic income protection and social benefits, with the emphasis placed on contractualization, i.e: the state and the individual work together to find suitable work. In contrast to the liberal activation welfare type, in the universalistic welfare type activation applies to all citizens and not select participants. It should be mentioned that no country fit either ideal type perfectly but have a mixture of elements of either type depending on the policy (Barbier & Ludwig-Mayerhofer, 2004, p. 429). However, Barbier and Ludwig-Mayerhofer’s typification meet the criteria of practicality specified by Arts and Gelissen (2002), in that they provide a topographical overview of the broad characteristics of activation policies. Moreover, this categorization of activation regime types corresponds with the Esping-Andersen’ work (and, largely, other scholars that have developed typologies over the years). Arts and Gelissen (2002) provide an extensive list.

In sum, activation policies oblige individuals dependent on unemployment benefits or social assistance to participate in work. Essentially, activation strategies are meant to ensure that the unemployed have better chances to find work. Under these schemes, governments place the emphasis on the individual’s responsibility to take active steps to secure his/her employability in exchange for receiving benefits. In a case of noncompliance, benefit sanctions can be implemented. European governments, under the influence of the European Union, embraced activation policies in an effort to increase labor market efficiency, productivity and provide means to an income for the unemployed (Eichhorst, Kaufmann, & Konle-Seidl, 2008). So, despite differences in welfare state ideology, European welfare states seemed to converge with regards to the implementation of activation policies to solve unemployment issues.

wage moderation and flexibility, active labor market policies, and employment-friendly and efficient tax and social policy” (pg. 309).

8_{Barbier and Ludwig-Maerhoffer (2004, pg. 428): ‘it is not possible to contend that a third, coherent,}

(18)

1.4 Convergence on activation policies

Holzinger & Knill (2005) offers a clear definition of the concept of convergence, namely: ‘any increase in the similarity between one or more characteristics of a certain policy…across

a given set of political jurisdictions…over a given period of time’ (pp. 768). The term

convergence can be observed in terms of policy goals, content, instruments, outcomes, and style (Bennett C. J., 1991). Moreover, there are generally four processes that cause convergence: (1) the emulation of successful neighboring countries, (2) elite networking (for example: the EES and OECD Jobs Study), (3) harmonization (for example: countries complying with ‘soft laws' because they want to be integrated into the European Union), and (4) penetration (for example: where countries are forced to go along with decision taken by external actors) (Bennett C. J., 1991, pp. 223-227). Some scholars point to elite networking motivations such as the EES and OECD Jobs Study as a reason for the relatively rapid spread of activation policies within the European Union (Van Vliet & Koster, 2011). Others have argued that it came as a result of globalization, deindustrialization and socio-economic changes. For an overview of other welfare state convergence scenarios by different authors see Bonoli, 2013; Starke, Obinger, & Castles, 2008; and Weishaupt, 2011.

Nevertheless, it is important to point out that although European welfare states converged on the same goal, there are differences in the degree of intensity and methods of implementation of activation programs/policies on a national level (Bonoli, 2013; Dingeldey, 2007; Gough, 2008; Lindsay & Mailand, 2003). In other words, there was activation but it happened differently for the different countries. For instance, the Nordic countries were the first to implement activation strategies such as educational and on-the-job training aimed at empowering the unemployed (Bonoli, 2013). These countries maintain a weaker workfare approach but have strong employment enabling elements such as the examples mentioned in the previous sentence (Dingeldey, 2007). Liberal welfare states focused on deregulation and liberalization of their economies, while conservative welfare states concentrated on early reducing the labor supply (Bonoli, 2013). Additionally, some countries relied more heavily on a marketization strategy9 (Van Berkel, 2009), and others spent more on activation than others. For example, spending on active labor market policies in the Netherlands in 1990 was 1.05 percent of the GDP. In the United Kingdom, this was 0.61 percent (Cook, 2008). Despite the differences in approach, what has remained central in the different welfare states is the

9_{The Netherlands made use of private companies, whereas other countries employed a mixed service}

(19)

heightening and promotion of individuals finding ‘paid work’ and joining the labor force (Barbier & Ludwig-Mayerhofer, 2004).

