Imagine America without unions. This shouldn’t be hard. In much of America unions have already disappeared. In the rest of America they’re battling for their lives.

Unions have been declining for decades. In the early 1950s, one out of three American workers belonged to them, four out of ten in the private sector. Today, only 11.8 percent of American workers are union members; in the private sector, just 6.9 percent. The vanishing act varies by region—in the South, it’s almost total—but proceeds relentlessly everywhere. Since 1983, the number of states in which at least 10 percent of private-sector workers have union contracts has shrunk from 42 to 8.

Following the 2010 elections, a number of newly elected Republican governors and legislatures in the industrial Midwest, long a union stronghold, moved to reduce labor’s numbers to the trace-element levels that exist in the South. A cold political logic spurred their attacks: Labor was the chief source of funding and volunteers for their Democratic opponents, and working-class whites, who still constitute a sizable share of the electorate in their states, were far more likely to vote Democratic if they belonged to a union. The fiscal crisis of the states provided the pretext for Republicans to try to take out their foremost adversaries, public-employee unions.

In Indiana, Governor Mitch Daniels signed a “right to work” law giving nonunion members who enjoyed the benefits of a union contract  the right to withhold dues to the union, making Indiana the first Midwestern state to pass such legislation. In Ohio, Governor John Kasich signed a bill repealing collective-bargaining rights for all public employees, but voters overturned that law at the polls. In Wisconsin, which had been the first state to extend those rights to public-sector workers, Governor Scott Walker also repealed those rights, but more selectively than Kasich: He kept them for police and firefighters. When outraged unionists and their allies mounted a recall campaign against him, Walker beat them back handily. In the nation’s capital, Republican senators and congressmen refused to confirm President Barack Obama’s appointees to the National Labor Relations Board, which adjudicates labor-management relations in the private sector.

Coming on the heels of the failure of the Democratic Congress of 2009–2010 to amend the National Labor Relations Act so that private--sector workers wouldn’t risk their jobs by forming a union, the Midwestern setbacks struck a growing number of commentators as labor’s death knell. Losing jobs as technology transformed workplaces, losing both jobs and middle-class wages as globalization transformed the economy, and blocked by statute and employer opposition from expanding—unions, some concluded, were history.

Within the labor movement, a number of leaders and activists quietly shared the same pessimism. They had invested in organizing with little to show for it. They had invested in politics but found that the Democrats they’d helped elect could not—or worse, would not—come to their aid. In 2008, they had seen the entire edifice of deregulated capitalism totter and almost collapse, plunging the nation into its deepest and most intractable recession since the 1930s. But unlike the ’30s, when workers flocked to unions, the current recession has only intensified labor’s downward spiral and business’s ascent. “What would it take for labor to come back?” one senior union staffer asked earlier this year. “This was the crisis we were waiting for, and it didn’t do it.”

For many Americans, the death of labor would doubtless seem the natural order of things, the dinosaur finally shuffling off to the graveyard. Unions have no presence in the hottest and hippest sectors of the economy, in high-tech, fashion, and finance. The public’s image of labor is a memory of a memory that’s anywhere from 50 to 100 years old—the Yiddish- and Italian--speaking seamstresses of the Lower East Side, the goons in On the Waterfront, and, for the historically sentient, George Meany puffing a cigar and damning the Vietnam peaceniks.

It doesn’t seem to matter that these images don’t conform to present reality. Today, there are millions more unionized teachers than unionized truckers. Of the six unions with more than a million members, two are headed by lesbians and one by an African American, a level of diversity in these troglodytic institutions not to be found on Wall Street or in Silicon Valley. A number of unions, particularly the Service Employees International Union (SEIU), play a central role in the political mobilization of Latinos, the group most likely to transform the American electorate. The AFL-CIO opposed the Iraq War and last year provisioned Occupy Wall Street.

But labor’s anachronistic image persists, and for a reason: It stubbornly represents blue-collar workers long after they’ve gone out of style and their numbers have diminished. It speaks for autoworkers and steelworkers, for the cutting-edge industries of 1935. To the young, even to most campus activists, unions are a holdover from their great-grandparents’ generation, speaking a language as incomprehensible as Old English: solidarity, shop stewards, seniority, strikes. Where are unions in the new economy? Can a union do anything for a temp? A part-timer? A software writer? A barista? Will anyone under 30—will anyone over 30—even notice if unions cease to be?

