In the 1930s and ’40s, the Congress of Industrial Organizations unionized American workers with an energy never seen before. But its peak years were short-lived, and the labor movement has struggled to reach such heights again.
John L. Lewis, leader of the Committee for Industrial Organization, addressing 10,000 textile workers in Lawrence, Kansas, on May 24, 1937. (Bettmann / Getty Images)
When I interviewed Professor Melvyn Dubofsky, I had prepared a long list of questions that I was hoping to get through. The first one, which I asked of all my podcast guests, was simply, “What was the CIO, and what is its primary historical significance?” He proceeded to talk uninterrupted for the next hour and twenty minutes, and what follows is a only lightly edited transcription of his answer.
—Benjamin Y. Fong
What was the CIO? It’s not easy or simple. Essentially it was an attempt or effort to bring unionization to areas of the economy where in the past unions had a history of failure. Before the formation of CIO, unions had failed to organize any of the sectors of mass production in the US and, from the final decades of the nineteenth century into the 1960s, mass production was the dominant sector of the US economy. Automobiles, electronics, and steelmaking employed enormous numbers of workers, the vast majority of whom did what was labeled non-skilled or semiskilled labor. But all of the mass production sectors also had a substantial minority of highly skilled workers. Without their skills, production could not succeed.
Before CIO came on the scene in the second half of the 1930s, most union members had their power in what could be called the labor market. Their power came from their skills. Without their skills in construction in other sectors of the economy, production could not function. Power came from their position in the labor market. In the mass production sector, the vast majority of employees had no labor market power. They were all readily replaceable. Their power came not in the labor market but at the point of production.
How you organize people is related to the power they bring to the situation. In the mass production sector, power basically belongs to the employers except for the minority of highly skilled workers. So how do you organize workers without power in the labor market?
The 1930s was the first time in a peacetime economy that there was a chance to organize such workers. The first part of the 1930s was the depths of the Depression. By 1935 or 1936, owing to a lot of New Deal initiatives, the economy was improving, employment was increasing. You had an administration in power in Washington, and in the states in which mass production was most heavily concentrated, that was more favorable to organized labor than ever before. The primary labor organization at that time, the American Federation of Labor, or the AFL, was dominated by unions of more skilled workers outside of the mass production sector in which their power came from their skills, the labor market.
The real clash was not between skilled and non-skilled. The clash was over whether it would pay off to unionize the mass production workers.
Given how hostile most American employers were to unions, the people in control in the AFL were very dubious about the ability to unionize mass production workers. They simply believed that workers without labor market power, workers who were easily replaceable, would be extremely difficult to organize. They thought they would be wasting their union resources, using money to do something that would fail.
There were within the AFL a couple of unions that had been very successful in reorganizing sectors of the economy, most particularly mining and clothing trades. Three notable unions were the United Mine Workers of America [UMW], the Amalgamated Clothing Workers of America (which was actually outside the AFL), and the International Ladies’ Garment Workers’ Union. The leaders of those three unions — most especially John L. Lewis, the president of the United Mine Workers — realized that this was the time to act. This was the time in which mass production workers could be unionized. Now the old story was that it was a split between the craft unions and the minority of unions that had less skilled workers. The real clash was not between skilled and non-skilled. The clash was over whether it would pay off to unionize the mass production workers. Would it be worth the effort?
Lewis, most especially, and the presidents of the two clothing workers’ unions — Sidney Hillman of the Amalgamated Clothing Workers and David Dubinsky of the International Ladies’ Garment Workers — all had experienced great success in unionizing workers in their industries in 1933 and 1934. They believed that what they had done the AFL could do in unionizing workers in mass production.
Their belief was that the way to organize workers in mass production was to act, to not worry about jurisdiction by particular skills. You could worry about jurisdiction after the workers were unionized. But the idea was to unionize the workers. Basically, by the end of 1934 and early 1935, they had come to the conclusion that they would try to urge the AFL to act to support and subsidize an organizing campaign in the mass production sectors. At the AFL convention in 1935, they attempted to make the unionization of mass production workers a policy that would be adopted. But within the AFL, the old guard resisted.
The construction trades and the machine trades (which had skilled workers) worried about their jurisdiction. If you organized the mass production sector en masse, skilled workers would not belong to the existing unions of their skill. The advocates of mass production said the point is to act, to organize, and only then worry about jurisdictional lines after success or victories. But the AFL in 1935 refused to act.
Then there’s the famous dramatic event at the convention in which Lewis, president of the UMW, clashes with Bill Hutcheson of the Carpenters’ Union and punches him. It was sort of a dramatization of what was about to happen. The CIO people decided to form a committee within the AFL, the Committee for Industrial Organization, without AFL authorization. They are going to go on on their own and try to organize the mass production sectors. The bulk of the organizers and the money are going to come from the United Mine Workers.
