EUROPEAN COMMISSION

Brussels, 4.3.2021

SWD(2021) 41 final

COMMISSION STAFF WORKING DOCUMENT

IMPACT ASSESSMENT

Accompanying the document

Proposal for a
DIRECTIVE OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL

to strengthen the application of the principle of equal pay for equal work or work of equal value between men and women through pay transparency and enforcement mechanisms

{COM(2021) 93 final} - {SEC(2021) 101 final} - {SWD(2021) 42 final}

Contents

Glossary

1.Introduction: Political and legal context

2.Problem definition

2.1.What is the problem?

2.2.What are the problem drivers?

2.3.Why is it a problem?

2.4.How will the problem evolve?

3.Why should the EU act?

3.1.Legal basis

3.2.Subsidiarity: Necessity of EU action

3.3.Subsidiarity: Added value of EU action

4.Objectives: What is to be achieved?

4.1.General objectives

4.2.Specific objectives

4.3.What is the baseline from which options are assessed?

5.SME test

6.Description of the policy options for EU level action

6.1.Options discarded at an early stage

6.2.Option 0: Status Quo

6.3.Option 1: Legislative action to create transparency for individual workers

6.4.Option 2: Legislative action to create transparency at employer level

6.5. Option 3: Legislative action to facilitate the application of and enforce the existing legal framework

6.6. Overview of the policy options

7.What are the impacts of the policy options?

7.1.Option 0: Status Quo

7.2.Option 1: Legislative action to create transparency for individual workers

7.3.Option 2: Legislative action to create transparency at employer level

7.4. Option 3: Legislative action to facilitate and enforce the existing legal framework

8.How do the options compare / preferred option

8.1.Effectiveness

8.2.Efficiency

8.3.Coherence

8.4. Comparison of options

8.5.Possible combination of options

8.6.Preferred Package

8.7.Hypothetical scenario of the potential economic impact of the preferred package of measures

8.8.Choice of legal instrument

9.How will actual impacts be monitored and evaluated?

Bibliography

Annexes

Annex 1: Procedural information

Annex 2: Stakeholder consultation

Annex 3: Who is affected and how?

Annex 4: Analytical methods

Annex 5: Initiative specific annex

Annex 6: Intervention logic

Annex 7: Subsidiarity grid

Annex 8: Job evaluation systems

Glossary

1.Introduction: Political and legal context

The right to equal pay between women and men for equal work or work of equal value has been a founding principle of the European Union since the Treaty of Rome in 1957. The requirement to ensure equal pay is set out in Article 157 TFEU and in Directive 2006/54/EC 1 (the ‘Recast Directive’), as complemented in 2014 by a Commission Recommendation on Pay Transparency 2 (the ‘2014 Recommendation’). Despite this legal framework, the effective implementation and enforcement of this principle in practice remains a major challenge in the European Union. In 2019, the European Committee of Social Rights identified the lack of pay transparency as one of the key obstacles to the implementation of the principle of equal pay 3 .

The European Parliament has repeatedly called for more action at EU level to enhance the application of the equal pay provisions 4 . Also the Council has asked for action both on the side of Member States and on the side of the Commission. In June 2019, the Council called on the Commission to develop concrete measures to increase pay transparency.

The European Pillar of social rights includes gender equality and the right to equal pay among its 20 principles 5 . The EU Action Plan for 2017-2019 on tackling the gender pay gap 6 specified that the Commission will assess the opportunities for improving pay transparency. Following European Commission’s President von der Leyen’s announcement in her political guidelines 7 to introduce binding pay transparency measures, the Commission reaffirmed its commitment to present this initiative in the Gender Equality Strategy 2020-2025 8 .

The present initiative follows on the Commission’s evaluation 9 of the relevant legal provisions (the ‘2020 evaluation’) and previous Commission work 10 . These assessments concluded that there was limited progress on enforcing the right to equal pay in Member States and evidenced, in particular, vastly distinct and largely inefficient systems operating in most Member States – making equal pay an inert legal provision. The evaluation pointed to a number of problems deterring victims of pay discrimination from enforcing their right and deemed to require further action: despite the 2014 Recommendation, a very limited number of countries put forward measures to improve transparency in pay setting systems within organisations; key legal definitions and concepts are not applied uniformly in practice and are insufficiently implemented across national legislations; victims have difficulties to claim their rights.

This initiative aims at tackling the persisting inadequate implementation and enforcement of the fundamental right to equal pay and ensuring the respect of this right across the EU by establishing transparency on pay. It is part of a broader package of measures and initiatives that focus on tackling the root causes of the gender pay gap, such as the adoption and implementation of the Work-Life Balance Directive 2019/1158, sectoral initiatives fighting stereotypes and ensuring better gender balance, and the proposed Directive on improving gender balance on company boards of large EU listed companies 11 . In addition, the initiative is coherent with the initiative aimed at increasing reporting by companies of relevant non-financial information. 12 It is also consistent with and supported by the EU Minimum Wage initiative 13 and the sustainable corporate governance initiative 14 .

While addressing the implementation of the principle of equal pay through pay transparency, this initiative takes into account different features of national social dialogue and collective bargaining systems as well as the autonomy of social partners and their contractual freedom. It also recognises the important role of social partners in addressing gender pay inequalities, including through pay transparency.

2.Problem definition

2.1.What is the problem?

The problem tackled in this initiative is the failure to realise in the European Union the fundamental right to equal pay for the same work or work of equal value, despite this right being enshrined in EU law for more than 60 years. The information and data presented in the 2020 evaluation 15 and in this chapter show that this failure persists 16 . 

The failure to realise the right to equal pay means that women and men may still be discriminated either directly or on the basis of pay structures that do not value the work of women and men equally, i.e. according to objective and gender-neutral criteria.

In many cases, discrimination or bias is the result not of deliberate discriminatory behaviour but rather of a failure to grasp, both by workers and employers, what it means in practice to pay men and women equally for the same work or for work of equal value.  17 This might not be straightforward, especially when it comes to ‘work of equal value’ since it must first be assessed what is the value of the jobs concerned and second whether the pay received by men and women for jobs of same value is not discriminatory on the basis of sex 18 . It is important to stress at the outset that under the current state of the law in the EU, the concept of ‘work of equal value’ only applies to different occupations defined as of equal value in the same organisation.

A number of legal cases illustrate the matter at stake. For instance, in a landmark case in Belgium regarding equal pay for the same work, a female researcher in the European Trade Union Institute did not receive the same automatic promotion going with seniority as her male counterparts. The Labour Court of Appeal of Brussels found the employer’s pay system opaque and, referring to the CJEU’s decision in Case 109/88 Danfoss 19 , concluded that there had been gender-based pay discrimination. 20

In another case, the French Supreme Court was faced with an equal pay claim from a female mushroom packer comparing her work with more highly paid male packers 21 . The Court noted that it was clear that women packers were systematically paid less than their male equivalents and that the employer could not produce any objective reasons for such difference between men and women doing the same work.

A UK landmark case from 2018 concerned the right to equal pay for work of equal value. In this case, about 30,000 claimants, mostly women working in ASDA supermarkets as shop-floor staff, brought equal pay claims against their employer on the basis of comparisons with the pay of male workers employed at depots as part of ASDA's distribution operation. The Court of Appeal of England and Wales held that the women were discriminated as the work of the shop-floor staff should be considered of equal value as the work in the distribution center. 22  

It was only possible for the claimants in the above cases to claim their right because they had the necessary information on the average pay levels of their male colleagues and because, in the UK case, they were able to group their 30,000 claims. Landmark cases like these only show a small top of a possible iceberg. Many women in the EU having salaries below those of their male colleagues doing the same work or work of equal value, do not necessarily know about it as they do not have the necessary information about average pay levels. This means they cannot enforce their right to equal pay. Lack of transparency thus hinders uncovering pay discrimination and allows discriminatory pay structures to remain in place.

Gender pay discrimination and the gender pay gap

Individual pay discrimination and systemic bias in pay structures are only one of the root causes 23 of the gender pay gap, besides other causes such as horizontal and vertical segregation. The gender pay gap accumulates to and remains at 14% in the European Union overall. 24

From a policy perspective, it would be useful to know in which sectors/occupations/types of employment gender-based pay discrimination is more widespread or whether pay discrimination affects some groups of workers more than others (e.g. among specific contract types or age groups). However, precisely quantifying the overall size of gender-based pay discrimination at that level is not possible as of today.

Indications on the existence of the problem of pay discrimination and gender bias in pay settings can be derived from available EU statistics on the gender pay gap at aggregate national level (e.g. the Structure of Earnings survey (SES) or other national data) and from the analysis of the related SES micro-data.

The ‘unadjusted’ 25 gender pay gap can be decomposed into two parts. A first component, the ‘statistically explained’ part, is the gap between male and female average earnings due to the differences in the average characteristics (sector of activity, age, occupation, full time versus part time, etc.) 26 of male and female employees.

The second, residual component measures the difference between the financial returns to men and women with identical characteristics. It depends on any other factor for which no observable variable was available, measurable or omitted from the analysis, such as pay discrimination. This second, ‘unexplained part’ includes gender pay discrimination and accounts for 2/3 of the gender pay gap in the EU Member States. 27 It is also the highest portion of the national pay gap in 21 out of 24 countries analysed in the support study 28 .

The same study shows that the relevance of the unexplained part is confirmed also at sectoral level. The unexplained part, including pay discrimination, is the largest factor responsible for gender pay differences at sectoral level. It is more important than the ‘occupation’ factor, which is the second largest cause. In other words, even when excluding the statistical impact on the difference in pay related to working in different sectors and occupations (i.e. women work in sectors and occupations which are generally less paid), the analysis of pay gap data still points to a potential problem with respect of equal pay (see Annex 5.2). This finding does not necessarily apply to all employers. The only way to ascertain whether pay discrimination actually exists is to recalculate the unexplained gender pay gap at organisational/employer level and verify whether these differences subsist and why.

Moreover, as explained above, under the current case law of the Court of Justice of the European Union, workers can enforce their right to equal pay only in comparison to workers within the establishment in which they work or, beyond such establishment, when the wage setting is derived from a single source (e.g. a parent company in a corporate group) setting the working conditions including pay. 29 As a result, the occurrence and the extent of pay discrimination can only be measured at employer level by comparing categories of workers doing the same work or work of equal value as defined by a given employer.

Other relevant indications of possible pay discrimination to be further explored at employer level, which the statistical analysis from the support study evidences at aggregate level, include:

·Pay penalties in hourly pay 30 for part-time jobs (compared to full-time). Since part-time jobs are predominantly held by women 31 , this indicates indirect discrimination: even if penalties apply in a gender-neutral way, in practice they put women at a particular disadvantage as regards pay. At employer level, it should be checked whether this difference is justified by some factor that cannot be captured at aggregate level (e.g. in case part-time is coupled with exemption from some specific tasks performed instead by the full-time employees in the same category).

·The pay gap is higher in managerial occupations and among graduate workers. The reason for this may be that women in such positions are not aware of the value of their work and do not request, negotiate or obtain their wages in the same way as men do. The peaks in high-pay occupation statistics highlight how much wage negotiation processes influence the gendered pay. At employer level, one should check whether the difference persists when everything else is equal (e.g. task performed, level of responsibilities, quality of deliverables etc.).

·Women earn less and work in lower paid occupations than men do, even if working in the same sector and with otherwise identical observable characteristics; this holds true for 70 out of 72 investigated sectors (see Annex 5.2). This puts into question the notion that women ‘have a preference’ for lower paid occupations and raises instead concerns that they are to some extent constrained into undervalued female-dominated occupations, and not only sectors. At employer level, one can verify whether this is the case by looking into job descriptions/definitions versus attributed responsibilities. In other words, checking whether e.g. the tasks performed are identical (‘same work’), but labelled differently, or require skills that are not properly recognised and valued (e.g. social skills) because of gender bias (‘equal value’).

·With regard to irregular payments (i.e. fringe benefits 32 ), women have a lower participation in job-related cash benefits, with particularly pronounced participation gender gaps in SE and NO and the UK. Moreover, these payments drop sharply in women’s late careers in terms of payment receipts (cash and in-kind benefits) compared to men. For those who receive them, gender gaps in monetary fringe benefits are much higher than gender gaps in (regular) wages. In 22 countries, the magnitude is 1.5 to 10 times higher for fringe gaps than for wage gaps. A large review of research found that employers systematically under-reward women performing relatively similarly to or better than men. 33 Again, the possible reasons for these differences might be found at employer level.

To sum up, differences in pay by gender drawn from available statistics, both at EU, national and sectoral level tend to indicate pay discrimination. These same differences might or might not turn out to be perfectly legitimate from an equal pay perspective when verified at an individual employer level. Nevertheless, in countries where pay transparency measures have been introduced, an important number of employers discovered that pay differences could not be explained from an equal pay perspective (see below).

Empirical evidence of pay discrimination

The empirical analyses also show indications that pay discrimination takes place even in countries with the highest gender equality ratings. 34 Evidence from countries that introduced pay transparency measures suggest that the latter allowed gender bias in pay to come to light (see box below).

Public perception of gender pay discrimination

Results of recent surveys on pay discrimination show a non negligible perception of gender pay discrimination. The majority of respondents (60%) in the public consultation carried out for this impact assessment, though not a representative sample, reported that they have experienced gender pay discrimination directly or know someone who has. In addition, a significant number of respondents (39%) think that women and men are not paid equally in their organisation. This corresponds to data gathered in surveys on discrimination more generally. 40 For instance, over a third (35%) of respondents across the EU think discrimination based on gender is widespread in their country – ranging from over 50% in France to 16% in Bulgaria 41 . When asked about gender equality in their company, only half of respondents in the EU believe that female and male employees are paid the same for equivalent positions by their employer – ranging from 68% in the Netherlands to 30% in Czechia 42 .

Legal cases

Successful legal cases 45 equally point to gender pay discrimination. The number of cases concerning pay discrimination is extensive in the UK 46 and France 47 , whilst being relatively rare in other countries. However, case law statistics most probably do not accurately reflect the size of the problem. In many cases, workers are not aware of the fact that they are underpaid on the basis of sex (lack of transparency). In addition, several Member States reported, in the targeted consultation, that these numbers are also under-representative due to procedural obstacles in claimants’ access to justice and to the prevalence of out of court settlements 48 . Among trade unions, the large majority of respondents highlighted that it is ‘difficult’ for their members to enforce equal pay rights in their country for various reasons while one third of employer associations identify that there is difficulty in assessing work of equal value without otherwise reporting difficulties to any significant level (See Annex 2).

Other circumstantial evidence

In practice, gender norms in the workplace, in particular on work valuation, have been shown to result in labour market discrimination against women 49 , and therefore in differences in pay, in various ways. Not all skills relevant for today’s modern service economy are valued in more traditional job valuations. For instance, an analysis of the retail sector in France 50 showed that occupational skills such as having a sense of human relations and ability to serve with care, skills that are highly relevant in retail, are mostly considered as innate and are therefore not valued in the evaluation of the position of retail workers. Also, the estimation of potential productivity of prospective workers that employers must make in order to take decisions at hiring/evaluation point 51 can be influenced, consciously or unconsciously, by gender bias; lacking specific information, employers may tend to rely on gender biased assumptions or experiences 52 . A bias towards more critical review of women, particularly as for personality aspects, 53 is especially prevalent for jobs traditionally held by men and has obvious consequences in self-perpetuating differences in pay. A gender bias has also been evidenced in wage/rise/bonus negotiations, showing that the frequency and outcomes of pay negotiations are not gender-neutral. 54 Gender norms influence both women employees (who tend to negotiate less 55 and have lower expectations as to salary level 56 ) and employers (e.g. expecting that women will accept lower pay offers than men; or perceiving their request differently).

Impact of intersectional discrimination

Finally, it is worth mentioning that the problem of pay discrimination is likely to be more profound for women incurring also other types of discrimination (intersectionality), for instance for women with a minority racial or ethnic background or a disability. 57 Other groups are also more likely to be affected by pay discrimination such as single parents (of which 85% are women) 58 and LGBTIQ people 59 .

Lack of data on the extent of pay discrimination

It is hard, at aggregate level, to know whether differences observed on the labour market (e.g. in terms of salary level, bonuses, promotions) result from employers’ discriminatory behaviour or from gender-biased pay setting practices as opposed to legitimate differences stemming from objective factors, e.g. different rewarding of merit 60 . Discriminatory behaviour is rarely openly stated; the reasons for differential treatment are often only in the mind (consciously or unconsciously) of the perpetrators and their victims 61 or they are hidden in pay structures which have not been evaluated for many years. It is therefore difficult to provide specific and extended data about pay discrimination under today’s rules. Promoting and realising pay transparency at employer level will increase the amount of data available and may thus enable to fine-tune the statistical analysis of the gender pay gap, providing a better understanding of the share which is justified by objective factors and the share which is not justified objectively, and thus likely to be linked to gender pay discrimination.

2.2.What are the problem drivers?

The intervention logic (see Annex 6) depicts the two key problem drivers, namely i) a market failure linked to the asymmetry of pay information at employer level and persisting gender bias in pay setting mechanisms and valuation of women’s work, ii) a regulatory failure linked to an inconsistent and inadequate application of key legal concepts, in particular the concept of ‘work of equal value’, and procedural obstacles such as a lack of proper remedies and an insufficient victims’ support. Both of these drivers leave room for pay discrimination to appear at individual but also at employer level.

1)Market failure - asymmetries of information on pay and persisting bias in pay

Pay discrimination can be viewed as a market failure caused by information asymmetries between worker and employer about market wages.

This asymmetry creates a problem for individual workers, as they do not have the information necessary to understand the relative value of their work 62 compared to that of their colleagues of opposite sex and therefore to detect sex-based pay discrimination when they negotiate wages 63 . In many instances, the information may simply not be available, due to a lack of transparency in wages and wage setting.

The problem of information asymmetry may be reinforced because of other factors. Workers may be reluctant to ask for such information because of fear of victimisation, or because of a prohibition or cultural reticence to discuss salary levels. Moreover, their own expectations 64 on pay can be gender-biased (i.e. stemming from gender norms 65 ). This might happen because women tend to compare their salaries to other women’s salaries 66 and might therefore simply not be aware of a gender wage inequality in their occupation 67 . Furthermore, people tend to compare themselves with workers belonging to the same job category but do not normally question whether the occupation is possibly only labelled differently or is of equal value to their own. Pay differences may therefore look justified while they are not (i.e. they cannot be justified by objective, gender-neutral factors). Finally, as indicated above, gender norms might also discourage women from self-advocacy 68 , lower the likelihood that they negotiate salary offers 69 and raises 70 or lessen the results of such negotiations 71 in comparison to those of their male colleagues 72 . Lack of pay transparency contributes to pay discrimination remaining hidden and its effects accumulate from point of entry into the labour market along the whole career.

This implicitly creates a problem also at employers’ level: the fact that discrimination is not known by individuals and therefore rarely challenged in practice masks the problem of discrimination at employer level, even in the case of employers who want to treat their employees equally on the basis of sex.

Regulatory failure – lack of legal clarity and difficulty to apply key legal concepts and lack of access to justice

The second key problem driver is related to a regulatory failure due to a lack of legal clarity of key legal concepts and the related difficulty to apply them in practice and a lack of access to justice to ensure the enforcement of the right to equal pay. This was clearly illustrated by the 2020 evaluation and other surveys such as a 2019 survey carried out among trade union confederations 73 . 

Lack of clarity of/difficulty to apply key legal concepts. Even though the concept of ‘pay’ has been clarified by the Court of Justice of the European Union, a number of national legislators (e.g. AT, FI, IT, LV, SE) have not incorporated such clarifications in national law 74 . In a similar vein, despite the Court’s guidance on the concept of ‘work of equal value’ 75 , the application of the concept in practice, by employers, social partners, enforcement bodies and courts, remains subject to varied and incoherent interpretations 76 . Stakeholders in the public consultation 77 stressed that the definition of ‘work of equal value’ elaborated by the Court of Justice should be visible in the law and that substantial guidance on how to evaluate work of equal value is needed, even for individual employers.

These problems lead to the fact that, at individual level, workers do not know what elements are relevant to assess whether they are paid equally on the basis of sex and to whom they should compare themselves as regards carrying out equal work or work of equal value. This lack of clarity deters victims from bringing claims and makes running a case even more difficult. Without clear criteria for assessing ‘work of equal value’, workers cannot exercise their right to equal pay effectively.

The above problems are a problem also at employer level. Employers and business associations still underline the difficulties in applying the concept of work of equal value. 78 This may allow unconscious bias in pay setting and gives room to gender pay discrimination. While gender bias is particularly relevant in the application of the concept of ‘work of equal value’, it also still exists as regards ‘equal work’. 79 A structural undervaluation of female work, related to specific occupations in the organisation or to women’s work in general, could influence wage structures due to institutional inertia: 80 in the absence of an external trigger, a legal obligation, or a very strong commitment at managerial level, there is no incentive for employers to evaluate the gender neutrality of their wage structures and revise them where needed, also because employers might simply not be aware that there could be a problem of gender pay inequality.

Lack of access to justice. The existing EU legal framework already makes efforts to support potential victims in bringing their claim, e.g. with protection against victimisation and rules on remedies and sanctions or penalties. However, the 2020 evaluation identified persistent problems with regard to the effective enforcement of the right to equal pay and procedural obstacles to this enforcement, leading to an insufficient protection of victims. It reported major problems related to the costs of litigation, the lack of a proper mandate for equality bodies to help potential victims of sex-based pay discrimination, the lack of possibility of collective action, the practical application of the reversed burden of proof, and the low level of compensation awarded. National experts reported the costs of court proceedings as a barrier to access to justice (AT, BE, EE, FI, HR, HU, LV, NL, PL) 81 . While legal aid is only available to (very) low income earners, in most countries the ‘looser pays’ principle is a disincentive for many victims to go to court. Compensation for material damages may be claimed in only 18 Member States and non-material damages may be compensated in 14 Member States. 82 Most national experts consider compensation or reparation of victims as a significant problem as remedies are too low and not dissuasive for defendants while they discourage victims from taking action especially when compared to litigation costs. 83

Nine Member States (AT, DE, DK, EE, HR, NL, PL, PT, SE) foresee no sanctions or penalties in addition to compensation. 84 Symbolically, the absence of sanctions or penalties suggests that society at large would not be concerned by gender pay discrimination.

The lack of enforcement and insufficient protection of victims were confirmed by participants in the public consultation.

2.3.Why is it a problem?

The Charter of Fundamental Rights of the EU prohibits discrimination, in particular on the basis of sex (Article 21), and states that equality between women and men must be ensured in all areas (Article 23). Equality between women and men is a fundamental value in the European Union (Article 2 TEU). Gender-based pay discrimination and the improper enforcement of the principle of equal pay is therefore first of all a breach of an individual’s fundamental right protected under Union law.

Article 23 of the Charter explicitly refers to equality between women and men in employment, work and pay. Article 153 TFEU lists equality between women and men with regard to labour market opportunities and treatment at work as one of the main objectives of the Union’s social policy. Article 157 TFEU calls on Member States to ensure that the principle of equal pay for male and female workers for equal work or work of equal value is applied. In practice this means that employers must have pay structures ensuring that women and men be paid equally for the same work or work of equal value. Gender-based pay discrimination hence breaches individuals’ social rights protected under Union law.

Besides violating fundamental and social rights, pay discrimination has negative economic consequences as it entails inefficiencies in how the labour market operates and lowers competitiveness in the internal market. From an individual worker’s viewpoint, insufficient protection from gender pay discrimination, even if only suspected, contributes to lowering personal expectations regarding prospective earnings. As such, it can distort decisions relating to employment and occupation. It also impacts workers’ decisions on time worked and career patterns, in particular when balancing household responsibilities with earnings’ perspectives; the expectation of a potentially lower income may influence women’s apparent own choice to engage in unpaid care rather than paid work. This happens even if the household would equally share housework and care (or have no care responsibilities) due to employers’ possible biased expectations of future caring responsibilities for women or the assumption that they are not the sole or main earners in a family. All this has negative consequences on poverty rates and ultimately on the gender pension gap (which stands at 30% 85 in the EU overall). Finally, personal beliefs regarding different levels of fairness in pay/pay structures may influence workers’ propensity to mobility across sectors or countries.

The above elements bring a clear risk of potential loss of productivity, due to suboptimal female labour market participation. Such suboptimal participation increases as a consequence of women’s increasingly higher level of education compared to men. 86 An economy cannot afford such a waste of talents, especially in an ageing society confronted with skills shortages.

For employers, pay discrimination has a hidden cost, as the motivation and productivity of employees 87   may decrease if they are – or suspect being – treated unequally. They may feel undervalued, disrespected, upset or even doubt their capacities and lose self-confidence. Pay discrimination may also result in a loss of talent and investment in people in case workers would decide to move to a different employer.

Furthermore, even if the guiding principle of equal pay for the same work or work of equal value is set at EU level, the differences in approach among Member States continue to be an obstacle, not only to workers’ mobility (see above) but also for employers to operate across Member States. Indeed, the very inclusion of the equal pay principle in the EU’s founding treaties (at the time the 1957 Treaty of Rome) explicitly aimed at ensuring a level playing field among employers in the internal market. 88 Ensuring a consistent equal pay between women and men is therefore inherently a market condition to safeguard fair competition among employers across the EU. Finally, without an EU approach there could be no action taken on matters of equal pay as employers might overlook the long term productivity gains and concentrate on a perceived risk of loss of competitiveness due to potential salary adjustments.

2.4.How will the problem evolve?

As evidenced in the previous sections, gender pay discrimination is a persisting phenomenon, even if its extent may not be evaluated precisely. Because it is hard to identify and evaluate by Member States, employers, workers, enforcement bodies such as equality bodies and courts, the situation is likely to continue to evolve only very slowly: if not identified, existing discriminations may persist. Without further policy intervention towards more pay transparency, it is likely to remain largely hidden, with workers deprived of tools to have their rights respected, employers lacking the incentive to analyse and revise their pay structures, and governments and policy makers unable to grasp the extent of the problem effectively enough to tackle it.

National legislative actions to correct this situation have been scarce and there are no indications that their rate would increase. This hesitation conflicts with an increasing openness 89 and demands for pay transparency from the general public as related to the broader concept of fair pay 90 , even though these have to some extent been embraced by certain employers 91 , notwithstanding potential costs 92 . The shift towards rebalancing the asymmetry of pay information between workers and employers remains, however, very slow overall. The 2014 Recommendation, which is per definition non-binding, received a limited follow up by Member States. As a result, there are no incentives for employers to act. In practice, the situation relies essentially on individual workers’ responsibility to become aware, collect information (if available) and take action against pay discrimination while balancing the cost-opportunity of doing so. This is unlikely to change on its own.

The support study carried out in the context of this impact assessment confirms that the unexplained part of the gender pay gap, which includes pay discrimination, is not likely to decline significantly over time without further interventions. 93  

The main factor that might shape future trends on the labour market, and therefore potentially impact on gender pay discrimination as part of the gender pay gap, is the impact of the COVID-19 crisis. The latter has already shown a gendered impact and a strong backlash on gender equality. 94   95 It also reversed the usual pattern of a recession: this time, it is women, rather than men, who experience larger employment losses and higher unemployment. Women are overrepresented in lower paid jobs which are the first and the most affected by redundancies in the crisis, for instance hospitality, retail, care, and personal services. 96 To the extent that these women might not find their way back into the labour market (also because employers might restructure supply chains by recurring to more automation in order to allow for social distancing), the gender pay gap might even decrease and give the illusion that gender pay discrimination will also decrease.

In addition, the share of pay discrimination in the overall gender pay gap may reasonably be expected to increase with the recession. An increase of precarious and atypical jobs 97 (e.g. due to more digitalisation) could reinforce this trend. Precarity risks increasing wage discrimination as it further limits negotiating power on the side of workers, which especially affects women. It may further reinforce the drivers mentioned in section 2.2 (market and regulatory failures). While these larger trends on the labour market are out of the scope of this initiative, their possible impact on the problem addressed in this initiative cannot be denied.

Besides transparency on wages and wage structures, there is no indication that the other problem drivers would be addressed. Even if the Court of Justice may shed further light on how to assess work of equal value, such guidance is unlikely to reach individual employers if this concept is not encoded in the legal framework and there are no practical tools to implement it in practice. Furthermore, there are no indications that the procedural obstacles to access to justice are addressed at national level, even though they have been known and acknowledged for many years.

Section 4.3 on the baseline scenario develops these points further.

3.Why should the EU act? 

3.1.Legal basis

The Union’s specific right to act in this field is set out in detail in Title X of the TFEU related to social policy. Its right to act on matters of gender equality in employment and occupation follows from Article 157(3) TFEU. The Treaty provides that, although Member States shall ensure the application of the principle of equal pay for male and female workers for equal work or work of equal value (Article 157(1) TFEU), the EU shall adopt measures to ensure the application of the principle of ‘equal opportunities and equal treatment of women and men in matters of employment and occupation, including the principle of equal pay for equal work or work of equal value’ (Article 157(3) TFEU).

Article 157(3) TFEU is the legal basis for the Recast Directive and for the 2014 Commission Recommendation on pay transparency. It should therefore also serve as the legal basis for legally binding pay transparency measures, which support the implementation and better enforcement of the principle of equal pay under Article 157 TFEU and the Recast Directive. Unlike for an ‘internal market’ legal basis, the existence of an internal market problem is not a prerequisite to the availability of Article 157(3) as a legal basis: the pursuit of equal treatment is sufficient in that respect.

Article 157 TFEU governs a specific subject matter (lex specialis) compared to Article 153 TFEU, which provides a legal basis for all measures implementing the principle of non-discrimination on grounds of sex in the field of ‘equal opportunities’ and in ‘matters of employment and occupation’. The lex specialis prevails over the general norm. Article 157(3) TFEU therefore constitutes the proper legal basis for this initiative.

3.2.Subsidiarity: Necessity of EU action

The principle of subsidiarity (Article 5(3) TEU) requires that the Union shall act only and insofar as the objectives of the proposed actions cannot be sufficiently achieved by the Member States, either at central level or at regional and local level, but can rather, by reasons of the scale or effects of the proposed action, be better achieved at Union level.

The EU principle of equal pay was established by the founding Treaties as a fundamental support for the functioning of the internal market. Equal pay between women and men for the same work or work of equal value ensures fair competition for companies which have to comply with a similar fundamental social requirement, thus creating a level playing field for companies operating in the internal market. In the words of the Court of Justice (1976):

The aim of article 119 [now 157 TFEU] is to avoid a situation in which undertakings established in states which have actually implemented the principle of equal pay suffer a competitive disadvantage in intra-community competition as compared with undertakings established in states which have not yet eliminated discrimination against women workers as regards pay. 98

The Treaty itself thus confirms that harmonised requirements facilitate cross-border operations and create equal conditions for investment in Member States. This objective cannot be achieved by individual Member States as national measures would inevitably distort competition and create market barriers and could not achieve the envisaged level playing field.

While the principle of equal pay initially had primarily an economic function aiming at avoiding distortions to competition, in 1976 the Court of Justice recognised, in addition to its economic goal, the social objective of Article 119 EEC and its horizontal direct effect 99 . The latter effect means that the principle of equal pay may be relied upon before the national courts and that these courts have a duty to ensure the protection of the rights which this provision vests in individuals, even if there are no implementing provisions at EU or national level. 100

Later on, the Court added that the social goal prevails over its economic function and that equal pay is a fundamental right. 101 Articles 2 and 3(3) TEU include the right to equality between women and men as one of the essential values and objectives of the EU.

Article 23 of the Charter of Fundamental Rights of the EU confirms that equality between women and men must be ensured in all areas, including employment, work and pay. Articles 8 and 10 TFEU add that the EU shall aim to eliminate inequalities, to promote equality between women and men and to combat discrimination based on sex in all its policies and activities. The TFEU also provides that, although Member States shall ensure application of the principle of equal pay for male and female workers for equal work or work of equal value (Article 157(1) TFEU), the EU shall adopt measures to achieve that aim (Article 157(3) TFEU).

It is important to note that Article 157 is mandatory in nature and that the prohibition of discrimination, as confirmed by the Court of Justice, applies not only to the action of public authorities, but extends to all agreements which are intended to regulate paid labour collectively, as well as to contracts between individuals. 102

The EU acted to implement the Treaty principle, in particular, through the 1975 Equal Pay Directive, subsequently merged into the Recast Directive. The evaluation of the Recast Directive showed that further action was needed, in particular in order to create transparency on pay and pay settings. This was done through the 2014 Commission Recommendation on pay transparency. The 2017 evaluation of the Pay Transparency Recommendation showed that it had not brought about the necessary change (see below). This was confirmed by the 2020 evaluation, which reiterated that EU-level action remains necessary to better enforce the right to equal pay through pay transparency. Member States and social partners did not so far take sufficient measures to combat such discrimination.

Pay transparency measures as outlined in the 2014 Recommendation were implemented only in some Member States, and the fragmented level of implementation raises doubts on the effectiveness of non-binding measures.

The measures introduced by some Member States 103 vary broadly in effectiveness, and many Member States did not take any action in this area. There might be several reasons for this, depending on the perceived relevance of the equal pay issue, that may cause reluctance to impose possible costs on employers that might put them at a competitive disadvantage vis-à-vis foreign employers who are not applying similar norms (yet).  

Member States are therefore unlikely to be able to address the problem on their own based on the existing non-binding measures. The existing legal framework will therefore not bring about significant improvements in pay equality without a new impetus.

3.3.Subsidiarity: Added value of EU action

Action at EU level would ensure that pay equality between women and men, enshrined in Article 157(1) TFEU, is effective and that all citizens can claim their rights according to the harmonised minimum standards applicable in all Member States.

EU action does not only create a harmonised minimum standard protecting workers (social objective), it creates equal market conditions for companies operating in the internal market, thus preventing unfair competition (economic objective). Only a coherent and comprehensive European approach can ensure a level playing field for market operators in all Member States and exclude possible unfair competitive advantages that could be derived from pay discrimination (see above). As shown in the report on the evaluation of the 2014 Recommendation, national measures regarding pay transparency are fragmented, scarce and most often set lower thresholds than those proposed by the 2014 Recommendation. Their variety, different levels of ambition, and the progressive nature of the changes bring unnecessary complexity and uncertainty and constitute barriers to operating across Member States.

EU action establishing pay transparency and related enforcement mechanisms would ensure that workers across the EU can experience the same level of protection of the equal pay right, and that employers can operate across Member States with lower risks/uncertainties. It would realise a fundamental and social right while preventing businesses from competing on an uneven playing field and therefore improve the operation and competitiveness of the internal market.

4.Objectives: What is to be achieved?

4.1.General objectives

The overall objective of the initiative is to improve the implementation and enforcement of the principle of equal pay for equal work or for work of equal value both as a fundamental right and social objective to be achieved in the European Union and as an economic objective linked to the full realisation of gender equal conditions on the internal market. It does so by creating transparency both at the individual worker and at employer level, based on a facilitated implementation of the legal concepts of equal pay for equal work or for work of equal value. It balances the responsibility for taking action between workers claiming their EU right and employers ensuring these rights are respected.

4.2.Specific objectives

The initiative more specifically aims at:

1.Empowering workers to enforce their right to equal pay: the initiative aims at bringing instances of pay discrimination to light through transparency, giving workers, especially female workers, the necessary information to act upon them. This objective could be supported by addressing the difficulties relating to the application of the key legal concepts relating to equal pay and the inadequate access to justice.

2.Addressing systemic undervaluation of women’s work at employer level: the initiative aims, through transparency, at correcting biases in pay setting mechanisms that perpetuate the undervaluation of work done by women. Such undervaluation occurs when women’s skills are not or not sufficiently valued in pay structures. This objective could equally be supported by addressing the difficulties relating to the application of the key legal concepts relating to pay and the inadequate access to justice.

At the same time, pay transparency measures will help to collect more information and data on the existence and extent of pay discrimination.

4.3.What is the baseline from which options are assessed?

EU action through the 2014 Recommendation and the European Semester.

Under the baseline scenario, the 2014 Recommendation would remain the main action at EU level to address the problems set out above. In addition, the Commission could continue to issue Country Specific Recommendations (CSRs) in the context of the European Semester. In principle, CSRs have the potential to address the phenomenon of the gender pay gap but only from a global perspective and highlight its root causes in Member States. In 2019, the pay gap was addressed in the country reports of 11 Member States 104 and a country-specific recommendation on the gender pay gap was issued to one Member State 105 . The CSRs can contribute to tackle some of the root causes of the overall gender pay gap e.g. by supporting measures to strengthen work-life balance policies, counter sectoral segregation, or even suggesting wage transparency.

The above actions are, however, unlikely to bring about the desired change and fulfil the objectives set out in sections 4.1 and 4.2. First of all, as shown above, the Recommendation has had limited follow-up in Member States and while a few Member States have recently moved towards more transparency (e.g. PT, ES), most Member States are not undertaking any action or action remains very limited (e.g. only for very large employers, only voluntary, no structural measures) and does not create the necessary level playing field. In regard to action through the European Semester, given the many issues addressed in the European Semester, the CSRs cannot reach the necessary level of detail to address the implementation of the principle of equal pay at employer level nor to reinforce victims’ protection or support. In addition, the European Semester is being reshaped towards a focus on the monitoring of Recovery and Resilience Plans and relevant challenges, in particular in the context of the green and digital transition. Hence, while country specific recommendations could be a useful policy tool, they could only complement legislative measures on pay transparency and, more generally, strengthen a comprehensive approach in tackling the gender pay gap.

Trends in the labour market (see Annex 5, section 6). Before the COVID-19 pandemic, women had largely reduced, if not reversed, some differences in the average characteristics relating to pay compared to men: they have surpassed men regarding the level of education and are moving, though slowly, into areas that were traditionally men-dominated such as construction, transport and automotive services. These elements, which were important in explaining differences in average pay in the past, may be expected to contribute less and less in further reducing the relative pay differences. Today, the justified pay gap represents a third of the total gender pay gap; in the future, a further reduction in the gender pay gap may be expected to result mainly from changes in the unexplained component, including gender pay discrimination. At the same time, trends in wage setting – e.g. moving away from collective towards more individualised payments linked to new forms of work – are increasing the scope for discretionary pay and therefore could put women further in disadvantage (as they are more likely victims of pay discrimination) (see section 2.1). As a consequence, it seems unlikely that without any positive intervention through binding measures, the rate of reduction of the unexplained part of the gender pay gap – which covers possible discrimination – could accelerate. 106

Effects of the COVID-19 pandemic. The profound changes in the socio-economic landscape 107 , though still to be fully manifested, have already shown a gendered effect (see section 2.4). The pandemic has had a strong impact 108 in sectors in which women are overrepresented 109 , namely retail trade, transport, hospitality and tourism. Women have also been more affected in their work-life balance, due to increased care responsibilities. 110 While the boost towards smart working might increase women’s possibilities for entry into and retention within higher level jobs, there is also a risk that these might lead to more precarious and atypical jobs (e.g. replacing employment contracts of some teleworking workers with outsourcing of services via online platform work). The shift towards more automation to allow social distancing could also, once the pandemic will be over, push in the same direction. It is also still unclear what will be the effect of policy efforts from the EU 111 and Member States to limit the economic impact of the pandemic and sustain the subsequent recovery, and to what extent these might take into account a gender-sensitive perspective. 

5.SME test

The specific impact of all policy measures on SMEs has been screened ex ante for all options. The aim of the analysis was to check whether SMEs 112 would be disproportionately affected and, where relevant, to include mitigating measures in the design of the policy options.

The measures envisaged are aimed at protecting a fundamental right. They should therefore in principle apply to all workers, independently from sector, employer size or type of contract. As a large majority of workers work in SMEs, 113 it would not be consistent with the purpose of this initiative to exclude all SMEs from its field of application, as their exclusion would not allow to reach the goals of the initiative and would undermine a fundamental right. Nevertheless, the measures may be designed in such a way as to achieve the objectives of this initiative without imposing a disproportionate burden on SMEs.

Consultation of SME stakeholders

82% of respondents from business/employers’ organisations and associations in the targeted consultation agree that pay transparency measures should be tailored to the size of the organisations. At the dedicated hearing, for instance, SMEunited (the association of Crafts and SMEs in Europe) requested the exemption of SMEs from a directive on pay transparency at EU level, especially as regards small and microenterprises (see Annex 2).

Assessment of alternative mechanisms and mitigating measures 

Taking into account the principle of proportionality, concerns expressed by stakeholders, and technical feasibility of some individual measures, it seems advisable to modulate the measures according to the size of the employers. This was also the conclusion of the impact assessment for the 2014 Recommendation and is in line with what is done at Member State level. In addition, specific measures may be added to shield SMEs and in particular micro-enterprises from increased business risk linked e.g. to exposure to legal costs. Furthermore, SMEs will benefit from the guidance and methodologies developed to assess work of equal value. Member States could also provide as mitigating measures ready-made templates with info line support or off-the-shelf software, since these were found to reduce costs for employers (Eurofound, 2020).

The definition of the exemption thresholds

The thresholds applied to tailor the design of the measures in this initiative are directly derived from the 2014 Recommendation. Similar thresholds are found in national legislation of a number of Member States. 114  

The definition of the thresholds is linked to the number of workers. 115 The reason for not referring to other criteria often used to define small and medium sized employers, such as turnover or annual balance sheet (e.g. based on Commission Recommendation 2003/361), is that the focus and scope of the proposed Directive is worker-oriented, based on the enforcement of an individual fundamental right, and it uses statistical methods for the implementation of which only staff headcount is relevant. Moreover, this reference simplifies the implementation by avoiding the complexity and administrative burden in verifying the compliance criteria by Member States and employers.

The proposed measures are adapted depending on the size of the employers, to find a balance between the interest of all workers in seeing their right protected and minimising costs and burden on employers, even if the latter would also benefit from an increase in productivity. While a number of considerations are taken into account to define the thresholds, the ultimate decision on the exact threshold to apply in this initiative is a question of political determination and a choice made on this trade-off. Focusing on larger employers also follows a cautious approach. It will allow to collect more data on the extent of gender pay discrimination and therefore bring more arguments for or against extending it further to smaller companies. The details of this tailored design of the measures are explained in the description of the options and analysed further for each measure in Section 7. SMEs could always on a voluntary basis adopt more far-reaching measures than those set out for them in this initiative in order to build a more attractive employer profile. Member States could also decide to extend the scope of the measures envisaged by this initiative.

6.Description of the policy options for EU level action 

The policy options described in Sections 6.3 through 6.5 below can be broadly described as follows:

-Option 1 aims at creating transparency at the level of individual workers, mainly aiming at empowering workers (first specific objective of this initiative). It includes two possible levels of ambition, with Sub-option 1A presenting a higher level of ambition than Sub-option 1B;

-Option 2 aims at creating transparency at employers’ level, mainly aiming at addressing structural undervaluation of women’s work (second specific objective of this initiative). It includes four possible levels of ambition, with Sub-options 2A and 2B presenting a more ambitious level while Sub-options 2C and 2D go for a lower level of ambition;

-Option 3 aims at facilitating the implementation and enforcement of the existing legal framework, contributing to both specific objectives of this initiative.

The basic work done in terms of data gathering by an employer to create transparency either at individual or at organizational level is the same for most Sub-options under Options 1 and 2 (see Section 8.5). The options differ as to the way in which the information is made transparent.

The three options can be effective on their own, depending on the political decision to achieve to a greater or lesser extent the policy objectives of this initiative.

6.1.Options discarded at an early stage

Council Recommendation

This instrument has been considered as an option aimed at anchoring pay transparency measures at political level and enhancing Member States’ engagement on the matter. It would replace the 2014 Commission Recommendation providing general guidance to Member States on how to address the lack of pay transparency. Given the limited follow-up to the 2014 Recommendation, the involvement of Member States in the decision-making process, even as regards non-binding measures, might constitute a step forward.

A Council Recommendation has the advantage of being a flexible instrument that could be adapted to national contexts. Moreover, it could provide a basis towards a level playing field between Member States on how to address the lack of pay transparency, based on a consensus by Member States at Council level endorsing their political commitment. For this reason, it could presumably be more effective than the 2014 Commission Recommendation.

This option would, however, present several drawbacks. First of all, providing general guidance to Member States, in nature not binding, is very uncertain in terms of compliance. Furthermore, the scope of such initiative would most likely be limited to pay transparency measures only as this instrument is not suitable to introduce changes to enforcement measures aiming to ensure the effective implementation of the measures and enhance workers’ access to justice that are already at least partly covered by the 2006/54 Directive. Thirdly, implementation would remain uneven, with different modalities corresponding to different levels of ambition, and without ensuring a minimum level of worker protection. Fourthly, there would be no enforcement at EU level allowing to assess and ensure the effectiveness of any measures taken. Finally, the analysis of existing Council recommendations shows that such type of policy instrument is more appropriate to address areas covered by Member States’ exclusive competence with a purpose to undertake political commitments by all Member States supported by existing EU financial instruments and programmes 116 .

Conclusion: Since this option is not likely to effectively achieve the general and specific objectives of the initiative, it has not been considered further in this impact assessment.

