4. Challenges in promoting effective diversity policies

Currently, the evidence base on ‘what works’ is still relatively small and the success of programmes usually differ across different national contexts and groups targeted. Gathering data on diversity in the workplace is the first step in assessing whether diversity policies are effective. However, in most countries, the collection of data on diversity is limited (see Table 4.1). Most countries that have introduced requirements for firms to monitor composition and pay of their workforce have done this for data collection on gender, age and, to some extent, disability.

Data collection is most advanced for monitoring gender imbalances. In Sweden, for example, firms with 10 employees or more are obliged to provide annual reports on pay differences between men and women in similar positions (so-called pay audits). Firms with more than 25 employees have to provide a wage action plan. Similar schemes are in place in Finland and Denmark. In Austria, companies with 150 employers or more also have to provide similar reports, but they are not publically available and need to be kept within the company.

Generally, the availability of employee data on ethnic or migrant background is limited. Only in few countries are employers required to provide data on the nationality of employees. The Netherlands used to monitor ethnic diversity in the workplace under the so-called Wet SAMEN (1998-2003), which required firms with 35 employees or more to provide annual reports on the share of ethnic minorities in the firm (based on country of birth and parental country of birth, excluding Europe), their job levels and diversity policies the firm was implementing.

Overall, the limited evidence renders it difficult to assess whether i) some sectors might be more diverse than others, ii) how women and minorities are faring within companies, iii) whether diversity policies have an impact at the firm level, and iv) whether their effectiveness differs across groups. In addition, results from the HR survey show that only around 50% of firms have assessed the impact of the diversity policies they implemented. Furthermore, among those who had experienced difficulties in implementing diversity policies, 45% stated that measuring their impact was one of the main challenges.

In this context, ‘knowing what works’ and particularly getting a better understanding of how the same policy may affect groups differently, remains a key challenge. At the same time, improving data collection on diversity in the workplace clearly needs to take into consideration concerns around privacy and data protection as well as national legislation around what kind of sensitive data on diversity can be collected. In a context that allows the collection of diversity statistics, this implies that employees cannot be required to provide such data and that privacy is ensured, either through anonymous surveys or – in case data is linked to individual employees – by restricting access to these data only to staff tasked with monitoring.

Voluntary commitment and active engagement by employers are critical to ensure that diversity policies gain traction, go beyond window dressing and have an actual impact at the firm level. Both in the private and the public sector, employers can take a number of actions that render diversity policies more effective. By their nature, however, such policies rely on the goodwill of companies and therefore can run the risk of not being implemented widely, e.g. when diversity policies are generally not that widespread or well known or when they are seen as controversial. Therefore, it is crucial to determine when legislative frameworks are needed that regulate firm action.

Finding the right balance between voluntary commitments and mandatory policies is a key challenge because it not only requires the ‘right mix’ of bottom-up and top-down approaches, but also because mandatory policies may have different impacts depending on the type of diversity measure and the group concerned.

For example, quotas for women on boards in Europe were generally found to be effective given the rather strict sanctions that were attached with it, such as having to leave positions vacant. At the same time, the impact of quotas for people with disabilities has been mixed. This is partly due to how they were implemented, e.g. creating additional costs for employers or allowing firms to pay relatively minor fines in case of non-compliance. Thus, the impact of mandatory quotas varies according to groups and strongly depends on how the law is implemented and what sanctions are foreseen.

Furthermore, the impact of diversity policies can also vary across firms. Anonymous application procedures, for example, may be beneficial if there is discrimination in the hiring process. However, in companies that may prefer diverse candidates, anonymous applications can have the opposite result; by redacting candidates’ names, age or address, employers are prevented from favouring under-represented applicants when credentials are equal. This shows that implementing such procedures in every employment context is unlikely to have a positive impact, yet it also raises the question of how to determine in which firms such policies would be helpful.

Given that the impact of mandatory policies can differ across groups, firms and types of policies, there cannot be one regulatory policy framework applicable in every context. Therefore, providing general recommendations for when legislative frameworks are needed to regulate firm action is difficult, as such policies are highly context-dependent.

Another issue concerns the design of policies – while evidence points to the importance of closely targeting them to specific barriers faced by women and minorities, there is a limit to how fine-grained these approaches can realistically be.