Classically, convergence theorists believed that the socio-economic consequences of globalization would balance out any differences between industrialized nations (Esping-Andersen, 2010; Iversen & Cusack, 2000). Indeed, the trend within welfare states is that universal welfare systems are becoming more liberal, and liberal welfare states increasingly adopt more universal traits (Achterberg & Yerkes, 2009)10. More particularly, there has been progressive acceptance of workfare programs11 (Vis, 2007). What is not clear is how welfare states with differing ideologies arrived at ‘similar’ outcomes, namely workfare policies. To clarify, I do not refer to workfare methods but a way of thinking12. Theoretically, one would expect different responses from countries of different welfare regime types, as it is believed that different sets of ideas cause decision makers to act differently in the same circumstance (Hay, 2002). To be more specific, workfare, as opposed to welfare, resonates with the liberal activation type, and less so with a country that holds the universalistic activation type ideology. I reason this way because I believe the focus of the liberal activation regime is creating independence for the neediest, i.e. reliance on the market and creating non-reliance on the welfare state (re-commodificaton). For the universalistic activation regime, the attention is on securing a strong social security system and flexible labor market for all (Barbier & Ludwig-Mayerhofer, 2004). I am interested to see how actors of differing welfare state types come to implement strategies along the same workfare lines.

In essence, convergence is about moving from differences to arrive at one common position and implies development over time. However, the point of reference, within this context, is not differences within European welfare states but the variance from some former stage to another (Bennett, 1991), i.e.: the shift from welfare to workfare. The ideas presented in the EES were translated into different national strategies to achieve the same goal. Still, there has been an increasing tendency to utilize workfare approaches. In order to identify how differing European welfare states with diverse ideologies arrived at a similar approach, I want

10_{Weishaupt suggests that there is increasing convergence as it regards public expenditure on active labor}

market policies in European welfare states (Weishaupt J. T., 2011).

11_{The Netherlands had the ‘Melkertbanen’ in the 1990s, the United Kingdom had the New Deal in 1998,}

there was ‘the labour market support’ scheme in Finland in 1994, in Sweden the activation guarantee in 2000, Norway had the ‘arbeitslinjen’ in the 1980s and Germany had ‘Hilfe zur Arbeit’ in the late 1990s (Vis, 2007, p. 119).

12_{The workfare way of thinking focuses on getting people back into paid jobs by expanding services that}

help to increase employability, on the one hand, and obliging (through penalties) benefit recipients to make use employment opportunities, on the other (Dingeldey, 2007). So, in other words, participants are required to work in other to receive unemployment benefit. Anyone that is able to work has to do so before they can think of relying on the welfare system.

(20)

to consider the policy contents and styles (i.e.: how a social policy is established and how it is formulated) employed by the political actors in these countries, considering that all policymaking revolves around the exchange of ideas (Stone, 2012). Hay and Rosamund (2002, pp. 148) said ‘it is the ideas that actors hold about the context in which they find themselves

rather than the context itself which informs the way in which actors behave’ (as cited in Ellison,

2006).

1.5 The role of ideas and its influence on social policymaking

It is said that ‘institutions are carriers of ideas’ (Rothstein 2005, pg. 7, as cited in Schmidt, 2011). Welfare states are institutions that are made up of a number of different organizations that carry out the informal and the formal rules defined by that institution (North, 1990). These rules are made and changed by humans, and determine how we interact with one another in society. Therefore, a study of institutional change will inevitably revolve around human behavior. Considering the above, it is no surprise that scholars often turn to some form of institutionalism to explain political action (Schmidt, 2002). Historically, there are three forms: rational choice institutionalism, historical institutionalism, and sociological institutionalism. Decisions of political actors were often relegated to one of the historical institutions, arguing that behaviors of decision makers were the product of the structures, processes, and culture they come from. And, that by studying the context of these, we would be able to predict a political outcome (Hay, 2002).