6.2.Option 0: Status Quo

The first possible course of action would be not to undertake further action and let the situation evolve under the current framework (See sections 2.4 and 4.3).

6.3.Option 1: Legislative action to create transparency for individual workers 

This option aims at realising the specific objective of empowering workers to claim their right to equal pay. Indirectly, by raising awareness among employers when preparing pay information it could affect gender bias in pay setting mechanisms and valuation of women’s work, but such effect would depend on the self-motivation of employers to launch action in that respect. Two alternative sub-options (1A and 1B) are suggested below. They present different levels of ambition, balancing the rights of workers against the cost and burden on employers, especially SMEs.

Sub-option 1A: Empower workers by granting a right to receive pay information

Measure 1: Transparency of salary information prior to employment

This measure aims to ensure that the employer informs prospective workers about the initial salary offer or a reasonable range expected to apply to the position concerned. It would strengthen workers’ bargaining power when it comes to pay setting and their understanding of their paid position compared to other workers carrying out equal work or work of equal value in the organisation. It would also ensure that the employer is not allowed to ask about the applicant’s previous salary or to require them to sign non-disclosure agreements concerning the wages they receive. The measure would be applicable to all employers.

Measure 2: Employer’s obligation to provide all workers with individual information on pay compared to their category

The existing obligations under the Treaty require employers to have pay structures ensuring that women and men are paid equally for the same work or work of equal value. In order to comply with this existing requirement, employers should group workers in their organisation according to the carrying out of equal work or work of equal value. This measure would require the information on average pay levels in such groupings to be made available to workers insofar as it relates to them. Workers would receive information on average pay levels, broken down by sex, only for their own category of workers doing the same work or work of equal value within the organisation. On the basis of the information received, workers would be able, in line with the requirements set by the Court of Justice, to compare their position and job description against a comparator attributed to the same category; this would enable them to evaluate whether or not they may be victim of pay discrimination based on sex.

The information would be provided by employers to workers periodically (e.g. synchronised with performance evaluation performed usually every year), without any prerequisite request from the worker.

Exemption: the measure would exempt employers with less than 50 workers from producing such periodic information in the absence of a worker’s request. This means that workers in such organisations would still be able to obtain the information, however only upon request. To address the possible fear of victimisation, workers would be protected by a strengthening of the existing framework regarding the burden of proof (from the Recast Directive) by adding an additional alleviation of this burden regarding compliance with the pay transparency requirements set out by the initiative. Thus, in case an employer does not comply with its pay transparency obligations under the initiative, the burden of proof would be automatically shifted on the employer in case of a worker’s complaint. In addition, minimum standards on access to evidence would be set. Such alleviated provision would at the same time better reflect the case law 117 of the Court.

Sub-option 1B: Empower workers by granting a right to request pay information

Measure 1: Transparency of salary information prior to employment

Same as measure 1 of Sub-option 1A.

Measure 2: Right of worker to receive individual information on pay compared to their category – upon request

The right to information would be the same as in Sub-option 1A above, with the difference that the pay information under this Sub-option would be provided by the employer only upon an individual worker’s request (or by workers’ representatives or equality bodies). The measure would apply to all employers without exemption.

6.4. Option 2: Legislative action to create transparency at employer level

This option aims at increasing transparency about pay structures of employers and organisations and thus bringing to light any potential structural gender bias. It would have an indirect effect on the problem driver of asymmetries of information on pay as some information would be shared with workers or made public. However, this latter effect would be limited because of exemptions excluding a significant share of the workforce and the potential difficulty of an individual worker to position themselves with the information provided. Four alternative Sub-options are suggested under this option. As in Option 1, they balance, on the one hand, the fundamental right and the need for protection of workers and, on the other hand, the costs and burden on employers, especially SMEs.

Sub-option 2A: Equal pay certification

This option would introduce the obligation for employers to receive by an accredited certification body the Standard accreditation ‘Equal Pay Standard’ (Standard ÍST 85: 2012 – Equal Pay Management System – Requirements and Guidance), which is the only example existing in the world. The accreditation certifies, ex ante, the gender-neutrality of the employer’s pay structure. 118 The certification ensures that the difference in the average pay of women and men for the same work or work of equal value (assessed by the employer according to the standard) is solely linked to objective factors (including employers’ discretion based on explicit gender-neutral criteria regarding both the position and the performance of the employee). There are four main criteria (expertise/competence, responsibility, strain and working conditions). Each workplace must then define its own sub-criteria. This Sub-option fundamentally differs from the next approaches considered under Option 2, which are limited to ex post actions, aiming at uncovering pay inequalities in ad hoc cases.

Exemption: employers with less than 50 workers would be exempted from this measure.

Sub-option 2B: Employers’ obligation to carry out a joint pay assessment

This option would require employers to carry out an annual pay assessment in cooperation with workers’ representatives. Such joint pay assessment could also be conducted by an external audit firm. It would include an assessment of the employer’s pay structure and any pay differentials based on sex, with the aim of identifying any unjustified gender pay differences. If the assessment brings to light such differences, these would need to be remedied.

This option would trigger a systematic evaluation by employers on the extent to which their own policies and practices may be, even unconsciously, biased or discriminatory and how they may thus contribute to the gender pay gap, as opposed to factors justified or outside the employer’s control. This option therefore aims at uncovering the often unconscious systemic undervaluation of women’s work. It will require remedial action.

The main difference with Sub-option 2A is that all action would be at organisational, employer level, with no mandatory external or Government involvement/certification.

Exemption: employers with less than 50 workers would be exempted from this measure.

Sub-option 2C: Basic pay reporting combined with joint pay assessment

This Sub-option presents a less ambitious alternative, reserving the joint pay assessment to larger employers (measure 1), combining it with a less demanding requirement of pay reporting on the basis of easily accessible data for medium-sized employers (measure 2).

Measure 1: Employers’ obligation to carry out a joint pay assessment

The content of this measure would be similar to the one set out in Sub-option 2B. In light of the annual publication of a pay report (see measure 2), the assessment could be carried out on a less regular basis, every three years.

Exemption: employers with less than 250 workers would be exempted from this measure.

Measure 2: Employers’ obligation to report on average differences in pay between female and male workers

This measure would require employers to publish annually the information on the average difference in pay between female and male workers as a percentage of the average salary of male workers in their organisation. Reporting on average differences in pay between men and women in the organisation permits to see how such differences are distributed among the overall workforce within the employer, whether they are visible in regard to the basic wage or rather/also in regard to other, complementary or variable components such as benefits and bonuses and the proportion of male and female workers receiving such components. The information could be easily processed by employers on the basis of existing data. It would not contain precise information on categories of workers doing the same work or work of equal value; it can therefore only give a rough indication of possible pay inequalities. No remedial action by employers would be required. The information made accessible under this measure would ensure a minimum transparency of easily available information; as such, it would complement the more detailed pay information envisaged under option 1, which would only be available upon request. This measure can still contribute to achieving the policy objectives, since the publicity of the information and the possibility for workers and their representatives, equality bodies and labour inspectorates to ask for an explanation of any differences shown in the report is expected to work as an incentive for employers to analyse further the drivers of pay differences and act upon them.

The information provided under this measure is to be shared with the national authorities which could use it for monitoring compliance with the obligation to report, to assess aggregated data, and possibly to tailor further targeted policy measures e.g. by sector or geographical area.

Member States could alleviate the impact of this measure on employers by gathering the information by employer on the basis of administrative data, if available and interlinked.

Exemption: employers with less than 50 workers would be exempted from this measure.

Sub-option 2D: Strengthened pay reporting and joint pay assessment in case of pay differences which cannot be justified by objective, gender-neutral factors

This Sub-option further reduces costs and burden for employers by limiting the obligation to carry out a joint pay assessment only to those employers which have a problem of pay inequality in their organisation. Such targeted action would be based on a strengthened pay reporting obligation which gives a more refined view on pay inequalities than the pay reporting under Sub-option 2C above.

Measure 1: Employers’ obligation to report on the average difference in pay between female and male workers by worker category

This measure would be similar to measure 2 under Sub-option 2C above but would extend the reporting exercise by the calculation of the average difference in pay between female and male employees by categories of workers doing the same work or work of equal value. As such, it would provide for a more refined diagnosis of possible pay inequalities in the organisation. This latter information would only be made available to workers and their representatives (and equality bodies and labour inspectorates upon request). The reason for this is that the categorization of workers performing work of equal value is based on a combination and weight of criteria which are relevant to the specific employer concerned. Such information may more appropriately remain within the employer concerned or be shared only with designated public bodies; it does not have much added value for the general public. The overall average pay gap at employer level would be made public as in Sub-option 2C.

Exemption: employers with less than 250 workers would be exempted from this measure.

Measure 2: Employers’ obligation to carry out a joint pay assessment if pay reports show pay differences which cannot be justified by objective, gender-neutral factors

Under this Sub-option, a joint pay assessment would only be required if the pay report under measure 1 shows a difference of average pay between female and male workers at employer level of 5% in any category of workers doing the same work or work of equal value, unless the employer can justify this difference by objective factors. In case of disagreement between the employer and workers’ representatives on the need to launch a joint pay assessment, this obligation may need to be enforced through the competent enforcement authorities.

Exemption: employers with less than 250 workers would be exempted from this measure.

6.5. Option 3: Legislative action to facilitate the application of and enforce the existing legal framework

This option consists of a package of 3 measures aimed at facilitating the application of and better enforcing the existing legal framework. It would tackle the regulatory problem driver relating to the inconsistent and inadequate application of key legal concepts relating to the principle of equal pay (measure 1). It also addresses the problems relating to access to justice and deficient enforcement through addressing the procedural obstacles and victims’ support (measures 2 and 3).

Measure 1: Facilitate the application of the existing key concepts of ‘pay’ and ‘work of equal value’

This measure especially would ensure a common approach on the application and interpretation of the EU rules on equal pay. It would enshrine the main guidelines and objective criteria set by the CJEU 119 (which is of general nature and authoritative in all Member States) in the law. The application of such criteria would facilitate compliance with the existing obligation to ensure that the pay structures of all employers are designed to uphold equality of pay between women and men doing equal work or work of equal value.

The legal concept of ‘pay’ will be clarified to reflect that ‘pay’ comprises not only salary, but also its complementary components whether in cash or in kind, which the workers receive directly or indirectly, in respect of their employment from their employer. This clarification will ensure that the equal pay principle applies to all components constituting pay – not only its fixed amount, but also variable components such as bonuses, overtime compensation, travel facilities, compensation for attending training, payments in case of dismissal, overtime supplements, and gratuities paid at the discretion of an employer, statutory sick pay, statutory required compensation and occupational pensions. The principle of equal pay does not mean that all workers should be paid equally; it means that any pay differences should be based on objective criteria, not related to sex.

In regard to the concept of ‘work of equal value’, the Court stresses that the comparison of jobs should be made in the light of objective criteria, such as education, professional and training requirements, skills, effort and responsibility, work undertaken and the nature of the tasks involved 120 . The existing legislation would benefit from an incorporation of these criteria, as examples, in the law. In practice the application of this concept means that employers have to identify key criteria which are relevant to the organisation and its business model and based on which workers doing the same work or work of equal value get comparable pay. Again, it does not preclude employers to pay workers doing the same work or work of equal value differently, as long as such differences are based on objective, gender-neutral and bias-free criteria as for example performance and competence.

In order to comply with the current legal framework, all employers should be able to define which workers in their organisation are carrying out work of equal value. The application of the criteria to assess the value of work also assists workers, both women and men, who may have reason to believe their work is under-valued because of gender based discrimination, to obtain the necessary information to resolve the problem through negotiation or, as a last resort, to bring the matter to the equality body, labour inspectorate or national courts. In situations where human resource management is not formalised or clear criteria are not established by the employer, the evaluation of the categorisation of workers by value of the work within a given employer or organisation may require effort at managerial level.

It is important to note that the comparison of ‘work of equal value’, under the existing case law of the Court, has to be done between workers at employer (as opposed to sectoral) level. It does not compare similar jobs at sectoral level nor similar jobs between different employers except in a situation where pay structures are laid down centrally for more than one organisation or business within a holding company or conglomerate. So far, the concept has not been used under EU law to address horizontal gender segregation, i.e. the part of the gender pay gap linked to the fact that women are overrepresented in low paid sectors, or other root-causes of the gender pay gap. This initiative does not aim to change this situation at this stage.

Besides clarifying the concept of ‘equal value’ in the law, its implementation in practice would be improved by tools or guidelines developed at EU and national level. At EU level, such guidance would be foreseen as a flanking non-legislative measure under the Communication accompanying the legislative initiative. Flexibility would be left to Member States as to the choice of the tools or guidelines at national level; these could be developed by Governments and/or social partners.

The tools to support the application of the concept of work of equal value could consist, for instance, of gender-neutral job evaluation and classification methodologies (see Annex 5, Section 14) 121 or IT tools such as Logib 122 .  Job evaluation and classification systems, can be exploited as a basis for an equitable pay structure, though they usually go beyond matters relating to equal pay, including all matters relating to labour relations, such as hiring, training, health and safety, etc. Any disputes related to pay or performance evaluation can be settled conveniently by referring to the existing job evaluation mechanism. For job evaluation mechanisms to be effective in removing gender inequalities in pay structures, they must include criteria relevant to modern job descriptions and be formulated in a gender neutral way 123 , i.e. not influenced e.g. by traditional stereotypes.

Measure 2: Improved access to justice for potential victims of pay discrimination

This measure includes a set of tools aimed at improving worker’s access to justice in case of alleged pay discrimination, addressing specifically the obstacles raised in the studies and consultations carried out for the 2020 Evaluation and this impact assessment.

Victims’ representation (legal standing and representative actions) 

This measure would strengthen the provisions of the Recast Directive regarding the defence of rights (Article 17) in matters of pay equality. Equality bodies and workers representatives’ organisations would have the capacity to act not only in support but also on behalf of victims in any judicial or administrative procedure. The aim is to relieve the burden of bringing a claim from the individual victim of pay discrimination, who may be in a precarious and vulnerable situation and fear victimisation. The measure would include the possibility for equality bodies or workers’ representatives to join claims of several victims, with the aim of supporting the latter but also possibly address structural discrimination in pay structures.

An alternative would be to extend the right of action also to associations, organisations or other legal entities pursuing the enforcement of the right to equal pay.

Strengthened and new remedies (compensation and injunction orders)

This measure would strengthen the already existing right to compensation under the Recast Directive (Article 18) by requiring Member States to enable victims who have suffered harm caused by unjustified pay differences to obtain full compensation. The CJEU already clarified that measures appropriate to restore genuine equality of opportunity must guarantee real and effective judicial protection and have a genuine deterrent effect on the employer. 124 This measure aims at translating this general requirement into more specific minimum standards applicable to matters covered by the initiative. It would ensure that the worker is placed in the position they would have been in if no unjustified pay difference had taken place (including full recovery of back pay and related bonuses or payments in kind, compensation for lost opportunities, and moral prejudice).

Furthermore, under this measure, national courts would have the possibility to stop an infringement or to order the implementation of structural or organisational measures to comply with the obligations foreseen under the proposed initiative.

Limitation periods and legal costs

One of the obstacles raised in the context of gender pay discrimination claims is the deadlines for claims to be brought before the courts (limitation periods). If they are too short or start running before the victim may actually be aware of the existing discrimination, limitation periods may make it impossible in practice for the victim to exercise their rights. Under this measure, therefore, Member States would be required to comply with minimum standards set out by the initiative and applicable to matters covered by the initiative. Member States will determine when the period starts, its duration and the circumstances under which it is interrupted or suspended.

Probably the most important obstacle preventing access to justice is the financial burden of enforcing the equal pay right. Such burden could be alleviated by allowing successful claimants to recover proceedings’ costs while defendants who are successful in rebutting a pay discrimination claim would, on the other hand, not have the right to recover such costs from the claimant. The envisaged rule would be accompanied by an additional safeguard so that the claimant-friendly rule would not apply in case of claims brought in bad faith, claims which are clearly frivolous or when the non-recovery by the defendant would be considered unreasonable under the circumstances. An alternative could be that the claimant would always be allowed to recover proceedings’ costs in an equal pay case from the defendant whether successful or not.

Measure 3: Other measures to enhance enforcement and implementation

Penalties

This measure would require Member States to lay down administrative or criminal sanctions or penalties aimed at enforcing rights and obligations under the initiative. The level of sanctions/penalties should be effective, proportionate and dissuasive. The level of fines should include a minimum level and be set at such a level as to reflect the value of equal pay as a core element of the fundamental right to equality and equal opportunities, ensuring a deterrent effect.

Involvement of the social partners

Social partners at all levels (e.g. employer, sectoral, national) play a crucial role in ensuring the proper implementation of the right to equal pay. While the matter is high on the agenda of social partners in some Member States, it is hardly ever discussed in some others 125 . The initiative, with due respect to the autonomy of social partners and in accordance with national law and practice, would require Member States to take appropriate measures ensuring that the implementation of the rights and obligations under this initiative be discussed with social partners. This could be achieved through different policy measures to help develop active social partnership and inclusion of equal pay matters, especially ensuring proper implementation of the right to equal pay for work of equal value as a key element of collective bargaining. The participation of social partners is important, for instance, in case gender-neutral job evaluation and classification methods are established to prevent or identify and tackle possible direct or indirect gender-based pay discrimination.

6.6. Overview of the policy options

7.What are the impacts of the policy options?

7.1.  Option 0: Status Quo

See the Baseline (Section 4.3). 

7.2.Option 1: Legislative action to create transparency for individual workers

The policy sub-options identified in this section primarily address the first specific objective (see Section 4.2) of this initiative, i.e. empowering workers to enforce their right to equal pay. They aim at strengthening workers’ bargaining power when it comes to pay setting and their understanding of their position in regard to pay compared to other workers carrying out equal work or work of equal value in the organisation (not across sectors).

Sub-option 1A: Empower workers by granting a right to receive pay information

Measure 1: Transparency of salary information prior to employment

This measure is likely to reduce the risk of gender bias when salary is set for a newly hired worker. The measure is twofold: first, it provides workers with information on the range of the salary envisaged for the vacancy before starting negotiating their salaries 126 . As such, it strengthens workers’ bargaining position by ensuring they have all relevant information to negotiate on an appropriate basis. Secondly, it prohibits to ask for previous salary, something which has been shown to perpetuate gender gaps in pay over time, especially when changing job. Overall, this measure would ensure that parties to the salary negotiation are on equal footing as regards the information relating to the value (or range) assigned by the employer to that position. It would ensure that salary negotiations are unbiased by possible gendered expectations, which tend to work (more frequently) to the disadvantage of women, with consequences lasting during the entire career. Stakeholders’ views: the targeted consultation with social partners informed that the inclusion of expected salary in job postings is supported by the large majority of trade unions (78%) that participated in the consultation. It is also the most preferred option for action on the side of employer association respondents, though not reaching a majority (41%) among them. As for the prohibition to ask about previous salaries, in the targeted consultation with Member States, seven out of the 13 who replied to the question included this option in their ideal preferred package of measures, five included the prohibition to ask for previous salary and six the prohibition of confidentiality agreements. In the targeted consultation with social partners about 38% supported this option, as the envisaged measures are yet known only in a limited number of Member States (see below).

Costs: there are no direct costs related to the right to information on pay before employment, and the prohibition to ask about previous salary. Indirect costs could arise if the measure would lead to a limitation of the negotiating space for both parties, which is not the case. The measure only requires to share/receive information on the starting offer (or reasonable range) and would not constrain the negotiating power of either employer or worker. 127 Employers could be protected from the risk of having prospective candidates ‘poached’ in very competitive job markets by the possibility, in the absence of a vacancy notice, to inform about the initial offer only at the beginning of the interview. While this would to some extent reduce the effectiveness of the measure, it would still enable workers to come better informed at the negotiating table. This measure would only carry an opportunity cost if the employer had intended to benefit from a gender-biased salary offer, which would be discriminatory and therefore illegal. Finally, in general terms, it is also more efficient to avoid having good candidates competing for positions that they would eventually refuse because paid below their minimum expectations.

Benefits: Beyond rebalancing the information asymmetry and reducing the risk of gender bias in negotiations 128 , disclosing the pay range offered shows to prospective workers how the skills and job requirements are evaluated and valued (process transparency). It also motivates employers to analyse the pay setting criteria due to be applied when making the offer, and to ensure that these are gender neutral. The measure is expected to have a positive gender effect, as studies show that women negotiate better in transparent environments 129 . Finally, it contributes to building trust between workers and employers therefore promoting a better work environment. It is further reinforced by the prohibition to ask about previous salary that breaks a vicious circle of path dependency in salary setting.

The measures envisaged may be expected to gradually benefit an increasing number of workers entering the labour market or changing job. 130 They may lead to an increase in women’s salaries. Estimating such increase cannot, however, be predicted with reliability – it would need to be based on an estimate of the gender biased difference in salary 131 that would have been offered in the absence of measures applied to an also estimated share of job applicants for each employer. This measure may benefit most women in precarious situations, such as single mothers, and women potentially subject to multiple discrimination, two categories which can be expected to face higher gender discrimination (see problem definition). Finally, the consequent increase in women’s salaries may have a negative impact on the profits made by employers, profits which, however, were potentially based on illegal behaviour.

Legal change: Transparency of salary information prior to employment is currently available in 4 Member States (AT, LV, PT, SK) and the prohibition to ask about previous salary only exists in PT. 132 No concern has been expressed in the support study that introducing this measure would raise questions of feasibility.

Conclusion: This measure entails low costs in comparison to related benefits, remedying the asymmetry of information on pay prior to employment and preventing a perpetuation of an existing unjustified pay difference between women and men. The cost/benefit analysis is therefore positive, even if it entails a change in the majority of Member States. 

Measure 2: Employers’ obligation to provide all workers with individual information on pay compared to their category

In order to be able to meaningfully exercise their right to equal pay, workers must have the information necessary to assess prima facie whether they may be the victim of pay discrimination. This measure would achieve that objective, by making available the information on the average level of pay of their comparators (i.e. workers carrying out the same work or work of equal value). In order to best address a possible fear of victimisation on the part of workers, the information would be provided without workers having to ask for it.

To the extent that employers have a firm-specific grouping or classification available, the information to be provided can easily be gathered (see Section 6.3, Sub-option 1A). If no such grouping or classification is available, the employer would need to determine first which workers carry out the same work or work of equal value. The transparent classification of jobs and positions by the employer provides an objective picture of the pay structure and implicitly of the pay setting and progression mechanisms. As such, it brings to light possible biases on which employers could, but would not be obliged to act unless a worker were to show an unjustified gender-based pay difference.

Exemption regarding the automatic provision of information for smaller employers.