Perhaps not. But everyone will notice the consequences. Absent a substantial union movement, the American middle class will shrink. Absent a substantial union movement, the concentration of wealth will increase. Absent a substantial union movement, the corporate domination of government will grow.

That labor must take some of the blame for its troubles doesn’t let liberals off the hook. Time was when bolstering the power of labor within the economy—“the labor question,” as it was called in the Progressive Era—was central to the liberal project. But once the New Deal and the union upsurge of the 1930s and 1940s created the first middle-class majority in the history of the world, the labor question fell off the list of liberals’ concerns.

Liberals were right to privilege the struggles of African Americans, women, and gays. But over the past 40 years, labor grew weak while corporations grew stronger than ever before—so strong that their control of government now threatens most of the liberal agenda. Which is why we must turn again to the labor question, to the battle for economic power that is an inherent feature of capitalist democracy.

II. A Union-Free America

Here’s what happens if the dinosaur dies.

When unions vanish, ordinary Americans lose their right to bargain collectively for their pay and benefits. Even those who have never bargained collectively will feel the loss. Some years ago, when unions were big enough that their effect on the larger economy could be measured, Princeton economist Henry Farber concluded that the wages of nonunion workers in industries that were 25 percent unionized were 7.5 percent higher than they’d be if their industry were union-free. When unionized companies were common, firms that were nonunion had to mimic the wages and benefits of their unionized counterparts for fear that their employees would leave or, worse, organize. That was certainly the practice at General Electric and other largely nonunion giants.

Nonetheless, union workers generally maintained a 20 percent wage advantage over nonunion workers. The key to the wage advantage is the percentage of union membership in a given industry or market. In cities where nearly all the class-A hotels are unionized, as they are in New York and San Francisco, housekeepers make more than $20 an hour. In cities where roughly half of such hotels are unionized, such as Los Angeles, their hourly wage is about $15. In cities where all the hotels are nonunion, such as Phoenix, housekeepers make little more than the minimum wage, if that.

From 1947 through 1973, when union density in America was at its peak, real wages for nonmanagerial employees rose by 75 percent. From 1979 through 2006, as union density collapsed, real wages for nonmanagerial employees rose by only 4 percent. Unable to get a raise, American households maintained their standard of living during those years by women entering the workforce and by going into debt.

 Density is just one element of unions’ ability to raise wages, however. The other is strikes. We look back now at the three decades of broadly shared prosperity that followed World War II as a time of union-management concord, when executives made their peace with unions and unions didn’t rock the boat. In fact, more strikes occurred from the late 1940s through the early 1970s than before or since. When union contracts expired, workers and managers fought pitched battles over the terms of the next contract. The largest strike in American history came in 1959, amid the sleepy Eisenhower years, when 500,000 steelworkers stayed off the job for 116 days. It was through such expedients that workers compelled management to let them share in their company’s proceeds. But as density declined, unions’ ability to win strikes declined with it. By the late 1970s and early 1980s, unions were striking less to win raises than to resist management proposals to freeze wages and cut benefits. The weaker unions grew, the fewer their strikes. In the early 1950s, there were roughly 350 strikes in the United States every year. Over the past decade, there have been roughly 10 to 20 per year.

As unions shrank, inequality grew. From 1947 through 1972, productivity in the United States rose by 102 percent, and median household income rose by an identical 102 percent. In recent decades, as economists Robert Gordon and Ian Dew-Becker have shown, all productivity gains have accrued to the wealthiest 10 percent. In 1955, near the apogee of union strength, the wealthiest 10 percent received 33 percent of the nation’s personal income. In 2007, they received 50 percent.

Profits have been growing at wages’ expense. Michael Cembalest, J.P. Morgan’s chief investment officer, has calculated that reductions in wages and benefits were responsible for about 75 percent of the increase in corporate profits between 2000 and 2007.
Today, wages and benefits make up the lowest share of America’s gross domestic product since World War II. Wages have fallen from 53 percent of GDP in 1970 to 44 percent today. Profits have been growing at wages’ expense. Michael Cembalest, J.P. Morgan’s chief investment officer, has calculated that reductions in wages and benefits were responsible for about 75 percent of the increase in corporate profits between 2000 and 2007.

