Hide table of contents

This report was conducted within the pilot for Charity Entrepreneurship’s Research Training Program in the fall of 2023 and took around eighty hours to complete (roughly translating into two weeks of work). Please interpret the confidence of the conclusions of this report with those points in mind.

We are grateful to the experts who took the time to offer their thoughts on this research.

For questions about the structure of the report, please reach out to leonie@charityentrepreneurship.com. For questions about the content of this research, please contact Zuzana Sperlova at zuzkasperlova@gmail.com or Moritz Stumpe at moritz@animaladvocacyafrica.org.

The full report can also be accessed as PDF here.

Thank you to Karolina Sarek and Vicky Cox for their review and feedback.

Executive summary

The aim of this report is to explore the potential of corporate campaigns against retailers of silk products, as a promising intervention to improve animal welfare.

Up to a trillion silkworms are killed every year in order to produce silk, a luxury material popular in the fashion industry. Most silkworms are farmed in China and India, where around 10-47% of them die due to diseases. The worms that make it to the pupae stage, where they encase themselves in cocoons and start their metamorphosis, are then boiled alive in order to extract the silk thread from their cocoons. While most of the discussions on animal welfare issues surrounding this silk production process have focused on the killing of pupae, our research suggests that the majority of animal suffering for silkworms is caused by diseases during their lifetime. Predominantly because of this factor, we expect silkworms to lead net negative lives, which means that reducing the number of silkworms farmed would be a net positive for animal welfare. This is of course conditional on their sentience, for which strong evidence is lacking and unclear. We account for this uncertainty in our analysis.

Corporate campaigns for animal welfare have impacted the food industry, for example through cage-free egg and broiler campaigns, and the fashion industry, for example through anti-fur campaigns. As of now, there are no corporate campaigns focused on silk or any other insect species. Based on the proven track record of corporate campaigns in different areas and some evidence that consumers and retailers can be made to care about insect welfare and other issues related to silk production, it seems plausible that corporate campaigns against retailers of silk products could be a promising intervention. However, we remain very sceptical about the sympathy that the public may extend to insects like silkworms. While silk campaigns may use the images and videos of boiling cocoons and worm farms to their advantage in generating public pressure, such imagery is unlikely to be as emotionally impactful as that of caged hens or other vertebrate animals. In addition, there is a lack of fully adequate alternatives to silk, which might hinder retailers from moving away from silk products. The alternative options are either of lower quality and mostly synthetic, like nylon and lyocell, or they rely on innovative technologies that are yet to be scaled up and commercially viable, like spider silk. The experts we interviewed emphasised the need to offer alternatives to encourage retailers to transition away from silk.

Overall, corporate campaigns targeting silk retailers appear to be a promising intervention, given the notable success of similar campaigns in other domains and the ongoing shift away from animal-based materials in fashion. Besides direct animal welfare issues, silk campaigns could leverage concerns about sustainability and human rights issues associated with silk production.

The expected impact and cost-effectiveness of such campaigns crucially depend on the strategic selection of target countries and retailers. A significant unresolved uncertainty in this report is determining the share of the total silk market attributable to specific retailers. We recommend consulting industry experts or commissioning market research to quantify the exact amount of silk that can realistically be targeted with one campaign. This would reduce the large width of our cost-effectiveness estimate, which currently lies between 0.1 and 1,311 welfare points per dollar, with a geometric mean of 12 welfare points per dollar. The lower cost effectiveness and remaining uncertainties lead us to the conclusion that further targeted research is necessary before deciding for or against recommending this intervention for a new charity.

Should further research endorse this intervention, we think it should be implemented by a newly incubated charity. The existing charities working on corporate campaigns that could expand their focus to include silkworms are either not strongly impact-focused or have a narrow focus on very different farmed animals.

1     Background

1.1    Cause area: Insects, farming, and fashion

1.1.1 Moral relevance of insects

Invertebrate animals are generally considered to be among the ‘neglected species’, a group of animals whose welfare concerns are rarely being addressed and considered. Rethink Priorities has done a number of reports on invertebrate welfare, including a 10-part report series exploring the philosophy of pain in non-humans, existing literature on the topic, estimates of cognitive capacity by taxa, and several other topics (Schukraft, 2019). They conclude that invertebrate welfare is a cause area worth prioritising. More than 99.9% of animals on earth are invertebrates, while less than 1% of Effective Altruism (EA) funding goes towards their welfare (as of 2019).

Insects are an important subgroup of invertebrates. There is no consensus among animal experts and environmentalists on whether or not we should give insects moral relevance. However, Faunalytics (2016) notes that:

“Time after time, societies have overturned widespread and strongly held views about whose welfare matters. The history of moral progress should remind us that what’s obvious isn’t always right. We must be humble enough to at least consider the possibility that we ought to care about insects.”

The question of sentience in insects is largely neglected in the literature. Rethink Priorities (Schukraft) has compiled the available evidence for invertebrate sentience in a 2019 report. The uncertainty of determining sentience in these species is high:

To develop accurate cost-benefit models that can be used to allocate resources across the animal welfare movement, we need to take the possibility of invertebrate pain and pleasure seriously. But determining whether invertebrates have the capacity to experience pain and pleasure in a morally significant way is an extraordinarily complex and difficult undertaking. There is tremendous uncertainty at virtually every level at which one might investigate the matter. We don’t expect to conclude with high confidence that invertebrates do or do not experience morally significant pain and pleasure. Such an outcome is too ambitious. Rather, our goal is to clearly map out the problem so that we can begin to systematically reduce key uncertainties in a cost-effective manner.” (Schukraft, 2019)

While it seems clear that we are still left with tremendous uncertainty, Faunalytics (2016) recommends approaching this uncertainty by assigning a non-trivial likelihood to insects suffering. Even if the likelihood assigned to it is extremely low, the number of insects that exist is so significant that the scale of the problem is still large. Section 3.2.3 of this report will tackle questions of moral weight, welfare ranges, and probability of sentience in more detail.

There are currently a small number of initiatives working on addressing the problem of insect welfare specifically. Open Philanthropy has funded insect welfare research with grants in recent years (2022a, 2023a). The Insect Welfare Research Society which was founded recently (2023), is also sponsored by Open Philanthropy, and has an academic focus:

The Insect Welfare Research Society is an academic-led research society founded in March 2023. The society aims to support 1) the global insect welfare research community and 2) the incorporation of evidence-based information on insect welfare into policy and practice by relevant stakeholders. We also support welfare research on understudied terrestrial arthropods that are not insects (e.g., spiders and other arachnids, myriapods like millipedes, etc.) and on understudied aquatic invertebrates (e.g., shrimp/prawns, horseshoe crabs, etc.)” (Insect Welfare Research Society, 2023).

1.1.2 Insect farming

When it comes to the specific topic of insects being farmed, the numbers of individuals raised for food and animal feed annually reach 1-1.2 trillion, with between 79-94 billion insects being alive on farms on any given day. The largest producers are Thailand, France, South Africa, China, Canada, and the US. This seems to be a very fast-growing industry, with a lot of new investments and many startups. The most commonly farmed species and their estimated numbers are summarised in the table below (Rowe, 2020). 

Table 1: Number of farmed insects

Industrial insect farming is a relatively new practice, and a lot of producers seem to aim to replace fishmeal with insects. The numbers in the table above are likely to multiply massively if this becomes standard practice. 

It appears that more grants and initiatives are being brought forward, mostly by the EA community. One of these is the newly launched (2023) Insect Institute (funded by Rethink Priorities). Their mission states:

We address challenges and uncertainties related to the production and use of insects for food and feed. We aim to assist this novel industry, policymakers, and other interested parties by providing evidence-based information surrounding the rearing of insects and the creation of a food system that promotes public health, animal welfare, and sustainable protein production” (The Insect Institute, n.d.).

Besides animal and human feed, there are other purposes insects are being farmed in smaller numbers - cochineal are used for red dye in food and cosmetics, honeybees are used for food and pollen products, cockroaches are farmed for traditional Chinese medicine, and silkworms are used for their cocoons, which are turned into silk.

There is an even larger number of insects alive in the wild. The opinions on whether people should intervene in nature vary among experts (Faunalytics, 2016). Despite the lack of consensus, it might be worth looking into how to limit human-caused suffering for wild insects. This is out of the scope of this report, however.

1.1.3 Animal use in fashion

There are many different ways animals are still used in the fashion industry today. A number of animals are used for wool - most famously sheep, but also goats for cashmere and Angora rabbits for their famously soft wool. Leather, mostly from cows, is another very commonly used material for shoes, clothing, and furniture. Birds such as ducks or geese are used for their feathers (down) for jackets and bedding. Fur coats from minks, foxes, rabbits, and many other species used to be commonly used and despite many modern alternatives, the fur industry still exists. Silkworms are used for their cocoons, which are then reeled into silk.

The species most commonly used in the fashion industry are listed in the table below, together with their estimated numbers.

Table 2: Number of animals farmed for fashion

SpeciesMaterialNumber of animals (in existence on an average day, if not indicated otherwise)Material produced (tonnes)
SheepWool1.2 bil1.2 mil (clean wool)
Cashmere goatsCashmere wool100 mil20 k
CowsLeather~1 bil9 mil
Angora rabbitsAngora wool~50 mil2.3-4 k
Geese and ducksDown feathers~3 bil270 k
Various mammalsFur

~100 mil

(killed annually)


420 bil -1 tril

(killed annually)

200 k

Historically, animals and plants (e.g. cotton) were the only source of material for human clothing. In the last century, the share of animal-derived materials has decreased due to the development of many synthetic fibre alternatives, such as polyester, nylon, viscose, and other mostly polymer-based materials (Wikipedia, 2023). A lot of the animal-derived materials are nowadays considered luxury (especially cashmere, silk, and angora wool).

The fashion industry is associated with many welfare issues and concerns. There have been various corporate and policy commitments to better animal welfare or even bans on production of various materials.

1.2    Topic area: Silk

1.2.1 Silk production process

Silk is the material that silkworms produce to build their cocoons. The production of silk involves a complex and intensive process that makes use of the metamorphosis of silkworms into silk moths.

The process begins when a female silk moth lays between 300 and 400 eggs, after which she dies. These eggs take roughly ten days to hatch into tiny larvae. These larvae are then housed in designated boxes where they are given mulberry leaves to eat. Over the span of about six weeks, they undergo significant growth, multiplying their original weight by 10,000 times and consuming mulberry leaves that weigh 50,000 times their original weight. Once this growing period is over, the silkworms start the cocooning process. They attach themselves to a stationary object and spend between three to eight days spinning their cocoons (Rowe, 2021).

Once the larvae encase themselves in cocoons, they are most commonly killed by boiling them alive to extract the silk thread. Boiling is the prevalent method because it preserves the integrity of the long, continuous silk thread. If the pupae inside were to mature into a moth and break free, it would damage the thread, significantly decreasing its commercial value (Animal Ethics, n.d.; Rowe, 2021; Tomasik, 2016). Although there are various species of silkworms, the slaughter methods described above generally apply to all of them (Rowe, 2021). It is unclear how many silkworms exactly have to be killed in order to produce a certain amount of silk. Estimates vary somewhere between 3,000 and 15,000 worms per kilogram, but the most detailed and up-to-date report on this issue by Rethink Priorities argues that the correct number should be close to the lower end of this range (Animal Ethics, n.d.; Faunalytics, 2016; Rowe, 2021; Tomasik, 2016). Many of the dead pupae are recycled as feed for fishes, pigs, and sometimes even for human consumption (Animal Ethics, n.d.). A select few silkworms are not killed and are allowed to mature into moths, ensuring the continuation of the silk production cycle by laying eggs for the next generation (Animal Ethics, n.d., Rowe, 2021).

The vast majority of silk production is located in Asia, especially China and to a lesser extent India. In 2017, China accounted for ~80% of global production, India for ~18%, with only ~2% distributed among other countries (Rowe, 2021). Consumption patterns are more distributed and will be investigated in more detail in the geographic assessment later in this report.

1.2.2 Animal welfare concerns

Most of the discussions on animal welfare issues surrounding silk production have focused on the killing of pupae via the process described above. Accordingly, people have argued over the sentience of pupae during this stage of the metamorphosis. Tomasik (2016) provides a good overview of the different factors that have been cited in that debate and concludes that huge uncertainties remain. Rowe (2021) also emphasises large uncertainties: “Currently, the evidence that silkworms suffer during slaughter is extremely limited, so it is hard to come to a definitive position on the welfare consequences of boiling silkworms specifically.”

However, Rowe (2021) provides a very helpful reframing of the issue, illustrating that most of the animal suffering in silk production does not come from the killing of worms at the pupae stage, but rather from diseases that affect worms during their lifetime. The reasoning seems worth citing in full here:

“Even if the slaughter is painful, which is not currently borne out in available evidence, it is likely a very brief experience, lasting just a few minutes, or around 0.02% of the average silkworm’s life, compared with disease taking up around 70%. So, any intervention to reduce silkworm farming will most likely primarily be benefiting silkworms who will die on farms due to disease loss.”

Estimating the extent of suffering caused by silk production, Rowe (2021) cites the following numbers:

“I estimate that in 2017 (the most recent year comprehensive data is available) that between 420 billion and 1 trillion silkworms were killed or died due to disease on silk farms. I further estimate that silkworms lived a total of between 15 trillion and 37 trillion days on farms, of which at least 180 billion to 1.3 trillion days involved some degree of potentially negative experience, such as being slaughtered or suffering from a disease (note that slaughter only counts as a small fraction of day in this estimate, since it only lasts a few minutes). [...] While only 61 billion to 170 billion of these worms die due to diseases and pests, diseases cause up to 99% of the cumulative days of potentially negative experiences I identified on farms, (with the remaining 1% caused by slaughter).”

The fact that most of the animal welfare concerns for silkworms affect their lives prior to slaughter means that it is much more relevant whether silkworms are sentient over the course of their lifetime than only during the pupae stage. Silkworms being sentient before the pupae stage is also not a given, but seems less controversial. Rethink Priorities has done highly relevant research on this issue (Fischer, 2022, 2023). For now, we will not go into detail on this complicated topic, but will assume that there is a non-trivial likelihood that silkworms are sentient. Given the enormous number of animals involved in silk farming, even a small likelihood of sentience justifies investigating how welfare can potentially be improved.

In section 3.2.3, we will go more into detail on questions related to the quality of lives that silkworms live under current farming conditions. Apart from the slaughter and diseases mentioned above, we need to consider further aspects of their lives in order to estimate whether the lives of farmed silkworms are net positive or negative on average. We will also build on Rethink Priorities’ work and integrate estimates on the welfare range and probability of sentience of silkworms in our cost-effectiveness analysis.

1.2.3 Existing alternatives to conventional silk production

As the traditional silk production process raises significant animal welfare issues, some have explored alternative methods like Ahimsa silk. Ahimsa silk, inspired by Jain principles and also called peace silk, allows the moth to emerge from the cocoon before harvesting the silk. Though this method spares the moth's life, the resulting silk is of inferior quality, resembling wool more than conventional silk. (Animal Ethics, n.d.)