The grouping of workers into categories based on the content or value of the work they are attributed to do requires a minimum number of observations to be effective. 133 This minimum number is influenced by the specific field, the organisation of work in the organisation and the proportion between women and men. In general, one can expect that smaller employers may have less room for differentiating between categories of workers or may possibly have categories with no possible gender split. In Member States that have such type of measure in place, the threshold varies very widely, between 10 and 500 workers - most often it is set below or at 50 workers. Studying the preparation of measures under national law shows that the applicable threshold is mostly the result of political decision-making. For this initiative, it may reasonably be assumed that the useful headcount making this measure meaningful, also in statistical terms, may be set at least 50 workers. Requiring smaller employers to pro-actively classify their workers and provide regular information may be considered as disproportionate. The ‘50+ workers’ threshold is also the one retained in the 2014 Recommendation. Using such threshold, the right to receive relevant information on pay automatically would apply to 50% of the EU27 workforce and on average 1% of EU27 enterprises.

To protect the right of workers in organisations with less than 50 workers, it must be ensured that the latter always have the right to request the information on the comparators that is relevant for them. Nevertheless, additional protection seems to be needed since experience in Member States (e.g. Germany) shows that workers frequently hesitate to request information on their comparators for fear of victimisation (e.g. being perceived as uncollegial or pushy). Although workers would have the right to request pay information, they might be reluctant to do so. Therefore, their right to request is accompanied by a strengthened mechanism regarding the shift of the burden of proof in case of pay discrimination. Under this mechanism, would an employer not reply to a worker’s request, this would trigger a presumption of gender pay discrimination and the burden to prove that there is no discrimination would automatically shift to the employer. Though this would not dispel fear of victimisation, it might provide more motivation for workers to act.

Data protection. During the targeted hearing with social partners, employers’ organisations raised certain concerns about data protection, especially on the conciliation of potential transparency obligations with their obligations under the EU General Data Protection Regulation (GDPR). These concerns were not shared by trade unions (see Annex 2). Disclosure of private data is allowed if required by a legal obligation with a legitimate aim – here the enforcement of the equal pay principle. 134 Nevertheless, as the culture around pay secrecy varies across Member States and in order to address the concerns, safeguards may be needed. First of all, disclosure of personal data of an identifiable individual could be avoided while protecting workers’ right to information. This could be done by disclosing the information through a third party, such as an equality body, bound to maintain the confidentiality of the data. In addition, the initiative would permit to limit the use of the information obtained for purposes of making a claim relating to equal pay only.

Stakeholder views: EU citizens who replied to the public consultation largely favour this measure (88%), as do trade union organisations (89%). Among employers’ organisations, only 18% support the measure. The main concerns expressed by employers relate not to the measure itself, but to the possible difficulties relating to the grouping of workers in categories of equal work or work of equal value (see Option 3) and data protection. The consultation with Member States revealed that a majority (59%) of them thinks this measure would be the most effective and would best support a better enforcement of the equal pay principle across the EU.

Costs: For employers, the direct cost would arise from assembling the necessary information and communicating it to workers in a transparent manner. The communication cost itself would be very low: automatic information can be easily added to the payslip and punctual replies to requests are even less costly overall. The overall communication cost for employers in EU27 135  was estimated at between 70 and 137 million EUR, i.e. an average cost per employer between a minimum of 32-135 EUR and a maximum of 53-262 EUR depending on size (see Annex 5, section 11 for details by Member State/employer size). These costs are recurring.

If the employer has not yet compared the value of work carried out by workers in its organisation, the direct cost of this measure might be more substantial. Such cost would, however, only arise from the non-respect of the current legal obligations under the EU Treaty and Recast directive. In that case, the preparatory work might be the most time-consuming and therefore costly activity under this measure, though the bulk of this cost would be one-off. The time would depend on the level of preparedness of the employer, the complexity of the organisation and the tasks performed in the different positions 136 .

In practice, the employer will build on the existing pay structure, hence no employer is likely to start entirely from scratch. Employers must review the criteria used to reward the different positions (not workers) and then group workers accordingly. The only constraint is that these criteria must not be gender biased and must apply to all workers. They can however be defined taking into account discretionary elements as long as these elements are objective and free from gender bias. Collective agreements may already provide for classifications that can further inform the process and therefore their coverage is another factor influencing how demanding the process would be and how many employers would be concerned. In addition, as set out in Option 3, the initiative could propose tools to support the classification exercise, thus limiting any burden which the measure may entail.

For employers with 50 workers or more, this classification would need to take place pro-actively in order to deliver the required information to workers. For employers with less than 50 workers, such classification would only need to be carried out in case of a request and would be limited to the category to which the requesting worker belongs.

The possible additional estimated average cost of setting up the grouping of workers would be between 217 and 1080 EUR per employer depending on employer size (taking into account that several Member States already have tools available to support the classification). It is not possible to reliably estimate the number of employers that would have to incur these costs and therefore the total cost at EU level. This action is necessary only the first year and only requires adaptations in the following years.

For Member States: the costs of this measure is negligible. It was estimated in the 2014 Recommendation impact assessment as negligible 137 , linked to adding an additional task consisting of regular monitoring of employers’ compliance and dealing with possible complaints.

Unintended effects: Granting access to information on pay could produce a ‘disgruntlement effect’, i.e. a decrease in productivity from workers who find out to be paid unfairly. Would the difference be large, workers might decide to leave the employer with a turnover cost to organisations estimated at 100-150% of the salary of the worker concerned 138 . Other concerns include upward pressure 139 on wages or pay compression 140 , and potential tensions and conflicts 141 , which may contribute to increased rates of absenteeism. While these effects cannot be excluded, it should be noted, that even without this measure, workers could in any case 142 suspect pay discrimination or misinterpret informal information on salaries. This underlines the importance of addressing possible unjustified pay differences and establishing effective communication on what justifies differences. A PayScale survey of 70,000 US employees in 2015 showed that when people know why they earn what they earn through open and honest discussions about pay, they are less likely to quit their job  143 .

There are only two empirical studies in the EU on this topic with divergent findings. Bennedsen et al. (2019) found some evidence of ‘disgruntlement’ after the introduction of pay transparency measures in Denmark, though there was no effect 144 on firms’ profitability because the reduction in productivity was offset by the lower wage growth of male employees. Gulyas et al. (2020) 145 found that in Austria, introducing pay transparency laws led to an increased retention rate of workers, pointing towards higher job satisfaction. There is no evidence on the extent to which these results can be generalised, although there is evidence that pay transparency has a positive effect on motivation (see section 2.1 and below).

Benefits: The most important benefit for workers is empowerment as they have the necessary information to assess whether they are paid in a non-discriminatory manner compared to other workers in the same organisation carrying out equal work or work of equal value – and a fortiori to enforce their right to equal pay. As to employers, while no remedial action is required unless a claim about unjustified pay differences is made, the measure may nevertheless invite them to look at their pay structures as regards the category/ies of work of equal value for which information is given.

Conclusion: This measure has a positive impact on empowering workers to enforce their fundamental right to equal pay by requiring employers to provide information automatically to workers. While being effective in reaching the first specific objective, it would be less effective in removing bias in pay structures. In addition, it would impose a burden of communicating information to a great number of organisations, including those that may not have a problem of pay inequalities. Therefore, while having a positive impact on protecting the fundamental right to equal pay, this Sub-option may be disproportionate, particularly taking into account the lack of hard data on the extent of the problem. Also, the current economic crisis as a result of the COVID-19 pandemic already puts enterprises under a lot of strain; imposing a broad measure covering all employers, including those that may have no problem, would be difficult to justify at the present time.

Sub-option 1B: Empower workers by granting a right to request pay information 

Measure 1: Transparency of salary information prior to employment

As in Sub-option 1A.

Measure 2: Right of workers to receive individual information on pay compared to their category upon request 

Under this sub-option, the right to information would only be enforced upon request from a worker. The type of impact of this measure would be the same as for Sub-option 1A; the difference would mainly lie in a lower administrative burden on employers, depending on the number of worker requests; there would be no requirement to pro-actively inform each worker on an annual basis.

Costs: It can reasonably be assumed that the communication cost 146 for a single request would be proportionally higher than the cost of a single request under Sub-option 1A/ measure 2, as there would be no economies of scale. The total cost per employer would, however, remain below the cost of the automatic right to information, at least up to the point where the loss of economies of scale in replying to single requests would make it more efficient for the company to automatically provide the information to all employees.

An accurate estimate of the number of requests is not possible 147 . Estimating this number requires hypotheses on the percentage of employees that would request the information. This depends on many factors, e.g. the level of awareness about gender pay differences, the perception of gender pay differences being justified or not , the availability of relevant comparators, the fear of victimisation etc. The number would also depend on whether or not this measure is combined with Option 2. Considering that the overall cost is not likely to remain substantial, an estimate is not provided. As a reference, in Germany, the cost of a single request (under the German law) was estimated at 20 EUR 148 and a request was filed by 4% of those entitled to ask, i.e. workers in organisations with more than 500 employees (i.e. for a company of 500 employees, the cost of the requests would be of 400 EUR).

For Member States: the costs of this measure, as in measure 2 sub-option 1A, would be linked to monitoring compliance and negligible.

Benefits: The benefits of this sub-option are, at individual level, of the same nature as in sub-option 1A, insofar as the worker has the possibility to obtain the information relevant to assess potential gender pay discrimination. However, under this Sub-option the responsibility, including risk of victimisation, left to the individual worker is larger as it is the worker who needs to initiate action to uncover gender pay discrimination. This could be countered, however, by alleviating the worker’s burden in accessing justice, especially by further alleviating the burden of proof as in sub-option 1A.

Legal change across Member States: The right to request information exists, in various forms, in 10 Member States, either directly from the employer and/or through a third party. 149 No major institutional or legal barriers to the implementation of this measure have been highlighted.

Conclusion: This measure has a positive impact on empowering workers to enforce their fundamental right to equal pay. Compared to Sub-option 1A, it shifts the responsibility of triggering transparency on the worker. As such, it may be considered less effective than Sub-option 1A. On the other hand, it gives greater consideration to the proportionality arguments and the lack of knowledge about the size of the problem: it is less demanding for employers and would be more coherent with the current economic recovery policy. In order to counter the lower effectiveness for workers, it could be accompanied by an alleviation of workers’ burden in accessing justice.

7.3.Option 2: Legislative action to create transparency at employer level

The policy sub-options identified in this section aim at realising mainly the specific objective of addressing a systemic undervaluation of women’s work at employer level by reducing the scope for gender pay discrimination attracting management’s attention to possible gender pay inequalities in wage structures. This recognises that pay inequalities do not only concern individual workers comparing their pay to co-workers of the other sex carrying out the same work or work of equal value, but can be embedded in a systemic way in the pay structures of an organisation, which may not value all relevant skills, and/or do so in a discriminatory manner.

Sub-option 2A: Equal pay certification

This certification, as developed in Iceland, ensures that ‘women and men, working for the same employer, are paid equal wages and enjoyed equal terms of employment for the same jobs or jobs of equal value, unless such differences can be justified by relevant considerations’. (IST 85).

Small employers’ exemption: The application of the standard in SMEs may be hampered by the lack of formal human resource practices (e.g. where each member of staff carries out multiple tasks and there exists no HR management) or may require hiring specialised staff. The measure may therefore involve proportionally more costs. In Iceland the measure applies to employers with more than 25 workers – a threshold which also applies for the obligation to develop equality plans at employer level – and follows a very strong and sustained political commitment at political level. To limit the burden which certification may bring for smaller employers and in coherence with the other envisaged measures, we rather propose a threshold of at least 50 employees, which allows to cover 50% of the EU27 workforce. Again, also in Iceland it appears that the decision on the threshold ultimately was a political decision, balancing workers’ coverage with employers’ burden. This balance also depends on the distribution of workers by employer size, which explains why a higher threshold may be justified in the EU than in Iceland.

Costs – The requirement currently exists only in Iceland. It was introduced without a formal impact assessment (See Annex 5, section 9 for details) and the implementation is still being rolled out, so there is no comprehensive evaluation yet. Empirical data are also scarce. The estimates of costs are based on preliminary information from employers having already applied the scheme, as shared in a Mutual Learning Seminar organised by Icelandic authorities in 2019.

For national administrations. This option is the most demanding among those proposed. As the standard was developed by Icelandic Standards, national administrations/social partners incur no costs beyond negligible adaptations to each country’s specific labour market and labour laws. However, a public authority must be tasked to manage the scheme, i.e. organise the accreditation of auditors, keep the companies’ register, possibly control the use of a logo, monitor application and levy possible sanctions in case of non-respect of the standard. This function could be carried out by an existing body or department in a ministry or an entirely new body, which could presumably also take up additional tasks related to gender equality in general. Extrapolating the cost for a voluntary labelling scheme computed in the impact assessment carried out for the 2014 Recommendation, the cost of setting up a new dedicated body may be estimated in current prices at 190,000 EUR 150 . This cost, if incurred, would come in addition to the costs estimated for monitoring compliance, i.e. a one-off cost of around 400,000 EUR for setting up a database and website and a recurrent maintenance cost of around 50,000 EUR annually. Running costs would depend on the number of new employees assigned to these functions.

For employers - A certification is more demanding than a pay reporting or a joint pay assessment presented under the next Sub-options, especially in terms of time invested and burden on the part of employers 151 . Interestingly, according to unpublished information from a survey among the first employers that introduced the standard in Iceland, it appears that employers at the beginning of the process had overestimated the costs but underestimated the time and efforts needed for the implementation 152 . The main reported challenges were the job classification, the implementation of procedures, the lack of time for implementation, and the certification process itself. The financial cost was not among the main challenges cited by employers; it was named only on 8th place out of a choice of 11 factors. Employers that already apply other management quality standards can build on those and reduce costs. Moreover, after the initial cost, the costs are likely to be significantly lower in the following years.

The estimated cost is relatively high (and substantially higher than for the other Sub-options presented below): between a minimum of 2.2 and a maximum 3.4 billion EUR for all companies with at least 50 employees, i.e. an average cost per employer between a minimum of 5,791 and 13,136 EUR and a maximum between 8,301 and 24,512 EUR depending on the size of the organisation. For comparison, the estimated costs of an audit in the UK impact assessment on the national pay transparency measures was around 15,000 EUR. The impact assessment for the 2014 Recommendation estimated an actualised amount of between 1,500 and 5,600 EUR, depending on the size of employer, for applying a voluntary labelling scheme. Finally, the more employers apply the scheme the lower the consultancy fees are likely to be.

Benefits – Based on the experience from Iceland, there are indications of behavioural changes triggered by the certification. Most surveyed employers 153 are satisfied with the standard: it increased the quality of human resource management and pay systems and highlighted pay inequalities as well as the intersectional dimension of discrimination. Employers are able to improve their job evaluation 154 systems and correct undervalued salaries. 60% of the surveyed employers corrected salaries of specific people and 11% corrected it for wage groups. 1/3 made improvements to the evaluation system or reviewed the salary system, and ¼ changed job titles or reviewed perks and bonuses. One-third of respondents also looked at equality factors not strictly required by the standard, e.g. number of women administrators, appointments to boards and committees, gender ratio in job classification, training and other professional growth opportunities within the job. Finally, over 30% which did not previously have any written procedures or quality systems developed them according to the standard. These findings correspond to the objective of the Standard to have better organised and formalised remuneration practices. The whole process required the involvement of senior management and reinforced the priority given to equality within the employer. Other benefits included increased employee satisfaction and a reduction in staff turnover. We can expect that similar benefits would come from an application at EU level.

Conclusion: This option would in principle eliminate any difference (at employer level) in pay by gender not linked to objective factors (including gender-neutral factors specific to the position or to the worker). As such, it would in particular address a systemic undervaluation of women’s work and remove hidden gender bias. It would also be very visible and increase awareness of the equal pay principle. Theoretically, there should not even be a further need to empower workers in organisations covered by certification nor to increase access to justice as the absence of differences in pay linked to gender bias is certified ex ante. In order to guarantee the right to equal pay also in smaller organisations, this option could be combined with Sub-option 1B, especially Measure 2, which ensures that workers always have access to pay information relevant to assess whether they may be victim of pay discrimination. Nevertheless, it might be that the great visibility of the measure might already motivate smaller employers to comply voluntarily with the Equal pay standard, which would reduce the number of individual requests. To optimise its effectiveness, the certification could be supported by a clarification of the key legal concepts relating to ‘pay’ and ‘equal value’ (see Option 3).

The main limitation of this measure consists in the costs and the burden it brings for employers. Taking into account the difficulty in assessing the scale of the problem and at a time of economic downturn, 155 it makes it score low from a proportionality perspective. It requires a strong commitment from the management at employer level and a strong political commitment to pass and monitor the implementation of the certification. Iceland has a longstanding tradition in this regard: employers with 25 workers or more are already required to have a gender equality plan and boards of employers with 50 or more workers are required to have gender balance (60/40 at least).

Sub-option 2B: Joint pay assessment

The joint pay assessment involves a systematic revision of pay structures combined with remedial action where needed.

Exemption: There may be an issue in implementing this measure in smaller organisations as the computation of the average pay differences can be easily distorted by outliers - few employees at the top or at the bottom of the wage distribution. In light of this possible distortion and weighing this risk against the burden of producing the assessment, the initiative would limit the obligation of pay assessment to employers with at least 50 workers. Applying such threshold, the pay assessment would apply to 50% of the EU27 workforce and an average of 1% of EU27 enterprises.

Carrying out, on a regular basis, an assessment of the employer’s pay structure and any pay differentials based on sex, would ensure a regular analysis of the proportion of female and male workers in each category doing the same work or work of equal value, information on average pay levels in those categories, and an analysis of the reasons for any pay differences between women and men.

The effectiveness of the measure would depend on the follow-up action on problems brought to light in the assessment. The Eurofound study showed significant differences between employers and workers in terms of perception of actions taken (Eurofound, 2020), suggesting the importance of addressing and communicating the pay audit results with workers. It also relies on the relevant expertise available at workers’ representatives level to be able to assess the root-causes of possible differences in pay – which might be stronger than workers’ own assessment of the same information.

This measure provides for a diagnosis helping the development of targeted action on equal pay within the organisation and would contribute to bringing to light conscious or unconscious gender bias in pay setting and discriminatory pay practices.

Ideally, an assessment should take place on an annual basis. This would allow to include short term, seasonal and interim workers where relevant and would follow from the definition of ‘pay’ encompassing all the elements going beyond basic pay that are best accounted for on an annual basis. It would also reduce the costs as it may be expected that no relevant major changes take place from one year to the next one, thus allowing employers and workers’ representatives to easily build on the results and actions from the previous year. It also allows to link the exercise to regular the workers’ evaluation, normally carried out every year, which might result in a revision of job descriptions (and possibly of pay) or the introduction of new profiles at employer’s level. Nevertheless, despite all the above reasons for an annual exercise, it may be noted that, in those Member States that have introduced a similar type of measure, the decision on the frequency of the assessment (between 1 to 4 years) is typically subject to political decision-making.

Stakeholders’ views: pay auditing/assessment is among the most preferred options for trade union respondents to the consultation strategy (95.6%), but received limited support from employers’ associations (23.5%). The ILO highlighted pay audits as a platform for change in helping to expose pay differentials and to reveal the need for structural change in the workplace. It pointed to the value of providing practical support for employers to implement measures; making employer action plans public; involving employers’ and workers’ organisations; recognising the effective contribution made by collective bargaining; and strengthening complaint mechanisms. The European Women’s Lobby recommended the introduction of mandatory joint pay assessments in all organisations, regardless of size. Support to employers with up to 500 workers is recommended by the Business & Professional Women organisation in order to create a culture of reporting and auditing on gender issues in the workplace, including pay.

Costs: The Eurofound report found that this measure would be more time consuming and costly than other transparency measures while nevertheless remaining moderate. The costs range from 88 EUR per audit in Estonia (where it is done only in the public sector) to 820 SEK (80 EUR) in Sweden (according to the reply from Sweden to the targeted survey), and to 1,390 EUR on an annual basis for the compilation of a full gender equality plan including a pay audit in Finland.

The average cost per employer (all sizes) would be between 1,800 and 2,500 EUR in their first year. This cost might decrease in the following exercises depending on the learning curve in relation to the exercise and specific context for the employer. It should be noted that the costs of an internal assessment are manifestly lower than those related to an audit carried out by an external consultant (e.g. in UK, the costs of an external audit have been calculated at around 15,000 EUR). The recurring costs relating to any subsequent assessments are expected to be lower.

The total cost for the EU 27 is estimated at between 636 and 932 EUR million, again taking into account that some Member States already apply similar measures. (See Annex 5, section 11 - for the split by Member States and employer size).

For Member States: the costs of this measure would be linked to monitoring compliance and therefore negligible.

Benefits: The cooperation between employers and workers’ representatives ensures a supported approach and leads to common action. The communication of the results contributes to a better workplace environment and has an additional deterrent effect against non-compliance or failure to follow up. This measure triggers mandatory action on the part of employers to look into their pay setting practices and address any potential gender bias in pay structures. By encouraging proactive measures, it reduces the risk of legal action for the employer and the need to respond to individual requests, hence savings of time and money. Employers could make use of tools to assess work of equal value, such as gender-neutral job evaluation and classification systems, which would effectively contribute to the implementation of the joint pay assessment.

Data protection: see Option 1, Sub-option 1A, measure 2

Legal change across Member States: several models of pay auditing are currently available in 8 Member States (BE, DE, DK, ES, FI, FR, PT, SE). Their frequency ranges between 1 and 4 years. The degree of legal change required in these Member States would vary depending on what they already have in place. The measure would need to be set up in the remaining 19 Member States which do not provide for a joint pay assessment yet.

Conclusion: In a situation where no ex-ante certification would be organised at Governmental level, an annual joint pay assessment involving workers’ representatives and employers would be the second best option to create pay transparency at employer level. The overall positive scoring of this measure by experts interviewed for the support study from across the different assessment criteria reflects the significant expected effectiveness of joint pay assessments in removing gender bias in pay structures. The institutional assessment points towards a positive conclusion in spite of challenges in relation to stakeholder resistance. There is no indication of legal barriers, including on data protection concerns as the information would not be about individual identifiable workers’ data. Overall, all perspectives converge to a positive assessment with the scope to boost equality outcomes through this measure. At the same time this option would bring the same proportionality questions as mentioned in Sub-option 2A.

Finally, as with pay certification, this measure could be accompanied by an additional measure ensuring that the right to equal pay is ensured for those workers not covered by any joint pay assessment, i.e. in employers with less than 50 workers. This option could therefore be combined with Sub-option 1A, in particular Measure 2, which ensures that workers in such organisations have access to pay information relevant to assess whether they may be the victim of pay discrimination. In addition, its implementation would be strengthened if combined with measures aimed at the facilitation of the application of the key legal concepts relating to ‘pay’ and ‘equal value’ (see Option 3). Finally, as no ex ante screening of pay structures is organised by public authorities, this option could be further supported by measures aimed at improving access to justice and other enforcement and implementation measures (see Option 3).

