Preventive active labour-market policies

Classifying policies as preventive or curative depends of course on the definition of the underlying risk. In the context of public health policies for example, various levels of prevention are distinguished (Sinfield, Reference Sinfield2007): preventing disease, preventing serious consequences of disease and preventing complications, disabilities and pain. A similar distinction in terms of levels of prevention can be made in the context of unemployment (cf. McKinnon, Reference McKinnon2010) by distinguishing between preventing unemployment as such, preventing the depletion of human capital during unemployment spells (Bonoli, Reference Bonoli, Morel, Palier and Palme2009) and preventing long-term unemployment and multiple forms of social exclusion. In other words, a specific policy can be preventive, as well as curative, depending on the risk to which it relates. In this article, the risk of becoming unemployed is used as the point of reference in distinguishing preventive from curative policies. Curative ALMP are defined as policies targeted at people in situations of unemployment and aimed at (re-)integrating them into the labour market. Preventive ALMP are defined as policies that aim at preventing people from becoming unemployed in the first place. Although this article focuses on preventive ALMP directed at workers, these are not, by definition, limited to policies targeted at the employed. Policies that support school-to-work transitions in order to prevent young school leavers from becoming unemployed can be considered as preventive ALMP as well.

Preventive worker-directed ALMP can be further differentiated according to the time horizon that they focus on. Preventive worker-directed ALMP with a long-term focus are future-oriented and aim to promote sustainable employability and labour-market participation. Those with a short-term focus are aimed at workers confronted with an acute risk of becoming unemployed because of individual dismissal, firm reorganisations or firm closures. The latter type of preventive worker-directed ALMP has received most attention in recent years as a consequence of the crisis that presented many workers with an acute risk of unemployment.

The case for preventive worker-directed ALMP

Despite the bias towards curative ALMP in policy discourse and academic research, several authors define ALMP in ways that acknowledge the preventive potential of these policies. Kluve mentions ‘preventing unemployment for those individuals at risk’ (Kluve, Reference Kluve2010: 27) as one of the objectives of European ALMP. The definition of (passive and active) labour-market policies given by Auer and colleagues involves a broad definition of the target groups of these policies as well, and includes the underemployed and employed people looking for better jobs (Auer et al., Reference Auer, Efendioglu and Leschke2005). Mandl and colleagues argue that ALMP cover both those already in unemployment and those at risk of unemployment (Mandl et al., Reference Mandl, Hurley, Mascherini and Storrie2010). Embroidering on this way of defining ALMP, the academic debate about policies to prevent workers’ unemployment can be interpreted as an elaboration of the preventive dimension of ALMP. A logical further step is to classify these policies as preventive worker-directed ALMP. This can be further substantiated by looking at how the literature discusses preventive approaches concerning the risk of unemployment.

The social investment state literature contains numerous references to risk prevention as a core objective of social investment policies. Esping-Andersen argues that ‘a truly effective and sustainable social investment strategy must be biased towards preventative policy’ (Esping-Andersen, Reference Esping-Andersen2002: 5). Hemerijck makes a similar point when he writes that ‘Social investment should become future oriented, with policies aiming to prepare individuals, families, organisations, and societies to pre-empt new social risks rather than simply repair damage’ (Hemerijck, Reference Hemerijck2013: 37). This focus, in the social investment approach, on preventing rather than repairing damage is also mentioned by various other authors (Bonoli and Natali, Reference Bonoli, Natali, Bonoli and Natali2012; Daly, Reference Daly2012; Jenson and Saint-Martin, Reference Jenson and Saint-Martin2003). The flexicurity literature has a similar orientation towards risk prevention. The long-term and short-term oriented types of policies distinguished above are manifest in flexicurity policy recommendations that aim to support work-to-work transitions, either to prevent acute threats of unemployment or to promote labour-market mobility within and between firms (Bekker and Wilthagen, Reference Bekker and Wilthagen2008). As was mentioned before, short-term oriented policies aimed at preventing dismissals have recently received particular attention in the literature (Boulin and Cette, Reference Boulin and Cette2013; Heyes, Reference Heyes2013a). In a study of short-time working arrangements, which were one of the measures adopted in many EU countries to cope with the crisis and prevent dismissals, an ‘activation’ of these arrangements was recommended by combining temporary reductions in working hours – which is a preventive but not necessarily an active measure – with initiatives focused on investing in workers’ employability (Mandl et al., Reference Mandl, Hurley, Mascherini and Storrie2010). A focus on risk prevention also characterises the transitional labour-market approach (Schmid, Reference Schmid2006), although in this approach risk prevention refers explicitly to a broad set of transition risks and not to unemployment exclusively (Schmid, Reference Schmid2015).