Even though Ahimsa silk avoids the killing of silkworms, it is not without its problems. The worms are still held in captivity and face harm and diseases during the production process. The method also requires more silkworms to produce the same amount of silk, as fewer cocoons are suitable for reeling. Further complicating the ethics of Ahimsa silk is the condition of the moths that emerge from their cocoons. The majority of animals that emerge from their cocoons are not capable of flying and die very shortly after hatching, a circumstance that could potentially lead to more suffering than the traditional method of boiling (Animal Ethics, n.d.; Rowe, 2021). Tomasik (2016) further adds that promoting Ahimsa silk could unintentionally encourage consumers to revert to traditional silk. If the price of Ahimsa silk rises due to increased demand, ethically conscious but financially limited consumers might opt for the less expensive, conventional silk, perpetuating the cycle of animal suffering.

In sum, "It seems that Ahimsa silk is not necessarily a more humane alternative, unless one was only concerned about suffering during slaughter, which doesn’t seem to be the main source of suffering in silk worms" (Rowe, 2021). Therefore, Ahimsa silk should not be viewed as a comprehensive solution to the ethical concerns surrounding silk production.

A more viable path towards reducing animal suffering caused by silk production seems to simply lie in the replacement of conventional silk with alternative materials, like artificial silk, nylon, mercerized cotton, polyester, and many others (Tomasik, 2016). This would completely remove silkworms from the production process and thereby avoid the suffering caused to them in the process. We will go more into detail on alternative materials in section 3.2.1 and the interview with Nicole Rawling.

1.3    Intervention options

Promoting non-animal alternatives to silk is thus one intervention that could potentially alleviate animal suffering by reducing the need and thus scale of silk production. For instance, the Material Innovation Initiative (MII) is “exploring next-generation silk alternatives that are less harmful to both animals and the environment” (Material Innovation Initiative, 2021). Rowe (2021) also mentions the promise of developing and promoting alternative silks to reduce silk production and the ensuing animal suffering.

However, according to their analysis, the most effective route may be campaigns targeting retailers, particularly in countries of the Global North. Rowe (2021) calls this "plausibly the most promising avenue for animal advocates to reduce silk production." They refer to successful past campaigns by PETA that have led to retailer commitments to stop the sale of silk products.

Other potential interventions that Rowe (2021) deems somewhat promising lie in advocacy for policy changes that ban the use of silkworms or consumer advocacy to change consumption patterns. Such campaigns could benefit from the existing public interest in more ethical silk production and human welfare issues associated with the process. “One notable reason silk bans and commitments to end the use of silk have been successful is that the silk industry frequently is accused of having violated human rights, through human health issues, child labour abuses, and slavery on factories” (Rowe, 2021).

Improving hygiene on silk farms could theoretically also reduce suffering by decreasing the prevalence of disease among silkworms. However, Rowe (2021) is not optimistic about this avenue. They point out that farms already have intensive disinfecting routines, and further hygiene campaigns may not significantly reduce diseases. Moreover, such campaigns could unintentionally make silk cheaper, increasing demand and potentially leading to more silkworms being farmed.

In conclusion, while various interventions exist, campaigns targeting retailers in countries of the Global North appear to be the most promising route towards reducing animal suffering caused by silk production. This is supported by past successes in securing retailer commitments, as well as the multiple human-welfare angles that could strengthen the argument against silk.

2     Theories of change

2.1    Scope of the intervention

Based on the background given in the previous section, this report focuses on corporate campaigns against retailers of silk products in the Global North as a potential intervention for a new charity to run.

Capriati (2018) provides a good overview of how corporate campaigns work and what they include:

“Corporate campaigns consist of efforts aimed at shifting corporate practices towards systems that improve animal welfare. Corporate campaigns can use a variety of strategies. In some cases, organisations promote change from within the company by supporting aligned stakeholders or offering technical assistance. Alternatively, advocates can launch social media campaigns against companies that refuse to engage. These campaigns can give consumers – who are often unaware of the problem – information about animal treatment in the production chain, creating negative publicity for companies. This creates a PR incentive for companies to address the issue.“

This description highlights the dual nature of corporate campaigns. On the one hand, they are based on a collaborative approach, working together with companies to implement certain commitments. On the other hand, they usually also involve a more confrontational approach that pressures companies via negative publicity. While it is typical for these two strategies to be divided up among two different organisations (‘good cop and bad cop’), the theory of change (ToC) outlined in this chapter depicts both aspects of this two-pronged approach, in order to analyse all of the relevant steps involved in this intervention. When it comes to implementation, we need to keep in mind that it might be helpful or necessary to have multiple/different actors involved.

The ToC even makes a further distinction: public pressure can be generated by activist-led activities (e.g. protests), but can also come from the broader public, as they are influenced by public campaigns. The “upper” path (outputs and outcomes) in the visual ToC below concerns the broader public support, while the “lower” part represents the direct actions by activists.

2.2    Causal chain

The graph below shows the most important steps that need to be completed for corporate campaigns against silk retailers to have the desired positive effect on animal welfare. The full visualisation can be found here.

The numbers indicate assumptions that need to hold for the different steps to work out as intended. Their colours represent the level of uncertainty for this specific assumption/part of the ToC prior to a deeper evidence review, from red (high), to yellow (medium), and green (low). These assumptions are outlined in more detail in the following section, highlighting some preliminary thoughts and considerations, before the crucial uncertainties will be addressed with a more in-depth evidence review in section 3.

2.3   Outlining assumptions and levels of uncertainty

1. There is information about which are the most important silk product retail markets

For assumptions #1 and #2, we need to do a geographic assessment, finding the countries that seem most promising for our work.

Our campaign should focus on retailers and countries that account for large shares of global silk product consumption, in order for them to have the largest effect possible. As our quick search did not show comprehensive statistics on silk products (only raw silk which is different from end products), we assigned this medium uncertainty. It would be quite problematic, if we could not find a country or group of countries to work in. However, even if there are no official statistics, we should be able to get some guidance from industry experts or other sources.

2. There are indicators that allow us to evaluate which countries provide a good environment for running corporate campaigns

The countries we focus on should not only have the potential to affect large parts of the global silk market, but should also provide an environment that is conducive to running corporate campaigns. Factors that play a role here are civil liberties (i.e. no legal restrictions, freedom to demonstrate, etc.), the strength of the animal advocacy movement and concern for animals among the public, successes of previous campaigns, and many more. It seems like data for good indicators should be available and we are fairly certain that we could find countries where corporate campaigns can be run properly.

3. Information about the main retailers of silk exists and is easy to find

As written under assumption #1, our campaign should focus on retailers and countries that account for large shares of global silk product consumption, in order for them to have the largest effect possible. As our quick search did not show comprehensive statistics on silk retailers, we assigned this medium uncertainty. It is unclear whether we should target broad-market retailers (like H&M or Gap) for whom silk only constitutes a minor share of their sales, or more niche and high-end companies for whom silk as a high-quality material plays a larger role. It would be quite problematic, if we did not know which retailers would be most promising to target. However, even if there are no official statistics, we should be able to get some guidance from industry experts or other sources.

4. Retailers are reachable and when contacted they express willingness to discuss

Once we have settled on certain retailers to target, we need to be able to get in contact with them. In the best case, they are open to changing their policies towards silk use.

Currently many retailers focus on being branded as ethical and sustainable and have committed to various targets. There could be low-hanging fruits of retailers that are willing to collaborate on moving away from silk. Still, it is far from certain that retailers are willing to collaborate with us. This uncertainty does not seem crucial though. If retailers are not willing to collaborate with us right from the start, we can run pressure campaigns to get them to reconsider their stance.

5. Our campaign reaches the public

As described above, public campaigns are important in order to generate further pressure on retailers that are not willing to comply with our demands. It might be that this “upper” path of our ToC is not strictly necessary, as it is more important that our campaign directly reaches retailers and we might not need broad public support. However, public support should significantly increase our chances of success. This is not just nice-to-have, but is important. Still, it is not as important as the direct path towards retailers.

In order to reach the public, we need to be able to design a good and convincing campaign. We need the right skills and expertise to do so, but can also learn from the wide range of campaigns that have been run by other organisations before (e.g. fur, cage-free, broiler). The learnings from these campaigns might not be directly applicable to our use case, as there should be a big information gap among the public when it comes to animal suffering in silk production.

Still, given that we can draw on learnings from other campaigns and that public support is not strictly necessary, this assumption does not seem like a huge uncertainty in our ToC. Whether our public outreach is sufficient to pressure and convince retailers is covered under different assumptions below.

6. Retailers consider changing their practices

Retailers might consider changing their practices based on their general openness (see assumption #4) or pressure generated by activists. The pressure generated by activists might not be strong enough though to incentivise retailers to collaborate before they get hit with public campaigns.

The direct impact of our outreach and campaign on retailers seems to be the most important path in our ToC, more so than the public pressure indirectly caused by our public campaign. If retailers are not willing to collaborate with us, a large amount of public pressure would be needed to get them towards a commitment, which seems hard to achieve. It therefore seems like a major uncertainty that we need to investigate, how likely it is that retailers will collaborate with us.

7. The public campaign leads people to actively pressure retailers

As written above, public pressure is an important supplement to the pressure generated by activists directly, but not strictly necessary.

Such pressure can come in many forms and barriers are relatively low for people to sign a petition or post on social media. Previous campaigns on other issues have shown that support for animal causes and the wider animal movement is strong enough to engage the public. However, it is quite uncertain whether people can be persuaded to care enough about insects to take active steps. It seems like the campaign would have to be very strong in order to ignite public pressure on such an opaque issue.

In sum, this assumption is relatively uncertain, but not the most crucial one in our ToC.

8. The campaign has an impact on consumer behaviour and demand goes down

Our assessment for this assumption is very similar as for #7 above. The assumption is very uncertain, as there is conflicting evidence from other campaigns about behavioural change and effects on consumer behaviour are usually small and short lived. However, we assume that this is not a crucial path in our ToC to influence companies via corporate campaigns.

9. Retailers have strong-enough incentives to make the commitments

The idea is that the direct action by activists and the potential public pressure generated by the campaign will lead retailers to make commitments against the use/sale of silk products. The question of whether this will be successful is probably the most crucial issue in our ToC. Without retailer commitments, most of the potential impact of our campaigns would be lost.

One important issue here is whether the (public) pressure generated via the previous steps is sufficient to drive retailers towards making commitments. This relates to the previous assumptions and it seems quite uncertain that pressure can be strong enough on a more niche issue like silk.

When asked to stop selling silk products, retailers might not just stop selling such products but might look for alternatives to replace them. For this scenario, it is highly relevant whether retailers have adequate, high-quality alternatives to use instead of silk. There seem to be some alternatives available, either cheaper (lyocell, nylon, etc.) or more luxurious new alternatives created by startups (spider silk, cactus silk, etc.), but the question of how adequate they are and whether retailers would be willing to use them instead remains uncertain. If cheaper alternatives already exist, we need to understand why corporations are not already using them and whether this indicates that silk is hard to replace. We could look into how transition from other animal materials happened in the past or why corporations might be hesitant to switch to alternatives.

10. Retailers have strong-enough incentives to follow the commitments

Once (or if) retailers have made the commitments, it is important that they also follow them in practice. This can either happen automatically (i.e. the corporations directly follow their commitments) or it may require sustained pressure on the retailers. As for making the commitments, this depends on the strength of the campaign and the retailers’ initial willingness to move away from silk products.

In any case, it is important for this intervention that retailers are followed up on their commitments. This has also been an important part of corporate campaigns in different areas (e.g. cage-free eggs). Enforcement rates for those campaigns seem to be relatively strong, but we will need to analyse this further.

11. Reduced demand by the retailers translates into reduced production

Once the targeted retailers stop or reduce their use of silk, this should have an effect on production. This is predicted by economic theory and simple logic and should be somewhat uncontroversial. However, it is an important question whether the retailers we would tackle use enough silk to make a significant difference on producers. This is not a given, but since we would select the most influential retailers for our campaigns (see assumption #3), this should be the case.

Another uncertainty concerns indirect effects. If certain retailers stop selling silk products, consumers who want to buy such products might simply shift their consumption to other retailers who still offer them. This is a significant worry, but also one that is not unique to silk. Campaigns against fur products are subject to the same uncertainty. Something similar could also be said of cage-free eggs, as consumers who want to buy cheaper products (which typically come from caged farming), could shift their consumption to other companies that still source such products. The worry about such a shift occurring should be larger in the case of silk, as this concerns an outright ban of a whole category of products, which certain consumers might be actively looking for. At the same time, it also seems likely that many consumers would simply buy an alternative non-silk product, as they are not actively looking for silk products specifically and are mostly led by what is offered by retailers. Overall, this is a potential effect that needs to be closely monitored, but we do not expect effects to be massively different than for other product categories that have been the target of corporate campaigns. We therefore mostly rely on evidence from such other campaigns (cage-free, fur, etc.).

One could also argue that one retailer stopping or reducing its demand might simply lead to a decrease in silk prices, which incentivises others to increase their demand as a result. However, this argument does not seem too convincing. The increased demand from others would once again increase prices, leading to a new market equilibrium, at somewhat higher prices and lower demand. Basic economic logic of supply and demand should hold. 

12. The production of silk will not be replaced by ‘ethical silk’ that has similar issues (disease) and lower yields (as it lets the cocoons hatch on their own)

It is conceivable that retailers are not willing to move away from silk completely, but want to switch towards more humane or sustainable methods. As outlined in section 1.2.3, the issue is that such alternatives are unlikely to lead to less animal suffering. Retailers or the public might not share our ethical assumptions though and think that what matters is only the killing of silkworms, not their farming conditions (and the suffering associated with that).

Considering the full scope of our intervention, this seems unlikely and we should be able to mitigate potential negative effects. Our campaign and outreach to retailers would make clear that other methods are not in fact an acceptable alternative.

13. Less silkworms being farmed is a net positive for animals

Finally, we need to understand whether a reduction in production would be a net positive for animals. This is only the case, if silkworms are sentient and the silkworms’ lives are net negative on average. A reduction in farming might also be a positive development based on other moral perspectives. 

This assumption seems likely to hold, given the welfare issues around silkworm farming outlined in section 1.2.2. However, as this is a crucial assumption for our ToC, we should investigate this in more detail. There are significant uncertainties around the sentience of silkworms as well as the quality of life for silkworms that do not suffer from diseases and whether their lives might be net positive.

3     Quality of evidence

The assumptions highlighted in the previous section need to be scrutinised in more depth by conducting an evidence review. This section is split into two parts. First, we focus on the steps and assumptions that are mostly under the control of the organisation that is running the intervention, trying to find evidence for or against the claim that the charity can get retailers to make and uphold commitments against the use or sale of silk. Second, we analyse the next steps that result from these commitments, investigating the assumptions that need to hold in order for the implemented commitments to have the intended positive impact.

3.1    Evidence that a charity can make change in this space

When trying to assess the evidence that a charity can achieve the desired change, i.e. getting retailers to make and implement commitments, we need to investigate assumptions #1 to #10 in our ToC. In order to focus our evidence review on the key questions that need to be investigated, we decided not to analyse every assumption separately, but rather to aggregate or group assumptions together wherever this seemed helpful. The table below gives an overview of the different assumptions and how we decided to aggregate and address them in this section.