Sub-option 2C: Basic pay reporting combined with joint pay assessment

Measure 1: Employer obligation to carry out a joint pay assessment

This measure would trigger a transparent revision of pay structures as set out under Sub-option 2B but only on the part of larger employers. In order to create at least some degree of transparency in medium-sized employers, it would be complemented by a pay reporting as referred to in measure 2.

Exemption: Considering that employers already produce pay reports (measure 2), action may already be triggered by stakeholders, and given the costs involved of a pay assessment, this additional measure would be limited to those employers expected to have increasingly formalised human resources in-house, i.e. usually employers with more than 250 employees. This threshold is the one usually used to refer to larger organisations in the European Union. 156 It is also in between the thresholds set at national level for similar measures. 157 The study of preparatory work at national level shows that the threshold is ultimately the subject of political decision. Finally, an integral part of a pay assessment is an analysis of pay structures and therefore also of the job evaluation and classification system in the organisation, in order to ensure that all relevant skills are valued and that they are valued in a gender-neutral way. This brings benefits mostly for employers of a certain size. 158  Exempting employers with less than 250 employees from the obligation of conducting the joint pay assessment, while further reducing the administrative burden from employers (as compared to Sub-option 2A), would limit the coverage of the measure to 33% of the EU27 workforce and an average of 0.2% of EU27 enterprises. 159

Costs: The total cost of the measure for employers in the class size ‘250+ workers’ in the EU27 for ensuring the joint pay assessment can be estimated between around 90 and 125 million, 160 i.e. an average cost between 1,829 and 2,175 EUR per employer (the detailed calculations per country and employer size are in Annex 5, section 11). One could expect that these costs would decrease following a first assessment, as subsequent assessments may be based on the previous one. Also, in order to reduce costs, the assessment could be carried out every three years, considering that the pay reporting would fill in the gap between the joint pay assessments.

For Member States: as in Sub-option 2B, the costs of this measure would be linked to monitoring compliance and therefore negligible.

Benefits: As in Sub-option 2B but for less employers and workers.

Measure 2: Employer obligation to report on average differences in pay between female and male workers

The information under this measure gives a general idea about the situation regarding pay equality in the organisation. In countries where such information is published (e.g. UK, DE), it creates peer pressure, raises awareness and triggers action. It gives a basis for workers’ representatives, labour inspectorates or equality bodies to ask questions and explanations if important gender pay differences come to light.

Publishing data about average differences in pay between men and women at employer level would not in itself directly contribute to reaching the objective of empowering workers to claim their right to equal pay, but it offers sufficient details to motivate workers to request more information should they suspect gender pay discrimination. This, in addition to reputational concerns - or as a result of questions by trade unions, labour inspectorates or equality bodies - could therefore trigger action on the part of employers. It could more specifically serve as a trigger for a systematic revision of pay structures from the perspective of discrimination based on sex, even if such follow-up action would not be mandatory.

The effect of this measure could be strengthened by ensuring a centralised publication of the data by the Government, as done, for instance, in the UK. 161 The gathering of the data by the Government would further allow public authorities to assess and monitor the impact of pay transparency measures at aggregate level. It would also allow them to analyse the data to detect differences per region, per sector, etc. and to develop targeted policy measures.

The national experts who contributed to the support study were particularly positive about this measure in terms of fitting within existing institutions and its likely effectiveness (behavioural feasibility).

Two important elements affecting efficiency need to be taken into account when shaping this measure:

·Publicity on the results and follow-up/monitoring: publicity is an important mechanism to promote change and awareness of the gender pay gap. In addition, it motivates employers to act taking into account branding, image, and capacity to attract talent. There should be follow-up action in case unjustified pay differences come to light.

·Sanctions and/or positive incentives should be in place to ensure that employers provide the information.

Exemption for smaller employers: There are statistical reasons to exempt smaller organisations from pay reporting. In small and very small employers, the computation of the average pay differences can easily be distorted by outliers or by few employees at the top or at the bottom of the wage distribution. Because of this possible distortion and weighing this risk against the burden of producing the report (even if the latter is low, see below), the initiative would limit the obligation of reporting on the pay gap to employers with at least 50 workers. The impact of such a limitation on the coverage of workforce and employers concerned is the same as set out under Sub-options 2A and 2B.

Stakeholders’ views: The majority of respondents (78.4%) to the public consultation think pay reporting would be effective to better enforce the principle of equal pay. 162 Furthermore, making the pay report available to the public at large is expected to be effective by between 65% and 78% of respondents from all groups, with the exception of only a limited minority of employers/business associations and organisations supporting this measure (20%). The Member States’ consultation shows that a large majority of respondents (82%) favour this measure and selected it as the first most effective one to better enforce the principle of equal pay. No particular concerns were raised.

Costs: For employers – The reporting consists of three different steps: (1) information gathering, (2) computing the average pay differences, (3) writing and clearing the report for publication. Data on salaries are already available to the employer and the computation itself is not difficult; the most demanding task is the clearance process for the publication of the results and possibly developing a supporting narrative.

The impact assessment for the 2014 Recommendation assessed the average time needed for (broader) pay reporting at 40 hours per year. This assessment is consistent with the estimate for Germany of about 3.5 working days for a report obligation for an employer with more than 500 workers 163 and with the estimate of 23 hours in the impact assessment of the UK government for a lighter measure. Finally, the mini-survey of employers carried out for this impact assessment showed that 7 out of 11 employers (mostly with over 250 workers) expect a cost below 1,000 EUR, noting that digitalisation and data availability simplify data processing.

Our estimation of costs for the EU27 is consistent with these figures and is of an overall cost of around 131 and 214 million EUR, i.e. an average cost between 315 and 500 EUR per employer. 164 This decreases in the following years to between 20 and 36 million EUR, i.e. an average of less than 100 EUR per employer (the detailed estimated amounts per Member State are available in Annex 5, section 11).

For Member States: The gathering and publication of the reports by Member States, including an analysis of the aggregate data, would entail costs on the part of Member States. The minimum costs are related to building a database to monitor compliance. This can be estimated as one-off cost of around 400,000 EUR for project and delivery of the website and recurrent maintenance of around 50,000 EUR annually. 165 An employee should be assigned full time to monitor compliance. The annual median earning of public employees in the EU is around 32,000 EUR. Member States could generally rely on existing administrative entities and/or hire new staff. Setting up a new dedicated body would be the most expensive option: the costs of setting up an equal pay competence centre in Estonia was estimated at around 1 million EUR over 4 years (including IT investments and administration costs, information, communication and other support activities). Such a body would be likely to carry out broader tasks than simply collecting and monitoring data, most likely it would also deal with additional gender equality policies.

Finally, there are no costs involved for trade unions/workers’ representatives in regard to the implementation of the measure. There may be work involved by these organisations in the follow-up to the publication of the pay gap information, in particular when questions arise regarding a specific employer or for the purpose of wage negotiations. However, any such follow-up action is voluntary. Nevertheless, it should be noted that pro-active monitoring of the published data by trade unions/workers’ representatives would significantly increase the effectiveness of this measure.

Unintended effects: Employers’ accountability for the existence of average gender pay differences in their organisation would make somewhat lower managerial wage discretion and could even influence hiring/firing decisions. IT applications allowing to assess the impact of any prospective hiring on such differences are already available for free 166 . The effect could be to reduce the discretion of the employer, also given the larger publicity of the information. For the same reason, however, more accountability could induce employers’ voluntary implementation of gender equality plans, also considering that workers’ representatives would have relevant information as basis for discussion.

Benefits: Pay reporting has intangible benefits by making employers more conscious of equal pay issues, with an added peer pressure element that could motivate a stronger focus and awareness on the gender pay gap not only among workers but also among other stakeholders e.g. NGOs, national and social media. Depending on the monitoring and action by relevant stakeholders, the measure may be expected to trigger action at employer level. However, it should be noted that the main weakness of this measure, if adopted on its own, would be the lack of pro-active remedial action by employers in case the data were to show possible pay inequalities in the organisation. Follow-up action would only be triggered in case of questions by workers, their representatives, or other stakeholders.

Legal change across Member States: several models of pay reporting are currently available in 11 Member States (AT, BE, DE, DK, ES, FR, IT, LT, LU, NL, PT). Some changes may be required in these Member States to adapt their current reporting obligation to that foreseen in this initiative. The measure would need to be set up in the remaining 16 Member States.

Alternative sub-option: Pay transparency realized through action by national administrations

It may be feasible, in some Member States, that national administrations are able to gather and interlink the necessary data allowing for a computation of the gender pay gap per employer. Where that is possible, such administrative action could replace the obligation for employers to do so.

Technically, this is only possible if administrative data matching employers’ (firm level) to workers’ (individual level) data, including benefits in cash and in-kind, are available 167 . It would require interlinking data from several public administrations (such as tax inspectorates, social security offices, etc.). According to a study carried out by the OECD, some countries should already be able to carry out this task 168 . All 20 surveyed Member States, except Czechia, record firm-level information; 12 countries already link firm-level with individual-level data 169 and five expect to be able to do so shortly (Austria, Bulgaria, Ireland, Italy and Spain). According to this study, Czechia, Slovenia and Slovakia are not yet able to link the data, and have no specific plans to do so soon. Despite this study, in the targeted consultation only six Member States out of 20 confirmed that their existing national registers (tax, social security, etc.) would allow to compute the gender pay gap at organisational level 170 . Computation by national organisations already happens for instance in Estonia and Denmark 171 . In the public consultation, Member States found pay reporting done by organisations more efficient than reporting by administrative authorities. Almost 80% of employer organisations replying to the public consultation found it ineffective for the labour inspectorate/equality body to perform automated checks based on employment, tax, and/or social security registers.

In addition, the computation of average pay differences at employer level by administrative authorities may have an additional drawback. It would leave employers in a reactive role, just in case they would be flagged in the system. If employers are involved in the reporting themselves, stronger buy-in and commitment from the employer’s management level may be expected, eventually leading to a mind shift at employer level. Also, employers may wish to control and handle the data themselves, particularly in light of the relative small burden that calculating the average gender pay differences at employer level entails.

Nevertheless, in light of the concerns expressed by employers’ associations as to the possible burden on employers in the initiative and in light of the fact that certain Member States may be able to alleviate such burden by achieving the results of this measure through administrative computation of the pay gap at employer level, the initiative could allow the flexibility for Member States to impose this as an obligation on employers or carry out the task at governmental level.

Overall conclusion of Sub-option 2C

Under this option, the regular joint pay assessment would be limited to large organisations of at least 250 workers and to a tri-annual exercise in order to limit burden on employers. A lighter obligation would be added for those organisations of at least 50 workers. This lighter annual exercise, based on available human resource data, would create a framework that would bring to light potential pay inequalities in organisations. Even if the data gathered would not be very detailed and may not trigger follow-up action from employers in the absence of action triggered by stakeholders, the publication of such data would give indicative information to stakeholders and may be expected to create peer pressure among employers. Therefore, this Sub-option would be somewhat effective in empowering workers and relatively effective in removing bias in pay structure. It would represent a proportionate measure having limited cost, imposing a limited burden on employers and therefore being relatively coherent with current economic recovery policy.

For all workers to have the possibility to access the specific information relevant to establish gender pay discrimination, the pay reporting and assessment measures above could be accompanied by the measure envisaged under Sub-option 1B, especially Measure 2.

As in Sub-option 2B above, the joint pay assessment could be supported by a clarification improved implementation of the key legal concepts relating to ‘pay’ and ‘equal value’ (see Option 3). As no ex-ante screening of pay structures is organised by public authorities, this option could be further supported by measures aimed at improving access to justice and improved enforcement and implementation (see Option 3).

Sub-option 2D: Strengthened pay reporting and joint pay assessment if pay differences which cannot be justified by objective, gender-neutral factors

Measure 1: Employer obligation to report on average differences in pay between female and male workers by worker category

The strengthened pay reporting envisaged under this option would require a slightly heavier investment on the part of employers as they would need to calculate, in addition to the average gender pay differences in their organisation overall, more specific pay differences among workers carrying out the same work or work of equal value. However, it should be noted that this measure requires processing the same information as envisaged under measure 2 in Sub-option 1A and 1B, but for all categories of workers. Depending on the situation, the employer may therefore already have the information available. Compared to the basic reporting under option 2C, access to this information is more valuable to precisely identify possible unjustified pay gaps and the reasons behind such gaps.

Exemption: Because of the heavier burden on employers and in order to remain coherent with the other thresholds chosen for the pay transparency measures, this measure would exempt small and medium sized employers, i.e. employers with less than 250 workers. This measure would therefore cover about 33% of the EU27 workforce and an average of 0.2% of EU27 enterprises. Again, ultimately the threshold is subject to political decision.

Costs: The total cost for the employers would amount to around 26 and 50 million EUR, i.e. between a minimum of 379-508 and a maximum of 721-890 EUR per employer depending on the size. These amounts are higher by class-size than those for Sub-option 2C measure 2 (basic pay reporting for 50+), but they are lower as regards the total cost since a larger share of employers would be exempted. These costs would also be reduced in the following years.

For Member States: as in Sub-option 2C, the costs of this measure would be linked to the gathering and publication of the reports by Member States, including an analysis of the aggregate data and monitoring compliance.

Benefits: The measure gives a more precise insight on the side of employers on possible discrimination and bias of which they may not have been aware (valuation of women’s work). As such, the measure creates a better enabling environment to trigger change, building an environment of transparency and trust. This would have a result on the employer’s reputation, making it more attractive for new applicants and current workers (retention of talent), and pointing to higher job satisfaction. 172 As an indirect effect it could also push the management to reflect on gender equality policies more generally. For workers, the measure gives them the information necessary to assess whether or not they may be subject to discrimination. This measure actually provides workers with the information foreseen under Option 1 in an automatic way, only in companies with more than 250 workers. For Member States, long-term intangible benefits relate to raising awareness and stimulating political and social debate on gender equality issues at employer level. The main weakness of this measure remains, however, that on its own it does not ensure follow-up action on the side of the employer, as it does not impose corrective measures. For that reason this sub-option envisages targeted joint pay assessment to be applied where unjustified pay differences have been discovered.

Data protection: see Option 1, Sub-option 1A, measure 2.

Measure 2: Employer obligation to carry out a joint pay assessment if pay reports show pay differences which cannot be justified by objective, gender-neutral factors

To ensure follow-up action in those organisations that show a problem of pay inequalities through the pay reporting conducted under measure 1, pay reporting would be accompanied by the obligation to carry out a joint pay assessment, for those employers showing a difference of 5% in any category of workers doing the same work or work of equal value which cannot be justified by objective factors. This measure would build on the pay report, not duplicating transparency obligations, and would remedy the main weakness of measure 1 in a targeted manner. The 5% criterion ensures statistical significance. In statistical terms 5% level gives an indication whether a phenomenon is worth paying attention to as something likely to occur beyond a mere coincidence. In most areas of scientific research it is conventionally set at 5% level.

Exemption: This measure would exempt employers with less than 250 employees, for similar reasons as set out under Sub-option 2C. Assuming that voluntary follow-up action may have been triggered by the publication of the pay report (measure 1), the joint pay assessment would only be imposed as a mandatory measure for statistically significant differences. As such, the measure would target follow-up action to those large employers where the pay report shows a manifest problem in regard to pay equality.

Cost: Based on the explanations on method followed to calculate the cost of joint pay assessments (cf. supra), the average cost of such assessment can be estimated between a minimum of 1,180-1,724 EUR and a maximum of 1,911 and 2,266 EUR depending on size. These costs are lower than the average cost of option 2B, as they are partly covered from the pay reporting under measure 2D1 for which employers would already have collected and published part of the information. This cost may be expected to decrease in any subsequent exercises or not be necessary at all, should the unjustified difference in pay be eliminated as aimed to by the joint pay assessment.

For Member States: the costs of this measure would be linked to monitoring compliance and may be considered negligible.

Benefits: The main benefit of this measure would be that follow-up action is targeted to only those employers where problems of pay inequalities are brought to light by pay reporting. See also Sub-options 2B and 2C above in relation to this measure.

Overall conclusion on Sub-option 2D

The pay reporting envisaged under this option would be more effective than under Sub-option 2C, as it would bring to light not only possible pay inequalities in organisations, but also unjustified pay differences in specific categories of workers doing the same work or work of equal value. While this information may trigger voluntary follow-up action, a mandatory joint pay assessment would be required only in case of manifest problems of pay inequalities. By limiting the obligation on employers and focusing more stringent intervention on employers which have a problem relating to pay inequalities, this Sub-option presents the most proportionate solution, taking into account current limitations regarding the measurement of the scale of the problem and the strain on companies in this time of economic downturn. The limitation of both measures to employers with at least 250 workers, while limiting their scope, would limit to the maximum extent the cost and burden on employers.

However, in order to ensure the adequate protection of all workers also in organisations with less than 250 workers, this sub-option could be combined with one of the Sub-options of Option 1; this could especially ensure that all workers have access to the information necessary to assess whether they are victim of gender pay discrimination.

In addition, in order to increase its effectiveness, the joint pay assessment could be combined with Option 3.

7.4. Option 3: Legislative action to facilitate and enforce the existing legal framework

This option aims to address the problem of inconsistent and inadequate application of key concepts relating to equal pay. It does so by allowing workers and employers to more easily point to possible comparators of the other sex doing work of equal value. Access to justice and enforcement of the existing legal framework would be improved not only to tackle individual instances but also systemic discrimination. As such, it would contribute to addressing the undervaluation of women's work and to empower workers to claim their right to equal pay, even without pay transparency.

Measure 1: Facilitate the application of the existing key concepts of ‘pay’ and ‘work of equal value’ 

Stakeholders’ views: The public consultation showed broad consensus that the concept of ‘work of equal value’ requires clearer definition and improved implementation if pay discrimination legislation is to be effectively implemented and enforced. This could be done, it was highlighted, by ensuring the gender neutrality of job evaluation and classification, a measure that had the highest consensus in terms of effectiveness across all groups of respondents (80%). However, the targeted consultation with social partners revealed antagonist views between employers’ organisations and trade unions. While a broad majority (87%) of trade union respondents view job evaluation and classification as effective, especially in light of the direct involvement of social partners, the majority (58%) of employers’ organisations view this as ineffective.

Costs: clarification of the ‘pay’ concept and incorporating the CJEU criteria into the new legislative proposal would not bring, as such, a change to the legal framework, as these criteria already apply through the case law today; it would only enhance the visibility of these criteria in the law. Therefore, it would not entail direct costs. As for the support for companies in the implementation of the concept of ‘work of equal value’, flexibility would be left to Member States concerning the choice of the relevant tools. Any costs would largely depend on the present availability of supporting instruments and therefore cannot be calculated precisely. For instance, gender neutral job evaluation and classification systems can be implemented at company level (which may take a few days in a larger company) but equally at collective bargaining level (which then depends on negotiations which may take months or years).

Benefits: Legal clarity aimed at a uniform application of key concepts would significantly support the proper implementation of the current legal framework, reducing implementation gaps.

The availability of clear criteria in national legislation would help claimants to establish a valid comparator and evaluate whether or not they are treated less favourably than the comparator performing the same work or work of equal value. It would also allow employers to better categorize jobs based on objective criteria and ensure that such jobs are remunerated in a just and bias-free way. It is of particular relevance to note that in the consultations carried out for this impact assessment, employers’ associations raised concerns on the difficulty to assess what constitutes ‘work of equal value’, knowing that they already are obliged by law to apply that concept. It should be stressed that the assessment of equal value is, under the current legal framework, an assessment at company level, and not at sectoral level (which is out of the scope of the present initiative).

Legal change: No legal barriers have been identified for the implementation of this measure.

Conclusion: This measure could help workers to establish a valid comparator and could allow employers to ascertain that their pay settings are based on bias-free and objective criteria and may therefore contribute to achieve both the specific objectives of empowering workers and addressing the systematic undervaluation of work done by women. Furthermore, clarification of the ‘pay’ concept would clarify that the equal pay principle applies to all components constituting pay, whether in cash or in kind, which the workers receive directly or indirectly, in respect of their employment from their employer. These measures would benefit an improved implementation of the current legal framework, but could also be used to support the implementation of new pay transparency measures.

Measure 2: Improved access to justice for potential victims of pay discrimination

Victims’ representation (legal standing and representative actions): 

-Legal standing

Stakeholders’ views: More than three fourth (76%) of respondents from all groups to the public consultation found that strengthening the mandate of equality bodies as well as workers’ representatives to support victims would be an effective measure to improve access to justice for individual victims.

Costs: this measure consists of a legislative change at EU level which entails no direct costs. In those Member States where equality bodies and workers’ representatives do not yet have legal standing, the measure might increase to some extent their workload assuming that they might, at least in the medium term, be asked to handle more claims. However, combined with the measure below on representative actions, they could ensure economies of scale as they could act on behalf of a group of victims.

Benefits: Ensuring legal standing for equality bodies and workers’ representatives would contribute to removing one of the most important obstacles which victims of pay discrimination face when claiming their rights, especially the burden of carrying a legal claim and the fear of victimisation.

Legal change required: Equality bodies have the power to bring cases to court or to represent the claimant before courts or administrative bodies in only 8 Member States (DK, FI, HU, IE, IT, SE, SI, SK). 173 This measure would thus require a legal change in 19 Member States where equality bodies presently do not have the power to act on behalf of individuals. As regards, workers’ representatives, they have legal standing in most Member States; the measure would entail a change only in four Member States (DE, EE, FI, RO). 174

-Representative action

Article 17(1) of the Recast Directive contains a general requirement for Member States to ensure that judicial procedures are available for victims of sex discrimination in employment matters. Member States have mainly not implemented this provision, relying on their common civil judicial procedures laid down in national law. As a result, in most Member States victims of discrimination can only pursue an infringement of the right to equal pay on an individual basis. Experience shows, however, that structural biases in pay structures in a given organisation may affect several workers in the same way.

Costs: introducing this measure as such would not entail costs. It would enable economies of scale in the preparation and litigation of equal pay cases and may reduce coordination and transaction costs of bringing victims together for redress purposes. Qualified entities would experience procedural efficiencies from being able to assess equal pay claims in a single procedure, enabling them to bear the costs of preparing a single action. If the action is successful, it will of course entail costs, in the form of proceedings costs, sanction and/or compensation, for infringing employers. Again, however, such costs would be incurred only to comply with their current obligations under EU law.