In sum, preventing unemployment, promoting sustainable labour-market participation and employability and supporting transitions over the life course are considered core aims of new welfare arrangements. Gallie (Reference Gallie and Esping-Andersen2002) argues that the stigma of unemployment and the negative impact of unemployment on job acquisition strengthen the argument in favour of policies aimed at preventing unemployment. The focus, in preventive strategies, on (sustainable) labour-market participation and (sustainable) employability resembles the objectives of current curative ALMP. When it comes to the substance of the policies aimed at preventing unemployment, similarities with curative ALMP exist as well. Education and training, usually considered one of the cornerstones of curative ALMP (Kluve, Reference Kluve2010), are without doubt considered to be the number one priority in unemployment prevention strategies. In knowledge-based economies, low or obsolete skills are an important new social risk (Bonoli, Reference Bonoli2005), and lifelong learning is considered vital in dealing with this risk in a preventive way. Some authors argue that these policies should specifically be targeted at low-skilled people who are the most vulnerable groups in knowledge-based economies and should be aimed at enhancing their capacity to learn (Lundvall and Lorenz, Reference Lundvall, Lorenz, Morel, Palier and Palme2009). Obviously, education and training can be part of short-term as well as long-term oriented preventive strategies. Other preventive policies that are mentioned in the literature resemble what are generally known as ‘services’ in the context of curative ALMP (such as job mediation, guidance, removing obstacles for labour-market participation) and include lifelong guidance and lifelong career development (Sultana, Reference Sultana2013), policies aimed at providing support in work-to-work transitions (Bekker and Wilthagen, Reference Bekker and Wilthagen2008; Schmid, Reference Schmid2015) and policies that help people to cope with difficult circumstances that might jeopardise their employment (Sinfield, Reference Sinfield2007). We can conclude that, although curative and preventive ALMP address different target groups (the unemployed and the employed respectively), the distinction between the two policy types is blurred given the similarities that exist in terms of policy objectives and policy instruments.

A crucial theme arising from the literature on the shift from curative to preventive policies concerns the issue of how social protection and prevention are related. Various authors point out that according to the social investment perspective, ‘there should be less emphasis on “social protection” than on being preventive and proactive’ (Jenson, Reference Jenson, Bonoli and Natali2012: 28). The notion of shifting priority from social protection to prevention has been criticised, in the same way that curative ALMP were criticised for undermining income protection. Several authors argue that decent social protection systems are a core element of active and preventive welfare arrangements. They prevent risk-aversive behaviour and stimulate labour-market transitions and therefore should not be interpreted as opposite to active and preventive strategies (Schmid, Reference Schmid2015; Sinfield, Reference Sinfield2007; Sjöberg, Reference Sjöberg2008). The importance of this issue becomes specifically pronounced in debates about the impact of labour-market changes and the policy responses to these changes on dualisation and the differential treatment of various groups in the labour market (Emmenegger et al., Reference Emmenegger, Häusermann, Palier and Seeleib-Kaiser2012; Jessoula et al., Reference Jessoula, Graziano and Madama2010).