Over the years there has been much discussion about the concept of ideas, how to define it and its role in policymaking. Schmidt (2008) sums up a list of previous academic works where ideas have been presented as switches for interests, road maps, or focal points (Goldstein and Keohane 1993); strategic constructions (Jabko 2006); strategic weapons in the battle for control (Blyth 2002); narratives which shape understandings of events (e.g., Roe l994); ‘frames of reference’ (Jobert 1989; Muller 1995); collective memories (Rothstein 2005); or national traditions (Katzentein 1996). Despite neglect from earlier generations of scholars, welfare state scholars now acknowledge the importance of ideas for welfare state development. Ideas help to clarify principles, organize behavior, and order the world and, therefore, can influence the decision-making of political actors (Goldstein & Keohane, 1993). Moreover, ideas are central to change and the forming of new institutions (Hall, 1993). As Hay said, before a change in policy can happen, a change in the ideas that formed the policy has to happen first (Blyth, 1997; Hay, 2002).

(21)

To address ideas in policymaking, the fourth form of institutionalism called discursive institutionalism was introduced. This form of institutionalism is concerned with the substantive content of ideas and on the interactive processes that serve to generate those ideas and communicate those to the public (see Schmidt 2002, 2008, 2011). The concept idea is difficult to define, however, Béland and Cox (2011) offer a simple and clear definition: ideas are causal beliefs which communicates what people want, what they deem appropriate, legitimate and proper (Béland & Cox, 2011, p. 3). Schmidt (2002, 2008) offers us three general levels to think about ideas within policymaking: policies (i.e.: policy ideas), programs (i.e.: beliefs that support the policy ideas) and philosophies (i.e.: deeper philosophical beliefs that underpin both the policies and programs). These ideas can be divided into two categories: cognitive ideas (cognitive ideas justify policies and programs according to logic and necessity) and normative ideas (normative ideas justify policies and programs in terms of how appropriate they are in light of public values and principles) Schmidt (2002).

The strength of this form of institutionalism lies in its ability to explain change and support other forms of institutionalism. While all institutionalism forms are able to link structure and agency, according to Vivien Schmidt (2002), the three classical forms of institutionalism are unable to adequately explain change. Discursive institutionalism goes beyond looking at the structures and processes that shape institutions and takes into account the individuals whose actions and interests brought about certain policy decisions through their ideas (Schmidt, 2002). This suggests that institutional change can occur when the ideas, expressed through discourse, also change (Hay, 2002). Such a change can come about as a result of socio-economic pressures (for example economic crisis), which encourages a rapid acceptance of ideas that would previously be unacceptable (Vis, Van Kersbergen, & Hylands, 2011). Another strength lies in discursive institutionalism’s ability to build on the other forms of institutionalism. Schmidt argues that the other three form of institutionalism can be treated as taken-for-granted, common-sense ideas which serve as background assumptions to further build theory on or to investigate a problem (Schmidt, 2008). When discourse is added to any of the three, it creates a stronger framework to build from (idem).

Schmidt (2002) goes further than just talking about ideas within the wider context of the three forms of institutionalism, acting as fillers to support other forms of explanation (Blyth, 1997, p. 231). Instead, she focuses on discourse. Discourse ‘serves to explain political events,

to legitimate political actions, to develop political identities, to reshape and/or reinterpret political history and, all in all, to frame the national political discussion’ (March & Olsen 1995,

(22)

by content and by usage. Content refers to the policy ideas and values that are presented, and usage denotes the interactive process of how a social policy is formed and communicated to the outside world (Schmidt, 2002). Thus, discourse is defined as the interactive process of the exchange of ideas (Schmidt 2002, 2008, 2011). Schmidt further conceptualizes this by the following interpretation: “whatever policy actors say to one another and to the public more

generally in their efforts to construct and legitimate their policy programs” (Schmidt, 2002, p.

169). She provides two ways to operationalize this: analysis of coordinative discourse and communicative discourse. The former concerns itself with communication among policy actors and the latter is communication towards the general public.

Considering the above, I argue that ideas also played a role in the transition towards activation. Pressure to fund an increasing welfare state, because of social-economic factors such as high unemployment, the changing labor market and social changes (aging and changes in the make-up of the traditional family), encouraged politicians to seek out other alternatives to the costly protective welfare state they held in the past. Ideas of social investment ushered in an era of activation. The European Commission provided European nations with an activation strategy (the EES) to fight the high unemployment that was so prevalent in European welfare states in the 1990s. The notions presented in the EES had to, ultimately, be translated into appropriate national programs and social systems. Before this could happen, these nations had to embrace the idea of activation, i.e.: accept it as cognitively and normatively acceptable (Schmidt 2002, 2008). As Blyth said, a paradigm shift is an idea that gains salience and becomes institutionally embedded because it fits within the ideological structure of a country (Blyth, 1997, p. 232). Both the political actors within the coordinative discourse arena, and the members of the public within the communicative discourse arena, judge the cognitive necessity and normative appropriateness of the ideas presented, and are persuaded (or not) by these ideas. Whether political actors or the public are persuaded by the necessity or appropriateness of a policy often depends on how an idea is framed (Béland, 2005). Framing is, as the word denotes, choosing to see some things (within the frame) and excluding others (outside the frame). This creates criteria of categorization, i.e.: what is acceptable or not, which ultimately structures the ideals that influence people's behavior (Stone, 2012).

