Nate Holdren uses the historical record to refute the idea that union expansion and wage growth are only possible in times of economic prosperity.
In a recent editorial in the art magazine The Brooklyn Rail, philosopher Paul Mattick scoffed that the RWDSU unionization campaign at an Amazon facility in Alabama “seems to have been a case of the ‘labor movement’ at its purest, with the union promising no more than representation (and dues collection), with nary a specific word about working conditions or wages.”
Getting a clear handle on what happened in that campaign is important, but Mattick’s take doesn’t help anyone do so. His implication that unions just want dues is startling, as it sounds like every union-busting consultant I’ve ever heard. Given that to be a Marxist, one must be able to separate Marxism from its (many and very bloody!) worst moments, I found it odd for Mattick to imply that the labor movement is representable by its worst moments (assuming this is one). It’s also simply factually incorrect for Mattick to say that the campaign didn’t talk about pay and conditions: it did! The website discusses unionization as a way to raise pay, and it discusses issues including workplace safety and job security. Mattick is simply factually mistaken.
In a rhetorically skillful line, Mattick writes that “[t]he idea that Amazon’s victory was due to the efforts at intimidation the company undoubtedly practiced is ridiculous, given the history of successful union drives in the 1930s and ’40s in the teeth of armed (and shooting) police and murderous company goons.” This sounds really good, if you don’t think much about it. But look more closely. The claim amounts to saying “some people could do a thing under terrible circumstances, therefore the idea that merely bad circumstances make the thing harder to do must be false.” That’s specious nonsense. That said, it is reasonable to ask, “how is it that we have a labor movement today which loses under today’s circumstances, when we had a labor movement once upon a time that won under much worse circumstances?” That’s a worthy question, though hardly a new one. Unfortunately it’s not a question that Mattick’s article does much to help us to investigate. I don’t claim to have a fully worked out answer. My impulse is to say that it lies in the history of the changing set of laws for integrating unions into US capitalism, and how that integration in turn has shaped unions. Informative works about aspects of that history include Joe Burns’ Reviving the Strike, Charles Romney’s Rights Delayed, Staughton Lynd and Daniel Gross’ Labor Law for the Rank and Filer, and Robin Cartwright’s writings here at Organizing Work. (To be fair, E Jones’s article that Mattick quotes from is also worth reading, not least for its refusal of widespread left pieties about the CIO.)
Mattick derides the “wistful wish for the revival of labor unions” as a “fantasy solution” to today’s problems because “present conditions are very different from those of the great union drives of the past.” To support this assessment, he reaches back to the 1930s, writing that “results for the new Congress of Industrial Organizations unions (the American Federation of Labor unions were barely hanging on) were decidedly mixed until preparations for war began.” The takeaway point from the comparison between the CIO unions in the early 20th century and the present, for Mattick, seems to be that the context is different today from the past such that “there is little place for unions to insert themselves as brokers and profiteers of labor peace” nowadays, even if such room once did exist. The reason there is so little place is tied to “[t]he disappearance of the productivity growth that marked the post-war period [which] rules out the possibility of increased wages.” Our rulers are out of carrots, apparently, such that “the political class find themselves at a loss to elaborate policies other than (…) austerity.” The state cannot promote unionism, allegedly.
I think Mattick is wrong in multiple ways. He is wrong to imply unionization in the 1930s and ’40s could only really succeed because of the boost to production provided by “preparations for war,” as I will elaborate on below. That said, even if capitalism needed war to supply higher wages and make unionization possible, there still seems to be no shortage of potential for war in 2021. Climate change promises to produce more of it, and to produce the functional equivalent of war as well, by destroying a great deal and so creating opportunities for capitalists in the aftermath of that destruction.
In any case, I think Mattick is wrong that unionization only grew due to the economic opportunities provided by “preparations for war.” Mattick quotes some numbers about membership in CIO unions — 1,350,000 members in 1940 and 2,850,000 in 1941 — which come from Walter Galenson’s The CIO Challenge to the AFL (they’re from table 19 on page 587.) Galenson estimates the CIO had 1,580,000 members in 1937, its first year independent from the AFL. Its numbers fluctuated up and down, having fallen to 1,350,000 in 1940, as Mattick quotes. Exactly why those numbers fluctuated is unclear to me. I suspect there were many different causes. The CIO consisted of multiple unions in multiple industries and was engaged in a political fight with the AFL, and the late 1930s were a tumultuous time period in which the institutions of industrial relations were themselves in flux. Investigating what the CIO was and was not capable of and why, and what its politics were, as part of understanding the present analytically and programmatically — these are very worthy matters of inquiry in my view. I don’t think Mattick’s remarks aid that inquiry. If anything, the force of his editorial points in the opposite direction, toward treating these matters as largely settled rather than as worth addressing seriously.
