(Mis)taking social responsibility? Implementing welfare state reform by private and non-profit organizations

ABSTRACT

This article relates institutional theory to the concept of organizational legitimacy with cognitive, moral, and pragmatic dimensions, to analyse how a Dutch national policy reform – aimed at expanding the social responsibility for sick leave and disability toward non-state organizations – is understood and enacted locally. Based on interviews in 52 organizations, the study highlights that implementing welfare state reform is predominantly based on pragmatic reasoning, and justified by specific moral and cognitive interpretations. The findings reveal that implementing reform is active institutional work by – paradoxically – restricted local agency, with disciplinary effects on a narrow range of actors.

KEYWORDS:

• Welfare state reform
• institutional theory
• organizational legitimacy
• social responsibility
• sick leave policy

Introduction

Reforms and new policies are continuously on the agenda in public administration and management (Christensen and Laegreid 2017; Hammerschmid et al. 2016; Pollitt and Bouckaert 2017). At the core of the reforms is regulation as a specific form of governance, ranging from formal authoritative rules to mechanisms of social control (Jordana and Levi-Faur 2004; Scott 2014). The reforms search for efficiency and effectiveness, as approached by deregulation, marketization and privatization, and for collaboration with private and non-profit actors, greater accountability, justice and fairness (Lane 1997; Osborne 2010; Pollitt and Bouckaert 2017). Such a mix of policy goals creates a context in which executive organizations have to accommodate a plurality of demands (Kraatz and Block 2008; Denis, Ferlie, and Van Gestel 2015). Yet, public policy literature has barely focused on the internal considerations of organizations when interpreting and applying macro policy aims and standards (Hupe and Hill 2016; Pandey and Wright 2006), in particular when delegating public policy implementation to private and non-profit organizations (Brodkin 2016; Butler and Allen 2008; Mariani and Cavenago 2013).

This article explores how a national policy reform is framed and connected to organizational interests and activities. We contribute to the still limited understanding of how macro-ideas and policies are diffused across institutional levels and how organizational actors and their immediate context play a role in this process (Bitektine and Haack 2015; Powell and Colyvas 2008; Zilber 2016). Our analysis is informed by the notion of organizational legitimacy (Suchman 1995), a key concept in institutional theory (Brown and Toyoki 2013; Suddaby, Bitektine, and Haack 2016), defined as the perceived appropriateness of an organization to a social system in terms of rules, values, norms and definitions (Deephouse et al. 2017). As many organizations operate in a highly pluralistic and complex social environment (Denis, Ferlie, and Van Gestel 2015; Kraatz and Block 2008; Greenwood, Magán Díaz, Xiao Li et al. 2010; Osborne 2010), they relate to different stakeholders and audiences who frequently have conflicting expectations and perceptions (Cloutier et al. 2016; Waldorff 2013). The concept of organizational legitimacy with three dimensions – cognitive, moral and pragmatic – can contribute to understanding the impact of institutional pressures, while describing the complex relationship between organizations and their environment in which public reforms attempt to intervene to create change (Deephouse et al. 2017; Suddaby, Bitektine, and Haack 2016).

The empirical case is the national policy for sick leave and disability in the Netherlands that shifted responsibility for reducing sickness absence and disability claims from the national to the organizational level. The delegation of welfare state’s social responsibility for sick leave and return-to-work to individual organizations has been advocated by the OECD (2010, 2015) to prevent illness of employees and actively facilitate their return to work and workplace inclusion. Since the Netherlands is known as an extreme case in regulating sick leave and disability with employers being responsible for the first 2 years of absence; the Dutch case is particularly relevant in exploring how institutional pressures at the macro-level are translated locally. Our study is based on in-depth interviews with HR directors, managers and health experts of 52 organizations across industry, non-profit services, and commercial services. To investigate why the outcome of the national policy reform is framed in a particular way at the local level, we explore how the private and non-profit organizations interpreted the ideas behind the delegated national welfare policy.

The contribution to the literature is twofold. First, applying the framework of organizational legitimacy with three dimensions – cognitive, moral, and pragmatic – shows how organizations react in a selective way on certain elements of the national policy shift, while other, often more crucial elements are ignored. We thus highlight how implementing macro institutional pressures into organizational practices is active institutional work (Lawrence, Suddaby, and Leca 2009), with disciplinary effects on a narrow range of actors. Yet, in contrast to an overly ‘agent-centred’ focus in the institutional work approach (see for criticism, e.g. Kaghan and Lounsbury 2011; Landau, Drori, and Terjesen 2014; Willmott 2011), our study also reveals that local actors embedded in their specific organizational context, have limited agency in applying new ideas or policies. Second, we show that the frictions between the cognitive, moral and pragmatic legitimacy in implementing institutional pressures stress the discrepancy between the public policy’s wider societal aims and how organizations interpret and enact the policy. Although literature has emphasized the role of cognitive perceptions and moral values in re-framing institutional pressures (Greenwood., Suddaby, and Hinings 2002; Suddaby, Bitektine, and Haack 2016), we show that enacting institutional pressures is a far more pragmatic process of interpretations and actions. In the light of the request for a more complete evaluation of the concept of organizational legitimacy (Suchman 1995; Suddaby, Bitektine, and Haack 2016), we thus provide a fuller picture by underlying the importance of pragmatic legitimacy and showing the relative impact and interplay of the three dimensions in organizational responses to institutional complexity.

The paper is structured as follows. Building on literature of institutional theory and organizational legitimacy, we discuss how welfare state reform may become elaborated and manifested in private and non-profit organizations. Next, we describe our research method, and present our findings. We show how a national policy change for sickness absence and disability aimed at expanding the social responsibility of private and non-profit organizations is justified differently at the organizational level, where it is interpreted according to managerial aims of control and cost-reduction, and used to discipline a narrow range of organizational actors. Based on our findings, we discuss consequences for institutional theory, and the shaping and implementation of institutional pressures in welfare state reform.