In sum, macroeconomic factors, such as globalization, aging, unemployment, etcetera, create situations where political players are forced to act. By studying the discourse of these actors, one can trace how social policy takes shape. In explaining the explanatory benefits of using ideas through discourse to understand political change, Vivien Schmidt discusses how fundamental it is to show that discourse matters by analyzing if and how these ideas are

(23)

presented in a way that re-conceptualizes the interests of political actors, reshapes institutional ideas, and/or reframes culture (Schmidt,

2008).

In this dissertation, I am describing the role ideas play in the adoption of workfare activation policies in differing welfare states. By looking at discourse surrounding policy content and styles, I not only identify if and where similarities lie, but also show the framing of ideas, which demonstrate how ideas influenced the adoption of workfare activation policies in different welfare states. I have the following expectations:

(1) The discourse surrounding activation in the countries under observation will be framed in such a way in order to be cogitatively and normatively acceptable in order to win approval;

(2) The liberal activation regime will frame activation as a way to gain independence of welfare markets, while the universal activation regime will focus on creating or maintaining equality, and encouraging personal contribution to society.

(24)

Chapter 2: Research design and methods

2.1 Problem definition and research design approach

In this thesis, I aim to answer the question: How do ideas, through discourse, influence

the adoption of workfare activation policies in different welfare regimes? I address the puzzle

of the convergence among welfare states to embrace workfare activation policies despite them having differing welfare state ideologies. Dissimilar welfare regime types are expected to handle differently in similar circumstances (Hay, 2002). However, over the years there has been a tendency to implement similar workfare activation policies among European welfare states. Using the activation regime typologies specified by Barbier & Ludwig-Mayerhofer (2014), I look at two countries in this study that represent the liberal and universal activation regimes, namely the United Kingdom and the Netherlands respectively. To answer my research question, I employ a within case research method called the crucial case research design by looking at a most-likely case and least-likely case. Ultimately, I use process tracing, which is a case study research technique that uses evidence identified from studying the case to make inferences about causal explanations about that case (Bennett & Checkel, 2014, Toshkov, 2016). As it regards the testing of ideational theories, it is important to look for evidence that demonstrates that political actors were influenced by their ideas and that those ideas were not solely the result of exogenous factors (Jacobs, 2014). I do not make a causal inference about the relationship between ideas and workfare activation policies. I am describing how ideas play a role in gaining acceptance. I, therefore, looked for sufficient and necessary conditions contained in the coordinative and communicative discourse of political actors. Firstly, it is necessary that the coordinative and communicative discourse that I analyze show evidence of ideas playing a role in the adoption of the policy. Secondary, to be sufficient, this evidence should show the liberal activation regime communicating ideas such as "work first", "make work pay", and "gain independence from welfare". Likewise, the universal activation regime should communicate along the lines of: "providing opportunities for the underprivileged", "solidarity", an "equality".

2.2 Conceptualization and operationalization of variables

(25)

In this section, I break down the concepts that I use in this study so that there is clarity about my interpretation of them, and how I measure them. Returning once again to the research question, I identified the following variables that are to be measured in this study, and present them in the table below:

 ideas/ discourse (independent variable),

 workfare activation policies (dependent variable),  welfare state (independent variable).

Variables Definition/

Conceptualization

Dimensions Indicators/Examples

Ideas Beliefs which

communicate what people want, what they deem

appropriate, legitimate and proper (Béland & Cox, 2011, p. 3).