Table 3: Assumptions and crucial considerations on whether the charity can get retailers to make and implement commitments

NumberAssumptionHigh-level questionPriority
1There is information about which are the most important silk product retail marketsCan we find promising countries to focus our campaign on?Not reviewed here, as this will be tackled in the geographic assessment in section 5
2There are indicators that allow us to evaluate which countries provide a good environment for running corporate campaigns
3Information about the main retailers of silk exists and is easy to find.Can we find the main retailers to target (based on their amount of silk sales and expected willingness to change)?Relatively low uncertainty, thus low priority; could be beneficial for later stages to already try to identify some retailers
4Retailers and brands are reachable and when contacted they express willingness to discuss
5Our campaign reaches the publicCan we engage the public to pressure retailers and reduce demand for silk?Relatively high uncertainty; this step is not strictly necessary for our ToC to hold though and only works as an amplifier
7The public campaign leads people to actively pressure retailers
8The campaign has an impact on consumer behaviour and demand goes down
6Retailers consider changing their practicesCan we get retailers to adjust their stance on silk and sign commitments based on our outreach and public pressure?Main uncertainty in our ToC; should be highest priority in the evidence review
9Retailers have strong-enough incentives to make the commitments
10Retailers have strong-enough incentives to follow the commitmentsCan we get retailers to follow through on  their commitments?Relatively high uncertainty, as this is an important assumption to hold

Below, we will tackle the four high-level questions identified for this part of the evidence review (excluding the first question, which will be tackled in section 5). We decided to do this in chronological order instead of priority, as this should make it easier for readers to follow the process through our ToC.

3.1.1 Finding retailers to target

The first uncertainty we are trying to resolve in this evidence review is whether an organisation that wanted to implement the intervention could identify promising retailers it should target. To maximise the expected impact with its intervention, the organisation should focus on retailers that sell large amounts of silk products and seem like they could be convinced to change their attitudes towards these sales.

An interesting data point here could be how the Open Wing Alliance (OWA) selects the companies it targets with its corporate campaigns for cage-free eggs, as such campaigns have been very successful (Bollard, 2017) and bear similarity to our intervention. Unfortunately, we could not find any public statement on the OWA’s approach in selecting companies. However, from The Global Manufacturers Report released in 2023, it is apparent that OWA’s targets are among the largest companies worldwide, suggesting that they target companies based on their size and influence in the industry (Open Wing Alliance, n.d.).

The significance of targeting large retailers is further supported by campaigns against fur farming. In one of their research newsletters, Open Philanthropy analysed the fight against fur farming and also stated that “advocates do best when they focus on the few hundred retailers that sell most animal products” (Open Philanthropy, n.d.).

More detailed evidence on how companies are selected for corporate campaigns comes from PETA, one of the most experienced campaigners in the fashion space. PETA is known for their on-the-street campaigns, associated with throwing paint at fur-coat wearers, but their tactics have changed in the last decade or two, focusing more on corporate outreach and collaborations with brands. Their strategy involves identifying industries or suppliers engaging in malpractices, then researching their clientele and supporters. When selecting which companies to target, they consider factors like the amount of the animal-derived material sold by the company, whether this company is likely to influence other companies, and whether the company’s customer base feels strongly about animals (for example young female customers typically care more about animal issues). For instance, in 2019, when an investigation into alpaca wool revealed horrifying welfare concerns, PETA identified brands (H&M, Gap, Anthropologie) that are both large and influential and likely to consider an alpaca ban, based on their previous promise to stop selling mohair products during a campaign. (Testa, 2020)

Summarising the evidence from corporate campaigns in related areas (cage-free eggs, fur, alpaca wool, and other campaigns run by PETA), we find that there is an established way of finding retailers to target.

This seems highly applicable to anti-silk campaigns. As our research for this report did not yield any strong and useful statistics on who the biggest sellers of silk products are, PETA’s approach outlined above should be the best approach, starting with producers and suppliers (primarily in China and India) and understanding who their biggest customers are. Key market players are highlighted in reports like this one or this one. Talking to industry experts should also be a promising avenue for finding out who the key retailers are. The research conducted for this report suggests that the key targets might simply be the biggest retailers of apparel, home decor, and similar key uses for silk. While these retailers are not exclusively focusing on silk products, they are so large that the amount of silk sales and their influence on other companies probably make them more promising targets than smaller retailers that might specialise in silk or luxury products in general. Broad-market retailers also seem more likely to agree to switching away from silk products, as silk is not as vital to their business. Even though this path is suggested by our research here, we are uncertain of this claim and suggest that a potential charity running this intervention investigate in more depth which companies use and sell the largest amounts of silk. Some more insights on the question of whether to focus on broad-market or luxury retails are also given in section 3.2.2 and are discussed in our interviews with Jonas BeckerPJ Smith, and an expert on fashion campaigns. For now, it seems likely that top targets will come from lists like this one, showing the top 20 fashion retailers in Europe. A quick sanity check revealed that the majority of such retailers should be selling silk products (e.g. Zara, H&M, Zalando) (H&M, n.d.; Zalando, n.d.; Zara, n.d.), with companies that have already pledged to stop silk sales, like Asos (The Guardian, 2018), being exceptions rather than the norm.

3.1.2 Engaging the public to pressure retailers

The second uncertainty we are trying to address here concerns the “upper path” of outputs and outcomes in our visual ToC, namely if and how strongly a campaign against the use/sale of silk can leverage public pressure to secure the desired commitments. Corporate campaigns usually involve a mixture of direct contact with the companies, and public campaigns to enhance the pressure on those companies. The public-facing part of corporate campaigns is usually driven by activist groups, but leverages the general public to exert greater pressure on the targeted companies. Our ToC shows two ways in which the public can pressure companies: Either via direct action that threatens the companies’ image or by changing their demand behaviour.

Our evidence review finds that changes in demand caused by the campaigns are very likely not a significant factor in convincing companies to make commitments. For instance, boycotts, which can be seen as a specific form or part of corporate campaigns, have been shown to have negligible effects on revenues (Institute for Policy Research, 2017). Analyses and explanations on the key mechanisms of corporate campaigns by groups like the OWA and PETA did not mention demand as a significant factor (Beck, n.d.; Testa, 2020). While the voluntary commitment of Asos to ban mohair, cashmere, and silk seems to be based on gradually shifting attitudes and demand patterns of consumers, and thereby shows that demand does play a role on companies’ decisions, demand changes do not fit with the short timelines and intense pressure which usually characterise corporate campaigns (The Guardian, 2018). We therefore did not investigate the likelihood of assumption #8 (that the campaign has an impact on consumer behaviour and demand goes down) to hold true, as we generally expect this not to be a major factor.

The much more important way in which the public can be leveraged is direct action against the companies’ public image. “While boycotts rarely hurt revenues, they can threaten a company’s reputation, especially by generating negative media coverage” (Institute for Policy Research, 2017). Bollard (2017) highlights that engaging public pressure via social media, online petitions, calls, or street protests is an important part of campaigns when companies do not collaborate or make commitments. The same is highlighted by Testa’s (2020) analysis of PETA’s campaigns, as PETA can draw on the backing of millions of followers and supporters. It thus seems very likely that support from a broad group of people from the general public boosts public campaigns. However, all of these pressure tactics are usually planned and coordinated by the campaign team. It does not seem to be an important aspect of the ToC that public groups organise themselves ‘organically’ to put pressure on the companies, but rather that they align themselves with the existing campaign (Beck, n.d.).

Applying these tactics to the case of silk retailers should be straightforward, using a public campaign to leverage the public to exert pressure on the companies. However, one key uncertainty is whether the public cares enough about silk production issues to raise their voices. While support for cage-free vs. caged egg production seems to be strong across a majority of countries (Sinclair et al., 2022), we could not find opinion polling data on public attitudes towards silk. One positive piece of evidence for consumers caring about insects comes from a 2012 controversy around Starbucks using cochineal as a red rye in its drinks. As the company faced negative publicity, it was willing to move away from the use of this animal-derived ingredient. However, it is important to note that the strength and success of public pressure in this case might have come from people’s disgust reaction to consuming insects more so than actual welfare concerns (Shute, 2012; Witrak, 2012). Similarly, a campaign against silk products could broaden its messaging beyond the mere animal welfare concerns and also highlight the human-welfare concerns related to silk production (e.g. modern-day slavery) (Rowe, 2021). In general, it seems very unlikely though that silk campaigns can garner similar levels of support as cage-free campaigns, due to the generally lower levels of moral concern for insects (Faunalytics, 2016). 

In sum, public support and backing has been an important part of pressuring companies, using tools like direct emailing or social media. However, a new organisation running campaigns on a more niche issue like silk might not have enough followers or public reach to exert strong pressure on companies. It might be that such an issue would be better fitted within a bigger organisation that is able to mobilise its supporters for this issue. While specialisation is usually a plus for charitable programmes, this is a watchout here and we need to consider whether this intervention might be better placed within a larger organisation that addresses animal farming for fashion generally, like PETA. We further explore this topic again in section 9.

3.1.3 Convincing retailers to sign commitments

The previous question on whether public support would be strong enough to put sufficient pressure on companies leads directly into the next and largest uncertainty in the ToC: Can we get retailers to adjust their stance on silk and sign commitments? As outlined in the ToC, this depends both on the direct cooperation with retailers (‘good cop’ strategy) and the effect of pressure campaigns (‘bad cop’ strategy).

To understand the likelihood of convincing retailers to make commitments, we want to look at the success rates of previous corporate campaigns and adjust these rates based on the specific factors related to silk products. Unfortunately, we could not find any numbers on success rates in the literature. However, we were able to use some of the expert interviews we conducted for this report (see section 4) to come to a prior estimate for success rates that we could then update based on further research. So in this specific case, we decided to already apply insights from a later part of the report to this earlier section, as other evidence for base rates was hard to come by.

Jonas Becker, the expert we consulted about corporate campaigns for broiler chickens, was able to provide us with highly relevant insights on this issue (see section 4.2 for the full notes from this interview). He noted that all of the campaigns that they have run over the last years in Germany have at least resulted in a compromise. When asked to give an estimate of success rates, he said that 50% would be a good conservative estimate for the broiler chicken campaigns. While silk is a very different product from chickens when it comes to the production process and animal welfare concerns involved, we use this estimate to inform our prior here, in the absence of better evidence. Given that his estimate is a conservative one and that all campaigns ended at least in a compromise, his statements suggest a prior likelihood of success around 60%. Our interview with a fashion campaign expert provided an additional perspective that we used to inform our prior (see section 4.3 for the full notes from this interview). This expert was reluctant to give a quantitative estimate of success rates. However, their experience from the fashion industry seemed less optimistic than what Jonas reported and from their remarks it seems unlikely that they are able to convince more than half of the companies they target. Combining these two perspectives, we decided to set our prior estimate for the probability of success at 40%.

This prior needs to be updated based on further considerations and pieces of evidence that seem relevant for corporate campaigns against the use/sale of silk products. The table below lists all of the most important factors we found and how they update our estimate, while the remaining part of this section describes our perspective on each of them in detail. We then conclude with our updated numerical estimate of the likelihood of success.

Table 4: Crucial considerations for the likelihood of success of corporate campaigns and their impact on our prior estimate

Lack of public concern for silkworms


Success of other fashion campaigns


Success of other corporate outreach tactics for less charismatic animals


Existing initiatives by retailers to shift away from silk


Public support for other insect issues


Potential to use related issues in campaigns


Lack of good alternatives


Lack of public concern for silkworms

The most important difference between corporate campaigns focused on silkworms and chickens is probably the difference in moral concern assigned to these two animal groups. As Faunalytics (2016) highlights, insect suffering has not been widely acknowledged as morally relevant yet. In the case of silk, the lack of empathy-provoking footage to use in public campaigns might be a huge impediment. For cage-free campaigns, “advocates created YouTube and Facebook videos to alert companies’ consumers to the suffering of hens in their supply chains. This tactic worked especially well because the reality of hens’ treatment conflicted with the images that companies had portrayed, leading consumers to feel deceived” (Bollard, 2017). Employing such video footage will likely be much less effective in the case of silk production. The lack of moral concern for the suffering of silkworms and the difficulty of inciting such concerns is thus the major factor we consider in our update here. (----)

Success of other fashion campaigns

Next to the corporate cage-free campaigns, we have also found positive pieces of evidence on corporate campaigns that different organisations have run in the fashion space. PETA reports that their campaigns against angora wool have been successful, where their eyewitness video footage of rabbits screaming while their wool gets ripped out led to over 400 brands to ban angora wool, with many major international brands among them. PETA has also run campaigns against alpaca wool with major companies like H&M, Gap, and Anthropologie (PETA, n.d.-a). An article by The New York Times illustrates PETA’s approach and the companies’ responses, but does not provide good evidence on the success of these campaigns (Testa, 2020). It suggests though, that corporate campaigns are a promising way to get commitments from fashion retailers. Next to PETA there are other organisations pushing for better animal welfare in fashion, such as The Humane Society (Humane Society International, n.d.; The Humane Society of the United States, n.d.), Four Paws (n.d.) and the Fur Free Alliance (2023). The fight against fur is a particularly insightful example that can act as a relatively good analogy to silk (despite obvious differences in the kinds of animals and farming practices involved). Fur production has been declining over the past years, with production down by 60-70% in most regions between 2014 and 2021, and many retailers have stopped selling fur products (Humane Society International, 2023). The Fur Free Alliance (2023) has coordinated the work of a wide variety of organisations on this issue, also running a programme directly targeted at retailers, called Fur Free Retailers (2023). Open Philanthropy has argued that corporate campaigns against fur have been successful and played a major role in this development (Open Philanthropy, n.d.). In sum, the evidence from other fashion campaigns seems very positive. Even though Faunalytics (2016) writes that silk might be the new fur, given the differences between those previous campaigns and the cause of silk, we only take this as a medium positive piece of evidence. (++)

Success of other corporate outreach tactics for less charismatic animals

There is evidence that business-oriented tactics can work in other, less obvious areas as well. Other organisations working on less salient issues, like shrimp welfare, have also found corporate engagement and outreach to be promising (Boddy & Jimenez, 2023). As we have not evaluated these approaches rigorously though, we only take this as a weak positive update. (+)

Existing initiatives by retailers to shift away from silk

Outside of campaigns, some retailers have already voluntarily shifted away from silk based on welfare concerns. One of the biggest examples is Asos, who pledged to stop sales of mohair, cashmere, and silk products from 2019 (The Guardian, 2018). This provides at least an anecdotal piece of evidence that the public and retailers can be motivated to care about silkworm welfare. (+) This example also suggests that it might be more cost-effective to run campaigns that push for banning all animal materials, instead of narrowly focusing on silk. We are very uncertain about this, as “focusing on a single issue” has been found to be important factors contributing to the success of cage-free campaigns (Bollard, 2017). This is a question that could be explored further.