Benefits: collective claims are a way to facilitate actions that would not otherwise have been brought because of procedural and financial barriers or fear of victimisation. In particular, allowing collective complaints can reduce the fear of reprisals and victimization and reduce the overall costs for complainants. 175  Furthermore, collective claims have the potential to improve compliance with pay transparency measures especially regarding businesses sensitive to reputational damage. They also allow more easily to uncover systemic discrimination and create visibility of equal pay and gender equality in society as a whole. Representative actions therefore have an important impact on employers' awareness and willingness to act preventively. This measure combined with the other proposed enforcement measures may be expected to have a preventive and deterrent effect and reduce costs linked to legal action.

Legal change: as collective action is currently only possible, in discrimination cases, in seven Member States (DK, ES, FR, IT, LV, NL, SI), 176 this measure would bring change in the legal framework of 20 Member States.

While it would be useful to extend the right of action also to associations, organisations or other legal entities pursuing the enforcement of the right to equal pay, the wide variety of such organisations and the interests they represent call for caution. Such alternative was therefore not retained.

Strengthened and new remedies (compensation and injunction orders)

Costs: The introduction of these requirements does not entail costs as such. Strengthened rules on compensation and court orders may entail costs for non-compliant employers, as would be the case with any non-compliance with the law. Some minimum costs of informing judicial authorities may be incurred; these may be covered, among others, by trainings such as those organised by the Academy of European Law (ERA) funded by the European Commission.

Benefits: a strengthened requirement to compensation will provide incentives for victims of gender pay discrimination to seek justice and uphold their right to equal pay. Court orders to stop infringements or impose remedial action will allow to address infringements of the pay transparency rules and systemic undervaluation of women’s work. In addition, these measures would contribute to dissuade employers to infringe the workers’ right to equal pay and motivate them to pro-actively undertake action.

Minimum standards on limitation periods

Costs: The introduction of this requirement does not entail any costs as such, except for training of judicial authorities which may have to apply the new rules (see above).

Benefits: for victims to have sufficient time to gather the necessary evidence to make their case and to take action once they become aware of the potential pay discrimination. Under the current framework some victims may already have moved on to another job and many other victims decide simply not to act because of the unsurmountable difficulties involved in taking the necessary action within the required time limits. Action on limitation periods requiring minimum standards would therefore enhance an effective access to justice.

Legal change: Some Member States (DE, EE, LV, SI) 177 have very short time limits to bring a case to court, from 30 days to 3 months. Other Member States have longer limitation periods from 3 years (AT, BG, CZ, HU, LT, PL) to 5 years (BE, FR, NL). 178  

Support to carry legal costs

Stakeholders’ views: respondents in the public consultation conducted for this initiative identified legal costs as a relevant enforcement problem in relation to pay discrimination. A large majority of respondents (92%) considers the provision of support to be important in order to lower these costs and enable action in pay discrimination cases.

Costs: Some minimum costs of informing judicial authorities may be incurred (see above).

Benefits: The possibility for prevailing claimants to recover the proceedings’ costs while not having to bear the costs of a successful defendant would alleviate one of the most important barriers faced by victims to enforce their rights and have access to justice. This rule would not apply to claims brought in bad faith, which are clearly frivolous or when the non-recovery by the defendant would be considered unreasonable under the circumstances. This rule would be innovative in most Member States, as it deviates from the normally applicable loser pays principle 179 . However, it is justified in the specific case of pay discrimination because of the particular vulnerability of victims in these cases.

The envisaged alternative that the claimant would recovers proceedings’ costs from the defendant whether successful or not would pose questions of fairness and proportionality and was therefore not retained.

Overall conclusion of Measure 2:

The measure could contribute to achieving both the specific objectives of empowering workers to enforce their right to equal pay and addressing the systemic undervaluation of women’s work at employer level that would potentially come to light. Furthermore, the measure may be effective in tackling the problem driver of insufficient access to justice and would trigger action to remedy systemic discrimination overall with clarification of what constitutes full compensation and providing for injunction orders in case of infringements.

However, while these measures may contribute to improving the existing legal framework, their effectiveness would remain rather limited if applied on their own because potential victims of gender pay discrimination would still lack crucial information on pay to bring gender pay discrimination claims to courts.

Measure 3: Other measures to enhance enforcement and implementation

Penalties

Costs: Strengthening rules on sanctions/penalties would require some minimum costs of informing the authorities that have the power to sanction under national law (see above).

Benefits: The proposed measure would strengthen the existing minimum standards on sanctions/penalties regarding gender pay discrimination across the EU and further the deterrence effect for employers engaging in illegal behaviour regarding pay setting practices and gender pay discrimination. At the same time, it would have a preventive effect in stimulating employers to comply pro-actively. 

Conclusion: this measure could be effective in tackling the problem driver of deficient enforcement of the right to equal pay. It would contribute to achieving the specific objective of addressing the systemic undervaluation of women’s work at employer level by sanctioning the infringement of the equal pay principle by employers and stimulating employers’ compliance. However, its effectiveness would remain limited if applied on its own because potential victims of gender pay discrimination would still lack crucial information on pay to be able to bring gender pay discrimination claims to courts.

Involvement of social partners

The effect of this measure requiring Member States to take appropriate measures ensuring that the implementation of the rights and obligations under this initiative be discussed with social partners would depend on the culture of social dialogue in the Member States. If it would lead to equal pay matters being taken up in collective bargaining, it would concern the share of the workforce that is covered by collective agreements. This percentage varies largely in the EU. In 11 Member States the largest share of workers in the country are not covered by collective agreements. Collective bargaining coverage ranges from 7% in LT to 98% in FR and AT. 180 The level at which discussions take place also influences the impact of the measure, with industrial and national level being the most common modalities. If it is at employer level, social partners could discuss employer-specific implementation and identify targeted solutions to reduce gender-based pay inequalities. As a result, action at all levels of social dialogue is useful. Such action would increase in effectiveness if supported by the information made available through pay transparency measures.

Expert views. The study carried out for this impact assessment showed that this modality has the most positive score of all by experts in terms of effectiveness in addressing the problem of pay inequalities. The experts’ scores on behavioural feasibility or effectiveness also underline positive outcomes. However, despite the overall positive scoring of this measure across the different perspectives, the importance attached to the autonomy of social partners on the one hand and the weak bargaining culture in some Member States on the other hand lead to a greater spread of opinions than for some other modalities (see Annex 5, section 8 and EY, 2021). Institutional and legal experts, social partners and individual Member States raised these concerns. In order to respect the autonomy of social partners, the initiative would not impose an obligation of result but only an obligation on Member States to take appropriate measures ensuring that the implementation of the rights and obligations under the initiative is discussed in social dialogue, without prejudice to the autonomy of social partners and in accordance with national law and practice 181 .

Stakeholders’ views: the obligation to include social partners in the implementation of equal pay matters was viewed as being effective to better enforce the principle of equal pay by 81% of respondents from the public consultation. The social partners’ targeted consultation shows that employers’ organisations and trade unions have opposite views regarding regular equal pay bargaining. A large majority of the trade unions respondents (85%) answered that this measure would be effective to better enforce the principle of equal pay while 64% of employers’ organisations answered that this measure would be ineffective. However, outside the context of collective bargaining as such, the involvement of social partners is generally considered positive.

Costs: In accordance with the impact study for the 2014 Recommendation, the average annual administrative cost of this measure for employers may be assessed as inexistent since the obligation only concerns social partners. For social partners the consideration of equal pay issues does not imply any significant costs, as it is part of their institutional discussions with no obligation of result. If the discussion would include matters such as the desirability of gender-neutral job evaluation and classification tools, this would require a minimum level of expertise for both parties. Several experts signalled that this is not necessarily always the case; some training would therefore be useful.

For Member States, this measure could require ensuring that social partners are involved in the implementation of the rights and obligations under the directive. The obligation could be monitored by labour inspectorates or by other competent authorities as part of the regular monitoring and enforcement of the equal pay legislation. Estimates on the time spent to monitor social dialogue are not available in the literature. If Member States would go beyond the obligation to ensure debate on the matter, and go into the monitoring of collective agreements themselves, it may be referred to the impact assessment for the 2014 Recommendation which reports that the monitoring of approximately 2,000 company level collective agreements would take 100 hours. That would mean an average updated amount of around 15,000 EUR per year per Member State. However, this would not be mandatory under the current initiative.

Benefits: The measure would raise awareness of equal pay matters among social partners, key stakeholders in ensuring protection of the right to equal pay. As a result of this measure, employers and trade unions could e.g. present a diagnosis of the situation, identify a list of possible measures to address equal pay issues and/or the discussion could feed into wage negotiations. In the context of this initiative, this measure contributes to addressing the systemic undervaluation of women’s work.

Legal change across Member States: In principle, this measure would not require legal change as social dialogue is available at national level. The measure would only require the inclusion of equal pay in the existing social dialogue and support the implementation of equal pay as such. As the 2020 evaluation identified, equal pay matters and pay audits have so far been explicitly included in collective bargaining only in 3 Member States (BE, FR and LU).

Conclusion: This measure has the potential to address the systemic undervaluation of women’s work at employer level with the development of active social partnership and inclusion of equal pay matters, especially ensuring a proper implementation of the equal pay right to work of equal value as a key element of social dialogue. It would not impose an obligation of result but only an obligation on Member States to ensure that the implementation of the rights and obligations under the initiative is discussed with social partners, without prejudice to the autonomy of social partners and in accordance with national law and practice.

8.How do the options compare / preferred option

8.1. Effectiveness

Option 0 would not be effective in achieving the objectives overall as no action would be initiated to tackle the problem drivers and issues highlighted. This means, at best, continuing at a very slow pace towards better enforcement of the equal pay principle.

Option 1 Sub-option 1A scores high on achieving the specific objective of empowering workers to claim their right to equal pay as it would allow workers to have the necessary and relevant information on pay and to be most effectively protected against potential victimisation. On its own, however, it would not address the problem driver related to the persisting bias in pay setting mechanisms and valuation of women’s work because of the lack of structural corrective measures. As such, it does not score well in relation to the second objective, even if it would create the conditions for potential remedial action; employers would be obliged to gather the information on a regular basis, at least in organisations with at least 50 workers, which might potentially incite them to analyse their pay structures more closely. In order to more effectively remove systemic bias in pay structures, this option could be combined with either of the Sub-options 2B, 2C or 2D.

Option 1 Sub-option 1B scores lower than Sub-option 1A but is still significant to achieve the specific objective of empowering workers to claim their right to equal pay as it would allow all workers to request the relevant information on pay to their employers. This option would reduce the administrative burden on employers by relieving them from the obligation to inform workers on a regular basis. Workers would be protected through a strengthened reversal of the burden of proof countering a possible fear of victimisation. Even less than Sub-option 1A, however, this sub-option would not address the problem driver related to the persisting bias in pay setting mechanisms and valuation of women’s work. Indeed, because of the lack of systematic gathering of the information by employers, the latter would be even less triggered to analyse their pay structured on hidden pay inequalities. For similar reasons as Sub-option 1A, this Sub-option could work better if combined with either of the Sub-options 2B, 2C or 2D.

Option 2 Sub-option 2A scores very high in effectiveness for the two main specific objectives and therefore also high on the general objective. It would contribute to empowering workers ex ante (rather than ex post, once a discrimination case arises), it would strongly address the systemic undervaluation of women’s work by an ex-ante screening of pay structures thus eliminating gender bias in pay setting practices at employer level. The envisaged exemption, however, would reduce its impact; in order to protect all workers, this Sub-option could be combined with one of the Sub-options under Option 1, for those organisations not covered by the certification.

Option 2 Sub-option 2B equally scores very high on achieving the two main specific objectives and thus on the general objective. The joint pay assessment is based on information needed by workers to assess possible individual discrimination which would be made available to them, even if the assessment would not be carried out ex ante. If the information is provided in a transparent manner allowing workers to place themselves individually in the context of the provided salary information, it could also score well on the first specific objective. In addition, the joint pay assessment with workers’ representatives would oblige employers, beyond any individual instances of discrimination, to focus the attention on potential gender bias in their pay structure and more importantly to take remedial action in case discrimination and bias are uncovered. However, the envisaged exemption would reduce the impact of this Sub-option; in order to protect all workers, it could be combined with one of the Sub-options under Option 1, for those organisations not covered by the joint pay assessment.

Option 2 Sub-option 2C scores lower than Sub-option 2A or 2B. Indeed, pay reporting as envisaged under this Sub-option would not give workers the information needed to assess possible discrimination in their individual case, but it gives a basis for workers, workers’ representatives, labour inspectorates or equality bodies to ask questions and explanations on overall gender pay differences in organisations and acts as an incentive for companies to assess those more in detail. Any systematic revision of pay structures from the perspective of gender pay discrimination would nevertheless depend on the willingness to do so by the employer based on peer pressure or questions as referred to above, except for large organisations with at least 250 workers that would be subject to a joint pay assessment. In order to be more effective on the first specific objective and to ensure the protection of workers in all organisations across the EU, this Sub-option could gain in effectiveness if combined with any Sub-option under Option 1, especially in those organisations not covered by the joint pay assessment.

Option 2 Sub-option 2D scores lower than Sub-option 2C but is still relevant to achieve the two specific objectives. It covers less employers but the inclusion of the additional key element in the pay reporting allowing to better uncover pay discrimination makes this reporting more useful than the one under Sub-option 2C. The limitation of the requirement to carry out a joint pay assessment to those organisations where problems would be evidenced, would permit to target action on the part of large employers where such action is mostly needed (relevant pay difference exceeding the statistical significance threshold). The envisaged limitation of the measures to organisations with at least 250 workers would, however, reduce the impact of this Sub-option. Again, given the exemptions envisaged and to ensure that all workers, also those employed in small and medium-sized organisations, have tools to enforce their right to equal pay, it would gain in effectiveness if combined with any Sub-option under Option 1.

Option 3 scores relatively high on achieving both specific objectives. It does so by allowing workers to more easily point to possible comparators of the other sex and employers to gain insight in their pay structures through a better understanding of what is ‘work of equal value’. Access to justice would be improved, not only to tackle individual instances of gender pay discrimination but also systemic discrimination. However, on its own this option would only partially achieve the objectives as it does not address the primary obstacle consisting in the lack of pay information necessary to assess whether one is victim of pay discrimination or whether there is gender bias in pay structures. Therefore, in order to contribute to the overall objective, it should be combined with other measures foreseen under Options 1 and/or 2.

8.2. Efficiency

Option 0 scores null in terms of efficiency because it would not have any more costs/benefits than the current framework.

Options 1 Sub-option 1A scores relatively well on efficiency in relation to the first specific objective as it has the most benefits/social impacts on workers and brings moderate economic costs for employers. Thanks to the exemptions tailored to the size of employers, the costs and burden are minimised for smaller employers. Even if workers would not automatically receive information in smaller organisations, an effective protection of their rights would still be ensured by the reinforced burden of proof mechanism and other access to justice tools. On the other hand, this option would still impose a burden of communicating information to a great number of organisations, also those in which there may be no problem relating to pay inequalities. The Sub-Option does not score very well on the second specific objective; it would create costs for employers without triggering a systematic revision of pay structures, even if it would create the necessary conditions for such revisions.

Option 1 Sub-option 1B scores better on efficiency in relation to the first specific objective. It achieves the objective of empowering workers, while maximally reducing the burden on employers. As such, it would constitute a more proportionate measure, in that information would only need to be provided in case of suspected problem. On the other hand, it would score lower in regard to the second specific objective, as it would seem unlikely to trigger any systematic revision of pay structures in organisations.

Option 2 Sub-option 2A scores lower than the other options on efficiency as the costs and burden of the option are much higher. Although the mandatory certification scheme has the potential to drastically reduce pay discrimination, its application to all organisations with at least 50 workers would include those that might not have a problem relating to pay inequalities. In addition, its application in Iceland has not yet been evaluated, therefore it is difficult to fully assess the rather high costs against the expected benefits.

Option 2 Sub-option 2B scores better than Sub-Option 2A but still lower on efficiency than Sub-options 2C and 2D. While it would have a strong influence on addressing gender bias in pay structures and has the potential to empower workers in organisations with at least 50 workers, it would still impose a noteworthy burden on medium sized employers, even if its cost would be lower than the cost of Sub-Option 2A. In addition, it suffers from the same drawback as Sub-Option 2A in that it would entail costs and burden on the part of all employers, including those which may not have a problem of pay inequalities in their organisation.

Option 2 Sub-option 2C. The limited average time of pay reporting added to the possible simplification of the data processing through digitalisation shows that the overall costs of pay reporting as envisaged under this Sub-option remains low. This is proportionate to the benefits it entails in terms of peer pressure and providing at least an indication of possible pay inequalities triggering questions from stakeholders. While the measure may involve more costs for Member States in regard to the monitoring of the data, they could generally rely on existing administrative entities. In term of efficiency, some Member States may be able to alleviate the administrative burden on employers by achieving the results of this measure through administrative computation of the pay gap at employer level. The joint pay assessment for employers with at least 250 employees would ensure that follow-up action takes place in large organisations; its costs remain moderate compared to the benefits it brings. However, the assessment would be imposed on all such organisations, even those that may not have a problem with pay inequalities, thus raising similar questions regarding proportionality as Sub-Options 2A and 2B above.

Option 2 Sub-option 2D would entail the least costs and burden for employers, ensuring transparency on pay structures in large organisations and providing targeted action where problems of discrimination and bias are detected. Benefits would be limited to workers in organisations with at least 250 employees. Nevertheless, this Sub-option could be considered the most adequate at this moment in time, in light of the lack of hard data on the extent of existing gender based pay discrimination on the one hand and the current economic downturn as a result of the pandemic on the other hand. It would constitute a first step, allowing to gather more data at least in large organisations and awaiting circumstances where employers could more easily carry the burden of more effective pay transparency measures. By laying down minimum standards targeting action only where necessary, this Sub-option would reach the second specific objective in a more proportionate manner. However, its efficiency in relation to the first specific objective would remain limited.

Option 3 scores relatively high on efficiency because the benefits and social impacts would be higher than the costs involved. The envisaged measures consist of legislative changes which entail no direct costs while providing strengthened protection and support to potential victims of discrimination and deterring potential infringers.

8.3. Coherence

In terms of coherence, all the options considered under this initiative are coherent with the social and economic goals of the EU as described in the Section 1 and Section 3.2 of this impact assessment. Together with other EU action such as the Work-Life Balance Directive, the proposed Gender Balance on Corporate Boards proposal, the EU’s Skills Agenda, it contributes to combating the root causes of the gender pay gap. It is coherent to the Gender Equality Recast Directive, to which it would constitute a lex specialis (see Section 8.8 below). As a targeted binding measure, the initiative would address the most problematic issue identified by the 2013 implementation report of the Recast directive – the implementation of the principle of equal pay between women and men. It does so by way of measures which already exist to a certain extent in a number of Member States. This proposal is also coherent with the initiative aimed at increasing companies’ reporting of relevant non-financial information 182 . It is consistent with and supported by the EU minimum wage initiative 183 and the upcoming sustainable corporate governance initiative 184 .

The initiative is coherent with the Commission's commitment to recovery from the economic crisis resulting from the COVID-19 pandemic. It seeks to achieve the proportionate balance between protecting workers’ rights and moderating the burden and costs of pay transparency for employers. Through a combination of measures, it may be ensured that a minimum level of transparency would exist in all organisations in the EU, even in small and medium-sized enterprises. The latter would thus not necessarily suffer in terms of image as attractive employers or ability to retain talent. Of course, they may always go for higher levels of transparency and remedial action if they wish to do so.

At different stages of the consultation process, including the consultation related to the 2020 Evaluation, Member States and stakeholders expressed some concerns, in particular as regards the possible interference of EU-level intervention with Member States’ different labour market models, different levels of social partners’ participation in the area of wage setting and related employment relations as well as the autonomy of social partners. All options are designed in a manner that effectively addresses the objective of the intervention while respecting different features of national social dialogue and collective bargaining systems and the autonomy of social partners. The initiative not only allows but would most desirably rely on social partners to be entrusted with the implementation of the measures and would allow for the necessary flexibility in this regard.

8.4. Comparison of options

The table below compares the different options taking into account the assessment criteria of effectiveness, efficiency and coherence.

Table 1: comparison of options

8.5. Possible combination of options

The above analysis shows that none of the options would reach the two specific objectives on its own. However, a combination of (Sub-)options would permit to reach the two specific and hence also the general objective of the initiative.

Option 3 is of limited relevance as a stand alone option. While it would improve better implementation and enforcement of the existing legal framework, it is not expected to improve the current situation tangibly if not accompanied by the possibility to gather information on a comparator and/or if employers are not involved in the gender analysis of the pay structure. This option should therefore be combined with one of the other options to be effective and to support them.

The basic work done in terms of data gathering by an employer to create transparency both at individual or at organisational level is the same for most Sub-options under Options 1 and 2. For this reason, most combinations of such Sub-options would lead to some economies in terms of overall cost and burden. The only exception is basic pay reporting (Sub-option 2C), where this obligation is much lighter as no active grouping according to work of equal value would be triggered; the calculation would concern only the percentage of pay difference between all women and all men working within the establishment.

In choosing the preferred combination of Sub-options, it is important to avoid unnecessary duplication of obligations. For instance, combining Equal pay certification (Sub-option 2A) or Joint pay assessment (Sub-option 2B) with Option 1A would add a communication burden on employers without adding much information for workers. The latter would indeed already have pay information as a result of the certification/pay assessment (except in organisations with less than 50 workers).

On the other hand, any Sub-option under Option 2 which would not lead to (all) workers having the necessary transparency on their individual situation in regard to pay, would need to be combined with either of the Sub-options under Option 1 in order to ensure that both specific objectives could be achieved. This would be particuraly the case for basic pay reporting (Sub-option 2C) where only the information on the overall gender pay difference in the company is provided.

8.6. Preferred Package

In light of the above, the preferred package should be a combination of measures creating transparency at individual worker level and at employer level. These measures should be accompanied by measures aiming at facilitating the application of the key concepts on equal pay and at strengthening access to justice.

The best combination would consist of Option 1B (ensuring transparency prior to employment and right to receive information on pay upon request), coupled with pay reporting on the part of large employers, including on pay differences among workers carrying out equal work or work of equal value and mandatory joint pay assessment in case of unjustified differences in pay between women and men in the same grouping of workers above a 5% threshold (Sub-option 2D). To be fully functional, these measures should be supported by Option 3 with its three measures.

This preferred package would empower workers to claim their right to equal pay and start creating an incentive to more systematically assess gender bias and discrimination in pay structures. It would achieve the objectives of the initiative while balancing workers’ fundamental right to equal pay with the possible burden and costs of the envisaged measures on employers. Even if Sub-option 2D only applies to employers with at least 250 workers, all workers’ rights would still be protected through their right to ask for information, reinforced further by improved access to justice. Moreover, the package may be expected to increase broader awareness around equal pay and could therefore prompt a voluntary move towards more transparency at employer level.