Lewis basically has decided that now is the time to act, that they can achieve success. He has support from a number of unions within the AFL beyond the clothing workers’ unions, and they form the Committee for Industrial Organization within the AFL. What happens next is that the AFL considers the CIO a subversive organization. The AFL orders them to follow decisions adopted by the AFL convention. The committee — Lewis, Hillman, Dubinsky, their allies — refuse. They say, now is the time to act. They’re even more convinced after the 1936 election in which the Democrats and the New Deal sweep the nation and have absolute majorities in both Houses of Congress, and a commitment by the [Franklin D.] Roosevelt administration essentially to support organized labor.
It’s immediately after the 1936 election that you have the famous sit-down strike in Flint, Michigan, which ties up the automobile sector, or at least General Motors. General Motors is basically shut down because in mass production, if you paralyze an essential sector of the industry or the firm, it’s unable to produce the final product. So this is where power at the point of production remains.
What makes the CIO’s initial victory possible is that the governor of Michigan, a New Deal Democrat, Frank Murphy, and President Roosevelt refuse to use either federal power or state power to undermine the strike. In fact, what the president and the governor do is pressure General Motors to negotiate with the union, meaning to some extent Lewis, and to recognize the United Automobile Workers as the representative of General Motors employees.
The strike goes from New Year’s Eve right into mid-February 1937 before the two parties come to an agreement. But in the aftermath of the agreement between General Motors and CIO, the first time a mass production employer concedes to a union organizing committee, suddenly there’s an upsurge throughout the entire mass production sector. Chrysler soon concedes. The rubber industry is organized. Back then, it was concentrated in Akron, [Ohio,] and Akron becomes a union town. The electrical appliance sector, General Electric and Westinghouse, begin to recognize unions. Only three weeks after General Motors conceded to a CIO affiliate, US Steel did the same in March, recognizing the Steel Workers Organizing Committee as the representative of its employees. So it was the enormous upsurge in organizing.
In the background, the AFL skilled unions are also having great success, but because they’re unionizing much, much smaller sectors of the economy, it’s not as notable. But there is a nationwide upsurge in unionization in the first half of 1937. The AFL and the CIO are still literally at war with each other because the AFL has decided now, gee whiz, the mass production sector is actually organizable, so we’re going to try to unionize those workers also. And so you have ongoing conflicts between CIO affiliates and AFL affiliates to organize the same workers.
Now, for a whole host of reasons, the CIO seems far more radical than the AFL. Why? Well, the mass production sector, long ignored by the skilled trades, had a history of being linked to left-wing influences. The two clothing unions, the Amalgamated Clothing Workers and the Ladies’ Garment Workers, had both been led by socialists from early on. After World War I and the Bolshevik Revolution in Russia, both those unions had significant communist minorities. The United Mine Workers always had a substantial socialist faction within the union, which at times had been dominant. Lewis, of course, was anything but a leftist, but Lewis was a believer in power, and the manipulation of power, and would use anybody who would help him obtain power. And so, the organizers that Lewis used in the CIO attempts to unionize mass production came heavily from the Left, socialist and communist.
It’s not until 1938 that the CIO holds a convention and becomes a separate organization, autonomous from the AFL. Until ’38, the CIO is still acting as a committee within the AFL, a committee that is considered schismatic by the AFL. In the first half of 1937, there’s a sense that labor is on the rise. It’s going to reorder the American economy, and the CIO is the spearhead of that movement to completely change what could be called power relations in the US economy; between employers or capital on one hand, and workers and labor on the other. Or at least, that’s how it looks going into the spring of 1937. That’s how the mainstream media perceived it. They thought that revolution might be on the way, and it upset them, and so mainstream media tended to emphasize the turbulence and violence of the organizing. It scared them.
The mass production sector, long ignored by the skilled trades, had a history of being linked to left-wing influences.
Politically, within the Roosevelt administration and among members of Congress, you can see emerging fear about the labor uprising getting out of hand, threatening the structure and stability of the nation. That’s how CIO looks in the early spring of 1937, a radical, threatening labor organization. But things change quickly. The Roosevelt administration, thinking that it had succeeded in ending the Depression, said it was time to moderate. Roosevelt, when he first ran for the presidency in 1932, had emphasized that his policies would lead to a balanced budget. But of course, to combat depression, the New Deal engaged in what at that time was aggressive deficit financing. Borrowed money to reinflate and stimulate the economy. Well, Roosevelt and his advisors thought they had achieved success, so now was the time to economize. To move toward budget balance, the New Deal held back financially.
