On May 17th, 2021 Amazon announced a new initiative aimed at improving the physical and mental wellbeing of its employees. Named WorkingWell, the program includes what the company calls ‘Health and Safety Huddles’, described as daily opportunities to engage employees on strong body mechanics, wellness topics, and ongoing safety education; ‘Wellness Zones’ that provide employees with voluntary stretching and muscle recovery exercises via accessible dedicated spaces; and AmaZen- interactive meditation kiosks to guide employees through mindfulness practices including meditations and calming scenes with sounds. In his April 2021 letter to shareholders, Amazon founder and then CEO Jeff Bezos (he has since stepped down as CEO while remaining Executive Chairman), announced that in addition to the WorkingWell program, Amazon is developing ‘new automated staffing schedules that use sophisticated algorithms to rotate employees among jobs that use different muscle tendon groups.’
The listed price for Amazon’s safety programs in 2021 was over $300 million, a hefty sum in much of the world but nary a drop in Amazon’s bucket. At the end of 2021, Amazon’s market cap was $1.7 trillion. A year into the pandemic the company reported its sales were up 44 percent; sales were up 67 percent higher in 2021 than they were in 2019. The first three months of 2021 saw $8.1 billion in profit, an increase of 220 percent from 2020. In the course of the pandemic Bezos’ personal wealth increased by 70 percent. Step back a decade and Amazon’s net income rose from $1.1 billion in 2010 to $21.3 billion in 2020.
As for the scope and efficiency of its online sales operation, consider at any given moment over 600 million items are up for sale and Amazon receives about 115 orders, basically a full delivery truck worth, every second. That’s 10 million fulfilled orders for a day. The roots of this was the creation of Amazon Prime back in 2005. It was an economic masterstroke. As of late 2006, Amazon had a total of five Fulfillment Centers (what the company calls its main warehouses), four in Kentucky and the original in Middleton, Delaware. By 2011, the number of Fulfillment Centers reached 21, along with some smaller facilities. With Prime taking hold, the building expansion really accelerated around 2021. According to the latest data on Mapping Amazon webpage (run by Good Jobs First), Amazon now has at least 500 facilities in its U.S. distribution network including 208 Fulfillment Centers, 107 delivery station networks for distributing, and dozens of Prime Now hubs processing two-day and same-day deliveries (along with ten airport hubs). The company claims it opened hundreds of facilities in 2021.
An estimated 60 percent of U.S. adults are Prime members–that includes 82 percent of households with income over $150,000. A survey by Consumer Intelligence Research Partners revealed 93 percent of Prime members keep their subscription after their first year, 98 percent do after their second year, and Prime members spend on average $1400 a year on Amazon purchases, compared with $600 by nonmembers.
If online shopping is what Amazon is primarily known for, it isn’t even close to the most profitable wing of the company. The pandemic may have made Zoom a household name, and brought Netflix and Slack to greater glory (Slack added 2.5 million users the first few weeks), they all are depend on Amazon Web Services (AWS). Essentially AWS lets any company outsource its storage and computational needs, saving it the expense of buying its own equipment. Most critical is the Infrastructure as a Service sector (IaaS), which supplies clients with servers, data storage, and networking hardware. The list of companies using AWS is endless: Reddit, Twitter, General Electric, Yelp. The CIA is a client. In 2013, the Agency agreed to spend $600 million to store its data on Amazon’s cloud. Apple itself forks over $30 million a month.
AWS is dominant in the IaaS sector with a 48 percent market share, more than triple Microsoft, its closest competitor. Revenue from AWS probably clears $50 billion a year. In fact, each dollar Amazon spends operating AWS generates ten times more profit than a dollar spent on its other ventures. It is probably no coincidence that it was Andy Jassy, AWS’s longtime chief, who Bezos tagged to be his successor as CEO.
As for some of those other ventures, Alexa has become a household word, eBooks have not proved to be quite the juggernaut they were envisioned to be, they have stabilized at around 20 percent of the book market, but Amazon usually has over 70 percent of it (along with 40 percent of paper book sales). With more product searches on Amazon than on Google, Amazon’s advertising business is large and growing- Amazon disclosed it generated $31 billion in sales in 2021. Twitch, Amazon’s video platform, attracts 15 million users a day. Amazon Prime Video trails only Netflix in streaming subscribers, and Amazon owns Whole Foods.