Public support for other insect issues

There are some scattered data points from other insect issues that can also provide relevant insights on the likelihood of success of corporate campaigns against silk. As mentioned in the previous section, cochineal use in Starbucks drinks has been an issue that received strong public attention and motivated Starbucks to change their policies. However, the difference between food and clothing is highly relevant here, as the disgust element that was present in the Starbucks case could not be leveraged for silk campaigns. Other insights can come from campaigns aimed at helping bees, like a UK government campaign to get people to take action to help these insects. While this does count as some evidence that society cares about insects and successful campaigns could be run, it is important to note that bees are a very different case from silkworms, as they are wild animals that have become a relatively beloved species among parts of the public (Lockwood, 2022). In general, most evidence we could find was on organisations or individuals advocating for wild insects, like Somerset Wildlife Trust (n.d.). Given that these other insect issues are quite different from silk production, we only take this as a weak positive factor. (+)

Potential to use related issues in campaigns

Campaigns against silk do not have to be restricted to a focus on animal welfare issues. The campaigns could still be effective, even if the public does not care about the welfare of silkworms. Rowe (2017) writes: One notable reason silk bans and commitments to end the use of silk have been successful is that the silk industry frequently is accused of having violated human rights, through human health issues, child labour abuses, and slavery on factories”. Leveraging these issues should increase the chances of success for corporate campaigns and we assign a weak positive effect to this factor. (+)

Lack of good alternatives

In any case, if retailers are supposed to shift away from silk products, they will need to have good alternatives. Highlighting lessons learned from the fur-free fight, Open Philanthropy describes how important it has been for companies to be presented with good alternatives:

"Giorgio Armani, Anna Sui, and Michael Kors all cited innovation in faux fur as a reason to drop fur, just as Dunkin' cited plant-based meat innovation in adding Beyond Meat’s sausage to its menu. Alternatives seem like an especially promising approach in China, which accounts for about 57% of global fur sales and 40% of all farm animals confined globally. A new group, the Material Innovation Initiative, wants to promote technological alternatives to fur and other animal-based fabrics, as GFI and PBFA are doing for alternatives to meat, eggs, dairy, and fish"’ (Open Philanthropy, n.d.).

Many synthetic silk replacements already exist, either in the form of more expensive novelty alternatives such as artificial spider silk and cactus silk, or in the form of cheaper replacements like rayon, viscose and lyocell. Silk is considered a luxury material and even though cheaper alternatives have existed for a while, they do not seem like a possible replacement for more expensive luxury products (Hayden Hill, 2023). The Material Innovation Initiative describes the situation around silk alternatives as follows: “Silk alternatives currently in the marketplace, such as polyester, viscose/rayon, and nylon, often fall short in terms of aesthetics, performance, and sustainability. Although a number of next-gen silk innovators exist, few are available at commercial scale” (Material Innovation Initiative, 2021). It appears that current alternatives are either not popular at the luxury product market, or not commercially developed. For that reason it might make more sense to focus on decreasing production rather than fully replacing it. The lack of very strong alternatives provides a medium negative update to our estimate on the chances of success. (---)

Further aspects

While we think that we have covered the most important factors in this section, there are more potential aspects to consider when it comes to estimating the likelihood of success of corporate campaigns against silk. Open Philanthropy lists a wide range of factors that have been drivers of success in cage-free campaigns (Bollard, 2017). A detailed analysis of all of these factors was out of scope for this report, but could be a fruitful avenue for future research. While some of the listed factors (e.g. the use of social media to pressure companies and providing technical support to companies in implementing their commitments) could be replicated for corporate campaigns against silk, others will probably be less replicable and might lead us to lower the expected likelihood of success.


Summarising the evidence gathered in this section, it remains uncertain whether corporations can be convinced to drop silk products from their offerings. On the positive side, evidence from other corporate outreach tactics (on other fashion topics and other ‘non charismatic’ species) is mostly positive and we found pieces of evidence that support the claim that consumers and retailers can be made to care about insect welfare and other issues related to silk production. However, we remain very sceptical about the moral consideration that people will attribute to insects and it also seems like there is a lack of fully adequate silk alternatives for retailers to shift towards. In sum, these aspects lead us to update our prior in a slightly negative direction (-), adjusting from 40% to 35%. Based on our significant uncertainty, we assign a relatively wide symmetric interval from our most pessimistic to most optimistic scenario and thus estimate the likelihood of success to lie between 10% and 60%

3.1.4 Getting retailers to follow through on commitments

Once the retailers commit to a ban, restriction, or phase out of silk products, the next important question is whether this commitment will be fulfilled as promised, and how long that will take. Our campaign cannot stop at the point where retailers make their commitments, but needs to make sure that we follow up on the implementation, as this has been an important part of other corporate campaigns.

To estimate the likelihood of the retailers following through on their commitments, we have looked at data from previous campaigns. Most available data on this specific issue comes from cage-free and broiler campaigns, which should be taken into consideration as we might expect there to be differences between campaigns for different species and the food and fashion industries.

Simcikas (2019) has reported on the impact of corporate campaigns, and they make a prediction on the follow-through rates for cage-free (48%-84%, mean 64%) and broiler commitments (1%-94%, mean 24%). Their report mentions estimates made by other researchers in the space, generally with 30-60% follow-through rates (AGuesstimate, n.d.). Looking at historical evidence of previous campaigns, an EA forum post titled ‘Will companies meet their animal welfare commitments?’ shows that most welfare commitments were met, but several were broken (Sainsbury’s broiler commitment), and several companies (such as Burger King) pushed their commitments back from the original timeline (Saulius, 2019). Another issue that was mentioned is the commitments being vague enough to give companies ‘wiggle room’ and avoid going through with the commitment as intended.

Looking at examples from the fashion industry, the reporting on follow-through rates is a little less clear. PETA for example has reported on many of their campaigns and then on the resulting sales ban or welfare commitments, but there does not seem to be an easily accessible database of commitments and their follow-through rates. Looking through individual reports of different brands, it seems like a significant number of the commitments went through (we would not expect follow-through rates to be significantly lower than in the case of the above-mentioned chicken commitments). After the angora wool campaign by PETA, over 400 brands committed to banning this material and followed through on the ban and a quick search on websites of some of these brands confirmed that no angora products were being sold (PETA, n.d.-b). Historically successful fur campaigns also caused some of the largest companies to stop selling fur (usually easily verifiable through a search of products on their websites). The RSPCA (n.d.) led a campaign with Humane Society International for a fur-free Britain, and some of the largest companies have stopped selling fur (Adidas, M&S, H&M, ASOS and others). It also seems that many companies are inclined to commit to welfare improvements and avoid specific suppliers, rather than fully ban a specific material, such as ASOS in the case of leather and wool (ASOS, n.d.), or H&M and Gap committing (Testa, 2020) to not use a problematic alpaca wool supplier. In the latter case neither of the brands committed to banning alpaca wool right away and claim to only source from good farms going forward.

Overall, the follow-through rate is not likely to be very low. Especially in the fashion industry, it is more easily verifiable whether the commitments are being kept by checking retailers’ product offerings. Our best estimate for follow-through rates is between 30% and 60%. The charity’s sustained pressure and follow-up work with the companies should ensure that enforcement rates are not lower than those known from other campaigns.

3.2 Evidence that the change has the expected effects

3.2.1 Replacement of silk by worse alternatives

Some retailers might want to explore welfare commitments where traditional silk is replaced with ‘ethically’ harvested silk, where the moths are allowed to hatch and cocoons are processed afterwards. This kind of silk is called ‘Ahimsa’ silk and comes with a variety of welfare problems, as described in section 1.2.3.

While the majority of silk comes from China (~80% in China, ~18% in India) (Rowe, 2021), it appears that Ahimsa silk is predominantly produced in India, where it was created by Kusuma Rajaiah (House of Wandering Silk, n.d.). The exact numbers of Ahimsa silk made by domesticated mulberry silk worms and wild silk produced by other species of silkworm are not easily available. It is possible that wild silk causes less harm to the silkworms compared to farmed silk (both Ahimsa and non-Ahimsa), as the non-domesticated species can have higher survival rates and are free to reproduce. However, there is little data available on the details of wild silk production and the amount of wild silk is low compared to farmed silk (Wikipedia, 2022).

The use of Ahimsa silk has been on the rise since when it was first commercialised in the 1990s. It is especially popular in the Global North (USA, Europe, Australia, Japan). While many sources talk about Ahimsa silk becoming more popular, finding its exact market share and history of growth seems difficult (Aishwariya, 2020). On the positive side, if this is true, it shows that consumers in those countries care about welfare issues of silkworms. However, it seems unlikely that Ahimsa silk could replace a significant share of the silk market, since the production process is less cost effective, with Ahimsa silk costing around double the price of conventional silk and the material also being lower quality. (Ganguly, 2008)

The issue of Ahimsa silk potentially replacing traditional silk seems like a moderate uncertainty. While the demand for Ahimsa silk is growing due to animal welfare concerns, it still constitutes a small share of the silk market. Moreover, if companies decide to commit to moving away from silks and cooperate with animal welfare groups, they can be steered away from Ahimsa silk. The issues surrounding Ahimsa silk can also be included in the campaign to bring awareness to the reality of the animal cruelty involved. Overall, we should be able to mitigate this issue.

3.2.2 Lowering production with lower demand

The global silk market makes up only 0.17% of the whole fibre market (Morletto, 2023). Market research estimates the size of the global silk market at ~$15 billion in 2021, projecting it to grow by 7.5-8% and reach $34 billion by 2031 (Allied Market Research, 2022). While we think that market research firms tend to overestimate growth rates, this at least provides an indication of the market’s general tendency for future growth.

The impact of the campaigns is dependent on whether the targeted brands make up a significant portion of the silk market and are therefore likely to influence production. Brands targeted in fashion campaigns in the Global North are usually larger companies ranging from affordable (H&M, Gap) and mid-range (Calvin Klein, Zara) to luxury brands (Gucci, Prada). 

There is data available on the largest exporters of raw silk (China, India), and on the largest importers (Japan, Italy, France, United States, Germany) (Global Commodities, n.d.). However, it seems more difficult to find what share of processed silk products are sold in which countries and through which brands. We checked the top ten European largest fast fashion brands (by revenue) and found that besides sportswear-focused brands and Primark, the famously low-tier brand, only ASOS did not sell silk (public commitment made in 2019) (Statista, 2020). Among the fast fashion names involved in silk were H&M, M&S, Zara, Next, and Debenhams (Statista, 2020). It is not clear how much of their product range includes silk, or how much of the silk market is taken up by these brands. However, we can easily conclude that fast fashion brands do provide affordable silk products and that targeting them would cover at least one part of the silk product market. 

Looking at the global leaders among luxury brands, silk products are sold by all top ten (LV, Dior, Gucci, Chanel, Rolex, Versace, Michael Kors, and others) (Thomas, 2023). Looking at the revenue brought in by fast-fashion vs. luxury brands, it appears they are around the same order of magnitude, with Zara bringing in over €23 billion in 2022 (Statista, 2023) and Louis Vuitton $18 billion in the same year (Thomas, 2023). We can assume that luxury brands sell significantly fewer products as their prices are much higher. This could mean that fast fashion brands sell a larger proportion of silk on the market, or at least cover a non-negligible share. If that is true, targeting them for corporate campaigns would mean targeting a significant share of the market and therefore any bans on sale would be more likely to impact (decrease) the production. However, silk products probably make up a larger share of total revenue for luxury brands. In general, there is still a lot of uncertainty about how much silk is actually sold by which brands and in which countries.

There is data available on the impact of fur sale bans on the production. A report from Open Philanthropy titled ‘What can we learn from the fur-free fight?’ came to the conclusion that banning sales is much more effective compared to banning production:

"Ban sales, not production. European fur production bans have done little to suppress supply or demand. But after the EU banned the sale of seal fur in 2009, the number of Canadian seals killed annually fell threefold, to fewer than 100K. Since the WTO upheld that sales ban, the EU could likely legally ban the sale of other cruelly produced goods. Advocates should push it do so, starting with eggs, meats, and furs that don’t meet the EU’s existing animal welfare standards" (EA Forum Archives, 2019).

Other examples (McFetridge, 2022; The Humane League, n.d.), this time from cage-free hens, show that corporate campaigns and the resulting commitments increased the share of cage-free hens in egg production. These interventions have clearly led to a decreased production or a wider change in welfare standards. While this does not automatically mean that any campaign can achieve this, there are clear precedents that it is possible.

A study done by Norwood and Lusk calculates how exactly the decrease by one unit of consumption translates into production (Norwood & Lusk, 2011). They found that not consuming one lbs of an animal product (various meats, eggs, milk) led to a 0.56-0.9lbs decrease in production. This further helps to show that reducing consumption can have a significant impact on production.

Overall, reductions in consumption seem to have a clear effect on production volumes. Based on the evidence collected here, especially the quantitative estimate given by Norwood and Lusk, our best guess is that production is reduced by 50-90% of the consumption volumes that are reduced by the targeted retailers.

3.2.3 Higher animal welfare with lower production

The question whether lower silk production then leads to less animal suffering hinges on the assumption that farmed silkworms lead net negative lives. There is a lack of empirical data that would allow us to assess the capacity of silkworms to suffer with great confidence. However, several attempts have been made to try to gather the available data and estimate the sentience of silkworms.

Rethink Priorities’ Welfare Range Estimate assessed several different species and their range of experiences, compared to humans. Silkworms were estimated to have 0.2% of the welfare range of a human (5th percentile 0, 95th percentile 0.073), when adjusted for their probability of sentience (Fischer, 2023).

Rethink Priorities’ report on silkworms concludes that the main source of suffering (if silkworms are sentient) is not caused by the death by boiling, but rather by the prevalent diseases, which can also cause 10%-47% mortality (Rowe, 2021). It seems like a lot of treatment with disinfectants and possibly antibiotics is already utilised, but there might be possibilities for improvement. It is likely that the intensity of disinfection is already high as farmers want to avoid costly losses (Tayal & Chauhan, 2017). We should consider that lives of silkworms that do not suffer from diseases might be net positive though:

“It’s also worth noting that silkworms who survive to slaughter could live otherwise positive lives. While there isn’t much research into silkworm quality of life, they mostly seem to live on beds of mulberry leaves eating prior to slaughter, with minimal disturbance that might lead to stress. However, a caveat is that research into nonlethal disease and injury on silk farms is basically nonexistent, since it is typically unrelated to commercial success” (Rowe, 2021).

Still, averaged across all farmed silkworms (those that do and do not suffer from diseases), we expect their experiences to be net negative (if sentient). In order to compare the cost effectiveness among different interventions, we conducted a welfare point estimate (available here for interested readers). We followed the report template used by Charity Entrepreneurship (CE) and estimated different characteristics of farmed silkworms to get to a welfare score on a scale from -100 (worst welfare possible) to +100 (best welfare possible). The welfare points of silkworms were estimated to be -19/100, meaning their experiences were overall slightly negative. However, many assumptions are included in the estimate and a lot of evidence is missing that could significantly impact the estimate. We are 70-80% confident that silkworms have net-negative lives on average.

If the production of silk went down, the worms would be unlikely to procreate successfully and survive on their own long term (Rowe, 2021). Therefore we do not have to consider whether the moths would be better or worse off in the wild, as that is very unlikely to happen. In this case we are then considering only two scenarios - one is replacing the silk production with a different farming style (Ahimsa silk, addressed in section 3.2.1), and the other is production of silk decreasing. If we take on the assumption that silkworm lives are net negative (based on our welfare point estimate), then lowering production should lead to less animal suffering, the ultimate goal of the intervention.

