A flop - despite BLM, Democrat and celeb backing

Labour’s latest Waterloo

The defeat of the campaign for union recognition at Amazon is not as straightforward as might be thought, argues Daniel Lazare

A union organising drive at an Amazon distribution centre in Bessemer, Alabama, was actually supposed to be the opening battle of a ‘pro-labour’ offensive on behalf of Joe Biden and the Democrats. But workers refused to play along, and last week voted down the recognition of a union by more than two to one.

The defeat has left the Democrats reeling. The union - part of the 1.3 million-member United Food and Commercial Workers - enlisted Black Lives Matter activists, politicians like Bernie Sanders and Stacey Abrams, and celebrities such as actor Danny Glover and rapper Killer Mike. Biden even contributed a video praising unions as petty bourgeois: “I have long said America wasn’t built by Wall Street, it was built by the middle class, and unions built the middle class.”

Referring to the 1935 legislation that for the first time gave unions a legal federal status, Biden added that “the National Labor Relations Act did not just say that unions are allowed to exist: it said we should encourage unions”.1 That is pure nonsense, as anyone the least bit familiar with US labour history will attest. But liberals said it was the most pro-union statement by any president in US history, and the UFCW was hopeful that it would put it over the top.

Despite all this - or could it be because of it? - the Amazon workers (85% of them black) turned thumbs down. Afterwards, a black worker named Lavonette Stokes offered the interesting observation that workers were “turned off by how organisers tried to cast the union drive as an extension of the Black Lives Matter movement because most of the workers are black”, according to The New York Times. As Stokes put it, “This was not an African-American issue. I feel you can work there comfortably without being harassed.”2

Could it be that the UFCW’s heavy emphasis on race rather than class may have backfired? If so, what does it say about the “correlation of forces” in terms of the US working class as a whole? Are indicators still pointing downwards, thanks to Bessemer? Or is there a hint of an uptick in the air?

The answer is the former, with the situation currently hovering somewhere in between bad and dreadful.


Why? The Democratic Party is as good a place as any to start. The party, obviously, is bourgeois through and through and so conservative that in many respects it is to the right of a standard European conservative party like the Tories. To the extent Democrats have forged alliances with the unions, it is not in support of true labour demands, needless to say, but of a cold war programme based on imperialism, suburbanisation and consumerism.

Unions are good, in Democratic eyes, if they help workers buy homes and cars, and thus contribute to the urban flight that has left millions of poor people - many, but not all of them, black and Hispanic - stranded in crime-ridden inner cities. They are bad, on the other hand, if they enable members to fight back on a class-conscious, working class basis, which, fortunately for the Dems, virtually no US unions do.

But since the collapse in union membership in the 1970s and 80s, the Democratic programme of middle class expansion has collapsed, as joblessness and social decay gnaw at the edges of a coast-to-coast exurban wilderness, consisting of big-box stores, fast-food restaurants and abandoned factories. In vast sections of the US, the only employers who still pay halfway decently are hospitals and prisons - which is why workers in the few industrial facilities that remain are generally not inclined to rock the boat.

Bessemer is a good example of what happens to those left behind. Located 18 miles southwest of the Alabama state capital of Birmingham - named after what at the time was the foremost industrial city in the world - it is itself named after Henry Bessemer, the Englishman who in 1856 invented the steel-making process that would dominate the industry for a century. It is rather as if China had named a couple of hi-tech centres after Silicon Valley and Steve Jobs. A thriving steel and railroad-car manufacturing centre by the turn of the 20th century, Bessemer was left high and dry by deindustrialisation some seven or eight decades later. Its population down 20% to around 27,000, it now has the highest violent-crime rate of any city of more than 25,000 people in the entire country and was recently named by a business website as the worst place to live in Alabama (itself the nation’s sixth poorest state).3

Working class residents can therefore be excused if they’re sceptical of Democratic assurances that happy days are around the corner. Likewise, they can be excused if they don’t think much of the UFCW’s organizing strategy. With more than two hundred distribution centres around the country, Amazon uses a swarm approach that allows it to quickly shift operations from one facility to another in case of a closure. Customers wouldn’t notice a thing, since a super-capacious highway system, the lowest fuel prices in the advanced economic world, and a fleet of some 50,000 trucks and vans all but guarantee that deliveries will continue without skipping a beat. Yet workers will never succeed in shutting an Amazon distribution facility down in the first place - short of a factory occupation, that is - since compliant state and local police can be counted on to limit picketers to a token presence at all entrances and exits.

So when it came to jobs starting at $15 an hour versus the empty promises of a union in bed with the Democrats, Amazon workers had no trouble figuring out which option is the more promising. The Old Confederacy is emerging as a major new manufacturing centre. Yet after major defeats not only in Bessemer, but at a Volkswagen plant in Chattanooga, Tennessee, in 2019 and at a Nissan plant in Canton, Mississippi, two years earlier, organised labour is making little headway.