The impact of institutional pressures: three dimensions of organizational legitimacy

Institutional theory seeks to explain how institutions – resilient social structures reflecting the wider societal expectations of appropriate behaviour – affect the actions of organizations (Scott 2014). Institutions, or institutional pressures, logics or demands, refer to the rules and regulations, normative prescriptions and cultural templates exerted on organizations in a given field (Pache and Santos 2010; Scott 2014). Scholars nowadays recognize that organizations face multiple, and often conflicting, institutional demands (Greenwood et al. 2011; Kraatz and Block 2008), and this institutional complexity is not a temporary situation but an enduring characteristic of many fields (Greenwood et al. 2011; Van Gestel and Hillebrand 2011; Voronov, De Clercq and Hinings 2013). Accordingly, scholars have become more concerned with understanding the strategies of actors, responding to (multiple) institutional pressures at the level of organizations (Oliver 1991; Pache and Santos 2010; Powell and Colyvas 2008; Thornton, Ocasio, and Lounsbury 2012).

In institutional theory, it is increasingly recognized that the spread of macro-ideas is dependent on local meaning created by actors in organizations (Lawrence, Suddaby, and Leca 2009; Wæraas and Nielsen 2016). Rather than a passive adoption of the same idea for identical reasons, the process of converting environmental stimuli into organizational practices is based on the interpretations by organizational actors in their context (Boxenbaum 2006; Sahlin and Wedlin 2008; Zilber 2016). In contexts, where national agencies lack coercive power over those who are involved in policy implementation in some way, e.g. private and non-profit organizations, institutional analysis is particularly promising (Baldwin, Chen, and Cole 2018; Smets, Aristidou, and Whittington 2017). However, most studies of institutions are still focused on the macro levels – the global, the societal, or the organizational field – rather than on the inner workings of organizations (Bitektine and Haack 2015; Powell and Colyvas 2008; Zilber 2016). Although it is widely accepted that new ideas or policies need to be justified and shared by organizations in order to become adopted (Boxenbaum 2006; Waldorff 2013), little is known about how (specific) actors interpret and shape institutional pressures within organizations (Pache and Santos 2013; Radaelli and Sitton-Kent 2016).

As DiMaggio and Powell’s (1983) already suggested, institutional pressures may in time generate similarity within an organizational field, a process called isomorphism, which is ‘a constraining process that forces one unit in a population to resemble other units that face the same set of environmental conditions’ (DiMaggio and Powell 1983, 149). This process is based on the central logic of contingency theory: organizations will adapt to their social environment in order to gain legitimacy in the same way as they adapt to their economic environment in order to gain material resources (Suddaby, Bitektine, and Haack 2016). We fully agree that legitimacy is then an outcome of continuous adjustments to maintain congruence between the internal ideas, norms, and interests of the organization and those in the external environment (Deephouse et al. 2017; Suddaby, Bitektine, and Haack 2016). However, as recent institutional research has shown, isomorphism does not exclude variety or change (Ashworth, Boyne, and Delbridge 2009; Wæraas and Sataøen 2014). Local agency may generate both similar and different organizational responses to institutional pressures (Sahlin and Wedlin 2008) and can imply organizational change as well as maintenance (Waldorff and Greenwood 2011; Wæraas and Sataøen 2014). In this article, we thus stress the role of organizations and its actors in responding to institutional pressures, without predetermined outcomes.

Organizational legitimacy

Our theoretical framework builds upon existing institutional literature on legitimacy. Parsons (1956) viewed legitimacy as an organization’s resemblance with an institutionalized value system. Goal attainment – in an economic sense – has priority, but ‘the conduct of the affairs of an organization must in general conform with the norms of “good conduct” as recognized and institutionalized in the society’ (Parsons 1956, 84). Later, Meyer and Rowan (1977) unfolded the distinction between organizations’ technical activities and institutional environment, and pointed to dominating ‘rationalized myths’ as sources for organizational legitimacy. Further, DiMaggio and Powell (1983) identified three mechanisms – coercive, mimetic, and normative, by which
organizations gain legitimacy. What we take from these approaches is the notion that organizations will need to take more rationalities, including the economic, into account in order to become legitimate. However, we do not share DiMaggio and Powell’s presumption that organizations will respond similarly due to similar challenges of political pressure, uncertainty and professionalism within their field. Instead, we are interested in exploring the nuances in organizations’ responses and their different underlying rationales without a predetermined theorization of how a particular organizational context will lead to a certain (isomorphic) response. That being said, we acknowledge that the range of variation may be limited, as national policy provides a common framework, within which local organizations develop their responses.

Accordingly, this leads us to apply the theoretical framework as developed by Suchman (1995), including the cognitive, moral and pragmatic dimensions of legitimacy. This framework provides for understanding the underlying rationalities for organizational behaviour including the taken-for-granted beliefs (cognitive legitimacy), norms of proper behaviour (moral legitimacy), and concerns of organizational existence (pragmatic legitimacy). All three dimensions of organizational legitimacy are relevant for investigating organizational behaviour (Suchman 1995). The first dimension of organizational legitimacy refers to cognitive perceptions, that is, to the acceptance of ideas and practices for their necessity or inevitability based on taken-for-grantedness (Scott 2014; Suchman 1995). The second dimension involves the concept of moral legitimacy that indicates what ought to be done locally in response to environmental changes (Greenwood, Suddaby, and Hinings 2002) and rests on an evaluation of how organizations perceive their responsibilities in promoting societal welfare, as stimulated by the wider environment (Suchman 1995, 579). The pragmatic dimension of organizational legitimacy refers to how organizational actors may use new laws and procedures for self-interested calculations of the organization’s most immediate audiences (Suchman 1995). In other words, pragmatic legitimacy may serve some actors’ interests more than others, and maybe based on rational economic aims of the organization more than legal or ethical responsibilities (Seing et al. 2015).