Why should the total membership figures for the CIO be taken as representative of unionism as such, rather than, say, looking at specific unions? For example, the United Auto Workers had 195,000 members in 1937, falling the next year, then climbing again. In 1939, their numbers were 165,000, significantly less than two years prior, to be sure. That said, the US Census reports that there were not quite 400,000 waged workers in the automotive industry in 1939. Relative to the industry’s workforce as a whole, the UAW was massive. By that year, Daniel Nelson reports that 15% of waged workers at General Motors and 53% at Chrysler were UAW members (page 16). Given that the UAW was only founded in 1935, this growth in the auto industry hardly sounds like a situation where unions could not get a foothold.
Mattick is mistaken when he writes “the American Federation of Labor unions were barely hanging on” by the late 1930s or early ’40s. According to Bureau of Labor Statistics numbers (177) the AFL unions were at a historic low in membership as of 1933 — 2,127,000 members, the lowest since 1916. That same year, the unionization rate for the US as a whole was 5.2% (BLS, 178). Galenson’s numbers that Mattick mentions show the AFL having rebounded to 3,422,000 members in 1936, including 800,000 members of CIO affiliated unions, as the break between the AFL and the CIO leading to CIO independence had not yet occurred. By 1940, according to Galenson, the AFL was at 4,247,000 members. This means in the second half of the 1930s the organization had grown significantly even after having lost almost a million members due to the departure of CIO-affiliated unions. That was a historic high in membership in AFL unions — hardly “barely hanging on.” In addition, membership in unions affiliated with neither the AFL nor the CIO grew from 600,000 members in 1938 to 1,072,000 in 1940. (Galenson, table 19.) Thus Mattick’s implication that unions could not really get a foothold before the war seems mistaken.
That said, it is true that unionization grew even more during and after World War Two. From 1940 to 1947 the national unionization rate increased from 17.7% to 23.9%. That’s a big increase, to be sure. But the rate of union growth was far larger in the years 1933-1939, when the unionization rate more than tripled, from 5.2% to 15.8%. (BLS, 177-178.) Daniel Nelson’s figures show the UAW having an 84% membership growth rate from 1935-1939, and a 22% rate from 1939-1946. Nelson’s figures show the Teamsters also having higher growth in the late 1930s than in the early 1940s (page 9). Thus from 1933-1939 the US economy had not yet benefited from “preparations for war,” yet some unions still grew. This means that to suggest there was a general stagnation or impasse for unionization in general followed by a war-facilitated boom in union membership is both historically inaccurate and not a good guide for assessing the prospects for unionization today.
Wages in manufacturing rose in those years as well. In 1932, when AFL membership was at a major low, wages in manufacturing in both so-called unskilled and skilled jobs fell to a multi-year low as well: $14.48 and $19.48 respectively, for men. These rates rose every year until 1940 by which time they had risen over 60 percent to $23.91 and $32.41, respectively. Of course, wages went up further from 1940 onward (BLS, 172). The point is that from 1933-1939, despite the Depression, before “preparation for war began” it was possible for wage rises to occur, and, in Mattick’s terms, “for unions to insert themselves.” (Those terms are unfortunate, again sounding like something out of a management consultant’s union-busting playbook.) It seems to me, then, that a comparison with the1930s and 1940s provides little to support the assertion that nowadays “the political class find themselves at a loss to elaborate policies other than (…) austerity” — at least not for any reasons of objective economic constraint. If anything, it seems to me that the 1930s and 1940s suggest a story of institutional malleability.
I find a similar account of institutional malleability in the face of capitalist class disorganization and working class mobilization in Marx’s account of the English Factory Acts, in Chapters 10 and 15 of Capital. Capitalism in mid-19th century England, Marx argued, was headed toward a catastrophic overwork of the working class with terrible human consequences. Workers’ mobilizations plus some far-seeing state personnel imposed new conditions on unwilling capitalists. To borrow Mattick’s words, “the money has to come from somewhere, and at this point some of it—if only to pay the interest on borrowing—has to come from the pockets of rich people. But rich people can hardly be expected to like this.” New regulation of capitalism resulted — the Factory Acts — which benefitted English capitalists in the long term, despite the recalcitrance of each individual capitalist in the face of those reforms. Rich people (manufacturers in particular, in this case) were forced by the state into a situation they didn’t want. And ultimately they benefited in the long term, as their class tends to from reforms. It seems to me that if we on the far left today rule out the possibility of such developments happening again in the present, we do so at our peril.
