Gender Discrimination in Fashion Supply Chains: What Should Companies Report?
Freedom from discrimination based on gender is a universal human right and one of the core labor standards of the ILO. It appears in several global instruments: the ILO’s 1958 Convention 111 (Discrimination in Employment and Occupation), its 2019 Convention 190 on Violence and Harassment, and the 1979 United Nations Convention on the Elimination of all Forms of Discrimination Against Women (CEDAW).
Reports of gender discrimination and sexual harassment in supply chain factories in the 1990s forced global apparel firms to include these issues in their codes of conduct for their suppliers. Virtually all corporate, industry, and multi-stakeholder codes prohibit discrimination and harassment and violence against women. More recently, and in part due to the ineffectiveness of these soft law approaches, these issues are now covered in the new due diligence legislation being enacted in Europe.
More from Sourcing Journal
Discrimination
This swing toward hard law has implications for global companies. How will they measure and demonstrate progress on these issues in their supply chains? How will regulators score gender discrimination in their due diligence schemes?
In-terms of pay discrimination, the simplest measure is the earnings differential between men and women doing the same job. The UN and ILO have observed that women do more for less in the apparel industry. Reports indicate that in several apparel-exporting countries, the gap in pay for equal work between men and women in apparel production ranges from 41 percent in Pakistan to 2 percent in Cambodia in 2019 and 2018, respectively. Another ILO study found an average gender pay gap of approximately 18.5 percent across several garment-exporting countries, of which only 4 percent could be explained by age, education and other factors, leaving 14.5 percent of the gap explained by gender discrimination.
Yet, we do not have systematic evidence. National statistics in many apparel-exporting countries do not publish data on gender-based wage differentials in the apparel sector. And despite banning discrimination in their codes, multi-stakeholder institutions such as FLA, Fair Wear and Better Work do not regularly collect gender-disaggregated wage data through their audits—a surprising omission as occasional studies by these initiatives show that women are paid less than men. (In one Fair Wear brand “‘performance check” a member brand calls on Fair Wear to provide more stringent requirements for data collection on the topic of gender.)
Leading global apparel brands and retailers also do not appear to collect gender-based wages as part of their audits. (Or if they do, they do not report it in the corporate responsibility reports on their websites). H&M, for example, provides data on country-wise average wages for their factories, which is more than most global brands divulge. Notably absent, however, is data on gender-based wage differentials. Adidas reports the percentage of Tier 1 suppliers in which wages surpass the legal-minimum but, again, does not differentiate those wages by gender. Levi Strauss & Co. as of September 2023 does not report wage or pay equity data for its supply chain (in an otherwise informative annual report). Patagonia, noted for its reputation as leading on women’s rights issues such as childcare benefits, publishes an annual summary of its suppliers’ living wage data but this is not gender disaggregated.
Social auditing firms are often required to collect wage data for their clients, and some occasionally report aggregate wage data in different countries, but do not break out that data in gender terms. Given the lack of data, it is no surprise that Cornell GLI’s analysis of more than 40,000 audits across multiple industries during the 2011-2018 period, uncovered very few cases of violations of gender discrimination.
To sum up, we have a relatively easy-to-measure aspect of a core labor and human right that is present in international instruments and multi-stakeholder codes of conduct. But we have almost no data from firms or their joint initiatives despite 25 years of private regulation.
Before we ask, “how much longer should this be allowed to continue?” we will look at data on a determinant of gender-based discrimination, harassment and violence.
Gender-based violence
Approximately 80 percent of the global apparel production workforce is female, yet the majority of supervisors are male, setting up the conditions for discrimination and, more acutely, gender-based harassment and violence.
But, as with pay discrimination, systematic data on gender-based harassment and violence is hard to come by. Given the risks, very few female factory workers will share with administrators, auditors or authorities instances of gender-based violence. As noted previously, the analysis of audits report very few instances of gender discrimination and harassment.
Yet, there is a lot of anecdotal evidence which has been documented by a variety of NGOs active in the garment sector. Asia Floor Wage Alliance (AFWA) for instance has published several reports highlighting the prevalence of gender-based violence in apparel factories servicing Walmart, H&M, GAP and other global brands. An ILO survey in Myanmar revealed that 42.5 percent of women across 14 garment factories have experienced sexual harassment at work, and a ILO Better Work survey in Indonesia documented that, prior to the program’s intervention, four out of five women apparel workers experienced some degree of sexual harassment in their factories.