Our discussion of preventive ALMP and the resemblances they bear to curative ALMP has thus far not paid attention to what are usually considered the most important characteristics of curative ALMP: their obligatory nature, the conditionality of income protection for the unemployed target groups of curative ALMP and sanctions in cases of non-compliance. In our opinion, it remains to be seen whether these characteristics constitute fundamental differences between both types of ALMP. Given the increasing emphasis on individual responsibilities in preventing and coping with risks (e.g., Lister, Reference Lister2003), the differences may very well turn out to be gradual. We did not find any explicit discussion of this issue in the academic literature. Nevertheless, we believe it cannot be ruled out that the shifting responsibilities trend will, at some point, result in a debate about the issue of whether entitlements to unemployment benefits should be made conditional upon workers’ efforts to maintain and improve their employability and avoid situations of unemployment.

Pluralistic management of unemployment prevention

Pluralistic risk management is another core issue frequently discussed in the new welfare literature. This issue is particularly salient in strategies aimed at preventing unemployment of workers and at promoting their employability in the firm and the labour market. Firstly, social partners and individual firms are expected to be more strongly committed to, and to have a stronger interest in, investing in employees than in the unemployed. Although the inclusive nature of this commitment and interest is highly contested where non-core and flexible workers are concerned (see below), generally speaking one may hypothesise that, compared to curative ALMP, the involvement of social partners and employers in preventive worker-directed ALMP will be easier to realise. At the same time, it has been pointed out that individual firms may be reluctant to invest in workers’ skill formation for fear of poaching from competitors (Hemerijck, Reference Hemerijck and Esping-Andersen2002; Van der Veen et al., Reference Van der Veen, Yerkes and Achterberg2012). This is one of the reasons why it may be advisable to oblige employers to invest in education and training, as has been argued in the context of reduced working hours schemes (Heyes, Reference Heyes2013b). Secondly, various authors have argued that policy mixes and public-private cooperation and coordination – i.e., pluralistic risk management – are crucial for making preventive strategies successful (Bekker and Wilthagen, Reference Bekker and Wilthagen2008; Hemerijck, Reference Hemerijck2013; Schmid, Reference Schmid2015; Van der Veen et al., Reference Van der Veen, Yerkes and Achterberg2012). This underlines the hybrid nature of policies aimed at preventing unemployment: they are a mix of social policies in the traditional sense of ‘policy of governments’ (Marshall, Reference Marshall1965: 7), industrial relations and collective agreements as a source of social policies (Trampusch, Reference Trampusch2007), and ‘social policies of the firm’ (Rein, Reference Rein1982). There is some empirical evidence that pluralistic risk management does indeed contribute to the successfulness of preventive approaches. A study of German shorter-working schemes concluded that these schemes were successful precisely because they combined government policies, collective agreements and firm-level measures rather than being merely state-subsidised reductions of working time only (Boulin and Cette, Reference Boulin and Cette2013). Mandl et al. carried out a comparative study of shorter-working schemes and found that these schemes (and flexicurity policies more generally) require ‘a climate of trust and broadly-based dialogue among all stakeholders’ (Mandl et al., Reference Mandl, Hurley, Mascherini and Storrie2010: 79), including the state, trade unions and employers.

