Executive summary

Despite comprising a large share of the workforce, women are a minority in corporate boardrooms across Europe. While progress has been made in recent decades among public (listed) firms, diversity lags behind in private corporations. This policy brief showcases evidence from the Gender Board Diversity Dataset (GBDD) – a new and unique data source which covers a comprehensive sample of European private and public corporations spanning multiple decades. Some of the key innovations of the GBDD relative to other datasets are the inclusion of private (non-listed) companies, a novel identification of the gender of board members based on linguistic and cultural heuristics, and a cross-country harmonization of firm-level data. These features make the GBDD a perfect tool for answering policy-related questions, and allow for cross-country and cross-sector comparisons. As such, the GBDD can help ensure that any policies aimed at promoting gender board diversity are grounded in scientific evidence.

The GBDD shows that women held only around 22% of board positions in the sample and, in many cases, were completely absent from boardrooms. The prevalence of firms with no women on boards amounts to 68% of all companies. From the policy perspective, addressing the causes of the total absence of women from boards is thus likely to be an important factor in promoting the overall gender board diversity. The GBDD also shows that certain sectors and countries differ substantially in terms of gender board diversity. Therefore, policies aimed at tackling sector- and country-specific barriers to entry for women could prove particularly effective. Finally, the GBDD indicates that despite an increase in the overall gender board diversity in the 2010s, the average number of female supervisory board members remained unchanged, which highlights the importance of looking at a broad range of measures when discussing diversity.

The GBDD is publicly available and can be downloaded here. The authors share the replication codes as well. All of the graphs shown in this policy brief can be reproduced using the publicly available GBDD. Anyone interested in the topic can utilize the dataset to gain further insights into gender board diversity in Europe.

Gender Board Diversity Dataset (GBDD)

The Gender Board Diversity Dataset (GBDD), created by Hubert Drazkowski, Joanna Tyrowicz and Sebastian Zalas, provides a novel cross-country perspective on women on management and supervisory boards over the past three decades. The GBDD is based on firm-level registry data circulated as Orbis, which we harmonize to ensure comparability across countries. A key feature of the dataset is that it covers registry information from both public and private (non-listed) companies. This makes the GBDD a very comprehensive source of information because the majority of board positions, as well as the majority of jobs in general, are in private, rather than public, companies. Data on private companies are scarce, and the GBDD is one of the very few data sources containing this information across European countries. The GBDD is based on a sample of over 28 million unique firms from 43 European countries observed on average for around seven years. It contains information about nearly 59 million individuals who sit on management and supervisory boards and covers the period between 1985 and 2020.

Another key component of the GBDD is the identification of the gender of board members, which is a key innovation relative other studies that use the Orbis data. While the original data do not specify the gender of individuals until 2010, they do include names and surnames, which we utilized to perform gender identification. By applying cultural and linguistic heuristics, we were able to uncover the gender of over 99% of the board members in our sample. For example, in some languages (e.g. Czech) surnames end with a gender-specific suffix, while in other languages (e.g. Polish) given names of women end with a vowel. The complete list of such rules has been compiled based on the World Atlas of Language Structures and Wikipedia entries for each of the languages in countries covered by the Orbis sample.

The GBDD is available online for everyone. It reports several measures of gender board diversity computed for countries over time, as well as for sectors in each country over time. As such, this is a unique source of information about gender board diversity in corporate Europe, and thus it can serve as a useful guide for policymakers and commentators. The data are available as a spread sheet and in data formats. Anyone interested in the topic can utilize the GBDD to gain insights into gender board diversity in Europe.

Background

While the labour force participation rate of women in Europe has been rising over the last few decades, and is approaching the participation rate of men, the proportion of company board positions held by women is still significantly lower. Despite these trends, the average share of company board positions held by women ranges from 10% to 30%, depending on the industry. This low representation of women in the corporate boardrooms has generated much heated debate and is a subject of ongoing research. It has also prompted several countries to mandate gender quotas for some companies (typically public companies or the largest limited liability companies). For example, in Norway, which was one of the first countries globally to introduce such quotas, large and medium-size firms are mandated to guarantee that both women and men account for at least 40% of board members. Given the increasing proliferation of mandated gender board quotas across countries, it is imperative that the public be aware of the main facts concerning gender board diversity in a broad set of companies i.e. those that make up the largest proportion of the economy, offer the majority of jobs, and often remain outside the scope of quota legislation.