The group-based approach to diversity and equality, while being the standard approach in OECD countries, may also run the risk of neglecting individuals that happen to fall outside of the groups identified as “diverse”. Another issue to consider is that identifying certain groups that are “in need of support” can perpetuate negative stigmas regarding the abilities of women and minorities, which in turn can aggravate their exclusion. For example, research has shown that subsidies which reveal a disability status to potential employers can have a stigmatising effect with regard to their productivity (Deuchert and Kauer, 2017[1]). Similarly, there is a possibility that women in firms with gender quotas are seen as less qualified and only hired because of their gender.

Despite these caveats, targeting diversity policies to group-specific hurdles and needs can still constitute a useful and oftentimes pragmatic policy approach. Considering that a given policy instrument can create different outcomes depending on the group and may even have adverse effects for some, a one-size-fits-all approach may not always work in practice. For example, flexible work arrangements that are only taken up by women because this policy did not include sufficient incentives for male employees may have adverse effects on the long run, e.g. regarding promotions and career advancement.

In addition, while policies would ideally be designed on the granular level, accounting for individual barriers as much as possible, it is clear that targeting is only possible to a certain degree, as such an approach is costly and inefficient. Policies cannot possibly account for all sorts of combinations of disadvantage and may become too narrow to reach a broad audience. Some group-based aggregation, therefore, presents are more efficient and pragmatic approach. Yet striking the right balance can be a challenge; the exact choice of whom to include is debatable and likely depends on the national policy context. Stakeholder engagement is critical to gain a better understanding of the challenges faced by different groups and should therefore systematically include civil society organisations and social partners in the consultation and evaluation process.

A common criticism is that diversity policies are often adopted as a means for image management and branding, with little regard for their actual impact.1 The term “empty shell policy” has been coined by Hoque and Noon (2004[2]) to describe the phenomenon that companies have diversity policies on paper, but do not implement them in practice. In an analysis based on the UK Workplace Employee Relations Survey, which is representative by industry and workplaces with 10 or more employees, they observe a gap between espoused, formal written policy and actual practice. Less than half of workplaces with diversity policies in place adopt corresponding supporting practices, e.g. parental leave, nursery, switching from full-time to part-time with regard to women. Moreover, employees often seem to have limited access where such practices do exist and importantly, this access appears to depend on their position in firm hierarchy.

These results are particularly concerning when viewed in light of the finding that the mere existence of diversity structures can induce high-status group members to become less sensitive to unfair treatment. In an experimental study, Kaiser et al. (2013[3]) find that the presence of organisational diversity structures causes high-status group members (White, male employees) to perceive organisations with diversity structures as procedurally fairer environments for underrepresented groups, even when in the experiment it is made clear that underrepresented groups have been unfairly disadvantaged. Managers and high-status employees not only become less sensitive to discrimination of underrepresented groups, but also react more harshly toward underrepresented group members who claim to have been discriminated against. Similar effects are found for judges who are more likely to rule leniently on discrimination cases when companies have diversity policies in place (Edelman et al., 2011[4]). Thus, diversity policies, no matter if mere window-dressing or not, can serve as a legal strategy for firms to shield themselves against claims of discrimination. In addition, assuming a ‘good faith’ effort also affects women and minorities themselves. In studies with fictional descriptions of firms and law suits, the mere existence of diversity policies in companies decreases women’s support for anti-discrimination legislation and their ability to detect sexist behaviour and lead ethnic minorities to assess complaints of discrimination more leniently (Brady et al., 2015[5]; Dover, Major and Kaiser, 2013[6]).

Thus, empty shell policies are not only ineffective in addressing diversity issues, but they may also have a detrimental impact; they can reduce the ability to detect and address persisting issues because the mere existence of diversity policies functions as a reassurance that enough is being done.

While little can be done to change a firm’s ‘true’ motivation, policy design can address a – possibly accidental – implementation of empty shell policies. This includes drawing up comprehensive plans for recruitment, retention and career development with specific, measurable objectives for each step. Such diversity plan are common in the public administrations of many OECD countries (Box 4.1). Monitoring progress over time as well as allocating responsibility for implementation and accountability for outcomes are other critical components to ensure that diversity approaches are more effective (Dobbin and Kalev, 2017[7]).