Depending on different moral philosophies, reducing the numbers of farmed animals can be seen as a positive improvement even if the lives of silkworms were on average net positive. Some views place more value on avoiding suffering (more important to prevent 10% of diseased silkworks from suffering than to keep 90% of net positive silkworm lives), and many people in the animal rights movement hold the opinion that any animal use is exploitative and inherently wrong.

In sum, reducing the number of silkworms in farming seems more likely to be a net positive than negative for animals, since farmed silkworm lives should be negative on average. However, this is conditional on the assumption  that these animals are sentient. There is a significant chance that silkworms are not sentient, such that our intervention would not have any effect on their welfare. We cannot resolve this uncertainty here, but account for it in our cost-effectiveness estimate for this intervention.

3.3 Summary of the evidence

This evidence review has clarified many uncertainties that emerged from our ToC and its assumptions. We list our main findings below:

  • It seems unlikely that people can be convinced to care enough about insect welfare in order to drive strong public pressure against retailers. However, there might be potential in including concerns around human rights and other issues related to silk production in the campaign. Also, the majority of commitments in other areas have been won by direct collaboration with companies and without the need for a public campaign. This means that commitments might still be won, even without widespread public support.
  • Our research suggests that current alternatives to silk are not adequate for retailers to shift towards them directly. Retailers would have to be asked to drop silk products from their offering without replacing them by products of similar quality.
  • The likelihood of success for corporate campaigns against retailers of silk products still seems relatively high. Corporate campaigns have a strong record of success in other areas and we do have some indications that they might also work in the case of silk. However, important negative factors are a potential lack of public support and alternative materials, as described above. Aggregating these different perspectives, we think that corporate campaigns against retailers of silk can have a success rate between 10% and 60%.
  • Enforcement rates of commitments should also be relatively high. Follow-through from previous corporate campaigns seem relatively strong and we do not have good reasons to believe that they would be lower in the case of silk. Our best estimate is thus for enforcement rates to be between 30% and 60%. The charity would have to do the necessary follow-up work though to make sure that this holds true.
  • There is a lot of uncertainty around the sentience and potential experiences of silkworms. We think that there is a non-negligible likelihood of silkworms being sentient. Based on the available evidence, we also think that the lives of farmed silkworms are net negative on average, mostly because of the strong prevalence of diseases.
  • We have good evidence from previous campaigns (cage-free eggs and fur), that corporate commitments have a significant impact on production. We estimate that for each unit not consumed, production falls by 50% to 90%.
  • It is unclear whether broad-market or high-end retailers would be a better choice as targets for our campaigns. We lean towards targeting the former, as those companies should be more willing to move away from silk and sales volumes might be comparable. The implementing charity would have to speak to industry experts to understand better who the biggest users/buyers of silk are.
  • Ahimsa silk is an alternative that has been branded as being more sustainable. However, the evidence suggests that this kind of silk is not in fact better in terms of animal welfare. The campaigns would need to ensure that the issues with Ahimsa silk are clearly described and that retailers understand that we will not accept this as an alternative.

While these findings seem highly relevant and have clarified many of the uncertainties from our ToC, we acknowledge limitations:

  • Our evidence review was relatively shallow, as we only spent ~50 hours on it in total.
  • Most of the evidence comes from successful cases like cage-free and fur campaigns, which might lead to our perspective being positively biassed. While we tried to account for differences between those cases and silk, it is possible that we have missed some crucial considerations.

The biggest remaining uncertainties revolve around the following issues:

  • The sentience and experiences of silkworms
    -> This is something that we cannot fully resolve in this report
  • The most promising retailers to target with campaigns
    -> Market research and insights from industry expert should help here
  • The most promising countries to target with campaigns
    -> Our geographic assessment in section 5 will tackle this issue

In the next sections, we will add further perspectives that should complement the picture that has emerged from this evidence review.

4  Expert views

Following our evidence review, we identified a variety of areas where it seemed promising to receive inputs from experts to deepen our understanding. The most important topic for us was to better understand the lessons learned from corporate campaigns in related areas (e.g. cage-free eggs, fur, angora wool) and how these might translate to silk. Another relevant issue to understand better was the availability of good alternatives to animal-based silk products. These topics are covered by the four expert interviews outlined in the following subsections. Two of these interviews (Nicole Rawling and PJ Smith) were conducted by CE previously and we could use these for our purposes. The other two (Jonas Becker and a fashion campaign expert) were conducted specifically for this report. We also considered including expert views from Rethink Priorities on silkworm sentience and welfare (e.g. Abraham Rowe). However, since our background and evidence sections already cited their work in depth, we decided against rehashing this content here.

4.1  Nicole Rawling, Material Innovation Initiative

Nicole Rawling from Material Innovation Initiative was previously interviewed by CE and we were provided with the recording and notes from that call. The interview focused both on the availability and development of suitable silk alternatives, and more generally on the potential for companies to move away from silk products.

Nicole acknowledged that the fashion industry currently lacks suitable alternatives to silk due to performance and aesthetic standards. She also opined that the main motivation for brands wanting to switch from silk is its significant environmental impact. Some options like polyester and rayon have gained popularity but come with environmental issues during production and disposal. For instance, polyester is often not viewed favourably due to petrochemical origins and the growing anti-plastic and anti-synthetic sentiment within the fashion industry. Nicole brought attention to synthetics' largely unrecognised impact on animals, particularly microplastics - suggesting that replacing silk with synthetics could potentially be net-negative for animals, although this is very uncertain.

This highlights the need for better alternatives that balance environmental and animal welfare concerns. MII is actively working to fill this gap by supporting the development of new animal-free and more sustainable silk materials. Drawing from her extensive experience in the for-profit sector, Nicole stressed the importance of possessing industry knowledge when attempting to bring about change in the fashion industry. She further pointed out the significant financial challenges faced by companies attempting to produce ethical alternatives to silk, including high R&D costs, challenging fundraising environments, and the cost of setting up new production facilities. As a result, it is difficult to say when the creation of a viable alternative to silk would happen.

Consumer demand for cruelty-free and sustainable silk could drive investment and innovation in alternative materials. However, Nicole also highlighted that consumer awareness about these issues is uncertain, suggesting a need for better education and awareness-raising campaigns. She suggested that promoting awareness of these harms could be a viable approach for a charity. Previous campaigns against the fashion industry, like PETA's, have focused on animal cruelty. Nicole sees potential in expanding these campaigns to highlight environmental and labour rights issues. However, the effectiveness of such strategies remains uncertain, particularly in the context of fast fashion brands, which remain profitable despite widespread knowledge of their use of slave labour or poor working conditions. Nicole also touched upon the cultural challenges in convincing people to switch to alternative materials, especially in regions where silk production is deeply ingrained in the culture and viewed as a luxury. She highlighted that lower-end and mid-tier brands have shifted from silk for cost reasons already, but luxury brands have not due to limited consumer concern about insect welfare.

Finally, Nicole acknowledged the large potential in silk campaigns. She emphasised the potential impact of even a minor reduction in silk demand, stating that even a 1% decrease could save a significant number of silkworms from the estimated 1 trillion killed annually for silk production.

4.2    Jonas Becker, Albert Schweitzer Foundation

Jonas Becker brings nearly five years of practical experience in corporate campaigns to the table, with a specialisation in broiler chicken welfare in Germany in association with the Albert Schweitzer Foundation.

In discussing the key components that contribute to a campaign's success, Jonas emphasised the importance of consumers having the choice and flexibility to switch brands based on ethical grounds such as animal welfare or sustainability. Another critical element is the gap between consumer expectations and the harsh realities of product manufacturing. This discrepancy can serve as a powerful lever to gain public support. Jonas also noted that the first step in any campaign should be corporate outreach, stating that public pressure campaigns are often secondary but become important if companies are not ready to commit to ethical practices.

Regarding the potential for a campaign against silk products, Jonas expressed a cautiously optimistic outlook. While acknowledging a general public indifference toward silkworm welfare, he mentioned that public opinion is malleable, pointing out how campaigns have successfully shifted attitudes towards chickens over the past decade. Jonas thinks that the brutal realities of silk production, where silkworms are boiled alive, can serve as a shocking revelation to the uninformed public. This 'shock factor' can be further supplemented by adding layers of human rights and sustainability issues related to silk production. However, even without public pressure, campaigns can often be successful. In his experience with chicken welfare campaigns, Jonas found that most successes came from direct corporate outreach without the need for public pressure.

When it comes to selecting which companies to target in a campaign, Jonas advised that the first targets should be retailers for which the likelihood of a successful campaign without public pressure is high. Such “easy wins” could provide crucial momentum for further campaigns. Market leaders, who may control up to 20% of the market, should be among the targets, but should be tackled only when there is a reasonable chance of success, since the outcome of those cases will heavily influence the ultimate influence of silk campaigns. He also leaned towards a focus on companies serving a broader market as opposed to niche luxury brands, explaining that silk will not be as costly for them to give up and that these companies have more to lose from negative publicity. Targeting companies is a sequential process, with the choice of the next target often depending on the success or failure of previous ones.

On the subject of commitments from retailers, Jonas explained that all campaigns that they have run have at least resulted in a compromise and that campaigns need to be run until this point. This can last from a few weeks to several years, but a good conservative estimate for duration would be around six months. Speaking from his experiences, Jonas suggested that a 50% success rate should be the minimum target to maintain the campaign's credibility. He also clarified that while the final commitments can be global, they are more likely to be concentrated in countries where the campaigns exert the most pressure. 

Finally, Jonas expressed concern about the potential negative impact of launching a new silk-focused campaign on ongoing campaigns. He stressed the importance of evaluating the current status of existing campaigns, such as those focused on fur, to determine whether resources can be reallocated to a new issue like silk.

4.3    PJ Smith, The Humane Society of the United States

PJ Smith is the Fashion Policy Director at The Humane Society of the United States. He was previously interviewed by CE and we were provided with the recording and notes from that call.

PJ underscored the powerful intersection of animal welfare and the broader sustainability movement. By aligning the campaign with environmental sustainability, it might be possible to generate more substantial corporate and consumer buy-in.

He thinks that education and awareness are significant factors. There's currently a lack of public awareness about the environmental and welfare issues associated with silk production. By educating consumers, it might be possible to change attitudes and increase demand for silk alternatives. However, PJ also warned about the challenges of promoting vegan fashion choices, which might be harder to sell to consumers due to less obvious animal welfare implications compared to vegan dietary choices.

Reflecting on the role of activism, PJ mentioned that sustained campaigns against companies selling fur had been effective, but long-term campaigns without progress might lead to resistance. He highlighted the importance of starting a dialogue with companies regarding their use of animal-derived materials, starting with less controversial issues before escalating to more significant commitments, as seen in fur-free campaigns.

A key component of any successful campaign will be the development and promotion of innovative, environmentally friendly alternatives to silk. Companies have shown an openness to explore such alternatives, but they need to be scalable and capable of replicating silk's specific properties. Thus, the development of innovative, scalable alternatives to silk could drive significant change.

PJ highlighted the role of high-profile fashion brands in promoting change within the industry, as seen when Gucci announced its transition to fur-free products. A similar high-profile declaration might be necessary to stimulate significant reductions in silk use.

Another key topic was the interplay of corporate behaviour and politics. PJ suggested that success at the corporate level could eventually lead to changes in public policy, as seen with bans on fur sales following successful fur-free corporate policies. The industry also needs to consider the potential impact of policy-driven changes, such as environmental taxes on companies with high ecological footprints, which could influence the use of silk.

In terms of future direction, PJ shared that the Humane Society plans to continue its policy advocacy work, with a focus on promoting alternatives and public education once fur is off their agenda. Finally, he urged for a stronger focus on fashion in animal welfare campaigns, given the industry's potential to significantly affect animal welfare.

In sum, the insights from this interview suggest that while there are challenges, a charity focusing on the reduction of silk use might be viable and could potentially have a significant impact on animal welfare and environmental sustainability. A charity focused on silk use reduction could potentially make an impact by employing strategies similar to successful fur-free campaigns, creating and promoting innovative silk alternatives, raising public awareness, and engaging in policy advocacy.

4.4 Interview with an expert on corporate campaigns in the fashion industry

We also interviewed an expert with decades of experience working on corporate campaigns against fashion brands. Due to their preference, this expert will remain unnamed in this report and their experience summarised briefly. The main takeaways from the interview were used to inform our decisions the same way as the other interviews.

The expert suggested possible campaign strategies, including focusing on the unnaturalness of the silkworms’ lives, disturbing images of boiled cocoons, using the existing public concern about bees and focusing on the environmental impact of silk. 

When selecting a campaign target, they consider it crucial to keep in mind the following:

  • Look at companies with strong environmental commitments and highlight the dissonance between their commitments and practice.
  • Look who the target consumers of the brands are (e.g. young people tend to care more).
  • Focus on individual brands instead of high street vs luxury. It might be easier for high street brands to reduce silk as it is in fewer of their products, but there are luxury brands like Gucci that appeal to young people and sell vegan-leather handbags.
  • There are some old-fashioned companies that do not have any bans and do not seem open to them at all, so it seems best not to waste time on them (e.g. Louis Vuitton).
  • Remember that brands often change strategy and managers and can completely change their commitments.
  • It might make sense to start with a collaboration (such as a new range of luxury silk alternatives) and if that works continue to push for more.
  • Strong commitments from industry leaders bring momentum, reframe the question (other brands suddenly have to answer why they are still using silk), and can lead to a tipping point.

4.5    Key takeaways from expert interviews

Aggregating the insights from our expert interviews, we highlight ten key takeaways below. In parentheses, we indicate which interviews these insights are taken from.

  1. Lack of suitable alternatives to silk: There is a current lack of alternatives to silk that meet both aesthetic and performance standards, with existing options like polyester and rayon facing environmental issues (Nicole Rawling).
  2. Environmental impact as a motivation for change: The significant environmental impact of silk production is a major motivation for brands considering a switch from silk. This factor is a potential focal point for campaigns (Nicole Rawling, PJ Smith).
  3. Consumer demand and awareness: Raising consumer awareness about silk's environmental and ethical issues is important to drive demand for alternatives. Current awareness levels are low, suggesting the need for educational campaigns (Nicole Rawling, PJ Smith).
  4. Financial and cultural challenges in producing alternatives: High R&D costs, fundraising challenges, and the cost of setting up new production facilities are significant barriers to creating ethical silk alternatives. Cultural challenges also exist, particularly in regions where silk is ingrained in the culture (Nicole Rawling).
  5. Campaign strategies and public perception: Campaigns need to emphasise the harsh realities of silk production and can benefit from adding layers of human rights and sustainability issues. Public perception of silkworm welfare is malleable and can be influenced by campaigns (Jonas Becker).
  6. Targeting and sequential approach in campaigns: Campaigns should initially target market leaders and companies serving broader markets. The choice of the next target often depends on the success of previous campaigns (Jonas Becker).
  7. Importance of corporate outreach and dialogue: Engaging in dialogue with companies about their use of silk and starting with less controversial issues can be an effective strategy. This approach has been successful in past campaigns (Jonas Becker, PJ Smith).
  8. Role of high-profile brands and consumer choice: High-profile fashion brands can play a significant role in promoting change within the industry. Consumer flexibility to switch brands based on ethical grounds is also crucial (PJ Smith, Jonas Becker).
  9. Potential negative impact on ongoing campaigns: Launching a new campaign on silk might affect ongoing efforts in other areas, such as fur campaigns. It's important to assess the current status of existing campaigns before reallocating resources (Jonas Becker).
  10. Future directions and policy advocacy: Continued focus on promoting alternatives, public education, and policy advocacy can help drive change in silk use and production practices. Success at the corporate level could eventually lead to changes in public policy (PJ Smith).

