In September 2017, Amazon announced its intention to build a second headquarters, dubbed HQ2. Over 200 municipalities in the US, Canada, and Mexico submitted proposals to host the site, with offers for everything from tax breaks to infrastructure improvements to promises of partnerships with high schools and universities in order to train a whole workforce for Amazon. Some even offered to rename their town Amazon. In the end, Amazon selected three locations: New York City, Arlington, and Nashville.
Labor and community organizations in each location quickly coalesced to oppose the developments and the use of public money to incentivize them. In New York, the coalition organized demonstrations; allied with key elected officials, like Senator Michael Gianaris; and trained community members in Queens — Amazon’s would-be neighbors — to act as spokespeople. The coalition hit the company on everything from its track record on workers’ rights to its role in the deportation machine. In the end, Amazon pulled out of the deal.
The story of the HQ2 fight shows that taking on a megacorporation like Amazon requires a broad-based effort involving workers, tenants, immigrant communities, and elected officials. Still, the victory in New York did not change the outcome in Arlington or Nashville — an indication of the challenge of organizing against the company without broad movement investment.
As we think about how to organize Amazon’s workforce, it is vital that we ask ourselves if our organizing matches the structure of the company — geographically, sectorally, and financially. Currently, it does not to the extent it needs to. Warehouse organizing is concentrated in just a handful of cities, like New York, Chicago, and Minneapolis, and usually at just a handful of facilities in the area. Excellent organizing has been done around some of Amazon’s divisions — like Ring — but vast pieces of the company — like Amazon Web Services, prescription drug delivery, and government procurement — have gone largely untouched. We know which pieces of the company make Amazon the most revenue and profit, but that has not translated into where we organize or led to a capital strategy outside of shareholder resolutions.
To truly take on a megacorporation like Amazon, we must go against it wherever it is — from rural areas to cities, from tech offices to warehouses — and we must build coalitions with whichever groups it harms. Community-labor coalitions are not new, but the number of groups directly or indirectly harmed by Amazon through its support of the carceral state and deportation machine, its contributions to climate change, and its role in gentrification is unprecedented. Amazon has negatively impacted the lives of a huge number of working-class people, and they are all part of our potential base. Finally, we must target our campaigns to extract maximum pain by creating bottlenecks in the company’s “value chain.” Doing so will take a scale of organizing and resources that our movement is not currently mobilizing but could (and should) be. It will also take patience; we must be willing to dig in for the long haul and accept that victory won’t look like clean majority NLRB elections. More likely, it will look like steady victories on things like wages, safety, and worker surveillance that are negotiated directly with the workers rather than delivered from on high.
Our organizing must operate at the geographic level that the company does.
Amazon has two major philosophies when it comes to its logistics network: mobility and redundancy. This means stopping work at a single warehouse, even a large one, will not necessarily stop production. Amazon does not rely on single warehouses for its products, and with its vast network of delivery vans, trucks, planes, and even its own gig platform, the company can reroute orders incredibly quickly. If workers shut down one warehouse, dozens of others with the same stock can fill that gap, and fill it without missing a two-day delivery window. The distributed organizing model — in which workers with support of unions and other organizations build a minimum level of support (20-30 percent, for example) across a specific geographic area — has promise here. Unions, workers organizations, or coalitions would coordinate between sites but workers could also take quick actions in their locations around issues like scheduling, unfair discipline, or safety. Groups like Amazonians United (AU) have had success in using a distributed model to build militant minorities in a larger group of warehouses. The organization has done quick actions like marches on the boss around company discipline, and it’s also organized a handful of regional walkouts, such as one in Chicago that won workers $3/hour raises. This month, workers in New York City and Maryland coordinated walkouts demanding a $3/hour raise.
This kind of militant organizing — combined with political and community coalition building — has great potential, especially with consistent minimum density across a chokepoint. The Inland Empire (IE) of California and the Lehigh Valley of Pennsylvania are two areas where Amazon’s — and the country’s — logistics and warehouses are concentrated. Forty-three percent of US imports move through the IE, and Lehigh is the logistics hub for the entire Eastern Seaboard from Maine to Virginia. These natural chokepoints provide leverage for a minority campaign spread out across warehouses in the region. For years, the Warehouse Worker Resource Center and the Teamsters have been working in the IE on the logistics sector as a whole. Through their organizing on warehouse floors, WWRC and the Teamsters identified rate and time off task (as quotas at Amazon are called) as the key issues for workers. They lobbied for a regulation, AB701, to disclose and prohibit quotas for warehouse workers, and then used the proposed regulation to create a common goal for distributed groups of workers to unify around, as well as a floor that they can enforce and build up from.
We must be clear about how the corporation actually makes its profit.
This will allow us to most effectively identify the corporation’s weaknesses and vulnerabilities, which we can leverage to bring them to the table. Take the example of the Asia Floor Wage Alliance (AFWA). AFWA is a coalition of unions and NGOs (notably women’s organizations) in Southern and Southeast Asian garment-producing countries that is working to establish a cross-country floor wage for the garment industry. Its member unions actively organize within garment production. However, bargaining power on the employer side doesn’t rest within the individual supplier companies but with brands like Gap, H&M, and, of course, Amazon (now the largest apparel company in the US). The power is with the purchaser and ultimate seller.