What’s causing this decline in workers’ ability to claim more of the nation’s wealth? It’s not that they’re less productive. According to a Wall Street Journal survey of the S&P 500, the nation’s largest publicly traded companies, revenues per worker, which were $378,000 in 2007, grew to $420,000 in 2010. Businesses now produce more with fewer employees, but even those workers who’ve kept their jobs haven’t seen their wages rise.***

Elite opinion insists that workers’ educational shortcomings are the root cause of declining incomes. In the years since 1979, however, the share of American workers with college degrees has increased from 19.7 percent to 34.3 percent, according to a new study by John Schmitt and Janelle Jones at the Center for Economic and Policy Research. Even so, the percentage of college graduates with good jobs, which Schmitt and Jones define as jobs paying at least $37,000 a year that provide health insurance and any kind of retirement plan, has fallen from 43 percent in 1979 to 41 percent in 2010.

So workers are better educated and more productive. What they lack is power. The current recovery is different from previous recoveries because few unions remain to press workers’ demands, and the unions that do remain lack the industrial or geographic density they need to win any gains or just hold the line for their members. This is most apparent in the Midwest manufacturing belt, where unions struggle to preserve at least some of the wage and benefit levels they enjoyed before they found themselves in competition with workers abroad and in the nonunion South. While veteran workers in unionized plants still make $26 to $32 an hour, new hires in companies like General Motors and Caterpillar make between $12 and $19 hourly, with contracts that lock them into these lower levels no matter how long they may work there. In 2008, average hourly wage and benefit costs in the Midwest were $7 higher than they were in the South; by 2011, they were $3.34 higher.

An entire region is downwardly mobile. An entire nation, but for the wealthiest 10 percent, is downwardly mobile. The share of jobs at the bottom of the economy grows as the jobs at the middle vanish. The United States now has the highest percentage of low-wage workers—that is, workers who make less than two-thirds of the median wage—of any developed nation. Fully 25 percent of all American workers make no more than $17,576 a year.

That’s what the disappearance of unions and the loss of worker bargaining power means to the economy.

Politically, it means that the liberal house loses its anchor tenant. Since the 19th century, and more particularly since the 1930s, unions have brought new groups into the American electorate. In the mid-1930s, as the organizers of the fledgling CIO fanned out across hundreds of mill towns, they also turned out those workers to vote for Franklin Roosevelt. That’s why Pennsylvania, which had not voted for a Democratic presidential nominee since 1856, moved into the Democratic column in 1936 and for most elections thereafter. In Southern California, beginning in 1997, labor brought so many new Latino voters to the polls that they flipped almost all of Los Angeles’s suburban congressional and legislative districts from Republican to Democratic, converting a purple state into a blue one.

Perhaps the unions’ greatest contribution has been their success at keeping their own white working-class members voting Democratic. Ever since the mid-1960s, when Democrats supported laws and programs to help African Americans, white working-class voters began to turn away from the party. Only the unions kept them in the Democratic fold. In exit polls dating back to the early 1970s, white working-class union members have voted Democratic at a rate 20 percent to 30 percent higher than their nonunion counterparts; among white working-class men, the margin is even higher. Unions’ election-time messages, whether conveyed by mail, e-mail, phone, or at the worksite, offer a persuasive counter-narrative to the Rush Limbaughs and Bill O’Reillys, who also focus their attention on white working-class men. The de-unionization of the white working class lets Limbaugh and O’Reilly go unanswered. It makes it harder for Democrats to win states like Ohio and Wisconsin.

Unions also lobby the presidents and members of Congress they help elect not just for their own causes but for progressive legislation generally. In 1965, the AFL-CIO played a crucial role in persuading both Democrats and Northeastern Republicans, whose districts were then home to tens of thousands of union members, to vote for Medicare, the Voting Rights Act, and anti-poverty programs. More recently, labor threw its weight behind the Affordable Care Act and the Dodd-Frank financial-reform law.