5     Geographic assessment

When planning the practical implementation of corporate campaigns against silk use, a charity would have to focus on certain retailers, as described in section 3.1.1. Just as importantly however, the charity would also have to focus its campaign on a certain country or group of countries. Even if a targeted retailer is active globally (i.e. in a wide variety of countries), the charity will not be able to challenge it in every one of these countries due to clear practical baƒrriers (lack of resources, cultural and language differences, different regulatory environments, etc.). Instead, it will need to focus on those countries where the intervention is likely to have the biggest impact, which are probably the retailer’s home country or largest and most important strategic markets.

For retailers that are active in many different markets, it is an open question whether a campaign in one or few select countries could lead the retailer to adjust its policies on silk products across its whole operations, or if it would only affect operations in the campaign countries. Such questions have to be considered carefully when deciding on the exact retailers and countries to focus on. In general, the combination of retailers and countries should be chosen in a way that maximises expected impact.

It seems most promising to us to focus campaigns on a smaller number of countries, maybe even only one country, and then address multiple retailers operating there. This should allow the charity to make use of economies of scale and run a more efficient campaign. In the next section, we thus conduct a geographic assessment in order to identify highly promising countries. The implementing charity could take these recommendations as a starting point to identify corporate targets in those countries. However, it also could be fruitful to start with identifying retailers that seem like promising targets and then decide in a second step in which countries they should be targeted. In general, it will be important to consider the interplay between retailers and countries when it comes to defining the focus on campaigns.

5.1    Geographic assessment

As specified in our description of the intervention’s scope, our geographic assessment is restricted to countries of the Global North. We follow UNCTAD’s definition of the Global North, which states that “North refers to developed economies, South to developing economies” (UNCTAD, 2022). In accordance with this definition, we use the countries that they list as developed in their classification (UNCTAD, n.d.). From this list, we remove minor countries that do not seem relevant for our purposes (e.g. Andorra, Liechtenstein, etc.) to arrive at a final set of 45 countries in scope.

Our geographic assessment is based on a weighted factor model, combining ten variables that seem relevant for gauging the promisingness of different countries. The full set of variables, their weights, data sources, and the countries’ detailed scoring can all be found in this spreadsheet. Interested readers are invited to read this spreadsheet in more detail, as we will only summarise the most important aspects of our assessment here.

The most important variables for our model were a country’s

  • market size for apparel,
  • amount of money spent on clothing per household,
  • number of farmed animal advocacy organisations per million inhabitants,
  • ranking in global silk product consumption, and
  • index of global influence/presence.

The detailed reasoning behind those different variables and their weights can be found in the spreadsheet.

Our model estimates that the most impactful countries for running corporate campaigns against retailers of silk products are the United States, Switzerland, the United Kingdom, Germany, and Japan. Further potential targets include Italy, France, and Ireland. It is out of the scope for this report to provide a detailed analysis for each of these countries and why their score might or might not be justified. We also did not research which are the most important retailers per country, something that would be crucial to analyse when deciding on the specific campaign targets. Here, we decided to only provide some preliminary thoughts as an input for further investigation.

  • Based on its huge market size and strong global influence, the United States would be a natural choice for corporate campaigns. “The United States is one of the world’s largest importers of silk garments and interior decoration fabrics and accessories“ (Global Commodities, n.d.). Consumption of finalised silk products is more important in the U.S. than the import and processing of raw silk.
  • Switzerland stands out with macroeconomic conditions and a large demand for luxury products. However, it is a relatively small market and might be too small to be a proper target for campaigns. A campaign in Switzerland might be combined with concurrent campaigns in Germany and France (see below), due to the close proximity in terms of culture, language, and geography.
  • Germany and the UK are both large and strong European economies and markets. Campaigns in those countries would affect large quantities of silk products and could also have significant spillover effects to other markets.
  • Japan is the only Asian country among our highest-rated countries, benefiting from a strong economy and large market for silk products.
  • Italy and France are both major European markets for silk, just as Germany and the UK. However, these two countries play a stronger role as processors of raw silk and less so as markets for sales to end consumers. For instance, Italy dominates raw silk imports to Europe because of its fashion industry (Blanchard, 2009; Global Commodities, n.d.). Since we care about end-consumer retailers though, this should not be the most important factor in our country selection and industries with a strong commercial interest in silk products might in fact hinder our chance of success.
  • Ireland is a smaller market and should probably be of lower priority than the countries named above. However, it could be a target for campaigns that jointly focus on the United Kingdom and Ireland.

5.2    Where existing organisations work

To our knowledge there are no existing campaigns against silk retailers in countries of the Global North (and South). PETA and some other organisations have some content on silk production, but it does not seem like they are actively running campaigns on the topic (PETA, n.d.-c).

We still think that there are organisations that could be interested to take on this topic and run campaigns against silk in the future. To identify which countries such organisations work in, we have used the members of the Fur Free Alliance (2023) as a proxy. We see that Bulgaria, Germany, Ukraine, the United Kingdom, and the United States are the countries with the most organisations in the Fur Free Alliance. The full list of countries and number of organisations can be found here.

6   Cost-effectiveness analysis

Our cost-effectiveness estimate shows a model of a charity running two corporate campaigns per year (an assumption we took based on the expert interview with Jonas Becker), for ten years in total, targeting retailers who sell silk products in countries in the Global North. The estimate includes a projection for the silk production for the next ten years (until 2034), which is assumed to correlate with the number of farmed silkworms. The benefit is expressed as the improvement in welfare points, achieved through a reduction in silk production. The numbers of farmed silkworms are based on a report by Rethink Priorities (Rowe, 2021). The impacts of the intervention itself are modelled from 2025, because of the expected delay between campaigning, commitments, and change in practice.

The lower bound value for cost effectiveness of our intervention is 0.1 WP/$, the upper bound value is 1,311 WPs/$. The geometric mean cost effectiveness is 12 WPs/$.

6.1    Effects

There is one primary effect of the intervention that we explored in our CEA, namely the reduction in the amount of farmed silkworms. As our welfare point estimate for silkworms shows, the expected welfare of farmed silkworms is negative, mostly due to the prevalence of diseases. With a decrease of silkworm farming, we expect improvements in welfare points. This was discussed in more detail in section 3.2.3. The effects are modelled over ten years from starting the intervention, which is the usual time period used in these assessments. This is probably on the pessimistic side, since it excludes a lot of the impact of the campaigns that start in the later years of running the charity and does not consider long-term effects.

To estimate the scale of the effects, we assume we can reach 0.1-1% of the global silk market when running one campaign. The estimate for how much of the global silk consumption we can influence is based on our understanding of the fashion industry. There are major retailers who have a large share of the market (~10%) in some countries and countries who are major consumers of silk should have a similarly large share of the global market. Assuming that the intervention will select for the most promising retailers and countries, we think we can reach around 1% of the global market in the best case scenario. There is also the potential that we can get the retailers to implement their commitments in multiple countries, not only the main targeted one(s), which adds to the potential share. In that sense, the upper bound should be realistic. The lower bound of 0.1% accounts for the fact that we might not be able to find such optimal scenarios (see uncertainties described in section 3.1.1) and that such big targets might be rare and we might run out of them over time.

We then multiply this share per retailer by two (assuming two campaigns per year, as described above) and discount the market share targeted based on the following factors:

  • The likelihood of success (10-60%, discussed in section 3.1.3)
  • The follow-through rates after commitment (30-60%, discussed in section 3.1.4)
  • The translation rate between decrease in consumption to decrease in production (50-90%, discussed in section 3.2.2)
  • An annual discount of our impact (1-10%, see the spreadsheet for our reasoning)

Aggregating all of these factors, we estimate the share of the global silk production we expect our campaigns to reduce in a certain year. Welfare point estimates are then used as described in section 3.2.3. The welfare points are discounted using Rethink Priorities’ welfare range estimate, which includes the likelihood of sentience in silkworms. The total amount of discounted welfare points (WPs) this intervention could bring is between 1.4 million and 2.8 billion in total between 2025 and 2034.

There is a possibility that the campaign could have other benefits that we did not include in our estimate. We think that running the first successful corporate campaign for insects could be a meaningful step encouraging further interventions in the space. It can also contribute towards more concerns among the public around insect welfare and possibly moral circle expansion. If this intervention proves feasible, it can be expanded to more retailers and more countries.

6.2    Costs

The total costs include a fixed cost to set up and run a charity for one year and the campaign costs itself. Our cost estimates are informed by previous interventions. The costs of the campaign are based on data from The Humane League, who have published their campaign budget over the last decade and include the number of staff working on the campaigns and the commitments won. Based on this, we assume a lower-bound cost of $100,000 and an upper-bound cost of $700,000 for one campaign.  Extrapolating these costs to 20 campaigns (two per year for ten years, as described above) and adding fixed start-up costs of $100,000, we arrive at a total cost estimate between $2.1 and $14.1 million.

One further consideration that might be worth exploring is if subsequent campaigns might have lower costs that the initial ones, since the strategy, messaging and background research might be reused.

7     Decision

In order to summarise the extensive research conducted for our report in a reader-friendly format, we decided to provide a table with a short Q&A, addressing the most relevant issues that emerged from our research and how we think about them.

Table 5: Summary of important questions and answers

What is the life of farmed silkworms like and would lower production volumes be a net positive?There is a lot of uncertainty around the sentience and potential experiences of silkworms. Based on the available evidence, we tentatively think that the lives of farmed silkworms are net negative on average, mostly because of the strong prevalence of diseases. Our CEA discounts for uncertainties around silkworms’ welfare ranges, including their probability of sentience.
Does the public care enough for insect welfare?It seems unlikely that people currently care enough about insect welfare in order to drive strong public pressure against retailers. However, public perception has been successfully targeted and shifted by previous campaigns and some of the experts we interviewed seemed cautiously optimistic about this. The image of silkworms being boiled alive could be leveraged for the campaign. There also seems to be strong potential for including concerns around human rights and environmental issues related to silk production in the campaign.
Is public pressure strictly necessary to win commitments?Pressure campaigns are an important tactic of corporate campaigns overall. However, the majority of commitments in other areas have been won by direct collaboration with companies and without the need for a public campaign. This means that commitments might still be won, even without widespread public support.
Are alternative materials readily available?Our research suggests that current alternatives to silk are not adequate for retailers to shift towards them directly. Retailers would have to be asked to drop silk products from their offering without replacing them by products of similar quality. There are ongoing efforts to develop better alternatives though.
What is the likelihood of winning commitments from targeted retailers?Corporate campaigns have a strong record of success in other areas and we do find some indications that they might also work in the case of silk. However, important negative factors are a potential lack of public support and alternative materials, as described above. Aggregating these different perspectives, we think that corporate campaigns against retailers of silk can have a success rate between 10% and 60%.
How likely are retailers to stick to their commitments?Enforcement rates from previous corporate campaigns seem relatively strong and we do not have good reasons to believe that they would be lower in the case of silk. Our best estimate is thus for enforcement rates to be between 30% and 60%. The charity would have to do the necessary follow-up work though to make sure that this holds true.
How would changes in demand by retailers influence production?We have good evidence from previous campaigns (cage-free chickens and fur), that corporate commitments have a significant impact on production. We estimate that for each unit not consumed, production falls by 50% to 90%.
Which retailers should the campaigns focus on?It is unclear whether broad-market or high-end retailers would be a better choice. We lean towards targeting the former, as those companies should be more willing to move away from silk and sales volumes might be comparable. The implementing charity would have to speak to industry experts to understand better who the biggest users/buyers of silk are.
Which countries are most promising for the campaigns?The most promising countries seem to be the United States, the United Kingdom, Germany, and Japan. France and Italy also could be valid targets. Switzerland and Ireland are countries that could be targeted as part of campaigns in their neighbouring culturally-similar countries.
What is the estimated cost effectiveness of such an intervention?Our cost-effectiveness analysis yields a broad range of outcomes, between 0.1 and 1,311 WP/$. The geometric mean cost effectiveness is 12 WP/$. These estimates are likely to err on the conservative side.
Are there any more “sustainable” silk production methods that retailers might want to focus on instead of a ban?Ahimsa silk is an alternative that has been branded as being more sustainable. However, the evidence suggests that this kind of silk is not in fact better in terms of animal welfare. The campaigns would need to ensure that the issues with Ahimsa silk are clearly described and that retailers understand that we will not accept this as an alternative.
How promising do corporate campaigns against silk retailers seem compared to other interventions to help insects?We did not look at other insect interventions in enough detail to be able to compare them. This might be an avenue for further research. Rowe (2021) thinks that “while there are promising interventions to reduce the suffering of silkworms, marginal resources in the insect space might be better spent in other areas, such as reducing the painfulness of pesticides, reducing the number of insects farmed for animal feed, and reducing the harms of cochineal farming. That being said, given the scale of silk farming, advocacy on this issue could plausibly be highly cost-effective when compared on a species-neutral basis to interventions to reduce vertebrate farmed animal suffering.”

Based on these considerations, we think that this intervention idea has enough promise to be considered further for implementation. However, in order to fully recommend this idea, we would need better information about the silk market shares of different retailers that could be targeted by the campaigns. For obtaining this information, we suggest further collaboration with experienced players like PETA, outreach to industry experts and stakeholders, and/or commissioning market research.

Conditional on the findings of such further research showing that there are highly promising retailers to target, it seems valuable for a charity to run a pilot intervention on this. The campaigns have quick and strong feedback loops (see section 8.2.3) and there should be strong learning value in running a first corporate campaign on an insect farming issue. The intervention could be shut down again fairly quickly, if the pilot does not prove to be promising.

In the following two sections, we explain what such an intervention could look like in practice and which crucial aspects need to be considered. We also make a recommendation on whether the intervention should be taken on by an existing organisation or whether a new charity should be started specifically for this topic.

8     Implementation of a new idea

In this section, we provide some more practical information that would be relevant for the implementation of this intervention idea.

8.1    What does working on this idea look like?

The way in which corporate campaigns are typically implemented has been described in section 2.1. The causal chain outlined in section 2.2 should also give a good overview of the steps involved. More details on corporate campaigns and how and why they work are given by Founders Pledge (2018) and Bollard (2017). The New York Times also provides a detailed account of PETA’s work with corporations in the fashion industry (Testa, 2020). Interested readers are encouraged to consult these resources to understand what implementing this intervention would look like in practice.

8.2    Key factors 

When implementing corporate campaigns against retailers in practice, we need to consider some crucial factors that might hinder our work. This section summarises our concerns (or lack thereof) about different aspects of a new charity putting this idea into practice. They only provide first indications and thoughts and should not be seen as in-depth research or advice.

Table 6: Implementation concerns


How concerning is this?