Thus far, the cognitive and moral dimensions have received more attention in research rather than pragmatic legitimacy (Deephouse et al. 2017; Suchman 1995; Suddaby, Bitektine, and Haack 2016). A recent overview of the use of Suchman’s most influential typology confirms that pragmatic legitimacy is still less recognized in institutional theory as a relevant category in understanding organizational behaviour (Suddaby, Bitektine, and Haack 2016). We aim for a more nuanced picture in this study, exploring all three dimensions and their underlying rationales. Moreover, the three dimensions of legitimacy can reinforce one another, but may come into conflict as well. Frictions among moral, pragmatic and/or cognitive considerations are most likely to arise when wider social institutions are in transition (Suchman 1995). Institutionalists may therefore perceive pragmatic legitimacy as more ‘strategic’ and less informed by institutional pressures. However, pragmatic legitimacy can be inspired by moral and cognitive considerations; or vice versa, pragmatic considerations can be used to show that the organization has ‘our best interests at heart.’ (Suchman 1995, 578).

Studies of organizations that operate in complex ‘hybrid’ environments show that actors develop strategies and practices that may align different institutional demands or logics over time (Reay and Hinings 2009; Skelcher and Smith 2015). Variation in responses is dependent on the nature of the institutional demands and the degree of ‘hybridization’: how much an organization represents conflicting demands (Pache and Santos 2010). Responses in complex environments with multiple demands may range from decoupling (prioritizing one logic over another), through compromise (modifying demands to restore balance) to combinations of competing demands by reconciling their intact elements (Pache and Santos 2013; Thomann, Hupe, and Sager 2018). We aim for studying how the different dimensions of organizational legitimacy may conflict or co-exist in an institutionally complex real-world setting of delegating collective social responsibilities to individual, private and non-profit organizations.

Methodology

Research setting

Our research is focused on understanding the interpretation and enactment by private and non-profit organizations of a delegated national policy. As a representative empirical case of this form of governance, we selected the shift in the Dutch national policy for sick leave and return to work since the mid-1990s. This shift is part of a wider strategy of ‘activation policy’ in the European Union, where a larger role of employers and their organizations in taking social responsibility for unemployment, sick leave and disability has been increasingly advocated (OECD 2010, 2015). The overall aim of activation policy is to increase labour market participation and reduce the number of social benefit recipients (Kluve et al. 2007). Within the EU, the Dutch reform is highly relevant while the Netherlands had relatively high levels of sickness absence and disability followed by a radical policy shift (Van Oorschot 2006). Historically, social benefits for sick employees were covered by collective funds, governed by employers’ associations and unions under the supervision of a tripartite board of central government, employer associations and unions. Individual organizations traditionally had a passive role in prevention and reintegration of ill employees.

Strongly growing disability numbers in the Netherlands, from 215.000 in 1970 to nearly one million people in 2000 (Van Oorschot and Boos 2000) led to reconsidering this traditional governance model. The Dutch government saw the high disability numbers as a lack of social responsibility of individual organizations. In 2001, a governmental committee (the Advisory Committee on Incapacity for Work in 2001) stated that a large group with minor labour restrictions are lost for the labour market while most of them could stay at work with due attention. The committee also saw the high level of drop-outs having a negative impact on the stability of the welfare state, while the system of social benefits was increasingly seen as controversial and unfair. As the committee explained in 2001: ‘Tensions and problems within the organization of work are not adequately addressed, but commuted at the expense of society.’ (4).

The far-reaching reform started from a different cognitive framework: instead of classifying illness as a physical or mental state, it treated illness as ‘behaviour’. As a consequence, illness does not necessarily imply that employees cannot work, and work can even be a remedy to illness. To underline organizations’ social responsibility, the Dutch government made each organization financially responsible for sick leave during the first 2 years, implying that employers are legally obliged to pay a minimum of 70 per cent of the salary of the ill worker. Financial responsibility was built up in two steps: since 1996, individual organizations have to pay the salary for their ill workers for 1 year (WULBZ Act, Law on the extended obligation to pay sickness benefits), and since 2004 for 2 years (VLZ Act). In 2002, the government also designed procedures for return to work in the Gatekeeper Act (WVP Act). Following this act it is legally required in each case of long-term sick leave (> 6 weeks) to have a medical diagnosis (by an occupational physician) and a plan for return to work (agreed upon by employer and employee within 8 weeks), with regular follow up meetings between sick-listed employees and their supervisors to monitor the plan and related actions. After 2 years of sick leave, the government can put sanctions if employer or employee has not done enough for return to work. The penalty for the employer is an extra year of salary payment; the consequence for the employee is that the salary can be (further) reduced. Organizations also need to retain workers with physical or mental restrictions that are able to earn at least 65 per cent of their former income (WIA, the Work and Income according to Labour Capacity Act 2006). With these regulative and financial incentives, it is assumed that organizations will reduce sickness absence and invest in prevention.

Data collection and analysis

The study was conducted in 2004–2010, after the new laws for delegating welfare policies on sick leave and return-to-work in 2002–2004 were introduced. So far, the national policy has not been altered and is still relevant. The 52 organizations in our study were selected for representing medium-sized and large organizations in three major economic sectors: industrial companies (18), non-profit services (14), and commercial services (20) across a diverse range of activities (see Table 1). During our research period, average sickness absence rates went down gradually from almost 5 per cent to about 4 per cent (CBS Statline), with rates traditionally being higher in public/non-profit services, lower in commercial services, with industrial companies in an intermediate position.

Table 1. Overview of participating organizations and respondents.