The preferred package would reach the objectives in the most proportionate manner possible as it allows to improve the practical possibility to verify and assert the respect of the equal pay principle for all workers while at the same time limiting costs to a maximum extent. It is not possible to provide an estimate of the total costs of the preferred package. As to the cost related to the individual right of information, the cost of a single request was estimated at 20 EUR. The total cost per company as for pay information will depend on the number of requests. The overall cost for pay reporting for employers would amount to around 26 and 50 million EUR, i.e. between a minimum of 379-508 and a maximum of 721-890 EUR per employer depending on the size. Where gender pay differences are small or absent, there will be no or little follow-up action (pay assessment) triggered - the average cost per employer to carry out the additional assessment has been estimated between a minimum of 1,180-1,724 EUR and a maximum of 1,911 and 2,266 EUR (expected to decrease in any subsequent exercises to respectively 843-1,232 EUR and 1,461-1,675 EUR). As described in previous sections, the cost for national administrations would amount to one-off spending of 400,000 EUR and about 82,000 EUR annually.

The choice for a rather modest policy intervention furthermore takes into account the lack of hard data on the size of the problem. The chosen pay transparency measures permit to target the more stringent obligations to companies with a gender pay discrimination problem. They may be expected to shed more light on the extent of pay discrimination in the future.

The choice is furthermore based on the uncertainty linked to the longer term economic impact of the COVID-19 crisis which puts employers already under significant stress. The latest economic forecasts indicate that recovery prospects are subject to a high degree of uncertainty, with downside risks to the outlook. 185 It is hence advisable to focus on instruments which, while giving individuals some minimum tools to claim their right, do not impose unnecessary costs and burden on employers at this time.

Once the impact of the current measures as well as the impact of the current economic crisis is clear, the opportunity for extending the scope of the measures could be assessed with the help of the additional data on pay discrimination collected through the first measures.

8.7. Hypothetical scenario of the potential economic impact of the preferred package of measures

The impact of pay transparency measures should ideally be measured through the reduction of pay discrimination and/or a reduction of gender bias in pay structures that would affect relative wages between women and men. As a suitable metrics of the scale of ‘gender pay discrimination’, or even of ‘gender discrimination’, is lacking, this impact assessment estimates – purely for illustrational purposes – distributional effects of a very modest reduction in the gender pay gap that may result from pay transparency measures. The assumption is that pay transparency will help correcting discriminatory gender differences in average wages and therefore reduce the share of the gender pay gap that is due to gender pay discrimination. Since this share, as mentioned, cannot be measured with precision, the estimates of a possible impact rely on experts’ judgement and other various, mostly qualitative, assessments of the possible reduction. More specifically, the hypothetical scenario is built based on estimates available in the impact assessment for the 2014 Recommendation for similar measures and on the qualitative assessment of experts consulted in the context of the support study for this initiative. Similar assessments have been carried out elsewhere, such as in the European Parliament’s study 186 and in studies assessing the impact of pay transparency measures carried out at national level 187 .

The qualitative assessment of experts confirms the possible scale of estimates presented in the impact assessment for the 2014 Recommendation for similar measures, namely:

·a right to pay information would have a positive impact in the order of 1-2 percentage points reduction of the gender pay gap;

·pay reporting would have a positive impact in the order of 2-3 percentage points reduction of the gender pay gap;

·joint pay assessment would have a positive impact in the order of 2-4 percentage points reduction of the gender pay gap.

In addition, if social partners are involved and this would lead to equal pay matters being included in collective bargaining, this would have a positive impact in the order 1 percentage point reduction of the gender pay gap.

Considering that the measures in the preferred package (combination of Sub-option 1B, Sub-option 2D, and Option 3), cannot have, and do not aim at having, an impact on all the root causes of the gender pay gap, but are meant to affect pay discrimination, which is one part of the unexplained component of the gender pay gap, the percentage change is applied to the latter only. In order to estimate potential distributional effects in a hypothetical scenario, we assume a conservative reduction of the unexplained part of the pay gap of 3 percentage points as a result of the application of the package of measures. Pay transparency measures would influence differences in pay by gender and therefore wages and the income distribution. This change assumes that the average speed of the reduction in the unexplained pay gap would double as compared to the change observed for the years 2010-2014. 188 This overall estimate is more conservative than what can be found e.g. in the assessment of the Recommendations of the Bauer report. 189  

Impact on household income distribution, risk of poverty and public budgets: The following analysis looks at the potential direct impact of a reduction in the unexplained part of the gender pay gap on the household income distribution (the at-riskofpoverty rates (per different groups) and on government budgets using the EUROMOD 190 model (See Annex 4, section 4 for more details).

EUROMOD is a static tax-benefit microsimulation model, which allows simulating the immediate effects of a policy change. In this modelling framework neither behavioral responses nor general equilibrium macroeconomic effects are accounted for; thus, results should be interpreted as a ¨morning-after¨ effect.

For this analysis, the simulations focused on the impact of an externally given targeted increase in women gross hourly wages, driven by the expected reduction of the gender pay gap, while keeping constant all other variables, including male wages. The estimated impact on household disposable income on poverty and inequality indicators and the budgetary implications, come both from this external shock in gross earnings as well as from the interactions of the tax and benefit system in each country.

The micro simulations obtained from the EUROMOD model evaluate the impact of closing the gender pay gap by raising women´s gross hourly wages and applying countries´ tax-benefit systems (as of 2019). Microeconomic analysis allows to take into account the initial distribution of wages, the shares of working women in each decile, the household composition, the structure of personal income taxes (e.g., progressivity of the personal income tax, joint assessment of taxable income, etc.) and the interactions within the tax-benefit system. We present the effects of the expected reduction of 3 p.p. of the unexplained part of the gender pay gap 191 , compared to the baseline given by the policy systems in 2019 (see Annex 4 – for a comparison with the impact of a reduction of 1 and 5 p.p.).

Main results (see further details in Annex 4):

Impact on gross earnings and inequality: Raising women´s gross hourly wages in order to close the gap would lead to an overall increase of total gross earnings of 6.9% on average at the EU level, and a reduction of inequality in market income (i.e. income before taxes and benefits) inequality for all Member States. Due to the interaction of the tax-benefit systems, the reduction of inequality of disposable income (i.e. after taxes and benefits) would occur for 18 Member States.

Impact on poverty: The at-risk-of-poverty rate would drop from the initial 16.3% on average in the EU27 to around 14.6%, with important heterogeneities across countries and by household types. The risk of poverty would be mostly reduced for single parent households, which are mostly women (85%). 192

Impact on public budgets: The reduction of the gender pay gap would generate a positive budgetary impact due to a rise in government revenues (higher collection of income taxes and social insurance contributions) and a small decline in social (cash) transfers (mainly explained by a reduction of means-tested cash benefits). At the EU population-weighted average level, the shrinking of the gap is expected to lead to a rise in government revenues from direct taxes and social contributions of about 7.5%, while the reduction of social transfers (cash benefits) would be of approximately 0.4% .

The results differ substantially across Member States. The growth in total market incomes and government revenues would range from around 4% to 14%. It would be particularly high in some countries, such as BG, EE, LT and LV (in between 10% and 14%), whereas it would be more limited in EL, IT, MT and NL (of about 5%).

Macroeconomic impact: The analysis of the behavioral impacts of a change in the gender pay gap could in principle be carried out with other types of models. For instance, a labour supply model could assess the impact of these changes in women gross earnings at the extensive margin (change in labour market participation) and intensive margin (change in number of hours worked for people already employed). The use of a general equilibrium model (which could estimate effects on other earnings, prices and gender pay gap) would however require a number of assumptions, which could make the macro-level effect of the reform difficult to evaluate. For example, accounting only for the wage increase would overestimate the effect at the macrolevel as it could not be excluded that male labour supply/participation would adapt to such a large increase in female earnings (e.g. within households). Firms might also just re-adjust wages on male workers in order to mitigate the higher wage costs.

Several studies have estimated the potential impact of a reduction of the gender pay gap on economic growth through different channels (See Annex 5, section 4 – for a summary review). The main channel is the increase in labour market participation. The European Added Value Assessment carried out by the European Parliament estimates a substantial potential effect of a reduction of the gender pay gap on the GDP: each percentage point reduction in the gender pay gap would translate into an increase in the EU gross domestic product (GDP) of 0.1 percentage point. 193 This impact was estimated as the result of the implementation of the Recommendation of the Bauer report 194 mentioned above. EIGE’s estimates of the effect on GDP of closing the gender pay gap are lower: they show a 0-0.2% increase in GDP per capita over the 2030-2050 period.

8.8. Choice of legal instrument

As regards the choice of the legal instrument, different types of acts were considered. A regulation would impose uniform obligations in all Member States, which would be most easy to monitor and apply across the EU. However, a regulation would make it more difficult for Member States to integrate the new measures taking into account the existing legal and administrative framework; it would also not allow flexibility to adapt the implementation in accordance with their specific labour market models. Room should be left for implementation especially considering the crucial role of Social Partners for a successful implementation.

A revision of the Recast Directive was discarded because of its wider scope and because its entire logical structure would have to be revised to include the new measures. Indeed, the new measures focus on one aspect of the Recast Directive, namely the right to equal pay, tailoring measures according to size of organisations; as a result, a specific directive dedicated to this matter, strengthening the general rules under the Recast Directive, is considered as more appropriate.

On the basis of the 2020 evaluation and this impact assessment, a new directive, laying down a framework to enhance the application of the equal pay principle through pay transparency and related reinforced enforcement mechanisms, is therefore considered to be a more appropriate instrument. A Directive allows for flexibility as to the means each Member State considers appropriate to ensure compliance with the obligations taking into account the national context. It would also be in line with the approach followed as regards similar obligations in the field of employment (posted workers) and discrimination (free movement of workers). Further development of non-binding measures, as mentioned in Section 6.1, is unlikely to bring pay inequalities on the political agenda and engage in setting the necessary framework to combat pay discrimination and bias in pay structures. This does not mean that the future directive may not be supported by non-legislative flanking measures, some of which may be directly relevant to a good implementation of the new rules (e.g. guidance on how to assess and define work of equal value based on the criteria set out in the new rules).

9.How will actual impacts be monitored and evaluated?

Monitoring of implementation. Transposition of the initiatve into national legislation will be monitored by the European Commission, in particular at the end of the transposition period and thereafter. The monitoring would look at legislative or non-legislative initiatives adopted by Member States beyond what is strictly required by the initiative and their effect. The Commission will evaluate the initiative 8 years after the end of the implementation deadline referred to in the initiative.

In addition, monitoring of the implementation of national legal provisions will be conducted through a body designated in accordance with the initiative. This national body may be part of existing bodies at national level pursuing similar objectives. Regular exchanges between these bodies would be organised at EU level in the context of the existing Advisory Committee on equal opportunities for women and men 195 (together with social partners and relevant NGOs represented in the Committee).

Quantitative monitoring of impacts. Ideally, progress towards achieving the objectives of the initiative should be monitored on the basis of a quantitative indicator of gender pay discrimination. However, the lack of data and methodological difficulties make developing such indicator a challenging exercise. 196 A feasibility study could be launched to investigate how to assess pay discrimination and the possibility to develop more robust indicators based on the additional information collected through the envisaged pay transparency measures.

In the absence of a pay discrimination indicator, a refinement of the GPG indicator at EU level could already support in-depth quantitative analysis. Such analysis should focus on the impact on people starting out in their careers and in their mid-career in order to capture changes in behaviour and outcomes. The aggregated information collected from data resulting from pay reporting could also be used with this purpose at national level e.g. by monitoring the evolution of data by employer/group of employers and checking the progression in the reduction of the gender pay gap over time. Depending on public availability of data, a selection (e.g. by size or sector) of employers could be monitored at EU level over time at aggregate level, possibly with in-depth analysis of volunteering employers. Finally, considering that the legal baseline is different across Member States, progress could be monitored with reference to tailor-made benchmarks, i.e. monitoring progress compared to the initial situation up to a common EU target. This could be done in coordination with the European Semester process 197 .

Qualitative monitoring of impacts could be organised as follows:

-Based on data collected in Member States, the state of pay discrimination may be evaluated based on the number and percentage of employers that will have to carry out a joint pay assessment. Such data will give an indication of suspected gender pay discrimination in the context of the pay reporting exercice set out by the initiative.

-Follow up analysis could be done on remedial action by employers, e.g. in terms of gender equality plans developed after the introduction of pay transparency measures or voluntary overall revisions of pay structures. 198  

-Data on the number of claims brought before enforcement authorities (e.g. equality bodies, courts) would also be a useful indication of the impact of the measures. This could be monitored through the European Equality Law Network and could be combined with a periodic survey of judicial and enforcement staff to test the extent to which awareness and skills to deal with these matters are increasing.

-The cooperation with social partners would allow collective qualitative information on implementation issues and to link the monitoring process of equal pay to the broader concept of gender equality in the workplace. The presence, quality and relevance of specific measures aimed at fostering equal pay in collective agreements e.g. the successful development and implementation of tools to facilitate the application of the concept of ‘work of equal value’ (e.g. gender-neutral job evaluation and classification systems) would be monitored through surveys or dedicated hearings with social partners. Possible recommendations from such surveys/meetings could feed into mutual learning sessions organised under the Commission’s Mutual Learning Programme on gender equality and capacity-building seminars.

-The increase in the level of awareness on equal pay issues among the general public could be monitored by looking at the frequency with which equal pay is mentioned in the press, parliamentary acts, company reports and other relevant sources.

-Finally, a Eurobarometer survey (such as the one from 2017) could show change in perceptions and attitudes.

The future evaluation of impacts, i.e. to assess factually the degree of success of the initiative, would require a timeframe of at least eight years from full implementation of the initiative and would require data gathered at Member State level. The review should ideally be synchronised with the release of SES data. 199 Different conceptual options are in principle available:

1. Counterfactual analysis – the golden standard to determine the specific impact of a policy measure. The literature shows some national examples regarding pay transparency measures 200 . Due to the granularity of the data required for this exercise, the analysis would need to be tailored to the specifics of the national legislation of each Member State and should therefore be carried out at national level. The Commission could support Member States through dedicated projects.

2. Indicators on transparency in wage setting: regression of wage levels on explanatory variables such as age, occupation, economic activity, etc. can provide an indicator (the coefficient of determination – R2) on the transparency of wage setting mechanisms in the different countries. Moreover, adjusting the gender pay gap for the above mentioned explanatory factors can give indications on possible inequalities between male and female earnings. Both indicators could be further refined, by collecting, possibly, a variable to account for career breaks.

Bibliography

Legislative acts:

Council, Directive 76/207/EEC, as amended by Directive 2002/73/EC, on equal treatment of women and men, 1976.

Council, Directive 86/378/EEC, as amended by Directive 96/97/EC, on equal treatment in occupational social security schemes, 1986.

Council, Directive 97/80/EC, as amended by Directive 98/52/EC, on the burden of proof in cases of discrimination based on sex, 1997.

Council, Directive 75/117/EEC on equal pay.

Deutscher Bundestag, Entwurf eines Gesetzes zur Förderung der Transparenz von Entgeltstrukturen, Gesetzentwurf des Bundesregierung, 2017. (Drucksache 18/11133).

Websites / articles:

https://www.payscale.com/data/gender-pay-gap  

https://jamstalldhetsfeministern.files.wordpress.com/2014/04/miljongranskningen_etapp2_nov2008.pdf  

https://www.sciencedirect.com/science/article/pii/S0007681317300356  

https://www.cipd.co.uk/Images/reward-management_2017-focus-on-pay_tcm18-34496.pdf

https://eige.europa.eu/news  

https://ec.europa.eu/info/policies/justice-and-fundamental-rights/upholding-rule-law/eu-justice-scoreboard_en

https://ec.europa.eu/growth/smes/business-friendly-environment/sme-definition

https://www.fastcompany.com/3065592/why-everyones-salary-should-be-revealed  

https://www.wgea.gov.au/data/wgea-research/gender-equitable-recruitment-and-promotion

https://www.wgea.gov.au/sites/default/files/documents/2019-04-4%20International%20reporting%20schemes_Final_for_web_0.pdf  

https://hbr.org/2013/08/a-fairer-way-to-make-hiring-an

http://publications.europa.eu/webpub/com/refit-scoreboard/en/priority/7/7-20.html

https://ec.europa.eu/info/publications/2020-commission-work-programme-key-documents_en

https://ec.europa.eu/info/law/better-regulation/initiatives/ares-2018-3415794_en   

https://ec.europa.eu/eurostat/web/products-statistical-working-papers/product/-/asset_publisher/DuuxBAj0uSCB/content/KS-TC-18-003  

https://www.ilo.org/global/about-the-ilo/multimedia/maps-and-charts/enhanced/WCMS_650829/lang--en/index.htm

https://ec.europa.eu/eurostat/web/products-datasets/-/SDG_05_20  

https://ec.europa.eu/info/policies/justice-and-fundamental-rights/gender-equality/gender-equality-strategy_en#gender-equality-strategy-2020-2025

https://www.aauw.org/resources/article/black-women-and-the-pay-gap/

https://www.payanalytics.com/faq

https://gender-pay-gap.service.gov.uk

http://www.bbc.com/news/av/magazine-41516920/the-skills-calculator-closing-the-gender-pay-gap  

https://web.stanford.edu/dept/radiology/cgi-bin/raddiversity/wp-content/uploads/2017/12/TheAbrasivenessTrap.pdf  

Case law of the European Court of Justice

Annexes

Annex 1: Procedural information

1.Lead DG, Decide Planning/CWP references

This impact assessment and the related initiatives are under the responsibility of the Directorate-General for Justice and Consumers (JUST).

The Agenda Planning Reference is PLAN/2019/5818.

The project has been added to the 2020 European Commission work programme 201 under the section 'A new push for European democracy', following the non-legislative Gender Equality Strategy 2020-2025, as binding pay transparency measures (legislative, including impact assessment, Article 157 TFEU, Q4 2020).

2.Organisation and timing

Work on the preparation of this initiative started during the autumn 2019. The Inception Impact assessment consultation was carried out between 06 January 2020 - 03 February 2020. The impact assessment was prepared with the involvement of JUST C.3 (Data protection) as well as the following Services through the Inter-Service Steering Group (ISG), chaired by the Secretariat General: DG BUDG, DG CLIMA, DG CNET, DG COMM, DG DEFIS, DG EAC, DG ECFIN, EEAS, DG ENV, ESTAT, DG FISMA, DG GROW, DG HOME, DG MOVE, DG RTD, JRC.

The Inter-Service Steering Group was set up in 2019. The first ISG meeting took place on 24 January 2020 and focused on a study to support an Impact Assessment of pay transparency measures, consultation strategy and on public consultation questionnaire. The following ISG meeting took place on 17 July. The interim report of the support study, the preliminary results aof the consultations and the outline of the Communication were discussed. Another meeting took place on 26 August to discuss a first draft of this IA. The last meeting took place on 11 November.

On each occasion, the members of the Steering Group were given the opportunity to provide comments orally and/or in writing on the draft versions of the documents presented.

3.Consultation of the RSB

This version of the impact assessment was submitted to the RSB on 23 December 2020. It took into account comments received from the Board on 25 September and on 17 November 2020. A hearing with the Regulatory Scrutiny Board took place on 23 September 2020. An upstream meeting with the Regulatory Scrutiny Board took place on 5 May 2020.

The RSB comments were taken into account in the following way:

Ø3rd RSB opinion:

General comments

Specific comments

Ø2nd RSB opinion:

General comments

Specific comments

Ø1st RSB opinion:

General comments

Specific comments

4.Evidence, sources and quality

DG JUST commissioned a number of studies in support of this Initiative, namely a dedicated support study used as main source for this Impact Assessment; a revised and expanded version of the Eurofound study on costs and benefits of existing pay transparency systems; a behavioural experiment and a literature review of behavioural economics carried by JRC(See Annex 5, section 5); additional material on administrative data sources and a simulation with the EUROMOD model of the impacts of a reduction of the GPG also by JRC; a review on the number of case law by the Legal Network/equality bodies.

The Commission consulted widely and received input from various sources for this impact assessment work.The consultation strategy included:

Consultation of high-level representatives of Member States in the High-Level Group on Gender Mainstreaming, on 29 January 2020 and 15 September 2020.

Targeted consultation of Member States through a separate specific questionnaire issued in March 2020.

Targeted consultation of social partners organised on the basis of a separate specific questionnaire issued in March 2020 (and a mini-survey for companies) and followed by a dedicated consultation hearing of social partners run by DG JUST and DG EMPL in June 2020.

A public consultation launched in March 2020, opened for 12 weeks. The consultation covers general awareness, experience and knowledge of citizens and stakeholders regarding pay transparency issues as well as views on the possible specific measures.

Relevant previous consultation activities:

In 2019, the Commission already carried out a number of consultation activities around the equal pay principle. This process started as part of the evaluation 202 of the relevant provisions in Directive 2006/54/EC implementing the Treaty principle on 'equal pay for equal work or work of equal value' (and in the 2014 Commission Recommendation on strengthening the principle of equal pay between women and men through transparency). It was also embedded in the preparatory work on the future EU Strategy on gender equality and related consultation. In particular:

From 11 January 2019 to 5 April 2019, a public consultation 203 was carried out for the Evaluation of the relevant provisions in Recast Directive implementing the Treaty principle on ‘equal pay for equal work or work of equal value’. Among others, the consultation's objective was to collect information, views and experiences on problems appearing from gaps and weaknesses of existing pay transparency measures at national and EU level. Furthermore, it sought evidence on the extent to which the 2014 Pay Transparency Recommendation had helped to reinforce the implementation of equal pay principle enshrined in Article 157 TFEU and Recast Directive. The respondents took the opportunity to share their views regarding forward-looking questions on relevant aspects of the transparency initiative including on the need for some further EU-level action on measures to address sex-based pay discrimination.

The Evaluation process also included a targeted consultation survey and semi-structured interviews with representatives of national equality bodies, competent public authorities, trade unions, employers’ representatives, labour inspectorates, NGOs and experts. The Evaluation resulted in an online discussion with selected experts from different Member States.

On 2019 International Women's Day, a public consultation 204 on the state of gender equality in the EU at present as well as about priorities for the future was launched in order to prepare the next Commission's gender equality policy framework. More than 1300 replies were gathered from a broad range of stakeholders, including academic/research institutions, business associations, company/business organisations, consumer organisations, EU citizens, non-EU citizens, non-governmental organisations (NGOs), public authorities, trade unions and others. They showed considerable public interest in introducing measures to further support the principle of ‘equal pay for equal work or work of equal value’, such as pay transparency. 205

A consultation of the High Level Group on Gender Mainstreaming and the Advisory Committee for Equal Opportunities for women and men took place through dedicated meetings in 2019.