However, the notion of pluralistic risk management is contested. Two issues are of particular importance in this debate. Firstly, the climate of trust considered necessary to make pluralistic risk management work is exactly what makes it highly vulnerable. Trust between the state and social partners or between trade unions and employers cannot be taken for granted (Bonoli and Emmenegger, Reference Bonoli and Emmenegger2010; Burroni and Keune, Reference Burroni and Keune2011), making concerted action involving the state, trade unions and employers not always likely and feasible. This raises important questions regarding characteristics of the institutional context in which pluralistic risk management may flourish or is likely to fail. The distinction, in theories about varieties of capitalism, between liberal and coordinated market economies is important in this context, of course. But, as Thelen (Reference Thelen2014) argues in a recent study, of equal importance are the trajectories of institutional change that have taken place in the era of liberalisation and, as Thelen emphasises throughout her study, these are diverse. Secondly, the balance between public and private responsibilities is at stake. In this context, two closely connected issues are discussed. On the one hand, the literature points at the governance capacities of public and private actors (Knill and Lehmkuhl, Reference Knill and Lehmkuhl2002). In other words, the distribution of responsibilities between the state, individual firms and social partners in managing social risks is not only an issue of the readiness of actors (or the lack thereof) to accept responsibilities, it is also an issue of their abilities to realise social objectives, as well as about the interdependencies between these actors in successfully managing social risks (e.g., Van Berkel and Leisink, Reference Van Berkel, Leisink, Leisink, Boselie, Van Bottenburg and Hosking2013). On the other hand, many authors argue that relying too strongly on collective agreements and firms’ human resources policies in addressing social issues will favour core workers at the expense of atypical, non-standard workers, exacerbating labour-market segmentation and dualisation processes as well as in-company creaming practices. Therefore, various scholars conclude that strong state involvement remains crucial in order to ensure that those most in need of preventive policies are not excluded from them (Bekker and Wilthagen, Reference Bekker and Wilthagen2008; Heyes, Reference Heyes2013a; Mandl et al., Reference Mandl, Hurley, Mascherini and Storrie2010; Schmid, Reference Schmid2015). In a similar way, some authors emphasise that strengthening social partners’ and firms’ responsibilities in dealing with unemployment prevention will result in stronger efforts in some companies or sectors compared to others. Small and medium-sized enterprises working under high competitive pressure are considered especially unlikely to invest in the employability of their workers (Schmid, Reference Schmid2015).

Summarising, although the importance of the joint involvement of the state, social partners and employers in developing preventive strategies is recognised, the feasibility of pluralistic risk management is far from self-evident. And even when institutional conditions for pluralistic risk management are favourable, the state is seen as having core responsibilities in preventing labour-market segmentation, both in terms of standard/core versus non-standard/non-core workers and in terms of competitively strong versus weak companies/sectors.

Policy substance: preventing unemployment

Generally speaking, until the financial crisis, preventive approaches in Dutch social policies mainly focused on the prevention of (long-term) sickness and disability. The privatisation of sickness benefits is probably the best-known example of this: since the early 2000s, Dutch employers have been obliged to pay sick workers at least 70 per cent of their wages for a maximum period of two years. The intention is that this will stimulate employers to invest in sickness prevention (by promoting healthy working conditions and healthy lifestyles) as well as in preventing long-term sickness absence (by promoting the reintegration of sick workers) (Van Gestel and Nyberg, Reference Van Gestel and Nyberg2009). The experiences in the area of sickness and disability prevention inspired debates about preventive approaches to unemployment, but eventually it was the financial crisis that triggered the introduction of the first policies aimed at unemployment prevention.

One of the preventive measures that was introduced at the start of the crisis was a variety of the short-time working schemes that have been introduced in many European countries (Mandl et al. Reference Mandl, Hurley, Mascherini and Storrie2010): the part-time unemployment benefits regulation as it was called in the Netherlands. The main objective of this measure was to retain workers in times of economic crisis and prevent their unemployment. It was based on the assumption that groups of workers threatened with unemployment in times of crisis would be needed again when the economy recovered (labour hoarding). What made this preventive policy exceptional from an international perspective was that it included a future-oriented dimension: it obliged employers to invest in workers’ employability by providing them with training opportunities while on reduced working hours. This obligatory element was absent in short-time working schemes in many other countries, where employers were stimulated but not obliged to provide training (Heyes, Reference Heyes2013b).

Whereas this policy intended to retain employees during times of economic crisis, a second policy aimed at promoting work-to-work transitions of workers who were going to lose their job. The Dutch government temporarily subsidised nine regional or sector pilots in this area. The objective of this policy was to promote a stronger role for employers and trade unions in finding new jobs for employees threatened with job loss, and to find out under what conditions pluralistic risk management may have a positive impact on facilitating work-to-work transitions. Some pilots were initiated by employers, whereas others were joint initiatives of employers and trade unions. The pilots used a large variety of instruments in supporting employees in finding new jobs, such as education and training, job guidance, mediation and outplacement services, job hunting and job carving (Visscher et al., Reference Visscher, De Groot, Van Eck, Van Gestel and Borghouts2012). These preventive instruments closely resemble those used as part of curative active labour-market measures.