N (52)	Sector & Organization	Respondents (N = 60)	Size in employee numbers (approx.)
	Industry
1	Chemistry/Pharmacy	Consultant Labor Affairs	12,500
2	Production of Bread and Banquet	Personnel manager	270
3	Touring car producer	Manager P&O	580
4	Car and truck manufacturer	HR manager	5,000
5	Sailing yachts producer	Head of Personnel and Organization	190
6	Cosmetic Industry	HR Account manager	230
7	Industrial firm	Occupational physician	12,000
8	Installation technology	HR director and a trainee	871
9	Cosmetic Industry	HR manager Social and Legal Affairs	480
10	ICT/Computer industry	HR manager	262
11	Industrial storage systems	Health and Safety coordinator	210
12	Research and production of seeds	HR officer	350
13	Print Industry	HR manager	3,000
14	Medical technology	Manager Social and Economic Department	24,000
15	Light electronics	Personnel Officer	450
16	Logistics/Transport industry	Manager P&O	425
17	Electronical industry/energy	Head of HR Department	3,100
18	Producer of transformers	Absence coordinator and manager P&O	317
	Non-profit services
19	Hospital care	HR manager	9,480
20	Health care	HR consultant and a Policy officer absence	2,500 (fte)
21	Health care	HR manager and HR team coordinator	2,300
22	Health care	Reintegration coordinator	3,200
23	Health care	Policy advisor HRM	3,000
24	Judiciary	Personnel advisor	271
25	Mental care services	Return-to-work advisor	1,000
26	Health care	Head of HR department	2,300 (fte)
27	Youth care	Policy advisor	371
28	Hospital care	HR manager	3,200
29	Crime fighting	Return-to-work advisor	1,400
30	Research and teaching	Head HRM department	6,000
31	Health care	HR manager	3,200
32	Academic hospital	Head of occupational physicians	8,500
	Commercial services
33	Financial services	HR business partner	28,000
34	Pension fund	Absence coordinator and HR advisor	2,900
35	Catering firm	Health manager	4,000
36	Law firm	HR manager	50–250
37	Law firm	HR manager	50–250
38	Law firm	HR manager	50–250
39	Lawyers and notaries	HR manager	300
40	Law firm	HR manager	50–250
41	Law firm	HR manager	50–250
42	Aircraft services	Head of employability services	30,367
43	Law firm	HR manager	50–250
44	Law firm	HR manager	50–250
45	Professional service firm	Senior & two junior consultants Labor Conditions	4,500
46	Professional service firm	HR manager Southern Netherlands	4,643
47	Professional service firm	Senior advisor HR	2,089
48	Professional service firm	HR manager	544
49	Employment agencies	Two staff members HR support	3,000
50	Hospitality services	Staff member P&O	200
51	Quality of life services	Project manager P&O	6,600
52	Law firm	HR manager	50–250

By interviewing directors and managers in HRM roles, and occupational health experts and reintegration specialists within organizations, we gained access to key-actors in the local management of sick leave and return to work in each of the 52 organizations. We held one interview in each organization, but some organizations sent several respondents, bringing their total number to 60. All respondents had in-depth knowledge of the national policy and legal demands, and could provide a detailed overview of how the delegated policy was interpreted and enacted in their organizations. The topics discussed were – following our research question – addressing how the respondents interpreted the cognitive, moral and pragmatic dimensions of the national policy, and how national ideas and regulations were converted into organizational policies and practices. The interviews lasted at least 60 min, to a maximum of 90 min, and were taped and transcribed verbatim.

The interview data were analysed using matrices and tabulations to search for patterns in the transcripts (Rubin and Rubin 2005), focused at the three dimensions of cognitive, moral, and pragmatic legitimacy. Once we had uncovered the main tendencies and relationships, such as passing new social responsibilities of organizations on to employees, we re-read the data descriptions to find any differing trends or exceptions (George and Bennett 2004). On the basis of the conceptual insights, we composed an initial outline of our results. During the research period, data analysis was built up by detailed interim reports, each comprising a subgroup of the organizations. We presented our findings to various audiences, including national policy makers, HR directors/managers and supervisors in organizations, occupational doctors, social insurance companies, and other researchers, and used their feedback to refine our conclusions.

Delegating a welfare state policy to private and non-profit organizations

Using the concept of organizational legitimacy, we apply the cognitive, moral and pragmatic dimensions to our case of delegating a public welfare policy to organizations across the three economic sectors: industry, non-profit services, and commercial services.

The cognitive dimension: understanding the core idea of the policy

The core idea of the national policy is that illness is no longer taken for granted as a legitimate reason for (full) absence from work. The delegated national policy includes the cognitive idea of treating illness as ‘behaviour’, with the possibility of doing (some hours of) work while being ill, for example, when recovering from a broken leg, a hip surgery or a mental disease. The vast majority of organizations in our study expresses their consent with the new cognitive frame. As reported by the senior HR manager of an international accounting firm: ‘Well, I think the approach is good: that you look at what someone can still do.’ (47). And a HR consultant of a hospital confirms: ‘Absenteeism is behaviour and can be influenced. If we say “come to work and try to do something again” or that we say “stay at home and feel comfortable”, that’s a big difference.’ (31). HR managers acknowledge that illness and return to work are issues to discuss between employee and employer, and support the idea in the national policy of individual organizations having an important role in dealing with the physical or mental complaints of their employees. A popular slogan to stress the behavioural aspect is ‘illness can happen, but absence is a choice.’ (28). HR managers also underscore the idea that employees should be encouraged to continue work during their illness, to be realized with ‘lighter’ tasks than usually, or by working less hours a day or week.

Interestingly, a few industrial companies claim that they had already embraced the core idea of the delegated policy and did not have to change much in their organization. As the consultant for labour affairs of a large industrial company expressed: ‘Our priority remained the same. I don’t believe our approach would be different without the new policy.’ (1). And a HR director of a firm in electronics confirms: ‘There are not really changes in our company. We have always been ahead of the law.’ (15). However, the majority of organizations in our study across the three sectors views the national policy idea as a reframing of their perspective on sickness absence. The organizations´ social responsibility for reintegration is perceived a cognitive change, as the health manager of a large catering company argues: ‘Compared to five years ago our culture has radically changed. If you have health complaints, it is not allowed anymore to stay home without communication. Openness and transparency have increased in the last few years, both for supervisors and employees.’ (35). This is confirmed by the Personnel Advisor of a Court of Justice, recalling: ‘When I started working here five years ago, the absenteeism rate was far too high and no absenteeism policy had yet been developed. In recent years, much has changed in this area and with success.’ (24).

