Can multinational buyers help curb river pollution by Bangladesh’s garment industry?

Bangladesh is the second-largest exporter of garments in the world. In 2024, over 4,000 factories exported garments worth approximately USD 38 billion to over 150 countries, including major global brands such as Uniqlo, which are regular buyers of Bangladeshi garments.

What is responsible sourcing – and why does it matter for Bangladesh?

As the fashion industry faces mounting pressure to make its supply chains more environmentally and socially compliant, there is growing attention to the need for multinational firms to ensure that their suppliers in developing countries maintain environmental, social, and ethical production standards – an approach popularly known as “responsible sourcing”. Against the backdrop of growing consumer demand for sustainable brands, the key question is: has sourcing from developing countries become more responsible? 

We explore this question using Bangladesh’s garment industry in an IGC study investigating whether exporting to international brands plays a role in mitigating the hazardous environmental impacts of industrial production. In particular, we focus on the water quality of rivers located near garment factories – Bangladesh presents a compelling case, as it is a riverine country. With over 900 rivers flowing within its borders, industrial water pollution poses a serious threat not only to public health but also to the ecological balance of the country.

How do brands ensure environmental compliance?

The rise of postmodern consumerism, particularly in Western countries, has shifted consumer-producer dynamics. Consumers now care deeply about brand ethics and corporate responsibility, and many are willing to pay more for brands that adopt responsible practices. This creates pressure on well-known brands to adopt ethical supply chain management. However, if brand loyalty makes demand less sensitive to price, brands can pass the cost of being "responsible" to consumers; as large buyers, they can also shift some costs to exporters in developing countries. 

We hypothesise that firms in developing countries that export to such brands contribute less to water pollution due to stricter environmental audits. Based on our factory visits in Bangladesh, we identify two main mechanisms through which brands ensure compliance. 

First, brands require suppliers to obtain environmental certifications such as OEKO-TEX (a global certification system for textiles and leathe r goods) and Zero Discharge of Hazardous Chemicals (ZDHC) certification. Certification agencies employ local experts who conduct regular and often unannounced audits, monitoring various aspects, including the installation and use of effluent treatment plants (ETPs). 

Second, many brands have liaison offices in Bangladesh with in-house audit teams that frequently inspect factories.

Linking customs data from nearby firms with riverine water quality

To investigate the causal relationship between trading with brand multinationals and local water quality, we construct a novel dataset by combining administrative customs data from the National Board of Revenue’s Automated System for Customs Data (ASYCUDA++) with surface water quality data provided by the Department of Environment, Government of Bangladesh.

Our dataset, which spans from 2014 to 2021 at a daily level, includes shipment dates, names and addresses of exporters and importers, and product details. To identify brands for our study, we use the Fashion Transparency Index, which ranks up to 250 fashion brands based on the extent of information they disclose about their social and environmental policies, practices, and impact. Data on river water quality are collected from 127 water quality monitoring stations located along rivers in Bangladesh from 2010 to 2019. We use dissolved oxygen, which is the amount of oxygen present in water, as our primary indicator of water quality, with higher levels indicating better water quality.

While our export data is available at the firm level, the water quality data is collected at the monitoring station level. Typically, multiple factories are located around each station. Thus, we consider firms located upstream and within 10 km of each monitoring station.

Figure 1: Locations of water quality monitoring stations and garment factories

Majorly exporting to multinationals leads to persistent improvements in water quality

We find that overall water quality deteriorates as the number of exporters within a circle increases. However, this negative association becomes statistically insignificant when only those firms that export to multinational brands are considered. These results suggest that trade with brands may help mitigate the adverse environmental effects of export production.

When we further investigate the causal impact, we find that major exports from local firms to multinationals increase dissolved oxygen levels in nearby rivers by 64%, indicating substantial water quality improvement. This positive impact on water quality persists over four years, and there is no differential effect during the period before the majority of firms near that station start exporting to brand multinationals.

Figure 2: Exporting to multinationals impacts local water quality

The right incentives can improve environmental quality despite weak state capacity

Responsible sourcing initiatives have been a growing trend in recent years, and in developing countries, where the state capacity to regulate and monitor is weak, pressure from global buyers can enhance environmental compliance. To incentivise exporters to comply with this pressure, even when the cost of compliance is high, it is likely that buyers support compliance by offering better contractual terms (for example, higher prices, larger volumes, longer-term agreements, and greater flexibility).

The findings of our study suggest that, with the right incentives, responsible sourcing can improve environmental quality in developing countries without compromising economic growth.

This research was presented as part of a session on "Greening Bangladesh's garment industry: Are we doing enough?" at LSE Environment Week.

Watch a recording of the webinar below: