In the United States, it is nearly impossible for workers to win a union contract against an employer that is determined to remain non-union. How much harder, then, to beat an employer that can hire $10,000-a-day anti-union consultants, get the United States Postal Service to place a mailbox in front of the plant for mail-in voting, and even change the pattern of stoplights to keep organizers from talking to workers at shift changes? This, of course, is what Amazon workers at plant BHM1 in Bessemer faced in their vote to win union recognition. In addition to the myriad obstacles that broken U.S. labor law places in the way of workers trying to organize against any company, megacorporations like Amazon, Uber, and Walmart also retain an awesome resource advantage. Beyond the shop floor, this resource advantage carries over into the political sphere. In California last year, Uber, Lyft, and Doordash outspent unions ten-to-one to overturn legislation guaranteeing employment rights to misclassified gig workers. Political and workplace democracy are both at a severe disadvantage against such agglomerations of wealth and power.
One response to the recent growth of corporate power has been a resurgent anti-monopoly movement. Notably for a movement often associated with consumers or small businesses, the current incarnation of anti-monopolism has won some support from labor unions. However, some of the pro-labor left remains skeptical. The most strident critic of antimonopolism, People’s Policy Project founder Matt Bruenig, maintains that only large companies have the resources to implement generous benefit plans, and that U.S. labor law makes it harder to organize workers at smaller companies in decentralized industries. Outside of Twitter polemics and magazine articles, there is also a longstanding belief in many parts of the labor movement that concentration benefits unions. Some trade unionists look back to the stable bargaining relationships unions once had with giant corporations like GM, US Steel, and AT&T, and see oligopoly as a major factor facilitating a more union-friendly labor relations regime. Others believe that concentration, by reducing the number of decision-makers that unions need to contend with, simplifies the task of organizing.
However, the evidence in recent decades points to concentration as harmful for workers and their unions. Virtually no mega-corporations have unionized since 1950, and rising industrial consolidation since the 1980s has been destructive to working conditions and union power alike. A look at the historical record shows that unions have had success and failures in both unconcentrated and concentrated industries, depending on institutional context, and that postwar union power in a handful of heavy industries rested on a specific set of institutional supports that have long since faded away. Given the structure of contemporary capitalism, in which large firms operate many dispersed facilities and contract out key functions to economically subordinate franchisees and contractors, trade unionism and antimonopolism are in fact complementary rather than opposing movements. Both can serve as important components of a progressive coalition to take power from corporations and democratize the economy.
The Golden Age of Oligopoly?
Recent evidence cuts against the notion that market concentration benefits workers. While it’s true that workers at big companies once received a substantial “large firm wage premium” over workers employed at smaller firms, the size of that premium has plunged by half since the 1980s, and it’s still falling. In addition, at least in local labor markets, rigorous empirical evidence shows that employer concentration holds wages down. Indeed, industries that have concentrated in recent decades have not followed the relatively egalitarian, but historically unique, paths of mid-century industrial corporations. Most notoriously, meatpacking companies in the 1980s pursued relentless consolidation hand in hand with brutal attacks on unions and workers. Since the 1980s, concentration in meatpacking has gone way up — four companies now control eighty percent of the market — while wages have gone down. In 1982, at the beginning of the industry’s consolidation wave, the United Food and Commercial Workers’ base hourly wage for meatpackers was $10.69, or $29.14 in inflation-adjusted dollars. The average hourly wage in the industry today is $14.23.
The ability of industrial workers to earn decent wages in concentrated industries in the middle of the last century was due to the unique institutional arrangements of the U.S. economy at that time, brought about in large part through the power of the U.S. labor movement. For a time, some workers did benefit from oligopoly. For the minority of workers employed in “basic industries” like auto manufacturing, concentration helped to facilitate a rough simulation of what is known as sectoral bargaining. Under sectoral bargaining, a single union contract covers all workers in an industry, taking wages out of competition and setting a floor in the labor market. Unlike in many European countries, U.S. labor law does not mandate sectoral bargaining, instead confining union bargaining rights to individual establishments or businesses. However, oligopolistic market structures allowed U.S. unions to simulate European-style sectoral bargaining through what was known as “pattern bargaining.” In pattern bargaining in heavily unionized industries, unions would first negotiate a contract with one firm in their industry’s oligopoly, and then extend the “pattern” to the remaining firms in the industry. As an approximation of European-style sectoral bargaining, it worked well enough for the thirty-or-so years that corporations honored the ritual. When corporations saw an opportunity to abandon the pattern in the 1980s, they were able to overcome labor resistance and do so, with the help of a newly repressive state. Without employer acquiescence to pattern bargaining, it is unclear what benefits concentration can offer to workers.