Yet, agreeing with the cognitive idea can be based on different motives. One is that since the national policy shift, employees’ health is increasingly recognized as an organizational risk that companies should reduce. As the health and safety coordinator of a medium-sized industrial company argued for his organization: ‘The moment you have to pay yourself – and that was the big turnaround here – then you have an interest in keeping sick leave as short as possible.’ (11). This is also expressed by the reintegration coordinator of a non-profit organization (22) who views ‘money’ as the most important incentive for their change. The Head of HR department of a large electronical industry even believes: ‘The employer is only sensitive to one thing, namely the money.’ (17). Pragmatic reasoning thus underpins cognitive legitimacy here. Whereas financial reasons and efficiency arguments seem to correspond more closely to private firms’ profit-making, competition and survival; they can also be found in non-profit organizations, especially when confronted with austerity (e.g. 27; 24; 29; 31). Secondly, HR managers identify the national regulations for the return-to-work process in the Gatekeeper Act as a key trigger for changing their mindset; the law structured their procedures for return-to-work, and they use it to discipline their supervisors for lowering absence rates. Although this motive is mentioned by many organizations, the non-profit organizations more often stress the impact of legal procedures. A third motive for HR managers to align with the cognitive idea is the aim for being a preferred employer. This motive was merely found in organizations (e.g. 24; 38) competing for attracting highly educated employees (e.g. lawyers, consultants).

In general, the cognitive response is thus clearly supported by pragmatic reasoning, illustrating the interplay between the dimensions of cognitive and pragmatic legitimacy. Yet, the rationales underpinning the pragmatic legitimacy differ – where private organizations primarily focus on economic fitness, non-profit organizations more often stress the impact of national procedures. The difference may relate to cultural cognitive variations at the level of organizations, where private companies are used to consider financial consequences given the market context, while non-profit organizations are used to take up legal rules in a bureaucratic context. Our case thus suggests a two-way interplay between dimensions of cognitive and pragmatic legitimacy.

The moral dimension: identifying social responsibilities

The moral dimension of legitimacy points to how HR managers interpret the responsibilities embedded in the delegated national policy. Although the HR managers in our study agree with the cognitive dimension underlying the new policy, they doubt if they are morally responsible for the problem of sick leave and disability and can – or should – take full social responsibility. As the HR manager of a medium-sized company in logistics and transport declares: ‘The government blames too much down to the employer.’ (16). A health expert of a large industrial company believes that societal problems are strongly pushed to individual organizations: ‘the system of social security shows austerity, the idea of solidarity is decreasing. The government wants to shift all responsibility to business from a cost perspective.’ (1). The HR director of a company producing yachts argues: ‘I think that the government has transferred a lot of responsibility; as an employer, you cannot bear responsibility for everything. Especially when workers are getting older and therefore are more likely to get sick and unable to work.’ (5). And the HR director of a company in installation technology agrees, saying: ’I understand the position of the government, but absenteeism is often not the fault of the employer. And what can you do then?’ (8). HR managers thus accept the policy change at the cognitive level, but are not convinced of their moral responsibility for employees with (temporary) health problems.

This tension between the cognitive and moral dimension of legitimacy in interpreting the national policy might explain why HR managers tend to frame the organizational responsibility first and foremost as that the employee should take more actions. The normative shift in the national policy that organizations should play a key role in return-to-work cases of (long term) illness is thus primarily interpreted as a responsibility for employees. According to a HR manager of a law firm: ‘I believe that the national policy has increased awareness, not so much among employers, but on the part of the employee, in the sense of “Oh yeah, I can be absent, but I really have to do something to get back to work and that will take more commitment than I’d thought.’ (36). This was confirmed by the HR director of a firm in electronics: ‘For employees, the most important effect of the national policy was that they became aware of being also responsible and that the government no longer has to pay for everything.’ (15). As a large company in commercial services explains: ‘The national policy requires us to restructure our actions in sickness absence management. This means that the responsibility is now for the employee, who often remained passive in the past.’ (42). Also the labour consultant of a large firm in chemistry believes: ’more and more emphasis is placed on the employee’s own responsibility. Employees should pay much more attention to their own health.’ (1). HR managers, especially in the private sectors, often argue that the responsibility has been placed too much with the employer: ‘The employer must continually incur costs in order to please the employee and prevent sickness absence, while the employee does not take his own responsibility.’ (41). It thus seems that the moral intention of the national policy to increase the social responsibilities of organizations is interpreted as primarily the employee’s concern.

To stimulate employees rather than the organization in ‘taking responsibility’, HR managers feel supported by the sanctions in the delegated national policy for situations that sick-listed employees are reluctant in their view to participate in the reintegration process. According to an HR manager in logistics and transport: ‘If someone refuses to work on his reintegration, we can take measures now.’ (16). And an HR director in a firm for installation technology: ‘Now the employee should really do his best. In the extreme case, you now have the possibility to dismiss people if they do not cooperate with their reintegration.’ (8). The HR project manager of a large company for quality of life services explains: ‘The national policy makes it simpler to create support in an organization with 6,600 employees. Because things are mandatory, it is easier to get behavioural change and a sense of urgency.’ (51). And the HR manager of a law firm confirms: ’I was happy with the national policy, you could just tell your employees based on the legislation: ‘you have a duty’. […] Of course you could have pushed them yourself as an employer, but it would have taken much longer. Now you have the pressure by national regulation.’ (39). Moral legitimacy is thus interpreted by organizational actors as the individual’s responsibility, and strengthened by referring to macro institutional pressures.

A consequence of shifting moral responsibility to individual workers is that the organizations in our study merely focus on individual measures for return-to-work rather than acting upon a strategy based on the investigation and prevention of recurrent reasons for absence and disability. A health and safety coordinator of a company for storage systems with hard physical work was concerned that: ‘Working conditions and safety do not receive enough attention.’ (11). And the HR manager of a large Police department (29) emphasizes that preventive measures are still in their infancy, especially for preventing recurrent reasons for either physical or mental illness. This is confirmed by most other organizations across our study. As far as initiatives on prevention are taken in the organizations in our study, they are often one-directional, aimed at changing employees’ behaviour, e.g. by offering a choice for healthy food in the canteen, or by subsidizing sports. Some organizations have contracts with health insurers for a ‘company care package’, e.g. health checks, where employees should take follow up actions. The emphasis on healthy individual choices is accompanied by a new rhetoric in organizations. As the reintegration officer of a non-profit organization indicates, they now prefer words like ‘vitality’ and ‘health’ in their new policy rather than ‘sickness’ and ‘absenteeism’, and the reintegration officer is called ‘vitality coach.’ (25). Organizations thus tend to perceive the delegated national policy as an incentive to change the worker rather than the workplace. Hence, the moral content of the delegated national policy seems to clash with the normative claims put forward by the HR managers.