Absence of women from boardrooms

A key insight emerging from the GBDD is that, despite women holding on average 22% of all board positions in a given industry, more than two thirds of all firms report no women in their boardrooms. Actually, 68% of all sectors across the European continent over the past several decades have not had a single firm with at least one woman in their boardrooms! Figure 1 shows the fraction of firms in a sector with no women in the boardroom. The x-axis shows the proportion of such firms in a sector, ranging from 0 (all firms in that sector have at least one woman on their boards) to 1 (women are absent from all corporate boardrooms in the entire sector). The y-axis shows the relative number of sectors in the sample for each of the observed fractions. For example, the spike at the value of 1 on the x-axis shows all cases in which an industry consists entirely of firms that do not have a single woman in their boardrooms.

This finding points to clusters of companies with potentially significant obstacles to gender board diversity. Since lack of representation could be considered as a significant barrier to diversity, policies aimed at promoting even minimal representation of women among board members could have a significant impact on overall diversity. Not least because, as evident in the GBDD, they would affect a substantial number of companies.

Our measure, which focuses on the fraction of firms without women, is new in the literature. Figure 2 shows the more frequently reported average share of women among all board members.  The horizontal axis shows the proportion of women among board members. Each bar represents the share of our sample, which comprises of industries observed in particular years and countries, with a given proportion women averaged across firms. The spike at the value of 0 on the x-axis shows cases with no women among board members.

There are substantial differences between industries and countries

The average firm-level share of women on corporate boards is only around 16% in the IT sector, while it is 35% in the education, health & care sector (EHC). Figure 3 shows two distributions of the average firm-level shares of women among board members, one for the broad IT sector and the other for the EHC sector. The distribution for the EHC sector is clearly shifted to the right relative to the distribution for the IT sector, which means that, across multiple countries and years, women tend to constitute a much smaller proportion of board members in the IT sector than in the EHC sector. Furthermore, the proportion of observations with no female board members (the spike at the value of 0) is much higher in the IT sector than in the EHC sector.

Breaking the data by country also shows significant differences. For example, firms with no female board members tend to be more prevalent in Poland than in Finland, which can be seen in Figure 4, where the distribution for Poland is shifted to the right relative to the distribution for Finland.

The above data suggest that there could exist a set of sector- and country-specific barriers to gender board diversity. Therefore, policies tailored to tackling those specific barriers could be more appropriate than blanket economy-wide policies.

Supervisory vs management boards

Another interesting feature of the GBDD is that it distinguishes between management and supervisory boards. While the key findings from the GBDD hold qualitatively true for both types of boards, there are quantitative differences between them. For instance, the average proportion of firms with no women on the management board is 74%, while that for supervisory board is 62%. This suggests that there might be significant differences in terms of barriers to diversity between the two board types. Therefore, each of them could require tailored policies.

Diversity generally increased over the last three decades

GBDD can also be used to assess how gender board diversity evolved over time. Generally, there was an increase in diversity in the 1990s, stagnation in the 2000s, and another increase in the 2010s. However, in the case of supervisory boards, the recent increase in the proportion of female board members was not accompanied by an increase in the number of women on supervisory boards. Whilst the full explanation of this observation would require deeper research, one possible explanation is that supervisory boards might have become smaller over time, with male board members accounting for most of the decline, thus mechanically increasing the share of female board members.

Conclusions

Despite the increase in gender diversity among company board members over the last three decades, women still comprise a smaller share of board members, and in many cases are completely absent from boards. While examining the reasons behind these observations is beyond the scope of this paper, the high prevalence of firms with no women on boards suggests the possibility that there exist significant barriers to entry for women, with this total lack of representation in many companies potentially being one of them. Policymakers interested in fostering an inclusive and fair society could focus their attention on understanding and removing barriers to board participation faced by women. Furthermore, identifying and tackling country- and sector-specific barriers to board diversity could be particularly impactful. The GBDD can be used by researchers and non-researchers alike to gain further insights into this topic, thus contributing to evidence-based policy making.