Labor’s other main contribution to the liberal cause has been—well, its contributions to liberal causes. Over the course of their history, to be sure, many unions have been estranged from progressives. The resistance of the building trades to racial integration and the support most unions gave to the Vietnam War are well known. But the most powerful union in the era of union power—the United Auto Workers under Walter Reuther from 1947 to 1970—functioned almost like a European social democratic party, seeding virtually every movement that emerged in the 1960s. The UAW supplied the money and the buses for the 1963 March on Washington and provided financial assistance to Cesar Chavez’s nascent farmworker union and its grape boycott. It gave start-up funds to Students for a Democratic Society, the National Organization for Women, and the first Earth Day. Today, the Service Employees International Union supports the immigrant-rights movement, while the AFL-CIO contributes to a range of progressive organizations.

No other progressive movement has anything approaching the level of resources that unions enjoy through their collection of member dues. In 2008, SEIU spent more than $60 million in its campaign for Barack Obama and congressional Democrats. This year, labor is likely to spend about $400 million on the election. Without assistance from unions, many liberal organizations will find themselves with severely reduced resources. They will be compelled to seek even greater support from foundations, which cannot legally fund overtly political activity. If unions fold, liberal America will not merely be weakened. It will be crippled.

III. Union Decline—A Whodunit

As conventional wisdom has it, unions declined because the industries in which they’d been strong could no longer afford them once globalization opened American markets to foreign competition. Unionized autoworkers and steelworkers couldn’t defend their pay levels against workers in China and Mexico who could turn out comparable products at a fraction of the American wage.

Indeed, during the past ten years, the growth of global markets has loosened the bonds between many of America’s largest corporations and the entire American economy. In 2001, 32 percent of the revenues of the S&P 500 came from abroad. By 2008, that figure had risen to 48 percent, as the rapidly growing middle classes of nations like China and Brazil began purchasing more. With rising markets abroad, U.S. businesses can afford to be less concerned with maintaining the purchasing power of American consumers—which means the wages of American workers.

But globalization by itself doesn’t necessarily lead to a weakened labor movement and declining worker income. If it did, unionized German manufacturing workers would not enjoy pay and benefits that exceed those of Americans even as their country has become the export giant of the Western world. Because unions are more powerful in Germany than they are in the U.S., and because German law requires large companies to divide their corporate boards equally between workers’ and management’s representatives, multinationals like Siemens, Daimler, and BMW have kept their most highly productive and best-paid jobs at home. Only where corporations have been free to structure globalization to their workers’ disadvantage—that is, in the United States—has it led to massive union decline.

Even so, globalization accounts for only a portion of labor’s descent in the U.S. Five years ago, economist Alan Blinder, a former vice chair of the Federal Reserve, calculated that roughly 30 million to 40 million American jobs could be offshored. People can make iPhones and blueprints and write contracts anywhere. Although most of those jobs wouldn’t be offshored, Blinder argued, anyone with such a job would sooner or later see his wages held down.

This still leaves more than 100 million American jobs that can’t be shipped abroad, whose wages can’t be undercut by the going rate in Shenzhen or Mumbai. Carpenters, cooks, supermarket clerks, and truckers have jobs that can’t be relocated. Yet unions have lost members in these place-specific sectors as well. In 1973, 40 percent of construction workers were union members. In 2004, just 15 percent were.

What drove this decline wasn’t offshore competition. It was growing employer opposition to unions, which was also the primary reason why unions could not expand into the growing service sector. Beginning in the 1970s, American businesses realized they could defeat unionization campaigns by exploiting the weaknesses of the National Labor Relations Act (NLRA), the 1935 law designed to protect workers’ right to organize, or by violating the act altogether, since the penalties for such violations were minuscule. (See sidebar, page 27.)

When a company fires a worker in the middle of an organizing drive, which the labor relations act explicitly deems illegal, the worker can file a complaint with the local office of the National Labor Relations Board (NLRB). If the board upholds her complaint, however, the most it can do is order the employer to rehire the worker and pay her back pay, minus anything she may have earned elsewhere during the months or years (the process is notoriously slow) since her firing. To get some perspective on how negligible such penalties may be, one just has to look at the 2007 unionization campaign at the Yale–New Haven Hospital, conducted outside the framework of the NLRA. An independent arbitrator ruled that management had committed numerous fair-practice violations and fined the hospital $4.5 million. During the decade of 2000–2009, by contrast, the total of all fines levied nationally by the NLRB for illegal punishment of workers for their union activity came to $36 million, or $3.6 million a year.