The pragmatic dimension: how to implement the policy

While the major reason for delegating the national policy to organizations was to improve labour market participation by preventing absence and retaining sick-listed employees, the financial incentives for employers are often viewed as a device to save short-term organizational costs in line with a pragmatic legitimacy. As one HR manager argued: ‘Since the national policy put the full cost of sick leave on to the employer, it became in our interest to restrict the absence period as far as possible.’ (11). The HR Director of a company in financial services stresses the dominance of pragmatic reasons and points to reverse effects: ‘I doubt if there is a vision. Ill workers cost money and thus the company wants to re-integrate them as soon as possible. [But] supervisors receive targets for reducing sick leave, which makes it harder to reintegrate people with labour restrictions than in the past.’ (33). Also, an HR manager in the printing industry (13) faces a tension between the aim for costs savings and for retaining sick-listed employees, while paradoxically, the Gatekeeper Act sometimes softens the road to dismissal. A few private organizations in our study even perceive sickness absence as a new option to make a profit. According to an HR director of an industrial company with almost 600 employees: ‘As an employer, we also have to make money and absence is an interesting area in this respect.’ (3). Although at the cognitive level, HR managers thus underline the national policy’s ambitions to improve the quality of sickness absence management and the reintegration of ill employees; they emphasize that the national policy requirements should be aligned with organizational interests based on pragmatic (financial) short-term reasons.

As a consequence, HR directors and managers across the study welcome the shift in the welfare state policy as an incentive to discipline their workers on sickness absence. HR managers frequently argue that particularly lower-skilled employees did not take enough responsibility for their return to work in the past. They perceive the delegated national policy as a support to increase control of employees’ sickness absence. Our study identified that the organizations convert the institutional pressures into different treatments for various categories of employees. For example, lower-skilled workers, such as administrative staff, cleaning personnel, or employees in the catering industry, are approached by control-and-sanctions, often based on scepticism about their health problems. According to an HR manager in commercial services: ‘I believe we should apply the rules more strictly, because employees call in sick too easily sometimes.’ (44). Higher-skilled workers, such as lawyers, university staff, doctors, and consultants, are treated in a much more subtle way, based on trust of the employees’ commitment to the organization. As an HR manager of a large private company explained: ‘The lower skilled people fit better with a control-model for sickness absence. For the higher educated employees we need a different approach while they have intrinsic motivation.’ (14). The selective treatment of various groups of ill employees is thus strongly related to how employees are perceived and categorized by especially private organizations, in contrast with applying equal treatment in implementing national regulations.

As a further consequence, and contrary to the aims of the delegated national policy for an inclusive labour market, the policy is interpreted as an incentive to select and test new workers for their health. Argued by an HR manager of a large firm in business services: ‘We now take someone from a temp agency to test whether he can handle the workload.’ (51). The delegated national policy thus seems to enhance the use of temporary jobs. Moreover, while at the national policy level the importance of labour market participation is emphasized, including jobs for people with labour restrictions because of (minor) health problems; most organizations in our study view employees’ health as an organizational risk and wish to reduce this risk in a context of market competition and/or public austerity. This means priority to adopting employees who are supposed to contribute to the organizations’ targets without any health problems. According to an HR manager of a catering firm, they only hire employees who are able to deal with the high workload: ‘If an employee has a problem with periods of extreme workload, it is hard to alter the workload pressure. We simply cannot use this employee.’(35).

Consequences of the national policy are not limited to an increase of temporary work to test and select new employees for their capacity to stand in the heat of the kitchen. HR managers in our study also point to the adverse effects of the push in the national policy to retain workers with limited physical or mental capacity; they believe that higher risks and responsibilities for individual organizations imply more, rather than less barriers, to employ workers with restricted work capacity. At first sight, the strict requirements seem to affect merely the lower educated workers. As the reintegration officer of a hospital tells: ‘We always try to adjust the job or find a new one within our organization, but for the lower skilled, like cleaning staff, this might be really difficult.’ (22). Mobility is even more difficult than in the past, due to more pressing performance targets for managers to reduce absence. As the HR manager of a printing company explains: ‘In the context of mobility, an employee is expected to be transferred internally every four to five years. But the pressure on productivity also puts pressure on this. Managers prefer not to have people with a disability.’ (13). Moreover, as an HR staff member of hospitality services explains, there is no budget for outplacement as an alternative: ‘Outplacement is being looked at, but this is more for higher positions. It is not possible to request outplacement for regular functions.’ (50). Pragmatic (financial) reasons thus are severe limits to take replacement seriously.

The impact of pragmatic legitimacy seems to vary mainly between low- and highly skilled employees, with the latter in a better position. As an HR manager in healthcare explains the context: ‘We notice that the possibilities for adapted/replacement work for low-skilled personnel are smaller than for highly skilled workers. This will become even worse as a result of automation.’ (22). Decreasing opportunities for re-integration, in particular for low-skilled employees, are confirmed by a multinational in industry: ‘re-employment in another position has become more difficult in recent years. Low-skilled jobs are disappearing due to outsourcing and automation.’ (7). Our data however show that similar problems arise with highly educated employees suffering from mental stress. The manager of a large firm in medical technology said about these workers: ‘Because of more mental strain, making up adjustments has become much more difficult.’ (14). This is confirmed by other respondents, for example, the head of occupational physicians of a large academic hospital (32). And a senior HR staff member of a large accounting firm argues: ‘Look, if your problem is that you can’t handle work pressure, then it’s very difficult to find a place where there is no work pressure anymore.’ (45). The Head of Employment Services of a large aircraft services company confirms that mental illness is a big problem, pointing out that more than a third of all cases of workers’ illness has a psychological cause (42). Also for employees with mental illness, options for re-integration have become more limited than in the past. A manager of a firm in medical technology argues that before the national policy change, his organization more often created new jobs to keep people with health restrictions on board. He argues: ‘Now the focus is clearly different. If an employee has problems with peak loads, it is difficult to adjust the work accordingly. Then there is no more room for this person in the organization.’ (14). Rather than realizing an inclusive labour market, the delegated national policy in a context of increasing competition and austerity seems to decrease the job opportunities for employees with labour restrictions. Paradoxically, the delegated national policy thus seems to work against its own aims to retain employees with minor labour restrictions in the labour force.