A dialogue between the European Commission and the social partners took place in May 2019 during a thematic seminar on the role, costs and benefits of pay transparency initiatives. 206  

Eurofound has compiled information on the ‘costs of pay transparency measures’ that focuses on EU countries that have already implemented pay transparency measures affecting employer level. 207 A small set of interviews (44 in total) was conducted with companies for each Member State present in the survey.

Annex 2: Stakeholder consultation

1.Consultation Strategy 

The objective of the consultation was to secure stakeholder inputs in relation to the proposed legislative initiative on strengthening the principle of equal pay between women and men through pay transparency. The stakeholders identified for the consultation were: the general public; and, specifically, the Member States; the social partners; and individual employers.

An online Open Public Consultation was targeted at the general public, over the period of 5 March 2020 to 28 May 2020. The questionnaire was available in electronic format only, and the official languages of the European Union (EU) were used, for answers and replies. The questionnaire included open-ended and closed questions. Some respondents sent individually formulated contributions by email.

Three targeted surveys were implemented to complement the Open Public Consultation. They addressed the Member States, the social partners, and individual employers. The engagement with the social partners was further complemented with 8 interviews with European level social partner organisations, 4 employers’ organisations and 4 trade union organisations. Information gathered during interviews are integrated here in social partners' sections. Finally, a dedicated consultation hearing was held in June 2020 with the participation of 25 employers and 25 trade unions’ representatives. The Commission also received 7 position papers afterwards. The Commission’s consultation with European, national and local social partner organisations in March 2020 on binding pay transparency measures resulted in mixed feedback on the need for the adoption of a legally binding measure on pay transparency.

The four surveys investigated views on the relevance, effectiveness, efficiency, coherence, and EU added value of pay transparency measures. Information on the general state of play in relation to national measures and legislation on pay transparency was gathered from the Member States and the social partners. Information about attitudes and experiences in relation to pay discrimination was gathered from the general public, social partners and individual employers. Information on the level of pay transparency in organisations was gathered from individual employers. Looking to the future, information on the effectiveness and impact of possible EU initiatives on binding pay transparency measures was gathered from the general public, the Member States, and the social partners.

Diverse information channels were used to maximise responses to the consultation, with information disseminated to: permanent representations (social attachés and counsellors) via the Council Secretariat; members of the High-Level Group for Gender Mainstreaming and of the Advisory Committee for Gender Equality; 87 social partners’ organisations; the European Network of legal experts in Gender Equality and Non-Discrimination, the European network of equality bodies (EQUINET); members of the EU Platform of Diversity Charters; the European Economic and Social Committee(EESC); experts of the Scientific Analysis and Advise on Gender Equality in the EU (SAAGE) network; and members of the European Women on Boards (EWoB) network.

Information on the Open Public Consultation was posted on the Facebook and Twitter accounts of the Directorate-General for Justice and Consumers, and the public consultation was referenced in a video disseminated on these accounts. Specific information was sent to individual journalists.

Six Member States, either in addition to the questionnaire or as an alternative, submitted position papers to clarify their views, and a number of international organisations and civil society organisations provided position papers.

2.Stakeholder Participation

All identified stakeholder groups were reached: general public; Member States; social partners; and individual employers.

The Online Public Consultation received 560 responses. 51.3% of respondents were women, 21.4% were men, 0.5% were ‘Other’, and 26.8% did not identify. The geographical distribution of responses was somewhat unbalanced: 39.3% of respondents were from Spain, with Germany following at a distance with 11.4% of respondents, and Italy with 9.5% of respondents. 4.8% of respondents were from Belgium, 4.1% from France, 3.6% from Poland, and 3.4% from the Netherlands.

Different types of respondents are evident: 67.1% of the responses were filed by EU individual citizens, 3.2% by non-EU citizens; 9% of respondents identified themselves as being from company/business associations; 7% as being from non-governmental organisations; 5.5% as being from trade unions; 3.8% as being from public institutions; and 2.5% as being from academic/research bodies.

The targeted survey of Member States’ authorities received 20 responses, from 17 Member States: Austria, Czechia (2), Denmark, France, Germany, Hungary, Ireland, Italy, Latvia (2), Lithuania, Malta, The Netherlands, Portugal, Romania, Slovenia, Spain (2), and Sweden. The authorities from Croatia, Estonia, France, The Netherlands, Portugal, and Sweden submitted off-line responses as an alternative or in addition to on-line responses. No responses were received from Belgium, Bulgaria, Cyprus, Finland, Greece, Luxembourg, Poland and Slovakia.

The targeted survey of social partners received 80 on-line responses and 4 off-line responses. 41.3% of the respondents were employer associations, and 58.7% were trade unions. The majority of respondents, in both cases, operate at Member State level (63.6% of employers’ organisations and 78.7% of trade unions). Nearly 50% of responses came from five Member States: Germany (26.3%); Finland (7.5%); Belgium Czechia and Italy (5% each).

The targeted survey of individual employers received 24 responses. Nine of these respondents operate in all Member States, 9 respondents operate only in one Member State, and 6 respondents operate in several Member States. 12.5% of respondents are associated with the manufacturing sector(s). 21% have between 50 and 249 employees, while 79% have at least 250 employees.

3.Analytical Methodology 

The preparation of the dataset involved the creation of four Excel databases to analyse the replies. The Excel files were imported in the Statistical Package for Social Science (SPSS) and elaborated using this professional statistical tool, which enabled checking for coherence and completeness, and control for duplications.

Five-point Likert scales on attitudes and expectations of respondents regarding impacts and effectiveness were systematically recoded in three-point Likert scales in the quantitative analysis, as responses were somewhat dispersed across modalities. All questions were analysed, highlighting, as relevant, differences between sub-groups. Differences had to be interpreted as they could be attributed to group distribution or to the natural variance in responses. Where statistically significant, the disaggregation by sub-groups was reported.

Inductive coding was used to derive themes from the answers in the qualitative analysis. This was conducted on qualitative information included in the questionnaires, on documentation received from stakeholders and Member States, on the analysis of documents received from the European Commission and on qualitative interviews. Clusters of the most recurrent type of answers served as a starting point for the analysis. Additionally, particular answers, albeit not recurrent, were included if considered particularly original and relevant by the experts involved. Essential messages were identified through a distillation of the responses.

Each questionnaire had a similar initial section of information on the respondent, while the questions in the other sections differed according to the respondents targeted. Responses were analysed, grouped by the themes they refer to, with: presentation of results (frequency and breakdown of responses); quantitative analysis for closed questions; and qualitative analysis for open-ended questions.

4.Results 

Public Consultation

The State of Play 

The gender pay gap is viewed as being particularly at issue in the private sector, in: micro (72%), small (71%), medium-sized (70%), and larger (61%) enterprises. Nevertheless 39% of respondents viewed it as being at issue in the public sector. 49% of respondents felt women and men have equal pay for the same work or for work of equal value in the organisation where they worked, however 39% felt that this was not the case.

Costs of litigation were identified as a relevant enforcement issue by respondents in relation to pay discrimination. A very large majority of respondents (92%) considered the provision of support to individual workers to be important in order to lower these costs and enable legal action in pay discrimination cases to be undertaken.

Future Measures 

The four measures presented, of the type not focused on strengthening victim rights, pay transparency measures, were each viewed as being effective by significant majorities of respondents: obligation to include equal pay matters in collective bargaining (81%); gender-neutral job evaluation and classification systems (80%); employers and employees’ representatives analysing pay levels and gender pay gaps in regular pay audits (79%); and regular employer reports on pay levels and gender pay gaps (78%).

In respect of the different types of respondents, this consensus was shared to a high level by individual EU citizens and non-EU citizens, and by respondents from trade unions, academic/research bodies, and public institutions. Agreement was at a lower level among respondents from company/business associations. Nonetheless, the numbers of this type of respondent that agreed with the effectiveness of the individual measures stood at between 40% and 50%.

In identifying the possible impacts of these pay transparency measures: 90% agreed they would raise awareness about equal pay issues; 88% they should be part of wider equal pay policies; 86% they would reduce pay discrimination; 77% they would have a positive impact on employees' motivation and productivity; and 72% they would improve the work climate. The same pattern of consensus is evident in this across the different type of respondents as for the effectiveness of prospective pay transparency measures, with low levels of agreement among respondents from company/business associations.

Respondents from company/business associations were more likely to agree that such measures might make it difficult to reward high-performing employees (78%), they limit employers’ discretion (76%), and they create a significant additional burden (74%). There was a low level of agreement from the other types of respondents that such impacts would result.

The four measures presented, of the type focused on strengthening victim rights, enforcement measures, were each viewed as being effective by significant majorities of respondents: employees have a right to access information on pay levels and gender pay gaps (84%); effective and proportionate penalties for companies and compensation for victims of discrimination (80%); reinforced mandate of labour inspectorates to intervene in support of victims (77%); and strong mandate of equality bodies to intervene in support of victims (76%).

The same pattern is evident in this consensus across the different type of respondents as with the first type of prospective pay transparency measures, with lower levels of agreement from respondents from company/business associations, where agreement ranged from 28% to 40%.

In identifying the possible impacts of these pay transparency measures: 91% agreed they would help enforce the right to equal pay; 88% they would be a powerful tool for equality bodies/labour inspectorates/social partners to support employees in defending this right; and 85% they would encourage employers to better implement the principle of equal pay. The same pattern of consensus across the different types of respondent is evident as for the first type of pay transparency measures, with respondents from company/business associations expressing lower levels of agreement.

There was significant consensus among respondents in their agreement on the effectiveness of the different modalities presented for implementation: the right to access information; job evaluation and classification systems; pay audits; and pay reporting. The only modalities that had low level of agreement related to sharing of information with employee on request (right of access to information, pay audits, pay reporting) and gender equality labelling (job evaluation and classification systems). In all instances, agreement from respondents from company/business associations was significantly lower than the overall consensus. In general, the measures were considered as suitable for all sizes of organisations, except pay audits which few respondents considered suitable for micro and small enterprises.

An analysis as to whether significant differences could be observed in individual citizens’ response patterns across the EU Member States found only minor cross-country differences in the response behaviour of individuals.

EU Member States

The State of Play 

Responses from the Member States in relation to the current situation were highly differentiated, reflecting different contexts across the Member States. Only 4 Member State respondents reported that current legislation was influenced by the EU 2014 recommendation. Six respondents reported that pay transparency measures had not been introduced due to the low priority for this issue on the policy agenda. Few respondents reported on the number of court cases on equal pay in the past year. However, most respondents did not think the number of cases was a valid indicator of the relevance of the gender pay gap issue. Some respondents noted that not all cases of discrimination end up in court, due to barriers of access to justice, and lack of information among employees.

Future Measures 

Members State respondents reflected some consensus in their agreement with the following options: employer reports regularly on pay level and gender pay gaps (82%); employees have a right to access information on pay levels and gender pay gaps (59%); employers who advertise a job vacancy to include information about expected salary or salary range (53%). The least favoured option was: forbidding confidentiality clauses on disclosing individual pay (35%).

There was some consensus among Member State respondents on the roles of equality bodies/labour inspectorates in supervising equal pay measures as being to: impose fines for non-compliance (75%); initiate equal pay cases on behalf of individuals (55%); represent victims in legal proceedings (50%); and pursue collective equal pay claims (45%).

In relation to the role of other national actors in implementing pay transparency measures there was significant consensus in agreement that organisations have a role in reporting on pay gaps (70%). There was some consensus in agreement that labour inspectorates have a role in performing automated checks on pay gaps based on employment, tax or social security registers (55%); and that employers’ organisations have a role in compiling pay statistics (50%). Only 25% agreed that statistical offices have a role in compiling pay reports; and 20% that equality bodies have a role in performing automated checks on pay gaps based on employment, tax or social security registers

In identifying the possible impacts of these pay transparency measures, there was significant consensus among Member State respondents: 100% agreed they would help enforce the right to equal pay; 90% they would contribute to raising awareness on equal pay issues; 90% they would reduce pay discrimination; 85% they would improve the work climate; 85% they are a powerful tool for equality bodies/labour inspectorates/social partners; 80% they would have a positive impact on employees' motivation and productivity; and 80% they would encourage employers to take action to better implement the principle of equal pay.

There was a low level of agreement that such measures: create additional administrative burden on organisations (50%); limit employers’ discretion (35%); and need to be tailored to size of organisation (40%). Only 25% agreed that such measures should be part of wider equal pay policies.

Social Partners

The State of Play 

The social partner survey points up different perspectives held by and experiences of trade unions and employer associations. Trade union respondents were more likely than employer associations to identify that gender pay discrimination is an issue and that pay discrimination legislation is difficult to use. Some trade union respondents noted the different levels of protection and of implementation of EU requirements on this issue across the Member States. These results were also confirmed by interviewes.

Most employer association respondents (81%) believe that there is a conflict between General Data Protection Regulation (GDPR) and possible pay transparency measures, while most trade unions believe there is not (57%). Most trade unions reported being often solicited by members in relation to gender equal pay (75%), while most employers had not often been solicited (79%). The comments received during the hearing as well as the analysis of the position papers received after the hearing confirm these different perspectives.

Future Measures 

Five legislative options at EU level were presented. The most preferred option for employer association respondents was for the inclusion of expected salary in job postings (41%), which was supported by 78% of trade unions. The most preferred option for trade unions was for regular pay audits (96%), which was supported by 24% of employer associations. A similar pattern is evident across the other options: regular employer reports (29% for employer associations vs 89% for trade unions); employee right to access information (18% vs 89% respectively); and forbid confidentiality clauses (18% vs 76% respectively).

Five options for binding pay transparency measures involving direct social partner participation were presented. Trade unions were significantly supportive of all options, bar certification related to job evaluation systems (47%, which compares to 21% employer associations) and self-regulatory measures linked to pay reporting (32%, which compares to 39% employer associations). The most preferred option for trade unions was for an equal pay plan (98%), which was chosen by 42% of employer associations, also their most preferred option. None of the measures were considered effective by a majority of employer associations.

In relation to standardised solutions, employer associations identified employers’ organisations compiling pay statistics for their members as the preferred option (42%), while this was the least favoured option for trade unions (87%). The most favoured options for trade unions was for the state to provide IT tools to organisations (87%); social partners to agree on/develop standardised tools (85%); and the labour inspectorate/equality body to conduct automated checks based on tax, social security or employment registers (79%).

When it comes to assessment of impact, a large majority of employer association respondents agree that pay transparency measures: create additional administrative burden on organisations (91%); limit employers’ discretion (82%); need to be tailored to size of organisation (82%); and might make it difficult to reward highly performing employees (79%). This assessment finds low levels of support among trade unions. 100% of trade union respondents agree that pay transparency measures: are a powerful tool for equality bodies/labour inspectorates/social partners; contribute to raising awareness on equal pay issues; and help in enforcing the right to equal pay. These impacts find low support among employer associations. 98% of trade unions agree that such measures should be part of wider equal pay policy, a view shared by 52% of employer associations.

In relation to raising awareness about equal pay legal provisions both trade union (96%) and employer association (82%) respondents felt awareness campaigns were the most effective. Employer associations felt an information webpage and online guidebooks would be effective (64%) as their next preferred option. Trade union felt training and guidance on pay reporting and auditing would be effective (96%) as their next preferred option.

Difficulties in implementation of pay transparency emerged in the existing experiences in Member States led to a request for the Commission to develop a definition of work of equal value in this new initiative on pay transparency.

As regards other obstacles in implementation, trade unions agree that measures should be applicable in all sectors with no exceptions or adaptation, as the ultimate goal is to go beyond gender segregation of labour and value female-dominated sectors more, as they are usually less attractive because of lower salaries and career opportunities. Within the group of employers’ organisations a resistance to have a uniform measure emerged. Additionally they mention the positive role that collective bargaining may have towards closing the gap. Social partners agree on opposing an initiative dictating on the content of collective bargaining.

In regard to the protection of privacy and the right to access salary information, the majority acknowledge that privacy can be protected with an accurate anonymisation of all pay data disclosed to employees. So this cannot be an obstacle to implementing a measure.

Finally as far as the added value of a European Union initiative employers' organisations argue that salaries are exclusively a national competence while trade unions mostly call for strong initiative from the Commission stating clear obligations can have the necessary impact to help reduce the gender pay gap in Europe. They argue that a measure protecting female workers and working to eliminate long-standing discrimination is long overdue, and to take advantage of this moment of change related to the COVID-19 crisis to build a more inclusive labour market.

Individual Employers

The State of Play 

Five respondents (21%) identify that gender pay discrimination may be an issue in their organisation, all of which were companies with at least 250 employees. These 5 respondents reported some improvements in internal policy of their organisation in the last five years.

13 respondents (54%) reported that their organisation had carried out at least one gender pay gap analysis. 10 of these respondents had at least 250 employees and 3 had between 50 and 249 employees.

All respondents reported on the pay information they make available. The level of pay transparency varies according to what part of salary is included.

§9 respondents provide a salary range at the point of hiring and information on average basic salary by category or position to employees, and 3 to employee representatives, without request.

§6 respondents provide information on average variable pay components in cash by category of employee or position to employees, and 4 to employee representatives, without request. 7 respondents have this information available on request.

§4 respondents provide information on average variable pay components in kind by category of employee or position to employees, and 2 to employee representatives, without request. 13 respondents have this information available on request.

§8 respondents provide information on pay raises by category or position to employees, and 3 to employees’ representatives, without request. 8 respondents have this information available on request.

§11 respondents (46%) reported an estimation of annual costs of this pay reporting: 7 respondents costed it as less than EUR 1,000, 3 between EUR 1,000 and EUR 6,000, and 1 above EUR 10,000.

Further Contributions: International Organisations

State of Play

The United Nations Entity for Gender Equality and the Empowerment of Women (UN Women) identified a set of initiatives it regards as good practice that it has collected from G7 and EU-based companies that have taken steps to address the gender pay gap, including, in particular, pay transparency measures. These include measures such as job evaluation systems, pay gap assessments and reports, annual assessment of pay equity and gender balance in company leadership, development and use of a pay assessment tool, and steps to ensure pay inequities in the labour market are not imported into an organisation.

Future Measures 

The International Labour Office (ILO) highlighted limitations of pay transparency measures that only target large enterprises, as this results in a limited number of workers being covered. It pointed out that legislation on pay transparency should be part of an ecosystem with proactive legislation addressing a range of interconnected issues including: increasing the prevalence of women in leadership; incentivising women to shift to higher-paid jobs; tackling violence and harassment; and redistributing unpaid care work, through improving care and family policies and services.

The ILO emphasised the importance of direct access to information on pay differentials as a means of: ensuring transparency; monitoring the pay gap; and as a basis for remedial action, including through the development of an equal pay plan. It highlighted pay audits as a platform for change in helping to expose pay differentials and reveal the need for structural change in the workplace. It noted that pay audits may be too aggregate to provide sufficient transparency, and the importance of involving trade unions or employee representatives in the process and making the results publicly available. It pointed to a value in: providing practical supports to assist companies to implement measures; making company action plans public; involving employers’ and workers’ organisations throughout the process; recognising the effective contribution to be made by collective bargaining; and strengthening complaint mechanisms.

Further Contributions: Civil Society Organisations

State of Play 

The European Women’s Lobby stressed that the concept of work of ‘equal value’ remains challenging. Criteria have not been provided in law to determine the meaning for work of ‘equal value’. This could usefully be a focus for a Europe wide study and would need to be addressed in pay transparency legislation. Make Mothers Matter identify the problematic of ineffective implementation and enforcement of equal pay legislation and identify the issue of definition of ‘pay’ and of ‘work of equal value’ as part of the reason for this.

Future Measures 

The European Women’s Lobby (EWL) pointed to the need to introduce mandatory pay audits in all companies, regardless of size. Smaller companies could be supported with technical assistance and financial support as necessary. It expressed confidence that GDPR would not be an obstacle to accessing information to ensure pay transparency. It emphasised the need for a focus on recruitment practices and non-discriminatory recruitment policies.

The Business & Professional Women organisation emphasised the need to support companies with up to 500 employees with training and financial and human resources, to create a culture of reporting and auditing on gender issues in the workplace. Training on methodologies should be compulsory in this area for employer associations and trade unions.

Make Mothers Matter underlined that more detailed reporting measures, along with a consistently applied enforcement system, is advisable.

Confederation of Family Organisations in the European Union (COFACE) emphasised that pay transparency measures need to include reference to the EU legislative framework in the area of work and family in order to guarantee that all workers have the same rights. There is a need to take action on the causes of the gender pay gap, including unequal sharing of care duties, unequal take-up of parental leaves, and lack of affordable childcare.

5.Way Forward 

A public appetite for tackling this issue can be identified from the consultation. It clearly identifies that an issue of fundamental rights is at stake, pay discrimination. Not all EU citizens enjoy this right fully. While a strong consensus is evidenced for action and for particular types of action, there is no full consensus on this. Respondents from companies/business associations generally did not share the same high level of consensus.

There are options which reach some level of consensus. These include: the obligation to adopt an equal pay plan at organisational level; employers’ requirement to report regularly on pay levels and gender pay gaps; employees’ right to access information on pay levels and gender pay gaps; and for employers and employees representatives to analyse information about pay levels and gender pay gaps. At the same time, the level of overall consensus evident behind the full range of proposals put forward in the consultation suggests that there is room for some further ambition in responding definitively to this issue that to-date has proven so resistant to change.

Annex 3: Who is affected and how?

1.Practical implications of the initiative

The Initiatives has potential implications for national administration, social partners, companies and all workers. Direct costs are mostly on the employer, but in large part non recurrent.

Table 1 summarises the main aspects of legislative options on pay transparency (see next page).

Table 1 – Summary of coverage, support by stakeholders, costs and main impacts of main pay transparency measures

2.Summary of costs and benefits

(1) Estimates are relative to the baseline for the preferred option as a whole (i.e. the impact of individual actions/obligations of the preferred option are aggregated together); (2) Please indicate which stakeholder group is the main recipient of the benefit in the comment section.

(1) Estimates to be provided with respect to the baseline; (2) costs are provided for each identifiable action/obligation of the preferred option otherwise for all retained options when no preferred option is specified; (3) If relevant and available, please present information on costs according to the standard typology of costs (compliance costs, regulatory charges, hassle costs, administrative costs, enforcement costs, indirect costs; see section 6 of the attached guidance).

3.Coverage of entreprises and workforce by employer threshold applied

Source: Own calculations based on Eurostat - Annual enterprise statistics by size class for special aggregates of activities (NACE Rev. 2) [sbs_sc_sca_r2] – 2017 or latest available or estimate