AFWA is thus building alliances with groups in the US and Europe to bring direct pressure to brands. AFWA organizes around issues that are high priorities for workers and that the brands are also keen to show they’re addressing, like gender based violence (GBV). After the murder of a Dalit garment worker named Jeyasre Kathinavel in Tamil Nadu in 2020, AFWA and Kathinavel’s union, the Tamil Nadu Textile and Common Labor Union (TTCU), launched a global campaign to bring justice to Kathinavel and her family. They arranged speaking tours of union leaders throughout the US and Europe and pursued legal pressure using International Labor Organization conventions and the Organization for Economic Cooperation and Development complaint process. They also partnered with groups in the brand countries to organize letter writing and demonstrations at brand locations.
Starting last year, TTCU began to sit down with brands like Gap and H&M. Negotiations are ongoing, but they are nearing supply chain-wide agreements on gender based violence that would get workers directly involved in training and oversight of managers within supplier factories, with the brands making compliance a part of continuing business with the supplier company. [Editor's note: since the publication of this article, H&M has signed a binding agreement.] In a practical way, AFWA now has access to these brands and their supply chain compliance officers as well as a better understanding of what kind of leverage they are willing and able to bring down on suppliers. This puts AFWA in a better position to negotiate over wages down the line. The GBV campaign also proves the core AFWA theory: that the brands are where the ultimate power in the garment factories lie and that, to avoid a race to the bottom, organizers should target the entire industry. The workers are bargaining with the company at the level the company operates, not through the narrow paths of the workplace or even country.
This organizing model can help us think about the most effective ways to harm Amazon’s bottom line. Amazon Web Services is not only the corporation’s most profitable arm but also how it earns lucrative government contracts. Worker organizers could create a crisis for Amazon by lobbying for an executive order imposing restrictions around labor and safety standards to government contracts for cloud computing. Doing so would reverse the dynamic of Amazon using AWS revenue to fuel its warehouse growth; in this case, warehouse conditions would affect the growth of AWS in the government sphere.
Our organizing must be as wide as it is deep in order to bring in all the groups that have a direct interest in fighting Amazon.
Amazon has its tendrils in so many parts of the economy that there is a huge potential base of people who could be mobilized to take it on. For example, in both the New York and Arlington HQ2 fights, immigrant groups organized against the company’s deep connections with ICE. In Joliet, Illinois, Warehouse Workers for Justice has been organizing a coalition around massive subsidized water usage by warehouses in the area, a dynamic that means residents are paying ever increasing water rates as Joliet may literally run out of water. There’s also Amazon’s partnerships with local police, notably through its camera doorbell, Ring. Amazon signs contracts with police departments for access to the cameras, which is buried in terms of use agreements, increasing police surveillance and investing the company in the police and surveillance states. Last year, the coalition that took on HQ2 in Arlington successfully got the county to cut its contract with Ring.
The Awood Center in Minneapolis offers another example of organizers mobilizing communities affected by Amazon, but in this case there’s a direct workplace connection as well. Awood (the Somali word for “power”) is a worker center based in the city's Somali community which has been organizing at Amazon since 2017 — the longest public organizing drive at the company. Somalians make up over 30 percent of the Amazon workforce in the Twin Cities while constituting 10 percent of the general population. Without these workers, Amazon can’t function. In 2018, Awood organized the first North American walkout of Amazon — bringing management to the table to negotiate with workers. They won the requirement that a white manager must speak to a Somali manager before making personnel decisions, that individual complaints be responded to in five days, and that management meets with workers every two months. In 2020, another walkout brought back Faiza Osman after she was fired for staying home due to COVID. One of the reasons for Awood’s success is that they negotiate not just over leave and safety but also prayer breaks — an issue that cuts across the workforce and has unified the Somali community behind the workers. The community’s support has allowed Awood worker leaders to be public in their organizing. Though some worker leaders have been fired, the community and fellow workers have rallied behind them, and Awood has continued to build.
Amazon’s massive environmental impact offers another opportunity for uniting groups beyond the warehouse floor. Tech workers at the Seattle headquarters have organized under the banner Amazon Employees for Climate Justice around Amazon’s shallow climate commitments, employing capital strategies like shareholder resolutions to push the company to cut emissions, particularly from logistics and cloud computing. San Bernardino Airport Communities in California (supported by WWRC and the Teamsters) and a coalition of groups in New Jersey, including Clean Water Action and Make the Road, are leading another effort to oppose new Amazon airhubs in already polluted Black and brown neighborhoods — where large numbers of Amazon workers also live.
Focusing just on unionizing warehouse workers risks missing the multiple ways that Amazon impacts working-class communities, underscoring the importance of building a broad coalition of community members and workers. The Athena Coalition was founded in 2019 to do just that: build a coalition that takes on Amazon’s role in the economy and our political system as a whole. This means winning in every arena that Amazon operates in and, ultimately, making it so the company can no longer exist as it is today.
Conclusion
Amazon built itself up deliberately, so we must also be deliberate in the way we organize the company. Distributed organizing can create bases in strategic areas that are not tied to singular NLRB election wins. Combined with geographic targeting, this organizing model sets us up to actually disrupt Amazon’s money flow — a strategy that would be especially effective if coupled with government policies hitting other parts of its business model. With Amazon poisoning communities and helping police and immigration services enforce a regime of terror, the imperative to organize the company is great. Amazon has shown itself to be ruthless, but it is also vulnerable to creative and expansive organizing. Winning at Amazon — truly breaking its power over workers and working-class communities — would be a bellwether of the strength of our movement. It’s a challenge we must meet if we are to bring about the changes our society desperately needs.