Low Concern

Access to information

Moderate Concern

Access to relevant stakeholders

Moderate Concern

Feedback loops

Low Concern


Moderate Concern

Scale of the problem

Low Concern


Low Concern

Execution difficulty/tractability

Moderate Concern

Positive externalities

Low Concern

Negative externalities

High Concern

8.2.1 Talent

The main roles required for this intervention if a new charity gets started would be:

  • Founder(s)/director(s) to run the charity
  • Campaign manager to plan and manage the campaigns
  • 1-3 employees to run the campaigns
  • External contributor, such as consultants on different topics

The roles do not need any rare skills and there is quite a large group of organisations working on similar organisations who train and hire people with such skills. A lot of the training can be done on the job.

Specific skills that might depend on the geographical location are language skills. For this, either a translator can be hired or the employees working on the campaign can be hired directly in the targeted countries. Local animal welfare organisations can be contacted for advice and collaboration.

8.2.2 Access Information

There are some areas where the access to information appears to be limited. Specifically, finding data on the amount of silk products sold by country and retailer seems difficult (see section 3.1.1), which impedes our ability to find the most relevant targets for campaigns. This can reduce the overall impact of campaigns and it might be hard to more accurately quantify the amount of silk products and therefore silkworms impacted by this intervention. Our experts explained ways in which this information gap could be circumvented though. Relevant stakeholders

One type of stakeholders are mission-aligned people and organisations that we could collaborate with or learn from. This includes organisations with experience in the fashion industry like PETA and Four Paws or with experience in corporate campaigning in other areas like those associated with the Open Wing Alliance. Access to these stakeholders should be easy, as we are working together on the same mission and knowledge-sharing is generally an important norm in the animal advocacy movement.

The public is another stakeholder we need to consider. They should be easily targeted by public campaigns, since advertising and marketing resources can simply be bought and other organisations have successfully targeted the public with their campaigns.

The hardest stakeholders to reach are probably those that are the target of our campaigns, the retailers of silk products. An expert on fashion campaigns described significant difficulties in getting some companies to cooperate. Jonas Becker seemed somewhat more optimistic. The fact that previous corporate campaigns have overall been quite successful makes us somewhat optimistic about this, but this needs to be considered carefully.

Another group of stakeholders to consider might be producers of raw silk or silk products. As described in section 3.1.1, we might want to contact such producers in order to learn more about the largest consumers of silk products and thereby identify the most promising targets for our campaigns. How willing and able such producers would be to talk about their customers and the destinations of their products is unclear and there might be significant barriers (e.g. non-disclosure agreements or business secrets).

8.2.3 Feedback loops

Corporate outreach and communication can be easily measured by the direct response and engagement (or lack thereof) of the targeted retailers. We can track their attitudes and the kind of commitments they are willing to make. Tracking follow-through rates on their commitments is significantly easier for the ban of silk products than for other campaigns that focus on welfare requirements (like cage-free eggs), as it can more easily be verified whether retailers are offering a certain product compared to checking the conditions in their supply chain. On the big picture level, we can also monitor the industry through revenue numbers and any commitments or sales restrictions imposed by retailers that were not the targets of our campaigns.

Trying to measure the impact of public awareness about the welfare issues around silk is more difficult, but also doable. We can measure the reach of different media and any signs of public pressure increase on the retailers. 

8.2.4 Funding

The funding for insect welfare is generally very limited. We have identified that Open Philanthropy has given out multiple grants to insect research and welfare, with $980,000 to insect sentience research in 2023 (Open Philanthropy, 2023a), $470,000 to Rethink Priorities for insect welfare research in 2021-22 (Open Philanthropy, 2022b), and further $270,000 for insect welfare field building in 2023 (Open Philanthropy, 2023b). The Wild Animal Initiative (WAI) has funded two insect initiatives with $63,000 and $58,000 respectively (Kerr, 2023). WAI is also funded by Open Philanthropy. Rethink Priorities launched two new initiatives in 2023, the Insect Welfare Research Society and the Insect Institute, both funded by Open Philanthropy grants (Insect Welfare Research Society, 2023; The Insect Institute, n.d.).

In the EA space, most if not all insect welfare and research funding thus comes from Open Philanthropy. Outside of this space, there are other broader funding streams. For example, PETA is the biggest campaigner against animal materials in fashion, and their funding comes mostly from individual donations. PETA gives out grants, but from the information available on their website, it appears that they mostly work on interventions and campaigns in house and only write out the occasional targeted grant, for example for specific research proposals. 

Open Philanthropy is also a major funder of corporate campaigns for farmed animals, with over $12 million given to Mercy for Animals over several years (Open Philanthropy, 2022b), $820,000 given to Equalia (Open Philanthropy, 2022c), $2.8 million to Animal Equality and with many more examples available on the Open Philanthropy website (Open Philanthropy, 2022d). 

In sum, most of our research directed us to Open Philanthropy, but there might be other funding sources that we are not aware of. Given the size of grants given out by Open Philanthropy to insect welfare causes and to corporate campaigns aimed at other animals, it seems like a corporate campaign against silk fits within their scope. 

Based on our cost-effectiveness analysis, we expect the annual costs to be between $200,000 and $1.4 million. The total funding requirements in our estimate are quite large due to the expectations of multiple campaigns in the projected ten-year span. However, based on other examples of Open Philanthropy’s funding, the scope of the interventions and the costs would be comparable with other grants given out to similar interventions. It might also make sense to start around the lower bound (less staff, one campaign initially) and increase the scope later.

Given the limited funding streams, there is a reasonable concern about obtaining funding, as it appears that without Open Philanthropy on board, this initiative might be difficult to fund. Gauging their interest in funding this intervention could be an important step when moving this idea forward.

8.2.5 Scale of the problem

The proposed intervention would only be able to target a small proportion of the whole silk market. As we are focusing on high-income Global North countries, the part of the problem the intervention can target is limited by excluding silk markets in other regions. The data to assess how much of the silk product market is covered by a specific retailer in a specific country is hard to access and more research into this would be beneficial for a more accurate CEA. 

The total scale of the problem is significant though and is projected to grow, with 420 billion to 1 trillion silkworms currently farmed per year, projected to grow to 750 billion to 1.8 trillion annually by 2034. Even accounting for the uncertainties around silkworms’ sentience and experiences (see section 3.2.3), the scale of the problem is large.

There is plenty of space to work on this issue, especially with no existing interventions targeting silk, and starting now can prevent more growth of the industry.

8.2.6 Neglectedness

As described in section 5.2, we are not aware of any organisations conducting corporate campaigns against silk. Insect welfare is generally a very neglected issue, as described in section 1.1.1. In total, silkworm welfare is thus a very neglected issue.

8.2.7 Tractability

Crucial considerations around tractability and likelihood of success are discussed in section 3.1.3 (and also to some extent in sections 3.1.2 and 3.1.4). We show that this idea seems tractable, but also highlight some concerns.

To summarise, we find that similar interventions in the fashion industry and other areas have been run with success and there seems to be a general public understanding that ‘there is something wrong with silk production’. Many retailers have transitioned away from other animal materials and some even from silk. There are other arguments in favour of stopping silk sales that can be used in the campaigns, such as the environmental footprint and forced labour. However, there are valid concerns around the lack of public concern for the welfare of insects and the lack of good alternatives to silk as a material. 

8.2.8 Externalities Positive externalities

In the fashion industry, often convincing one large retailer leads to a shift in the industry and in the mindset. Especially based on the information by an expert on fashion campaigns, the targeted brands are often selected by their influence on other brands and not necessarily by the amount of the animal material products sold. There could be a positive ‘cascading effect’ impacting the whole industry in multiple countries.

Another positive externality can be to have a first successful corporate campaign directed towards insect welfare. This can encourage campaigns aimed at insects farmed for food and animal feed and be the catalyst for more interventions.

In addition, reducing silk consumption and thereby production can mitigate the negative environmental and human rights issues associated with silk production. Negative externalities

A major potential negative externality lies in the uncertainty around silkworm sentience. If silkworms are not sentient, any resources dedicated towards the issue are wasted and would be better employed elsewhere.

Relatedly, there is some uncertainty around whether the lives of farmed silkworms are net negative on average. While we do think that this assumption is true (as described in section 3.2.3), we are far from certain. If farmed silkworms’ experiences are indeed more positive than we assume (either roughly neutral or even net positive), our campaigns could have no or even a negative impact.

Despite planning to include the issues of ‘ethical’ silk in our campaigns, retailers could still choose to transition to ‘ethical’ silk instead of finding other replacements or reducing silk product offering. This could lead to more silkworms farmed, as more worms are needed to produce the same amount of silk.

In addition, there can be negative consequences for other existing corporate campaigns in related areas. Attention is sparse and by asking retailers to commit to a ban of silk sales, we might lower the chances of success for other campaigns (against wool, leather, etc.).

Finally, if there is a strong counter campaign, silk could also be more strongly marketed (for example as a sustainable material), leading to a more positive image and increased demand for silk. It seems unlikely though that the counter campaign could be stronger than the anti-silk campaigns.

9  Directing funding towards and existing entity  

Outside of the EA space, PETA stands out as a prominent charity with decades of experience in fashion campaigns. Some of the drawbacks of their campaigns are that they do not appear to focus on quantifying their impact and optimise based on that. The advantages of PETA are its massive public presence, decades of experience, and thorough understanding of the fashion industry. We have not looked into whether PETA are open to working on specific projects, if given funding. However, this is probably not the most promising strategy, as their methodology and values diverge in certain aspects from the EA approach. Similar considerations apply to organisations like Four Paws or any other larger not specifically EA-aligned organisations with experience on fashion topics.

Out of the existing organisations in the EA space, The Humane League and Mercy For Animals are among the largest and best known campaigners. Their focus lies mostly in farmed vertebrate animals, with most of the commitments focusing on cage-free and broiler campaigns. While they have the campaigning experience and knowledge, both the fashion industry and silkworms as insects seem out of their current scope and strategy. They would probably be open for consultations around this topic, but directing funding from their campaigns and venturing into a different industry and species does not seem easier or better than setting up a new organisation. The same should apply to other EA-aligned organisations working on corporate campaigns, for instance as part of the Open Wing Alliance.

To summarise, there are no currently existing organisations that seem well suited to include silk campaigns in their scope. Starting a new charity to focus on this intervention should make more sense and it is likely that the existing charities are open to share their expertise, if useful.

10   Conclusion and next steps

Based on the findings of our research and after a decision-making meeting and discussion with other researchers in the animal welfare space, we slightly lean towards recommending this idea for a newly incubated charity (~3.7 on a scale from 1 (not recommend) to 5 (recommend)). The quite low cost effectiveness estimate and some major remaining uncertainties mean that we cannot confidently recommend this intervention at this point yet.

Instead, we have identified several promising next steps that could help reduce those uncertainties and lead to a stronger conclusion towards (not) recommending this intervention further. Firstly, we should commission a market study or talk to fashion industry experts to identify the exact silk market share of specific retailers in specific countries. Secondly, we should talk to PETA or other established players in the fashion space to learn more in depth about which specific retailers are more likely to collaborate based on their reactions to previous campaigns. This should give us a clearer idea of the market share we can target and therefore the cost effectiveness of this intervention. Another uncertainty we could address is funding availability. We could contact Open Philanthropy to gauge their interest in potentially funding corporate campaigns against the sale of silk products.

Conditional on the findings of such further research showing that there are highly promising retailers to target and funding streams available, it seems valuable for a charity to run a pilot intervention on this. The campaigns have quick and strong feedback loops and there should be strong learning value in running a first corporate campaign on an insect farming issue. The intervention could be shut down again fairly quickly, if the pilot does not prove to be promising.


Aishwariya, S. (2020, January). Interesting dimensions of Ahimsa silk (Vegan silk). ResearchGate. https://www.researchgate.net/publication/340279882_Interesting_dimensions_of_Ahimsa_silk_Vegan_silk

Allied Market Research. (2022). Silk market size, share & growth | Industry analysis report 2031https://www.alliedmarketresearch.com/silk-market

Animal Ethics. (n.d.). Silkhttps://www.animal-ethics.org/silk/

ASOS. (n.d.). Leather and wool products at ASOShttps://www.asos.com/customer-care/product-stock/why-have-asos-not-banned-leather-or-wool/

Beck, A. (n.d.). Alexandria Beck: How the open wing alliance is driving progress worldwide. Effective Altruism. https://www.effectivealtruism.org/articles/alexandria-beck-how-the-open-wing-alliance-is-driving-progress-worldwide

Blanchard, W. O. (2009). The status of sericulture in Italy. Annals of the Association of American Geographers19(1), 14-20. https://doi.org/10.1080/00045602909357092

Boddy, A., & Jimenez, A. (2023, September 25). Two years of shrimp welfare project: Insights and impact from our explore phase. Effective Altruism Forum. https://forum.effectivealtruism.org/posts/Qo3559TqP5BzoQyWX/two-years-of-shrimp-welfare-project-insights-and-impact-from

Bollard, L. (2017, April 11). Why are the US corporate cage-free campaigns succeeding?. Open Philanthropy. https://www.openphilanthropy.org/research/why-are-the-us-corporate-cage-free-campaigns-succeeding/

Capriati, M. (2018). Cause area report: Corporate campaigns for animal welfare. Founders Pledge. https://www.founderspledge.com/downloads/fp-animal-welfare

EA Forum Archives. (2019, August 28). What can we learn from the fur-free fight?. Effective Altruism Forum. https://forum.effectivealtruism.org/posts/x4gqDBNYGQf7GdGpB/what-can-we-learn-from-the-fur-free-fight

Faunalytics. (2016, October 21). Does insect suffering bug you? https://faunalytics.org/does-insect-suffering-bug-you/

Fischer, B. (2022, November 7). The welfare range table. Rethink Priorities. https://rethinkpriorities.org/publications/the-welfare-range-table

Fischer, B. (2023, January 23). Rethink priorities’ welfare range estimates. Rethink Priorities. https://rethinkpriorities.org/publications/welfare-range-estimates

Founders Pledge. (2018). Cause area report: Corporate campaigns for animal welfare. Founders Pledge. https://www.founderspledge.com/downloads/fp-animal-welfare

FOUR PAWS. (n.d.). Four pawshttps://www.four-paws.org/

Fur Free Retailer. (2023). The future of fashionhttps://furfreeretailer.com/

Fur Free Alliance. (2023). Recent news. Fur Free Alliance. https://www.furfreealliance.com/

Ganguly, S. (2008, May 23). Ahimsa silk weaves its way into global marketplace. The Economic Times. https://economictimes.indiatimes.com/ahimsa-silk-weaves-its-way-into-global-marketplace/articleshow/3064434.cms?from=mdr

Global Commodities. (n.d.). Producers and consumershttps://globalcommodities-silk.weebly.com/producers-and-consumers.html#:~:text=Japan%20is%20the%20largest%20silk%20consumer

The Guardian. (2018). Asos pledge to ban mohair, Cashmere and silk from 2019. The Guardian. https://www.theguardian.com/fashion/2018/jun/19/asos-pledge-to-ban-mohair-cashmere-and-silk-from-2019