National statistics in 2002–2009 show that the labour participation of handicapped people did not increase and even decreased (in full-time equivalents), despite the growing labour participation on average and despite the national policy aiming for improvement (Jehoel-Gijsbers 2010). The national regulation (in particular the WIA Act for disability benefits) also limited access to social benefits. As a health expert of a large catering firm recognizes: ‘The assessment for the WIA means that more than half of our people who become unfit for work have less than 35 percent wage loss and are therefore not entitled to a benefit, while they really can no longer work. There is a lot of inequality in this regulation.’ (35). The Dutch Parliament had exactly the opposite aim when debating this law, namely an optimal involvement of the working population, with an emphasis on what people with restrictions still can do rather than stressing incapacity for work (Parliamentary papers 2004–2005; 30 034, no 3).

Discussion and conclusion

This article studied a rather invisible form of passing welfare state responsibility on to private and non-profit organizations and out of public scrutiny. Our study examined the organizational responses to the delegated welfare policy in a wide range of sectors, and suggests that complex institutional pressures are interpreted by local managers to implement specific changes in organizations. We found that crucial elements of the delegated national policy (e.g. collective prevention of absence, an equal treatment of lower and higher-skilled employees, and social inclusion of employees with labour restrictions) were not put into practice while being at odds with the organizations’ own perspective on legitimate behaviour. Instead, organizations interpret the national policy reform with calculations of short term, pragmatic interests. The organizations accept the new cognitive frame in the national policy where illness has changed from ‘condition’ to ‘behaviour’, but interpret it in moral terms as a responsibility of the employee rather than the organization, and actively connect it to prioritizing the self-interest of the organization (pragmatic legitimacy). The delegated social responsibility for an inclusive labour market is interpreted by the organizations as a new risk that should be managed, and where possible, avoided. For pragmatic reasons, organizations justify actions that serve the organizations’ interest over the employees, and tend to pass organizational responsibility for prevention and return-to-work on to sick-listed employees. Practising the delegated policy differently for diverse categories of workers is another pragmatic and selective interpretation with disciplinary effects on certain (lower skilled or mentally ill) workers. Based on our findings, we contribute to the literature in two ways.

First, it has been recognized in the literature that legitimacy can be built on many sources and stakeholders (see, for example, Deephouse et al. 2017). Yet, there is still a focus on responding to institutional pressures as if these pressures were clear and uncontested rather than multi-dimensional and competing (Pache and Santos 2010; Waldorff and Greenwood 2011). This is visible in criteria for studying organizational responses to legitimacy challenges, such as to: ‘demonstrate adequate performance’, ‘show fit with social values’, and ‘conform to meaning systems’ (Deephouse et al. 2017, 41 Table 1.1). These responses may be explained by using the concept of isomorphism (DiMaggio and Powell 1983), where organizations respond in similar ways to common coercive, normative and/or cognitive pressures. Yet, our findings demonstrate how institutional pressures for welfare state reform are understood and enacted by private and non-profit organizations, where core ideas of the delegated policy are modified and practices are selective. Especially the private companies in our study tended to favour pragmatic legitimacy in terms of cost reduction and risk avoidance as guidance in their policy implementation. Our findings confirm results of a recent study in Sweden, where ‘employers had difficulties in taking social responsibility for RTW [return to work], in that economic considerations regarding their business took precedence over legal and ethical considerations.’ (Seing et al. 2015, 1760). Similarly, we found that organizations applied procedures differently depending on how sick-listed workers were valued related to their business. Thus, it is in the local process of implementing change that we discover which elements of the delegated welfare policy are pursued and which are not, or even reversely implemented (Edelman, Uggen, and Erlanger 1999).

While institutional theorists increasingly recognize the ambiguous, conflicting and evolving nature of social standards for appropriate behaviour (Greenwood et al. 2010; Thornton, Ocasio, and Lounsbury 2012); many barely focus on the internal considerations of organizations when interpreting and applying the often multi-interpretable standards. Among a few recent exceptions (e.g. Pache and Santos 2013; Radaelli and Sitton-Kent 2016; Waldorff 2013), a case study of the German hospital sector (Bode, Lange, and Märker 2017) showed how an ‘organized ambivalence’ between a welfare and a market orientation in shaping the institutional context was affecting the sensemaking of key organizational actors. We highlight how implementing macro institutional pressures into organizational practices is active institutional work (Lawrence, Suddaby, and Leca 2009) with disciplinary effects on certain actors. Yet, in contrast to an overly ‘agent-centred’ focus in the institutional work approach (see for criticism, e.g. Kaghan and Lounsbury 2011; Landau, Drori, and Terjesen 2014; Willmott 2011), our study also shows that local actors embedded in their specific organizational context, have limited agency in applying new ideas or policies. We demonstrate that an agent has agency to interpret and implement a policy but this agency is more aligned with policy selectivity than with a comprehensive implementation of the policy. This study shows how a common organizational quest for economizing and control, related to private organizations’ economic survival and non-profit organizations’ protection in a competitive and insecure environment is dominant over the welfare state aims for increasing labour market participation, social inclusion and equal treatment. However, what seems to be similar organizational responses to a new national reform, are in fact based upon different rationales for legitimacy.