It has long been accepted that a key way of reducing the risk of gender-based violence in apparel factories is to increase the number of female supervisors and achieve at least gender parity, i.e. the percentage of female supervisors is similar to the percentage of female workers.
Women make up the vast majority of apparel workers in Bangladesh and Cambodia but make up only less than a quarter of production supervisors, according to surveys of factories in those nations. AFWA documents the gendered hiring outcomes in garment supply chains in Bangladesh, Cambodia and India: in various departments such as fabric store, cutting, fusing/pasting production and finishing/packing—women formed less than 20 percent of the supervisory group.
We argue that global brands serious about reducing gender-based violence in their supply chains (as required by German and European regulation) must invest in helping their suppliers train more female supervisors. A relevant outcome measure is this: the ratio of the percentage of female supervisors to the percentage of female workers in their supply chain factories. This is a simple, clear and easy to understand metric that would show progress (or lack of it) over time on gender equity in hiring and promotion, as well as probability of gender-based harassment and violence.
As an example, we report below data on supervisors collected by a global retailer from its supply chain factories. While this retailer has factories in many countries, we selected two factories from each of 11 countries to illustrate the variation within and (very broadly) across countries in the percentage of female supervisors.
The data from Table 1 show considerable variation in the percentage of female supervisors both within and across countries. Apart from one factory each in Egypt and India, women workers constituted the majority of employees in these factories. Four factories of 22 total exceeded the gender parity requirement (where the percentage of female supervisors was higher than the percentage of female workers).
The percentage of female supervisors was very low in Bangladesh and Guatemala. But the factories in China, Haiti, India, Indonesia, Sri Lanka and Vietnam exhibit significant variation; some of the factories in these countries need to increase substantially the percentage of female supervisors to reach gender parity, while other factories need less of an increase. The hiring inequities are clear here, and we expect that the risk of gender-based violence is lowest in factories where the percentage of female supervisors is highest.
This data in Table 1 is from a leading global retailer (our NDA does not permit us to identify the company). But it is not clear that most companies are collecting this information. We are encouraged that some companies have started setting goals. Tesco, for example, has committed to ensuring that 30 percent of its Tier 1 supplier managers are women by 2025 as part of an agreement with the International Union of Food, Agricultural, Hotel, Restaurant, Catering, Tobacco and Allied Workers’ Associations (IUF). To achieve this goal, the company committed to collect and publish gender disaggregated data on wages and professional roles, including for those workers on flexible and informal work arrangements.
Conclusion
This brings us back to the data and accountability problem we ran into at the end of the pay equity discussion above.
Despite conventions, soft law instruments, voluntary codes of conduct, countless audits and countless reports deployed over the last 65 years, we still do not have clear, systematic data regarding supervisory hiring and wage differentials in fashion’s supply chains. The irony is that this data are easy to collect. Every factory maintains a payroll which auditors are supposed to audit for a range for wages, working hours and overtime violations. To calculate average wages and supervisor-worker ratios by gender takes a few clicks of a mouse in a payroll database.
What’s missing is liability for lead firms that fail to prevent or remediate these harms. There is no meaningful penalty (or, if you prefer, sufficient disincentive) attached to failing to act out their obligations to workers.
This may be about to change. The new raft of laws in Europe—particularly the Corporate Sustainability Due Diligence Directive in the EU (not yet in force) and the 2023 German Supply Chain Due Diligence Act—require that lead firms look for and prevent or remediate discrimination by gender in their home operations and in their supply chains.
The hope is that hard law will beget hard data on gender pay differentials and hiring, but much depends on what companies will be required to report. For example, will they be required to report on “soft” inputs such as how their training programs aim to minimize risk of gender discrimination? Or, will reporting swing to requiring hard data on actual outcomes, such as actual wages by gender. The easiest pay discrimination metric is to report, in percentage terms, is the ratio of female wages to those of males, controlling for occupation and seniority.
In our view, the success of this new suite of corporate accountability measures in Europe will depend to a large extent on the requirement that companies report annual (and historical) data on workplace and labor rights outcomes.
Reporting is, of course, insufficient. It not the point of the due diligence regulations. The point is corporate accountability and large-scale and lasting improvements in livelihoods, working conditions and labor rights for workers in global production. But reporting of outcomes data is a necessary condition for the brands and manufacturers working to meet the standard, and for the regulators, investors and unions trying to see which lead firms are doing what’s required and which are not.
Acknowledgement: We are grateful for the excellent research assistance provided for this analysis by Katherine DeRose, undergraduate student at Cornell University.