 Cognitive ideas (logical – does it fit within the institution)  Normative ideas

(appropriate – does it make sense)

Liberal ideas:

 Benefits are only for the neediest people

 The labor market should be the first source of support

Social democratic ideas:

 All citizens in need of support may apply for benefits  Benefits should

help participants to play an equal role in society

Discourse The interactive

process of the exchange of ideas (Schmidt 2002, 2008, 2011).  Coordinative discourse (communication among peers)  Communicative discourse (communication to the public) Liberal discourse:

 Phrases such as: “work first” and “make work pay”  Seek employment

first before seeking welfare

 Be independent of welfare/ you can make it on your own

Social democratic discourse:

(26)

 Duty to society so everyone that can work, should work  We need to provide opportunities for the underprivileged  Words: solidarity, equality Workfare activation policies Workfare: A work first approach that mixes labor market participation and social protection efforts.

 Stricter conditions  Efforts to employ the

least employable  Less generous benefit

provision

 Strict job search requirements  Penalties for

non-compliance

 Subsidized jobs for the youth or disabled  Lower benefits  Shorter benefits duration Activation: Programs that promote and assist the unemployed to return to work  Educational programs  Work programs  Entrepreneurship programs  Job training programs  Private sector subsidies  Start-up grants  Direct employment in public sector jobs  Job search assistance such as counseling and monitoring Welfare state A system of social protection provided by the government to assuage social risks  Liberal  Universal

Liberal welfare state:

Means-tested, non-generous benefits

Universal welfare state:

Universal benefits, generous benefits

2.3 Case selection

(27)

Case selection has two objectives: representativeness and variation (Gerring, 2007). The former alluding to how well the case(s) selected can be said to represent the population for which inferences are made, and the latter speaks to the strength of the causal relationship being studied. With the view to achieving these goals, I have elected to carry out a crucial case selection approach, which is based on how well the case selected fits the theory. Moreover, the most- and least-likely case design allows for me to select a case based on the independent variables or X (ideas/ discourse and welfare state regime), the dependent variable or Y (workfare activation policies), and/or the control variables Z (Gerring, 2007). The control variables in this case are factors that can also have a causal influence on the development of workfare activation policies, such as institutional factors (example: previously implemented policies, and partisanship), sociological (example: aging, and the changing family) and economic factors (example: economic crashes) (Bonoli, 2010; Pascual & Magnusson, 2007). I keep these factors constant and focus on discourse (Toshkov, 2016).

Because I wanted to find out how ideas influence workfare policies in disparate welfare states, I selected two countries, the United Kingdom and the Netherlands13, which represent the liberal and universal activation regimes specified by Barbier and Ludwig-Mayerhofer (2004). The United Kingdom is considered a liberal welfare state (Arts & Gelissen, 2002), and the Netherlands is considered a social democratic welfare state (idem). The United Kingdom's benefits system is means-tested to provide for only the most deserving and is not very generous. The Netherlands's welfare system provides state-funded universal benefits and is generous, with benefits that are close to minimum wage. Consistent with the abovementioned theories, the liberal activation regime places emphasis on commodification (i.e.: dependence on the labor market), where the universalistic activation regime places the focus is decommodification (i.e.: dependence on the welfare state in the form of social policies and social insurance once work falls away) (Mascini, Soentken, & Van der Veen, 2012). The two cases offered the opportunity to compare two welfare states of which one is, or rather should be, (theoretically) receptive, i.e.: the United Kingdom, and the other not receptive, i.e.: the Netherlands, to workfare activation approaches.

13_{I would have preferred to have included a third country, for example, France (which represents the conservative}

welfare state regime), in my comparison. This would have been a better fit with Esping-Andersen's typology and thus strengthen any inferences I make on the wider population of welfare states. However, though Barbier and Ludwig-Mayerhofer acknowledge that there may be a third activation regime typology to represent the conservative welfare regime identified by Esping-Andersen in the Three Worlds of Welfare Capitalism, they have not identified a third activation regime typology that fits.

(28)

Secondly, I looked closer at a specific workfare activation policy in these countries. I selected workfare activation policies in the 1990s that occurred around the time when there was a shift in thinking about the welfare state from passive to active and when the European Employment Strategy was released (Deeming, 2015; Mascini, Soentken, & Van der Veen, 2012; Verhoeven & Tonkens, 2013). In the United Kingdom, I study the New Deal for Young People, and in the Netherlands, I look at the ‘Wet Inschakeling Werkzoekenden 14_{)”, which were both} implemented in 1998 in their respective countries. Moreover, both the governments in the Netherlands and the United Kingdom were social democratic and came into power after a long period of conservative rule.