Still, the vast majority of Amazon employees and subcontractors work in online retail. In September 2021, Amazon announced plans to add another 125,000 employees in the U.S. It is now the second only to Walmart as the largest private employer in the U.S. On one hand there is no particular reason to single out Amazon for its treatment of workers. The Economic Roundtable, a nonprofit research group, recently surveyed more than 10,000 Kroger workers in Washington, Southern California, and Colorado and found that about 75 percent of them said they were food insecure, about 14 percent they were homeless or had been so in the previous year, and 63 percent said they did not earn enough income to pay for basic monthly expenses. Kroger is also one of the five largest private employers in the country.
For the past decade, Progressives have made the rather timid demand of a $15 minimum wage a central part of their platform and Amazon has been paying that for a few years. Still this shouldn’t disguise the brutal reality of working in an Amazon warehouse. Despite the hourly pay and meditation kiosks, the turnover rate can reach 150 percent a year. In October 2020, the Seattle Times reported that the turnover rate for Amazon’s front-line workers was at least double the industry average during the initial months of the pandemic.
Then there is the overall Amazon effect, which is quite real. The Prime business model has forced other major retailers to adjust. Target now offers free shipping on orders of $35 or more, and free two- day shipping on orders placed with a Target RedCard. Walmart offers the same. It is a safe bet this dynamic will only continue to expand.
The model not only grinds through workers like a buzzsaw but leaves them injured at an appalling rate. Data released last June by the Occupational Safety and Health Administration (OSHA) showed that since 2017 jobs at Amazon were more dangerous than other warehouse jobs. Walmart has never been considered enlightened when it comes to its workers, but in 2020 its rate of serious incidents was less than half of Amazon’s rate. In the years leading up to 2020 the numbers were worse (the pandemic prompted temporary measures that increased safety). From 2017 to 2019, the overall injury rate increased 21 percent from 7.5 percent in 2017 to 9.0 in 2019 (the serious injury rate was 6.5 percent in 2017, 7.8 in 2019).
A December 2021 report by the National Employment Law Project found that Amazon warehouse workers in Minnesota are more than twice as likely to be injured at work as other Minnesota warehouse workers. The report cites Amazon’s own records showing that the annual rate of injury at its Minnesota facilities is 11.1 percent, more than double the non-Amazon warehouse rate of 5.2. It also found that real wages for warehouse workers declined 14 percent in the years after Amazon began operating in the area.
Back in 2012 Amazon purchased the robotics company Kiva for $775 million. The reason for the acquisition was that Kiva just pioneered a robot that could move shelves around a warehouse by lifting them. If the conventional view is that technology liberates humanity from drudgery, this is far from the case for workers at Amazon. In a grim foreboding of the future, injuries are higher at the warehouses where the robots have been deployed than those where they have not been (54 percent higher in 2019 according OSHA data). The Kiva robots (along with Pegasus, an item-categorizing system) may serve workers some of the trouble of running around the warehouse since the robots follow markers on the floor and route shelves to workers to pick the correct items, but they also drive greater production speed and allow management to have greater surveillance power. If a lag between an exhausted worker performing repetitive tasks at high speed is too long, the time is logged as ‘time off task.’ Such time is tracked by computers and too much of it can result in a worker being disciplined or even fired. Of course, robots don’t get fatigued. According to data from the International Federation of Robotics, worldwide unit sale of ‘professional service robots’ increased 41 percent during 2020, while the money spent increased by only 14 percent- meaning the technology is getting cheaper.
A report from the Center for Investigative Reporting (for Reveal News) shows the effect of the machines:
The robots were too efficient. They could bring items so quickly that that the productivity expectations for workers more than doubled, according to a former senior operations manager who saw the transformation. And they kept climbing. At the most common kind of warehouse, workers called pickers- who previously had to grab and scan about 100 items an hour- were expected to hit rates of up to 400 an hour at robotic fulfillment centers.
It is inevitable that such conditions will stir worker organizing. Workers at the Amazon plant in Bessemer, Alabama are currently holding their second vote on joining The Retail, Wholesale, and Department Store Union (RWDSU). The union lost in a lopsided vote last April but filed a successful legal challenge with the National Labor Relations Board (NLRB), which ruled Amazon interfered with the first vote, particularly by installing a mailbox on the premises. The NLRB ruled the mailbox led workers to believe their votes were being monitored. Other company actions included getting the timing of a traffic light changed to give pro-union workers less time to canvass workers in their cars, outfitting temp workers, who are ineligible to join the union, with ‘vote no’ clothing to wear on the floor, and plastering the warehouse bathroom with anti-union flyers. The union originally filed for a 1500 worker bargaining unit. Amazon managed to get it extended to a 5,800 unit, forcing organizers to divert energy into getting more workers to sign union cards rather than solidifying the original 1500 bargaining unit. The counting of the second vote is scheduled to begin on March 28th. At that same time, workers at another Amazon warehouse in Staten Island, New York are scheduled to hold their union vote. This will be to join a self-organized union called the Amazon Labor Union (ALU). The ALU has filed a petition for an election at another nearby facility as well. The NLRB has confirmed that organizers at the facility have submitted signed cards for at least 30 percent of the potential union group.