But it is not only tired old union hacks who are to blame. There is also US labour law - a dense web of rules and regulations covering union-certification elections, compulsory arbitration and the like, whose purpose is to tie workers down Gulliver-style, so they can barely move a muscle. So complete were the controls back in the 1980s that the joke was that, while it might still be legal to strike, it was definitely illegal to win. But even that is gone nowadays, as bosses employ an army of legal muscle to squash work stoppages before they can even begin.

One study found that union militants stood a 19% chance of being fired if they participated in a union-organising campaign - an activity supposedly protected by federal law. Another found that bosses threatened to close down operations in 57% of union-recognition elections if workers voted pro-labour, threatened to cut wages and benefits in 47% as well, and fired workers outright in 34%.4 It is rather like being thrown out on the street for daring to vote communist in some dusty Spanish village in the early 1930s.

Despite polls showing that more workers than ever want to join a union, the private-sector membership rate has fallen from 29% in 1970 to 6.2% as of 2019 - a drop of close to 80%. If the federal minimum wage had kept up with productivity, it would now be $22 an hour instead of the measly $15 that Congress recently voted down.5


The Protecting the Right to Organize Act - touted by both liberal Democrats and the Democratic Socialists of America - will supposedly change all this by stepping up penalties for employers who penalise pro-union workers, by preventing them from arbitrarily reclassifying workers as independent contractors to deprive them of their union rights, and by allowing sympathy strikes for the first time since the 1940s. But at the same time, it would actually tighten the net by imposing compulsory arbitration at nearly every stage.

This is the opposite of what led to the National Labor Relations Act in 1935. A year earlier, the US had seen a ‘hot spring’, in which three general strikes erupted within a five-week period: one in a series of auto plants in Toledo, Ohio; another among dockworkers in San Francisco and other West Coast ports; and a third among truckers in Minneapolis. The first was led by the American Workers Party (an intermediate grouping promising “an American approach” to Marxism), the second was led by the Communist Party, while the third was organised by the Trotskyists of the Communist League of America, forerunner of the US Socialist Workers Party.

All were broadly Marxist in orientation, and all took place in the rough-and-tumble days before the bourgeois state locked labour in a fatal embrace. All were expressly opposed to government-supervised elections, compulsory arbitration and other corporatist measures. They marked the start of a massive class war that the ‘new deal’ subsequently scrambled to contain.

Contrary to Democrats and the DSA, therefore, it was not the federal government that liberated workers: it was workers, rather, who liberated themselves before the federal government could figure out what was going on. Any labour upsurge in the 2020s would have to follow the same path. It must be Marxist, it must originate with the workers themselves, and it must oppose Democrats and the rest of the federal apparatus, now more than ever.

To say something must happen, of course, does not necessarily mean that it will. Bessemer shows that workers think for themselves and that they are perfectly capable of doing the maths without help from the Democrats. Indeed, it suggests that the American working class is a vast terra incognita as far as the party is concerned - a place where people speak a different language, entertain different thoughts and are distinctly unimpressed by celebrity endorsements or racial appeals by ‘woke’ liberals. Democrats may think they know what is in the mind of frontline workers, but, as Bessemer shows, they do not have a clue. This does not mean that true class-consciousness is beginning to emerge; but it does mean that workers are watching and waiting, as they make up their minds. The upshot could be another bout of labour stagnation - or a 1934-style explosion of working class militancy.

Historical comparisons may prove useful. Strikes plunged during the boom years of the late 1920s and fell even more during the bust. With unemployment shooting through the roof, the first priority of every worker was to hold onto whatever job they still had, no matter how low wages might go. Only after the crisis bottomed out in 1933 did strikes begin taking off.6

Similarly, after peaking in 1970, strikes have fallen to negligible levels over the course of half a century. The capitalist winter has proved longer and deeper than virtually anyone expected and has wreaked havoc with labour organisations of every size and shape - from ordinary labour unions to the deformed or degenerated workers’ states of the Soviet bloc. The long decline has also wreaked havoc in the US, the headquarters of the post-Soviet imperial system, with workers paying a greater and greater price, as time goes on. Mass dissatisfaction is building, but what is impossible to determine is whether it will peter out in populism of an increasingly rightwing slant or lead to a burst of genuine class war.

The economy is key, just as it was in the 1930s. But, as long as the Biden administration keeps flooding Wall Street with government-supplied credit, the advantage will lie with the capitalist class. Only when the stimulus ends will the working class movement have a chance to intervene.

  1. youtube.com/watch?v=Pd7w6tS0BSA.↩︎

  2. nytimes.com/2021/04/09/technology/amazon-defeats-union.html.↩︎

  3. 247wallst.com/special-report/2019/07/22/worst-cities-to-live-in-every-state-4/2.↩︎

  4. See cepr.net/documents/publications/dropping-the-ax-update-2009-03.pdf; and files.epi.org/page/-/pdf/bp235.pdf.↩︎

  5. epi.org/publication/our-deeply-broken-labor-market-needs-a-higher-minimum-wage-epi-testimony-for-the-senate-budget-committee.↩︎

  6. ‘Work stoppages caused by labor-management disputes in 1946’ Monthly Labor Review May 1947 (available at jstor.org/stable/41818327).↩︎