The third policy that we discuss here concerns the establishment of mobility centres. These centres intended to provide a regional infrastructure for facilitating early interventions in cases of mass dismissals and for coordinating public and private efforts. In addition, services of the Dutch unemployment benefit agency could be mobilised before workers actually became unemployed, redirecting ‘curative’ public resources towards unemployment prevention initiatives. Since 2010, the mobility centres and their services have been integrated into the unemployment benefit agency through the establishment of employer services agencies.

The prevention of acute unemployment was the main aim of these preventive worker-directed ALMP, although the short-time working scheme also included a longer-term employability perspective. Although innovative in the Dutch context, these preventive policies did not represent a radical shift in dealing with unemployment. Some were temporary and their coverage – in terms of the percentages of people threatened with unemployment that were affected by these policies – was very modest. For example, the proportion of Dutch employees that used the short-time working scheme was 1 per cent of the labour force (De Groot et al., Reference De Groot, Friperson, Weda and De Jong2012), which is low in comparison with countries such as Germany (Schmid, Reference Schmid2015). Nevertheless, taking into account the incremental nature of many more fundamental policy reforms, these policies may turn out to be forerunners of a more significant shift towards preventive worker-directed ALMP. Recent policy reforms seem to confirm this. In 2013, a social pact was concluded between the Dutch government and social partners (Stichting van de Arbeid, Reference van de Arbeid2013). The implementation of the agreements in this social pact requires new legislation, of which the Work and Security Bill that was introduced into the Dutch parliament in November 2013, is the first step. The pact puts a clear emphasis on the prevention of unemployment and promoting work-to-work transitions, as well as on the joint responsibilities of the state, social partners and employers in developing preventive strategies. According to the pact, social partners will be made responsible, as of 2020, for policies concerning prevention, support, job-mediation and reintegration (that is, ALMP targeted at workers as well as unemployment benefit recipients), representing a clear shift from state-dominated to pluralistic risk management. In terms of preventive measures, the introduction of so-called transition payments in cases of dismissal is presented as an important innovation that should support work-to-work transitions of people threatened with unemployment. The entitlement to transition payments – an example of preventive ALMP with a short-term focus – is regulated in the Work and Security Bill.

The preventive approach and pluralistic risk management

As we saw, pluralistic risk management is a core feature of the preventive approach to unemployment in the Netherlands, which confirms the critical role the academic literature ascribes to pluralistic risk management in making preventive approaches both feasible and successful. A useful theoretical perspective that helps to explain the rise of pluralistic risk management in preventive approaches in the Netherlands, and that goes beyond the distinction between the varieties of capitalism, i.e., liberal and coordinated market economies, is provided by Trampusch (Reference Trampusch2007). Although Trampusch's research focused on the role of collective agreements negotiated between social partners as a compensation for welfare state retrenchments, her typology of institutional conditions that support or hinder collective agreements on social risk protection is also useful when we want to understand the involvement of Dutch employers and trade unions in policies aimed at preventing unemployment. Trampusch's typology distinguishes two dimensions: the timing of the institutionalisation of industrial and political citizenship, and the role of the state (active or passive) in promoting and facilitating collective agreements between social partners on social protection arrangements. In countries where industrial citizenship preceded political citizenship and where the state plays an active role, social partners are most likely to develop collectively negotiated social protection schemes. And it is exactly this group of countries to which the Netherlands belong (Trampusch, Reference Trampusch2007; Yerkes and Tijdens, Reference Yerkes and Tijdens2010). If we consider policies aimed at preventing unemployment as a form of social protection, and if we take into account the crucial role of pluralistic risk management in developing these policies, countries such as the Netherlands are likely candidates for providing a relatively favourable institutional context for the development of preventive approaches.