Second, we show that the frictions between the cognitive, moral and pragmatic legitimacy in implementing institutional pressures stress the discrepancy between the public policy’s wider societal aims and how organizations interpret and enact the policy. While literature has emphasized the moral and cognitive dimension, i.e. specifying problems and justifying solutions as crucial in the process of implementation (Deephouse et al. 2017; Greenwood., Suddaby, and Hinings 2002); our study emphasizes the importance of pragmatic legitimacy. We explain that enacting welfare state policies by individual organizations is a far more pragmatic process of interpretations and actions that meet specific organizational interests. Using the cognitive, moral and pragmatic dimensions of the concept of organizational legitimacy, we could thus provide a more comprehensive explanation of how and why a national transformation of welfare state policy is framed and connected to organizational interests and activities, responding to an enduring request (Bitektine and Haack 2015; Powell and Colyvas 2008; Zilber 2016). In addition, our findings highlight the interlinkages between the three dimensions by showing that this pragmatic economic reasoning is justified by specific moral and cognitive interpretations of national regulations (Suchman 1995; Suddaby, Bitektine, and Haack 2016). For example, the cognitive interpretation of illness as behaviour and the moral interpretation of the employee as key actor rather than the organization is used to justify the disciplinary action of intensified control of ill workers, in particular of lower-skilled employees. In the light of the request for a more complete evaluation of the concept of organizational legitimacy (Suchman 1995; Suddaby, Bitektine, and Haack 2016), we thus provide a fuller picture by underlying the importance of pragmatic legitimacy and showing the relative impact and interplay of the three dimensions in organizational responses to institutional complexity.

Practical implications

This study draws attention to the important issue of selectivity in policy implementation where agents are not in a position to fully commit to all dimensions and expectations of the policy. Showing the paradoxical nature of agentic capacities in the context of delegation of welfare state responsibilities, the concept of legitimacy helped to understand what is valued in implementation, and how selectivity in policy implementation is motivated. Our study suggests – cf Seing et al. (2015) – a hegemony of economic rationalities in delegated welfare policies for sick leave. However, if agents perform institutional work it may also be done to attenuate the predominance of the firms’ economic interests. Taking social responsibility in delegated welfare policies thus requires the commitment of strategic management in organizations for a more comprehensive policy, as well as more collective and effective support of the wider institutional environment.

Our findings highlight that the ‘message’ of a delegated national policy to individual organizations is open to other interpretations than a focus on an inclusive labour market and saving societal costs by prevention and reintegration. There has been renewed debate in the Netherlands and other European welfare states about national policies for sick leave and disability, stressing responsibilities of both the state and the profit and non-profit sectors (European Commission 2016). In December 2018, the Dutch government responded to a persistent request of the Association for SMEs with a proposal to limit employers’ obligations for sick leave expenses to 1 year and to cover financial risks and return-to-work obligations in the second year by private insurance companies (only for organizations with <25 employees). Beside lowering financial pressures for small organizations, there is however a growing urgency for a more strategic agenda for the wider group of organizations, to control the growing annual inflow of disability cases and reduce the enduring barriers for employees with minor labour restrictions to entry the labour market. After 2010, their situation has been further deteriorated during the economic crisis, and is not improved afterwards. According to the last national evaluation: ‘Despite the reintegration policy, trends in the labour participation of people with health impairments remain a concern. While the policy is aimed at creating an inclusive labour market, the participation figures in recent years point to a situation that is more akin to “survival of the fittest”.’ (Versantvoort and van Echtelt 2016, 163). To stimulate a positive contribution of organizations to an inclusive labour market and the reshaping of work settings in favour of prevention and employee well-being, it can be helpful to develop institutional pressures for more collective actions by organizations, and focused on long-term solutions (Quick, Macik-Frey, and Cooper 2007).

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Notes on contributors

Nicolette Gestel

Nicolette van Gestel is Professor of New Modes of Governance, and Academic Director of the Executive Master in Public and Non-profit Management at TIAS School for Business and Society, Tilburg University, The Netherlands. She is an academic member of the Social and Economic Council of the Dutch government, and visiting professor at Oslo Metropolitan University. Her research interests include public sector reform, decision-making in plural (network) settings, and the changing role and position of professionals in the welfare state. She currently works on the EU Horizon2020 project Co Production and Co Governance: Strategic Management, Public Value and Co creation in the Renewal of Public Agencies across Europe (COGOV). Recent papers have been published in Public Administration, Organization Studies, Human Resource Management, Personnel Review, BMC Health Services Research, Public Money & Management, European Journal of Social Work, Public Management Review, Scandinavian Journal of Management, and Perspectives on Public Management and Governance.

Susanne Boch Waldorff

Susanne Boch Waldorff is Associate Professor at the Department of Organization, Copenhagen Business School, Denmark, Program Manager of the Master of Public Governance, and Director of the Center for Health Management at CBS. She is member of the Center for Public Organization, Value and Innovation (POVI), and visiting scholar at TIAS School for Business and Society, Tilburg University, The Netherlands. Her research interests include the translation of reform and policy into organizational practices, public sector management, innovation, and professions. Drawing on institutional theory and organizational analysis, she published in the Academy of Management Journal, Scandinavian Journal of Management, Research in the Sociology of Organizations, Journal of Change Management, The Innovation Journal: The Public Sector Innovation Journal, BMC Health Services Research, and Critical Public Health.

Jean-Louis Denis

Jean-Louis Denis is Professor of Health Policy and Management at the School of Public Health, Université de Montréal and holds the Canada Research Chair on health system design and adaptation. He is senior scientist at the Research Center of the CHUM (CRCHUM) and visiting professor at the Department of Management, King’s College London. His current research focus on the role of medical doctors in health system change and improvement, on transformative capacities of health systems in Canada and in Europe, and on the development of learning health systems in academic health centers. Recent papers have been published in the Journal of Health Politics, Policy and Law, Organization Science, Academy of Management Annals, Milbank Quarterly, Administration and Society, Implementation Science, Journal of Public Administration Research and Theory and Perspectives on Public Management and Governance. He is co-editor of the Palgrave series on Organizational Behaviours in Health Care, and Associate Editor of BMC Health Services Research.