What was the result? In the latter half of 1937, the economy began to contract again. By 1938, it was entering what was soon called the Roosevelt Depression. What happened to the mass production sector? Demand for their products declined. What happens when you’re producing less? You need fewer workers, so you begin to lay off workers. What happens when the labor market loosens, when employers have less reason or motivation to deal with worker demands? They become much tougher in dealing with the unions that they’ve initially recognized. It becomes much harder to get a second contract. Now, firms are demanding givebacks. The sector of steel outside of US Steel, the so-called little steel companies (which are not little by any reasonable definition), refuse to recognize the Steel Workers Organizing Committee. The steelworkers respond by striking in the spring of ’37.
This time, when the CIO appeals to the Roosevelt administration, and to the governors in Pennsylvania, Ohio, Illinois, and Indiana, where the steel industry was then concentrated, they find they no longer have warm allies or friends. Part of the reason is the labor upsurge in the first part of 1937 had frightened enormous numbers of Americans. Politicians and office holders began to receive letters from ordinary citizens. Worrying about labor taking over the country, even New Deal Democrats begin to draw back from their support of labor. What happens? The little steel companies defeat the strike. Ford has no difficulty holding out against the United Automobile Workers. In a loosened labor market, it’s very difficult to win anything. The companies become much, much harder in their bargaining on top of that.
The CIO, especially the United Mine Workers, had expended an enormous amount of money; one, in supporting the Democratic Party in the 1936 election, and two, in organizing the mass production sector. They had a lot of new union members, but they did not have all that many dues-paying union members. Dues had to be collected individually from each member. It was not easy. Then as the economy tightened in late ’37 and ’38, union members were even more hesitant to pay dues. So as the treasuries of the existing CIO unions shrank, the CIO cut back. It laid off organizers. It had fewer paid officials who helped manage the new unions, and suddenly, the CIO and labor no longer seemed a rising power. The best historian of the CIO, Robert Zieger, said the CIO was, even at its peak, a fragile juggernaut. It looked more powerful than it actually was.
By the end of 1938, you have all kinds of forces working against CIO. In Congress, you now have, instead of a dominant New Deal majority, you have an alliance between Republicans and Southern Democrats. They’re investigating the CIO, the National Labor Relations Board, and accusing both of being communist. You have what amounts to an anti-CIO hysteria, and the AFL plays its part in encouraging that. The AFL was far less subject to losing members and power in the so-called Roosevelt Depression. Again, unlike the CIO, whose strength was concentrated in particular sectors of the country, the AFL had a presence everywhere because trucking was nationwide. Every state had local teamsters or truckers. Construction trades existed everywhere. The AFL had members and influence in every state, in almost every middle-sized to large city.
You have what amounts to an anti-CIO hysteria, and the AFL plays its part in encouraging that.
By contrast, the CIO was concentrated in the Northeast and the Midwest. It had no members to speak of in huge areas of the country, and so by the end of ’38, ’39, the CIO seemed to be a spent force — still having failed to organize Ford, or the so-called little steel companies, or many, many other firms in the mass production sector.
What caused the change? War. War in Europe ended the Roosevelt Depression. The US became the arsenal of democracy. The economy started to boom. Before long, the US would have conscription. What happens to the labor market? It tightens. Once again, workers have power, so employers in the mass production sector change their tune. The union organizes Ford, it organizes the little steel companies. It becomes dominant in the mass production sectors.
Now, there were firms that succeed in remaining outside of union recognition, but they do so basically by offering their employees everything unions win for their members, and sometimes more. Firms like Eastman Kodak, DuPont, and many others provide what could be called their own welfare systems. They provide health care, pensions, and all kinds of activities for their employees. They create what amount to company unions that aren’t challenged by their workers. They create little realms of their own.
Sanford Jacoby has written a wonderful book on how these firms remain nonunion and keep the sympathy and support of their employees, and the war is fundamental to everything. The same thing happened in World War I, but US participation in World War I lasted less than two years. As soon as World War I ended, employers said, “We go back to the preexisting system.” World War II lasted for almost five years, and during those five years, the federal government, through the [National] War Labor Board, made employers and their enterprises deal on a regular basis with unionized labor.
What happened was that five years of regular dealing with unions normalized the union presence in the mass production sector. After World War II, unlike World War I, employers and enterprises fought not to get rid of unions, but to place limits on union power. What you have for the next three to four decades is what can be called a Cold War in the mass production sector between enterprises and their unions. Now, the CIO comes out of the war with a substantial membership, with a presence almost everywhere in the mass production economy. But again, it remains smaller than the AFL and is not as present in the entire nation.