2.4 Data collection sources

In section 2.2, I explained that the coordinative and communicative discourse can be found in written sources. Thus, my principal data collection method was to investigate primary sources. These primary sources consist of the following:

 Official legislation of the New Deal for Young People and Wet Inschakeling

Werkzoekenden.

 Spoken and written texts by the political actors that were instrumental in bringing about the policy. These are Tony Blair (Prime Minister) and Gordon Brown (Chancellor of the Exchequer) in the United Kingdom, and Wim Kok (Prime Minister) and Ad Melkert (Minister of Social Affairs and Employment) in the Netherlands.

The discourse texts consisted of the following: o Speeches

o Political manifestos o Pre-election debates 15

o Governmental debates between proponents and opponents in the House of Commons/ Lords in the United Kingdom and the First and Second Chambers in the Netherlands

o Newspaper and television interviews

14_{In English, loosely translated, this is the unemployed activation act. It is the law governing the activation of}

young people and the long-term unemployed.

15_{I included pieces of evidence before the legislation, such as pre-election debates and political}

(29)

The following table outlines the specific pieces of evidence found for each data source.

New Deal for Young People Wet Inschakeling

Werkzoekenden

Speeches Tony Blair’s Leader’s speech,

Brighton 1995 (Blair, 1995)

(Communicative discourse)

Wim Kok’s Den Uyl-lezing in 1995 (Kok, 1995)

Gordon Brown introducing NDYP to the House of Commons

(HANSARD, 1997 a) (Coordinative discourse)

Political manifestos

Political manifesto Labour Party: New Labour Because Britain deserves better (Labour Party, 1997)

Political manifesto PvdA: Wat men bindt – Verkiezingsprogramma 1994-1998 (PvdA, 1994)

Election debates

Question Time - Tony Blair

Hustings, 24 April 1997 (YouTube, 1997 b) (Communicative discourse) Verkiezingsdebat 1994 (Youtube, 1994) (Communicative discourse) Governmental debates

Welfare-to-work strategy debate in the House of Commons

(HANSARD, 1997 b)

(Coordinative discourse)

Het wetsvoorstel Regeling voor de totstandkoming van een

gemeentelijk werkfonds voor voorzieningen ter bevordering van de toetreding tot het arbeidsproces van langdurig werklozen en jongeren (Wet inschakeling werkzoekenden) (Handelingen II 1996-1997, 1997 )

Welfare-to-work Strategy debate in the House of Commons

(HANSARD, 1997 c)

Het wetsvoorstel Regeling voor de totstandkoming van een

gemeentelijk werkfonds voor voorzieningen ter bevordering van de toetreding tot het arbeidsproces van langdurig werklozen en jongeren (Wet inschakeling werkzoekenden) (Handelingen I, 1997-1998, 1997 ).

(30)

Interviews Panorama: The Leader Interviews -

Tony Blair (07.04.97) (Panorama, 1997)

Werk, werk en nog eens werk' (27.05.95)

Interview met Minister van Sociale Zaken en Werkgelegenheid: Ad Melkert (Melkert, 1995)

Position paper Working paper No. 2

“Europe, the third way, die neue Mitte” (Blair & Schröder, 1998)

The United Kingdom

 Paper on the Third Way

The paper by Tony Blair and Gerhard Schröder can be characterized as coordinative discourse. The target of Blair and Schröder seemed aimed at European government, evidenced in the following statement: “We invite all social democrats in Europe not to

let this historic opportunity for renewal pass by” (Blair & Schröder, 1998, pp. 12)

 The pre-election interview with Tony Blair in Hustings

The Hustings interview was a platform for members of the public throughout Britain to ask questions to Tony Blair before the 1997 elections. This can be categorized as communicative discourse.

 Panorama interview

Tony Blair sat down to an interview on the television program Panorama with David Dimbleby in April 1997. This interview was very combative with Tony Blair being forced to defend his position on various issues. Because his audience was ultimately the general public, this text can be characterized as communicative discourse.