Whereas Trampusch's typology focuses on the diversity of institutional conditions in various countries, a recent study by Thelen (Reference Thelen2014) highlighted differences between the ‘traditional’ manufacturing sector and the ‘new’ service sector. And although Thelen points at these differences in the context of an argument on varieties of liberalisation, they may also affect the likelihood of the successful development of pluralistic risk management in both sectors. Until now, the issue of the impact of sector differences on the feasibility of pluralistic risk management has received little attention in the literature. But sector differences may have played a role, for example in the Dutch work-to-work pilots – only a few survived after state subsidies were withdrawn and those that did involved traditional industries with a long tradition of cooperation between social partners and strong sector institutions. Thus, institutional conditions for pluralistic risk management may differ between countries as well as between sectors of industry within countries.

In terms of the social consequences of pluralistic risk management, Trampusch (Reference Trampusch2007) points out that collectively bargained social protection arrangements may strengthen social inequalities between, among others, workers in ‘typical’ and those in ‘atypical’ employment. As we saw, this is a crucial issue in the context of preventing unemployment, as people in atypical employment are most strongly exposed to risks of unemployment and can be considered most in need of policies that promote their employability and support successful work-to-work transitions. This focuses attention on another aspect of the ‘active’ role of the state: the protection of vulnerable groups in the labour market and the degree to which the state intervenes to avoid a situation in which an increasing role of social partners and firms in providing social protection arrangements aimed at unemployment prevention results in processes of dualisation.

Pluralistic risk management and dualisation

In the academic literature, the Netherlands is mentioned as one of the countries where processes of dualisation have been modest, at least from an international perspective. Thelen's (Reference Thelen2014) recent study of varieties of liberalisation, for example, showed that the Netherlands has become less dualistic and more ‘solidaristic’ during the period from the 1980s to the late 2000s. According to Thelen, social policies played a role here: the Dutch government introduced policies strengthening the position not only of part-time workers – in the Dutch context, part-time work has increasingly become considered a form of regular rather than flexible work – but also of flexible workers, including temp agency workers. In other words, the Dutch state has taken an active role in introducing policies that intend to mitigate and even reduce dualisation between ‘regular’ and ‘atypical’ workers. Thelen's argument that the Dutch state assumes an active role in the dualisation issue also seems valid in the context of the increasing emphasis on unemployment prevention. Both the social pact and the Work and Security Bill specifically address the position of flexible workers. The social pact states that flexible workers have equal access to education activities in their firm/sector and the same entitlements as ‘regular’ workers to work-to-work transition facilities. The Work and Security Bill stipulates that workers become entitled to transition payments after having worked with the same employer for two years.

However, this does not mean that dualisation is no longer an issue in the Netherlands. Even the small-scale policies focusing on unemployment prevention, discussed above, showed that dualisation is a real threat. For example, the majority of employees using the part-time unemployment benefit scheme were experienced workers who had been working with their employer for over ten years, whereas workers on temporary contracts were under-represented (De Groot et al., Reference De Groot, Friperson, Weda and De Jong2012). In the context of preventing long-term sickness and disability, which as we saw is subject to pluralistic risk management as well, processes of dualisation between ‘regular’ and flexible workers are clearly visible (Van Berkel, Reference Van Berkel, Lindsay and Houston2013). This raises the question of whether state social policies are providing adequate protection for atypical workers. It also raises the issue of the effectiveness of social policies in changing sector and firm-level practices in dealing with various groups of workers; that is, the issue of the capacity of the state in successfully countering or preventing dualisation. In other words, even though the state may be committed to addressing the labour-market position of flexible workers and processes of dualisation between ‘regular’ and flexible workers, this does not necessarily imply that its policies are effective. Therefore, the debate on dualisation should not only focus on the role of the state and state social policies, but also on the implementation or ‘translation’ of these policies in sectors and firms (Van Gestel and Nyberg, Reference Van Gestel and Nyberg2009).