Evidence on the Dutch Wet SAMEN law has shown that, when employers were required to monitor their workforce and report on measures taken, it was associated with an increase in employment of the target groups (OECD, 2008[8]). This suggests that a mere requirement to report on measures and outcomes can be very effective, as it requires employers to reflect on the issue. Lastly, strengthening internal networks and empowering employees, e.g. through staff networks and mentoring programmes, can support a more bottom-up approach to improve accountability.

Diversity policies, and particularly more elaborate diversity strategies, are usually more likely to be implemented by large companies due to larger recruitment needs, bigger HR departments and possibly a stronger interest in branding than it would be the case among small and medium-sized enterprises (SMEs, i.e. companies with under 250 employers). Thus, for SMEs some diversity policies may not be feasible to implement, while for other, less costly approaches, the perception may still be that their implementation is too time-consuming and expensive.

Despite a large selection effect of firms that already have experiences with implementing diversity policies, results from the HR survey indicate that such policies are less common in smaller companies. Around one in three companies with less than 250 employees had implemented diversity policies, compared to around 45% in firms with 250-500 employers and 65% in firms with more than 500 employees. In addition, while around 40% of large companies stated that the topic has gained considerably in importance for them in the past five years, this was only the case among in one in five companies with less than 50 employees.

Despite these obstacles, promoting policies in SMEs is critical; in the OECD, they account for 99% of all firms and around 70% of jobs on average (OECD, 2017[10]). Furthermore, evidence suggests that discrimination tends to be most pronounced in small- and medium-sized companies (SMEs). Possible explanations for this are that these companies face higher stakes when hiring one staff member with an uncertain productivity level, or that SMEs have less experience with diverse workers and are therefore more likely to hold negative stereotypes (Heath, Liebig and Simon, 2013[11]).

While some measures such as outreach campaigns may not be feasible or useful for SMEs, particularly for small and micro enterprises, most diversity policies can in fact be implemented relatively easily as they do not require the introduction of an entirely new policy, but rather an adaptation of existing measures.

For example, recruitment procedures can be adapted relatively cheaply, e.g. by including statements in job descriptions encouraging under-represented groups to apply, not asking candidates for photographs or ensuring more diverse selection committees in interview panels. Furthermore, external partners are critical to support SMEs in implementing diversity policies that require more know-how or larger HR capacity. These can include employers’ associations, regional or city-level governments and public employment services as well as NGOs and civil society initiatives, providing, for instance, (online) information and trainings or helping with increasing the talent pool through outreach activities. Such efforts were implemented most strikingly in the Flemish policy of developing so called “Diversity Plans” with SME employers. Support was provided in co-operation with social partners: employer organisations established a “Diversity Service Point” delivering made-to-measure services to support diversity management in SMEs. Trade unions deployed “diversity consultants” who supported staff seeking to establish a diversity policy in their company and who worked closely with SME employers, advising them on how to recruit and make most of diverse workforce. Moreover, SMEs that sought to develop such “Diversity Plans” could request financial support from the Flemish Department of Work and Social Economy (OECD, 2016[12]; van de Voorde and de Bruijn, 2010[13]).

Addressing pushback against diversity policies and – possibly by extension – negative attitudes towards those targeted by these policies, is a critical step to ensure that diversity can be implemented effectively.

Seeing that in the EU only around one in three would support diversity training, monitoring of recruitment or monitoring of workforce composition in their workforce (Eurobarometer, 2015[14]), both policy makers and employees may have to anticipate adverse reactions when implementing diversity policies. In addition, research suggests that particularly among ‘high-status’ groups, zero sum beliefs, i.e. the idea that gains of one group necessarily come at the expense of another, are particularly prevalent and can create a sense of losing out or being treated unfairly (Wilkins et al., 2014[15]; Kidder et al., 2004[16]).

Therefore, proactive communication and an adequate framing of diversity is critical to mitigate negative reactions. Framing diversity policies as a means to avoid government sanctions or lawsuits, for example, has shown to lead to more negative reactions than framing diversity policies as part of a management approach that is implemented for business reasons (Kidder et al., 2004[16]). At the firm level, communication around the rationale for diversity policies should therefore emphasise that such policies are not introduced at the expense of other employees and highlight how these are beneficial to the firm overall. Executive leadership is important in this regard, also to highlight the business case. However, a mere top-down approach may also create discontent, particularly when diversity policies are then perceived as a something externally prescribed where staff had little opportunity to provide input. Thus, including staff in the development of such policies, providing opportunities for feedback and listening to concerns may help prevent negative reactions later on.