The grind of unionizing one workplace at a time within an operation as vast as Amazon is an uphill battle. The high turnover rate certainly works against union effects. It is also not impossible to imagine Amazon simply closing a warehouse whose workers vote for a union, just like Walmart did to one of its Quebec stores in 2014. Amazon’s AI software can shift orders between warehouses in the case of a strike or disruption at a single warehouse. No doubt the company will continue to push the anti-union envelope hard.
Last October, motorcyclist Justin Hartley was riding his bike in Virginia Beach when he was hit by a truck with an Amazon logo. The accident fractured his left wrist and left Hartley without a left knee. He is suing Amazon for $100 million. According to the lawsuit, the driver, Christopher Gill, admitted his eyes weren’t on the road but on the Amazon-provided GPS (Gill has issued a denial). Kevin Biniazen, the attorney representing Hartley, called the accident ‘preventable’ and said in a statement ‘The unrealistic expectations that are put on drivers are fueling these negligence cases. The driver was so entranced in making his delivery that he did not see our client.’
Whatever the merits of this lawsuit, it illustrates that the Amazon effect doesn’t end at the warehouse. What is known as ‘last mile delivery’, usually meaning the delivery of commodities to customers at home, is one of the costliest parts of the ecommerce chain. Around one-third of shipping costs occur in this final stage. Hence the industry has long called it the ‘last mile problem’ since it is time-intensive involving multiple stops with low drop sizes. A major factor of Amazon’s acquisition of Whole Foods was likely the adding of 440 additional refrigerated warehouses within 10 miles of 80 percent of the population. As is often the case, ‘problem’ can simply be industry-speak for ‘unions’, and in last mile delivery there are unionized workers at USPS and UPS (the largest private sector unionized employer in the U.S.). This is a main reason why mountains of venture capital funding are being poured into autonomous trucks.
For the time being, Amazon’s last-mile solution is using subcontractors. These take the form of two programs: Amazon Flex and Amazon Delivery Service Partners (DSPs). Amazon Flex is a simply part of the gig economy. Drivers are independent contractors and must either use their own vehicle or rent a suitable one. They get paid by completion of a delivery route, not by the hour. Last year the Federal Trade Commission fined Amazon $61.7 million for withholding tips from its Flex drivers.
DSPs are small ‘independent’ delivery firms, with between 20-40 vans that exclusively deliver packages for Amazon Prime customers. These are the omnipresent gray vans with the Amazon logo. They are contractors, not Amazon employees. The size of the companies are limited to 40 vans, both to make driver organizing difficult and also to keep Amazon’s power and flexibility over the delivery prices. Dealing with numerous small enterprises gives Amazon such leverage. By 2019, half of Amazon packages were being delivered by these two groups of subcontractors, a number that no doubt is now higher.
Like in the warehouses, Amazon is constantly finding new methods to surveil drivers. In early 2021, the company installed AI powered cameras (made by the tech startup Netradyne). The cameras are now in more than half the vans in the fleet. Drivers have complained that the cameras, ostensibly installed for safety, observe drivers so closely they record data beyond the drivers’ control such as getting cut off on the road or having to adjust a mirror. Such data is used to evaluate performance and give out any prizes or bonuses. In March 2021, Amazon paid $8.2 million to contract drivers in Washington State to settle a class-action lawsuit over wage theft. A month later, drivers in Florida filed a similar lawsuit.
While all this paints a bleak picture of an unstoppable company, hellbent on keeping workers unorganized and exploited, no entity is truly invincible. The fight is never over. If any of the current organizing campaigns are a success, they will likely spur action in other workplaces. Such is happening now at Starbucks. Amazon is massive, but where there is size there is vulnerability. The enterprise is both stretched out, dependent on hundreds of thousands of workers over a large space, and tightly organized at the time. Coordination between warehouse workers and delivery drivers is far from impossible. The pandemic has brought the fragility of global supply chains to light, along with the heroism of frontline workers. As more shopping is shifted online, the public consciousness of Amazon’s frontline workers will only grow leading to more public support. The leviathan can still be breached.