Aside from its lack of staying power, the postwar arrangement had a few additional shortcomings. First, it excluded the vast majority of workers, who were employed outside of basic industries in light manufacturing, services, and other sectors. Those excluded were disproportionately women and not white. Second, the arrangement was unstable. Labor’s forward progress was all but halted by an employer counter-offensive culminating in the Taft-Hartley Act, a 1947 law that outlawed secondary boycotts, legalized “right to work” laws, and prohibited foremen and independent contractors from unionizing. Meanwhile, almost immediately after agreeing to their first union contracts, big corporations started looking for ways to undermine and sidestep unions by outsourcing more and more work to temp agencies, independent contractors, franchisees, and spun-off suppliers. Giant corporations maintained their market shares and oligopoly power in final product markets, but they hired fewer and fewer workers directly, outsourcing labor to subordinate third-party contractors. For example, in the 1960s, auto companies began outsourcing parts manufacturing to captive non-union companies that they could dominate through their enormous buying (“monopsony”) power. Far from being content with oligopoly pattern bargaining, the United Auto Workers instead launched an antitrust campaign against the auto companies, alleging that their monopsony power over outsourced suppliers depressed wages. Supply chain monopsony of this kind has only gotten worse.
The Path to Pattern Bargaining
Pattern bargaining was of course never the final goal for unions, which aspired to the more ambitious goal of broad industrial democracy; their fall back to pattern bargaining represented a compromise in the face of corporate intransigence rather than a triumph. The post-1880s rise of large firms coincided in many industries, most famously in steel, with increasing concentration and the destruction of craft unions. While craft unions persisted in pockets where workers managed to defend their occupational privileges, pre-New Deal attempts to re-establish worker power by unionizing industrial workers in economically-concentrated mass production industries failed, almost uniformly so, in the face of overwhelming employer power. (The classic examples from the period include the failed uprisings at Homestead and Pullman, as well as the 1919 steel strike.)
The few union successes among industrial workers prior to the passage of the National Labor Relations Act came instead in unconcentrated industries, particularly garment manufacturing, coal mining, and trucking. During the Progressive Era, when unions largely operated outside the protection of the law, the tenacious United Mine Workers of America (UMWA), Amalgamated Clothing Workers of America (ACWA), and International Lady Garment Workers (ILGWU) unions were remarkably successful in organizing a high proportion of industrial workers in their industries in major geographical markets. Beginning in the early 1930s, Midwest and West Coast locals of the International Brotherhood of Teamsters also succeeded in taming the atomistic, unconcentrated industries of trucking and warehousing through a “leapfrog” organizing strategy that focused on secondary boycotts and isolating anti-union employers.
Of course, the mining and garment unions owed much of their success to the degree of solidarity inculcated in the tight-knit work and community lives of, respectively, isolated coal regions and dense ethnic urban enclaves. Meanwhile, the Teamsters benefitted from the strategic importance of trucking to all other industries. However, the decentralization of employers proved to have two key advantages for each of these unions. First, unions could play employers off each other, isolating and punishing anti-union bosses with selective strikes and boycotts. Second, decentralized markets allowed unions to play a governance role, positioning themselves as the stabilizing and regulating force in their industries. Employers found this almost as desirable as workers did. In markets too decentralized to support informal pattern bargaining, these unions instead negotiated formal “master agreements,” miniature sectoral bargaining schemes, that set a wage floor for all employers in a geographic market. In addition, by pooling funds across employers, they established multi-employer benefit funds, which financed union-owned housing projects, financial institutions, health centers, and even vacation resorts for their members. Most famously, the garment union multi-employer benefit funds built prototypical welfare states decades before the New Deal, casting doubt on Bruenig’s claim that only large firms can afford generous union benefits. Stable, profitable firms that can afford benefits need not be corporate giants.
The road that ultimately culminated in the pattern bargaining of the 1960s began in the 1930s, with the creation of the Committees for Industrial Organization (CIO). The leaders of the CIO, in particular Sidney Hillman of the ACWA, believed that the giant corporations in the basic industries of steel, auto, and rubber manufacturing were simply too powerful for unions to face on their own. Hillman believed that organizing them would only come with the help of substantial government support. The United Auto Workers (UAW)’s sit-down strike against General Motors at Flint in 1936, the catalyst for the CIO’s meteoric rise, is perhaps the most famous example of the vulnerability of monopolistic corporations to worker direct action at a single “choke point.” In fact, the UAW’s strategists timed the strike for after the election of liberal Governor Frank Murphy. Just as the union hoped, Murphy refused to use the National Guard to protect GM’s private property and evict strikers, leaving GM helpless in its efforts to restart production. Murphy’s action, combined with the recent passage of the National Labor Relations Act protecting union rights, proved pivotal in winning the strike and giving the CIO its first foothold in a concentrated, basic industry. The CIO’s existence remained tenuous at best, however, until rescued by World War II. With war mobilization came the heavy hand of the state, for once on the side of labor, which forced intransigent corporations to recognize CIO unions. Other monopolistic industries like telecoms succumbed to unionization just after the war, at the peak of U.S. labor militancy. Unions were thus eventually able to unionize monopolists, but only with a level of wartime-dependent state support not seen before or since. When new mega-corporations like Walmart, FedEx, and Intel, arose in the second half of the 20th century, unions were unable to rely on state support, and failed to organize or establish pattern bargaining relationships with the new corporate giants.