Proactive communication about the rationale of diversity policies is not only important for firms, but also on a public policy level. Framing diversity policies as one part of social policy, alongside with policies improving access to education, training and jobs, may help to steer a discussion away from ‘preferential treatment’ and instead move towards a discussion around increasing equality of opportunity for all.


[5] Brady, L. et al. (2015), “It’s fair for us: Diversity structures cause women to legitimize discrimination”, Journal of Experimental Social Psychology, Vol. 57, pp. 100-110.

[1] Deuchert, E. and L. Kauer (2017), “Hiring subsidies for people with a disability: Evidence from a small-scale social field experiment”, International Labour Review, Vol. 156/2, pp. 269-285.

[7] Dobbin, F. and A. Kalev (2017), “Are Diversity Programs Merely Ceremonial? Evidence-Free Institutionalization”, in The Sage Handbook of Organizational Institutionalism, Sage, London.

[6] Dover, T., B. Major and C. Kaiser (2013), “Diversity initiatives, status, and system-justifying beliefs: When and how diversity efforts de-legitimize discrimination claims”, Group Processes & Intergroup Relations, Vol. XX/X, pp. 1-9.

[4] Edelman, L. et al. (2011), “When Organizations Rule: Judicial Deference to Institutionalized Employment Structures”, American Journal of Sociology, Vol. 117/3, pp. 888-954.

[14] Eurobarometer (2015), Eurobarometer 83.4: Climate Change, Biodiversity, and Discrimination of Minority Groups, May-June 2015, European Commission.

[11] Heath, Liebig and Simon (2013), Discrimination against immigrants – measurement, incidence and policy instruments, OECD Publishing, Paris, https://dx.doi.org/10.1787/migr_outlook-2013-en.

[2] Hoque, K. and M. Noon (2004), “Equal Opportunities Policy and Practice in Britain: Evaluating the "Empty Shell" Hypothesis”, Work, Employment and Society 3, pp. 481-506.

[3] Kaiser, C. et al. (2013), “Presumed Fair: Ironic Effects of Organizational Diversity Structures”, Journal of Personality and Social Psychology, Vol. 104/3, pp. 504-519.

[16] Kidder, D. et al. (2004), “Backlash towards diversity initiatives: examining the impact of diversity program justification, personal and group outcomes”, International Journal of Conflict Management, Vol. 15/1, pp. 77-102.

[10] OECD (2017), Entrepreneurship at a Glance 2017, OECD Publishing, Paris, https://dx.doi.org/10.1787/entrepreneur_aag-2017-en.

[12] OECD (2016), Working Together: Skills and Labour Market Integration of Immigrants and their Children in Sweden, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264257382-en.

[9] OECD (2011), Public Servants as Partners for Growth. Toward a Stronger, Leaner and More Equitable Workforce, OECD Publishing, https://doi.org/10.1787/9789264166707-en.

[8] OECD (2008), Jobs for Immigrants (Vol. 2): Labour Market Integration in Belgium, France, the Netherlands and Portugal, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264055605-en.

[13] van de Voorde, M. and H. de Bruijn (2010), “Mainstreaming the Flemish Employment Equity and Diversity Policy”, in Equal Opportunities?: The Labour Market Integration of the Children of Immigrants, OECD Publishing, Paris, https://dx.doi.org/10.1787/9789264086395-10-en.

[15] Wilkins, C. et al. (2014), “You can win but I can’t lose: Bias against high-status groups increases their zero-sum beliefs about discrimination”, Journal of Experimental Social Psychology, Vol. 57, pp. 1-14.


← 1. Similar criticisms have been launched against ‘greenwashing’ policies that seek to brand a company as environmentally friendly to attract customers, whilst having little or no positive environmental impact.

Metadata, Legal and Rights

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area. Extracts from publications may be subject to additional disclaimers, which are set out in the complete version of the publication, available at the link provided.


© OECD 2020

The use of this work, whether digital or print, is governed by the Terms and Conditions to be found at http://www.oecd.org/termsandconditions.