Overall, Mattick’s editorial strikes me as a just-so story exhibiting a view that Simon Clarke has criticized: “Frustration with the limitations of the organised labour movement,” Clarke writes, “has frequently led socialists to look to relatively more marginalised groups and strata” (5). Such efforts have repeatedly not worked out, Clarke argues, “unless they are integrated into a broader labour movement, the only secure base of which has proved to be the trade union organisation that develops out of the struggle over the terms and conditions of wage labour, which cannot by any means be reduced to organisation on the basis of the sectional interests of particular groups of wage labourers” (6). The labor movement, in Clarke’s view, and I largely agree, is “for all its faults (…) the only collective expression of the interests and aspirations of labour, in hundreds of different ways, at every level and in every part of the world” (14), though I would stress the particular importance of new organizing by independent and explicitly radical unions like the IWW and the affiliates of the IWA.
I suspect that the proliferation of just-so stories like Mattick’s on the far left is a product of the far left being relatively socially weightless, largely cut off from any real collective expression of people’s aspirations for better lives. This particular kind of just-so story is inadequate not least because it doesn’t bear critical thought and, as I tried to show above, has little evidence behind it. It also fails politically to live up to what Clarke argues is “a responsibility to supplement the intellectual resources of the labour movement, to help to broaden its understanding and its horizons” (14). There is no shortage of would-be myth makers of the soft-left who want to encourage us to mistake capitalism with a few more social protections for actual socialism, or mistake state managers of capital accumulation for representatives of working class humanity and aspirations to a new society. Some of those would-be myth makers are in and around the official labor movement, including some advocates of the PRO Act nowadays.
Generally speaking, that movement, especially in relation to the system of labor law, works to represent workers as a kind of economic citizens — that is, as a kind of capitalism-compatible collective subject. Mattick implies this representation is external to the class. I would argue that it tends to arise from the class’s own vision and needs. Unions aren’t some con job misdirecting angelically anti-systemic workers. Unions are a vehicle for workers to express the interests that all subjects of capitalism have in capitalism’s continuation and in jockeying for position within the system.
We also have other interests in capitalism’s overthrow by revolution: workers’ interests are contradictory. Revolutionary working class politics is less an expression of a coherent set of pre-given interests than it is a process of collective action that selects from our contradictory interests — or, better, constructs new interests. Collective action short of revolution is a complex mix of these interests and processes of interest-construction, which is part of why I agree with Simon Clarke about why the far left should prioritize the labor movement, in order to be able to shape those processes for the better.
Just-so stories like Mattick’s, of capitalism in decline placing objective limitations on the ruling class’s abilities to do anything but make cuts and bust heads suggests we in the far left don’t really need to be able to respond to any of those misleaders and myth makers. Since those social protections and the victories of those capitalist politicians are impossible anyway, the story goes, no need to be able to criticize their shortcomings. In 1865, Marx wrote that the labor movement should reject “the conservative motto: A fair day’s wage for a fair day’s work!’” and instead “ inscribe on their banner the revolutionary watchword: ‘Abolition of the wages system!’” Much of Marx’s writing was dedicated to arguing why a fair day’s wage is still a terrible exchange. Mattick’s just-so story, on the other hand, implies that a fair day’s wage is impossible anyway, so no need to argue against it.
It seems to me that rather than assert the impossibility of events like the growth of the AFL and the CIO unions in the Depression, the far left should be able to articulate the limitations of those formations historically and the limits of their analogs today. (We should also be able to do so without using factually and politically questionable cliches about unions as third parties interested only in dues.) Working-class struggle tends to generate those analogs, so as class struggle heats up, dealing with those analogs will become more pressing. As Anton Pannekoek wrote, there is a “fundamental conflict between the self-emancipation of the working class through its own power and the pacifying of the revolution through a new sympathetic ruling clique.” Any step that workers take toward self-emancipation brings about not only a reaction from the heights of the ruling class, but also a rejuvenation of the forces of pacification. Just-so stories of capitalist decadence, secular stagnation, and so forth lead the far left to be unable to politically respond to (but, I suspect, fully able to socialize collegially among the reservoirs of) those sympathetic would-be ruling cliques. Should “the political class” begin to pursue approaches to social control and reproduction of capitalist social relations not predicated on austerity, the far left must be able to respond with more than just declaring those responses impossible.
Nate Holdren lives in Iowa. He is the author of Injury Impoverished and an infrequently updated blog.