What now?
The claim that it is easier, under present conditions, for unions to organize in concentrated markets with only a few companies appears plausible at first glance. As Bruenig points out, the default under American labor law is that unions must win majority elections at individual facilities before they can legally represent workers and negotiate contracts. Organizing every restaurant in Manhattan alone would take centuries under this process. Since concentrated markets reduce the number of companies unions must fight, shouldn’t concentration make organizing easier?
Unfortunately, the problems establishment-level bargaining creates for unions in unconcentrated markets are in some ways even more extreme in concentrated markets. Large corporations do not necessarily have large establishments. As socialists like Bruenig will be aware, Marx himself distinguished between the concentration of capital (the growing size of capitalist establishments, due to economies of scale) and the centralization of capital (the concentration of ownership in a few large firms). When antimonopolists propose breaking up large firms, they don’t mean tearing down large plants and setting up smaller ones in their place. They mean breaking up the ownership of those plants.
Since monopolistic firms encompass many separate establishments, the establishment-level bargaining problem reappears in concentrated industries, only against a stronger and better-resourced opponent. As labor lawyer Brandon Magner points out, monopolistic corporations typically employ workers through dispersed networks of individual plants and facilities, allowing large firms to easily shift production among their plants to avoid unions or punish militant union locals. Smaller firms lack this ability. This means that unlike the Flint sit-down strike, there is no strategic Fisher Body Plant #1 at Amazon that a union can target to bring the company to the table. Amazon can simply route traffic to other warehouses. The choice boils down to: would we rather fight a bunch of little monsters, or a massive many-headed hydra?
I got my start in the labor movement helping to organize industrial laundries, with the union now known as Workers United. In my experience, the smallest and the biggest firms each posed the toughest challenges, for different reasons. In an industry like industrial laundries with low barriers to entry, organizing could feel like playing whack-a-mole: undercapitalized, inefficient, low-wage sweatshops could and did pop up all the time, threatening union standards. However, giant corporations were no better for the union or for workers. National multiplant corporations, armed with much deeper pockets, could easily absorb anything we threw at them and launched costly anti-union campaigns. In contrast, small and midsized regional firms had limited resource advantages over the union, and we successfully organized more than a few of them. The midsize companies were the best employers: big enough to reach efficient size and afford decent wages, but not so big that they accumulated the immense power of the megacorporations to crush unions.
Trade Unionism and Antimonopolism: Complementary Movements
Antimonopolism and trade unionism share the same goal — fighting corporate power — and work best as complements rather than substitutes. Notably, both movements peaked between 1945 and 1975 and declined after 1980. They also share a common nemesis: neoliberalism. Antitrust enforcement was extraordinarily vigorous throughout the postwar years of high union density and robust wage growth. In fact, the era of high steel worker wages was far from an era of tolerance for market power: a 1960 merger between two California grocery chains was blocked even though the companies controlled only a 7.5 percent market share. This was no accident, as the architects of Roosevelt’s Second New Deal, which began in 1935, saw antitrust enforcement and collective bargaining as complementary policies in generating a high-wage, high-growth economy. Believing that giant corporations had sabotaged the economic recovery by charging prices that were too high while paying wages that were too low (an analysis shared by Marxist and heterodox economists like Paul Baran, Paul Sweezy, and Josef Steindl), they launched the greatest wave of trust-busting in U.S. history simultaneously with the implementation of the National Labor Relations Act that legalized unions.
The policies advocated by each movement today are complementary as well. Bruenig is right that the worst employers are often small, undercapitalized sweatshops. But the answer to this problem doesn’t lie in monopolies. Passing a $15 minimum wage and reforming the National Labor Relations Act to make it drastically easier to unionize would make it much more difficult for inefficient businesses — big or small — to utilize low wages as a competitive strategy. High wages and unions would penalize companies below minimum efficient scale but without encouraging them to amass more dangerous amounts of power. Meanwhile, antitrust policy can protect small suppliers from monopsonistic predation by massive buyers like Walmart or Amazon, allowing them to raise wages for their own workers. Antitrust can perform a similar function for franchisees dominated by powerful fast-food brands. As antitrust advocate Zephyr Teachout puts it, “we should make it easier to organize people, and harder to organize capital.” We need both movements as part of a progressive coalition if we are to democratize our economy and protect it from corporate power.