Well, Lewis, who’s a strange character, back in 1940, had broken with his associates in the CIO over the 1940 election. Lewis throws his weight behind Wendell Willkie and the Republican Party. Basically, Lewis says on the eve of the election to CIO members, “You have a choice. Vote for Willkie or for me.” Of course, almost without exception, CIO members vote for Roosevelt and the Democratic Party. Lewis, in the aftermath, marches the United Mine Workers out of CIO. He’s not welcome in the AFL either. During the war, the UMW and Lewis are on their own, and that’s where they are after the war, although Lewis is now playing footsie with the AFL and will go back into the AFL. In the meantime, the CIO is now led basically by Sidney Hillman, who had been Roosevelt’s wartime labor advisor. Phillip Murray, once Lewis’s lieutenant in the Mine Workers, becomes the president of the CIO.
The CIO at the war’s end has a substantial minority of unions that are led by communists or reds. The Transport Workers Union, the National Maritime Union, the United Electrical Workers, the International Longshoremen’s and Warehousemen’s Union — the four big unions that can be said to be led by reds. After the war there’s the conservative reaction. Republicans take over power in Congress after the 1946 election. That’s how you get the Taft-Hartley law.
And then in the 1948 election, you have a third party, whose candidate is Henry Wallace, who had been Roosevelt’s vice president for two terms. Wallace leads the Progressive Party. The Progressive Party gains the support of the CIO’s red unions. The CIO’s leadership is upset not because its red unions seem favorable to the Soviet Union and are outside of the Cold War consensus, but because they see Wallace and the Progressive Party as a threat to New Deal dominance. They fear that the Progressive Party will take support away from Harry S. Truman in 1948 and throw the presidential election to the Republicans.
The CIO at the war’s end has a substantial minority of unions that are led by communists or reds.
The CIO in ’48 had adopted as an organizational policy and principle support for the election of Truman and the Democrats. Its red unions had refused to follow policy, and that is really why the CIO acts against them. If they had supported Truman and the Democratic Party, I suspect at least in 1948 that the CIO leadership would not have acted to expel them. But then again, two years later, when the Korean War broke out, if the red unions had remained sympathetic to the Soviet Union, I suspect they would’ve been suspended and expelled. But by then, “Red Mike” Quill, the leader of the Transport Workers Union, had become, let’s say, non-Red Mike Quill.
But in any event, the CIO essentially had done as much as it was ever going to do. It was not in the position by ’48 or ’49 to conquer any new frontiers. It had enough on its hands managing relations with the automobile companies, the steel companies, the rubber companies, and the electrical appliance companies. In terms of total membership, the old AFL unions were dominant. Again, they had many more members, and they had influence, strength almost everywhere in the country. I would think the final act in the CIO’s story was its Operation Dixie campaign, in which it tried to unionize the South and particularly the Southern textile industry.
The South was the big problem, not only because Southern-based industries fought unions, but also in the South elections were controlled by a minority of voters. The Southern congressional delegations in the House and the Senate that were Democratic in name were Republican in sympathy and action, and so Congress was dominated by a Republican and southern Democratic coalition. There were some Southern Democrats that were outside that coalition, but they came primarily from the coal mining areas in Alabama, Kentucky, Tennessee, and Western Virginia. Where the Mine Workers had a presence, they elected not hardcore segregationist Democrats, but Democrats who were willing to work with the National Party more than with Republicans.
That’s why in Kentucky, Tennessee, and Alabama, three states where the United Mine Workers were especially represented, their representatives and senators were more linked to the National Democratic Party. But the South as a whole was the area that could break union influence and power. When Operation Dixie failed — and it failed abysmally — the CIO was left helpless as an independent organization.
So the CIO said, well, it’s really time now. Once again, they had played around in the late ’30s with coming to an agreement with the AFL, but now was the time to put labor together again nationally. Instead of having labor divided, now we need labor united. They go back into the AFL. The AFL gives them a separate industrial union sector, but they are a minority. They have fewer members. They remain a distinct factor, an influence in the AFL-CIO, but a minority factor, a minority influence. They’re not going to make policy.
The CIO people still have a different position than the dominant wing of the AFL has in foreign policy and domestic policy on a whole raft of issues. Now, the CIO unions participate in the Cold War, but the CIO faction, unlike the AFL, tries to maintain relations with European unions outside the Cold War consensus. At home, the Industrial Union Department inside the AFL from early on is the supporter of what becomes the civil rights movement and the public social welfare state, as distinct from what could be called the union private welfare state. The Industrial Union Department, the old CIO unions, have built up with their employers their own private welfare state through collective bargaining, but they still support what could be called a public welfare state for everybody. The old AFL unions don’t.
So you have these differences, but after ’45 and ’46, labor and the CIO conquer no new frontiers. What they do for three or four decades is maintain their own stability within the nation and engage in what amounts to a domestic Cold War that rarely is going to turn hot. In other words, employers aren’t going to rake the unions. The unions are not going to directly challenge employer or management prerogatives.