Guesstimate. (n.d.). ACE's 2018 CEE model for the humane leaguehttps://www.getguesstimate.com/models/11268?token=IRiIOLspBVVmv5QW0WzruAf2H7WyjrG2NhafFZRioEraKPmAPLur0nz6Zh7jdoWQPuxXiOwKoP3bbP-D7AL1OA

H&M. (n.d.). Silk shirt - White - Ladieshttps://www2.hm.com/en_us/productpage.1176678002.html

Hayden Hill. (2023, April 7). 14 ethical silk alternativeshttps://haydenhill.co/blogs/journal/silk-alternatives

House of Wandering Silk. (n.d.). Ahimsa silk - history & meaninghttps://www.wanderingsilk.org/ahimsa-silk-the-story

The Humane League. (n.d.). We want to change the lives of 6.7 million birds. here’s howhttps://thehumaneleague.org/article/cage-free-goal

Humane Society International. (2023). The fur tradehttps://www.hsi.org/news-resources/fur-trade/

Humane Society International. (n.d.). Protect all animalshttps://www.hsi.org/

The Humane Society of the United States. (n.d.). Cold noses, warm heartshttps://www.humanesociety.org/

The Insect Institute. (n.d.). Addressing the challenges accompanying insect farminghttps://www.insectinstitute.org/

Insect Welfare Research Society. (2023). About IWRShttps://www.insectwelfare.com/about-iwrs

Institute for Policy Research. (2017, March 29). Do boycotts work? Institute for Policy Research - Northwestern University. https://www.ipr.northwestern.edu/news/2017/king-corporate-boycotts.html

Kerr, C. (2023). Wild animal welfare research grants 2022. Wild Animal Initiative. https://www.wildanimalinitiative.org/blog/grants-2022

Lockwood, A. (2022, July 23). Advocating for insects: Why, what and how to campaign effectivelyhttps://www.surgeactivism.org/articles/advocating-for-insects-why-what-and-how-to-campaign-effectively

Material Innovation Initiative. (2021). Material innovation initiative issues the “What Makes Silk, Silk?” Reporthttps://materialinnovation.org/wp-content/uploads/Silk-Report-Press-Release-.pdf

McFetridge, S. (2022, February 11). Cage-free chicken campaign scores surprising success. WPDE. https://wpde.com/news/nation-world/cage-free-chicken-campaign-scores-surprising-success

Miller, R., Schiestl, M., Trevarthen, A., Gaffney, L., Lavery, J. M., Fischer, B., & Schnell, A. (2022). From pigs to silkworms: Cognition and welfare across 10 farmed taxa. https://doi.org/10.1101/2022.11.11.516141

Morletto, E. (2023, October 12). Silk market: Europe reasserts itself as a major producer. Luxury Tribune. https://www.luxurytribune.com/en/silk-market-europe-reasserts-itself-as-a-major-producer

Norwood, F. B., & Lusk, J. L. (2011). Your eating ethics: A guide to eating based on your beliefs and preferences. https://doi.org/10.1093/acprof:osobl/9780199551163.003.0008

Open Philanthropy. (2022a). Rethink priorities — Insect welfare researchhttps://www.openphilanthropy.org/grants/rethink-priorities-insect-welfare-research/

Open Philanthropy. (2022b). Mercy for animals — Corporate campaigns (2022)https://www.openphilanthropy.org/grants/mercy-for-animals-corporate-campaigns/#:~:text=Open%20Philanthropy%20recommended%20a%20grant,%2C%20Latin%20America%2C%20and%20globally

Open Philanthropy. (2022c). Equalia — Corporate campaigns (2022)https://www.openphilanthropy.org/grants/equalia-corporate-campaigns-2022/

Open Philanthropy. (2022d). Animal equality — Corporate animal welfare campaignshttps://www.openphilanthropy.org/grants/animal-equality-corporate-animal-welfare-campaigns/

Open Philanthropy. (2023a). Queen Mary University of London — Insect sentience researchhttps://www.openphilanthropy.org/grants/queen-mary-university-of-london-insect-sentience-research/

Open Philanthropy. (2023b). Rethink priorities — Insect welfare field-buildinghttps://www.openphilanthropy.org/grants/rethink-priorities-insect-welfare-field-building/

Open Philanthropy. (n.d.). What can we learn from the fur-free fight? https://us14.campaign-archive.com/?u=66df320da8400b581cbc1b539&id=edb84be486

Open Wing Alliance. (n.d.). The global manufacturers reporthttps://assets.ctfassets.net/ww1ie0z745y7/226x5wBdDLLnZSyqVzIhUm/2906c2cd5dab6e128414e268c8099931/OWA-23-campaign-cage-free-manufacturers-report-final-2.23-update.pdf

PETA. (n.d.-a). A look inside the angora rabbit wool industryhttps://support.peta.org/page/1846/petition/1

PETA. (n.d.-b). Down and silk: Birds and insects exploited for feathers and fabrichttps://www.peta.org/issues/animals-used-for-clothing/animals-used-clothing-factsheets/silk-birds-insects-exploited-fabric/

PETA. (n.d.-c). Some of the 425+ brands that have banned angora woolhttps://www.peta.org/living/personal-care-fashion/more-brands-ban-angora/

Rowe, A. (2020, June 29). Insects raised for food and feed — global scale, practices, and policy. Rethink Priorities. https://rethinkpriorities.org/publications/insects-raised-for-food-and-feed

Rowe, A. (2021, April 19). Silk production: Global scale and animal welfare issues. Rethink Priorities. https://rethinkpriorities.org/publications/silk-production

RSPCA. (n.d.). Fur Free Britainhttps://www.rspca.org.uk/getinvolved/campaign/furfree#:~:text=The%20fight%20against%20fur%20is,global%20leader%20in%20animal%20welfare

Saulius. (2019, February 1). Will companies meet their animal welfare commitments? Effective Altruism Forum. https://forum.effectivealtruism.org/posts/XdekdWJWkkhur9gvr/will-companies-meet-their-animal-welfare-commitments

Schukraft, J. (2019, June 12). Invertebrate sentience: A useful empirical resource — EA forum. Effective Altruism Forum. https://forum.effectivealtruism.org/posts/AMFuSWrsutBFraZtE/invertebrate-sentience-a-useful-empirical-resource

Schukraft, J. (2019, July 9). Invertebrate welfare cause profile. Rethink Priorities. https://rethinkpriorities.org/publications/invertebrate-welfare-cause-profile

Shute, N. (2012, March 30). Is that a crushed bug in your frothy Starbucks drink? NPR. https://www.npr.org/sections/thesalt/2012/03/30/149700341/food-coloring-made-from-insects-irks-some-starbucks-patrons

Simcikas, S. (2019, August 9). Corporate campaigns affect 9 to 120 years of chicken life per dollar spent. Rethink Priorities. https://rethinkpriorities.org/publications/corporate-campaigns-affect-9-to-120-years-of-chicken-life-per-dollar-spent

Sinclair, M., Lee, N. Y., Hötzel, M. J., De Luna, M. C., Sharma, A., Idris, M., Islam, M. A., Iyasere, O. S., Navarro, G., Ahmed, A. A., Curry, M., Burns, G. L., & Marchant, J. N. (2022). Consumer attitudes towards egg production systems and hen welfare across the world. Frontiers in Animal Science3https://doi.org/10.3389/fanim.2022.995430

Somerset Wildlife Trust. (n.d.). Action for insectshttps://www.somersetwildlife.org/what-we-do/restore-somersets-nature/campaign-nature/action-insects

Statista. (2020, February 6). Leading European fast fashion brands based on total revenue worldwide in 2019 (in million GBP)*https://www.statista.com/statistics/1094176/european-fast-fashion-brands-ranked-by-revenue/

Statista. (2023, March). Net sales of the Inditex Group worldwide from 2013 to 2022, by brandhttps://www.statista.com/statistics/456505/sales-inditex-group-worldwide-by-format/#:~:text=Zara%2C%20which%20includes%20the%20Zara,to%20over%2023%20billion%20euros

Tayal, M., & Chauhan, T. (2017, February 14). Silkworm diseases and pests. SpringerLink. https://link.springer.com/chapter/10.1007/978-981-10-3304-9_9

Testa, J. (2020, May 31). How PETA won its messy fight and took a seat at the table. The New York Times. https://www.nytimes.com/2020/05/31/style/peta-fashion.html

Thomas, C. (2023, June 7). 10 most popular luxury brands in the world, ranked – but which is no 1? South China Morning Post. https://www.scmp.com/magazines/style/luxury/article/3223144/10-most-popular-luxury-brands-world-ranked-lvmhs-louis-vuitton-chanel-and-dior-and-kerings-gucci

Tomasik, B. (2016). Insect suffering from silk, shellac, carmine, and other insect productshttps://reducing-suffering.org/insect-suffering-silk-shellac-carmine-insect-products/#Silk

UNCTAD. (2022). Handbook of Statistics 2022https://unctad.org/system/files/official-document/tdstat47_en.pdf

UNCTAD. (n.d.). Classificationshttps://unctadstat.unctad.org/EN/Classifications.html

Wikipedia. (2022, June 26). Eri silk. Wikipedia. https://en.wikipedia.org/wiki/Eri_silk

Wikipedia. (2023). Synthetic fiberhttps://en.wikipedia.org/wiki/Synthetic_fiber

Witrak, C. (2012, March 30). Starbucks admits its mistake in hiding the use of cochineal beetle extract from customers. Yahoo Finance. https://finance.yahoo.com/news/starbucks-admits-mistake-hiding-cochineal-192000461.html?guccounter=1

Zalando. (n.d.). GANT Scarfhttps://en.zalando.de/gant-scarf-burnt-sugar-ga351g00x-o11.html

Zara. (n.d.). ZW Collection Printed Silk Tunichttps://www.zara.com/uk/en/zw-collection-printed-silk-tunic-p08571743.html?v1=295443598&v2=2290840





More posts like this

Sorted by Click to highlight new comments since: Today at 2:01 PM

Thanks for sharing, Zuzana and Moritz!

This would reduce the large width of our cost-effectiveness estimate, which currently lies between 0.1 and 1,311 welfare points per dollar, with a geometric mean of 12 welfare points per dollar.

Given the maximum welfare of 100 welfare points per year, my understanding is that 1 QALY could roughly be interpreted as 100 welfare points. So I think your range is 0.001 to 13.11 QALY/$, whose geometric mean is 0.12 QALY/$. GiveWell's top charities create 50 QALY, saving one life, for around 5 k$, i.e. they have a cost-effectiveness of around 0.01 QALY/$. So I believe your preliminary range could be parsed as 10 % to 1.311 k times the cost-effectiveness of GiveWell's top charities in terms of direct effects.

Thanks for adding that helpful perspective!

We did not make the comparison to QALYs since the program looked at global health/development and animal interventions separately. So we only compared animal interventions to other animal interventions. My personal perspective is that animal interventions indeed tend do be much more cost-effective, which is why I focus on that area.

It might also be interesting to note that CE's own previous research (which we did not really look at when we conducted our own research) estimates ~20-30 welfare points per dollar, which is somewhat higher than our best guess of 12 WP/$.

This was comprehensive and I didn't read carefully so I might have missed something: I think the likelihood of success might be too high, or at least not accounting for all likelihoods of success that is relevant to the cost effectiveness result. So my comment on likelihood has 2 parts:

Part 1: I think (and maybe I missed where you included it) you might want to also include the chance of success of the organization or project itself. Many start-ups fail due to bad team dynamics etc. independent of how likely suppliers might be to sign on. Maybe a CE success rate could be used as a prior.

Part 2: A 10% to 60% chance of a supplier signing on seems very high. This has a "middle value" of 35% (not sure what distribution you used). Coming from a sales point of view, your pitch to these companies seems way harder than e.g. selling some new cash registers to their stores. And I think often the likelihood of a sale is <1%. So if you only have 10 large retailers to "sell" this to, then the chance of failing to sell to any of them at the 1% level is 99%^10=90% meaning there is only a 10% chance of success (if I got my math right). So depending on how likely you think you are to make a "sale" and how many prospects of sufficient size you have, you might be able to calculate another indication of the chance of success to inform your analysis.

It was fun to read and I love how work such as yours probably make me expand my moral circle a bit - even if the chance of sentience is low, if the worms are sentient there are so many of them that this feels like unnecessary "moral risk". Also, I hope I am mistaken and that the cost effectiveness is actually promising. I just wanted to provide what I hope might be constructive feedback as I got a bit of a feeling this intervention sounded a bit too unlikely to be cost effective. 

Thanks for reading our work and engaging, Ulrik! 
I think your concern is fair and well spotted, we do feel the highest uncertainty in our estimates of likelihood of success. 

Concerning part 1, as far as I know the CE incubation success rate is 80%, which considering our wide interval range is relatively close to 100%. I do agree that it seems to be an additional factor worth including, depending on whether we look at starting a new organisation or supporting an existing one with a good track record.

Concerning part 2, we updated on these numbers several times during the research process. One major source of optimism were our interviews with experts, whose experience suggested ~50% rate of success. We tried to look at past corporate campaigns, but a problem with using the existing data on other corporate campaigns is that the situation for silkworms as insects might be too different than when it comes to cage-free and broiler campaigns. The public concern, place of production and scale of the problem vary too strongly. There is also a long history of fashion corporate campaigns, but it is hard to find the numbers to estimate their average success rate. In our next steps section we highlight the need for a better understanding of the fashion industry and the retailer distribution and the benefits of consulting experts on the topic of fashion campaigns. We would expect this to significantly update our likelihood of success.

To sum it up, I mostly agree with your concerns. With our report we got to a point where we did not have the capacity to resolve these uncertainties in the time given, but know the next steps we would take to get to a more accurate CEA. 

Related discussion here:

Also linking CE's own previous research here. Note that we did not really look at this previous report when we conducted our own research and came to the topic with fresh eyes.

Great work! 

I skimmed and may have missed things, but I was surprised to see that "the majority of commitments in other areas have been won by direct collaboration with companies and without the need for a public campaign". 

Just to clarify, was your impression that companies were happy to collaborate and compromise with advocates because of their desire to branded as ethical and sustainable? That is, they might see a credible threat that if they don't compromise, a public campaign might be launched. And they want to avoid this because the campaign could stain the company's reputation no matter whether it ends up taking up the corporate commitment in the end, after public pressure? 

Thank you!

For fashion brands specifically, the "sustainability" or "ethical" label are a strong PR point for many (and no concern at all for others). This has been leveraged into collaborative changes with such brands in the past, especially by PETA, who used to run more "bad cop" campaigns but now lean more into collaborating. One reason for this can be the growing concern of the public, and specific subgroups of consumers who care or want to show they care about sustainability or animal welfare. Probably also the history of PETA's strong campaigns can lead companies to look for a collaborative approach and compromise, as you said under the threat of a public campaign. The high success of these campaigns also most likely lies in very strategic selection of the brands - PETA and other campaigners usually go for the "low hanging fruit" and target brands that have an image of caring about such issues and are likely to want to maintain it. 

Thanks both!

I just wanted to add to Zuzana's response that our basis for the statement in question is also based on what we have learned out about cage-free corporate campaigns. From our understanding, the threat of a "bad cop" campaign is often enough for companies to sign commitments. So yes Jojo, your interpretation is correct I would say.

Thank you both for your responses! Makes sense

Curated and popular this week
Relevant opportunities