Richard Freeman, the Harvard professor who is the dean of American labor economists, has found that in 1950, for every 200 workers who voted to join a union, only one was fired during the organizing drive. By the early 1990s, that figure had grown to nine. American businesses had figured out that illegally firing workers to deter their unionization might be the single greatest deal they’d ever find.

What had restrained business in 1950? In part, companies held back because America was a different country then. Its population and most of its major industries were concentrated in the heavily unionized Northeast and Midwest. The Sunbelt boom had yet to take place. When the South and Southwest began to grow in the 1960s, unions couldn’t gain entry there, impeded by right-to-work laws that the federal government had allowed states to adopt when it passed the Taft-Hartley Act in 1947.

But there was another factor. In the years following World War II, leading American businesses believed in or were resigned to Fordism, the doctrine that calls for compensating workers well enough that they can consume well enough, too. It took its name from Henry Ford’s 1913 decision to pay his workers an unheard-of $5 a day so they could afford to buy the cars they were making, although it actually required the great unionization campaigns of the 1930s and 1940s to raise workers’ incomes to the point where they could make such purchases.

But as the Sunbelt grew, the nonunion South gave birth to a harsher economic order. Wal-Mart replaced General Motors as America’s largest private-sector employer. Instead of paying its workers enough to buy new cars, Wal-Mart paid its workers so little they had to shop at discount stores like Wal-Mart. Founded in 1962, the company grew to the point that it could compel—on penalty of exclusion from its shelves—the companies that made and transported the goods it sold to slash their own costs, which invariably meant their workers’ wages went down as well. It was a system as all-encompassing as Fordism, but it was Fordism in reverse.

One way Wal-Mart, its suppliers, and many other companies lower those wages is to hire workers not as employees but as independent contractors and temps. Businesses thereby evade any obligation to provide workers with benefits and make it still more difficult for unions to organize. How do you unionize workers who aren’t employees of the company you’re organizing? Such employment relations aren’t generally covered by the labor relations act, and such employment arrangements are increasing. By most estimates, roughly 25 percent of all American workers fall into this contingent--worker category. In many cases, this is an employer ruse. FedEx, for instance, dictates the schedules, routes, pay rates, and conditions of work for its drivers but calls them independent contractors to thwart any attempts to unionize them.

IV. Whodunit (continued): We Done It

The dysfunctional state of labor law and the shifting practices of American business, however, aren’t the only reasons for labor’s decline. The shifting beliefs and practices of American liberals must take some responsibility for this decline as well.

Liberalism, like business, has changed since the middle of the last century. In the late 1940s, both liberals and leftists, thrilled by the advances in worker incomes and workplace democracy that unions had recently brought about, still viewed labor’s battle with capital as the linchpin of liberalism’s concerns. “Class conflict is essential if freedom is to be preserved,” historian Arthur Schlesinger Jr. wrote in his 1949 book, The Vital Center, “because it is the only barrier to class domination.” To Schlesinger’s left, sociologist C. Wright Mills extolled the labor leaders then guiding the nation’s new industrial unions in his 1948 book, The New Men of Power. Both Schlesinger and Mills were partisans of Walter Reuther and his associates at the UAW, who in 1946 had struck General Motors over the question of the control of capital. Reuther wanted GM to let both the union and the government have a co-equal say with management over wages, prices, and technology—a scaled-down version of the system then being put together in postwar Germany. GM rejected it out of hand, and the UAW settled for a hefty raise instead. But for Schlesinger no less than Mills, class conflict—and working-class advances—was central to the liberal project.

And then, it wasn’t. Unions, even Reuther’s mighty autoworkers, lacked the power to raise the question of corporate control ever again. Instead, the UAW’s subsequent strikes wrested from the auto industry a series of agreements that entitled workers to benefits, insurance, and vacations and that saw their wages rise with increases in productivity and the cost of living. Other unions followed suit, but the grand drama, the question of whether America would be a more social or more capitalist democracy, was off the table. Labor relations had been reduced to dollars-and-cents agreements between powerful companies and powerful unions. It was the end of ideology, wrote sociologist Daniel Bell. It was a matter of administering prices and wages in a stable oligopolistic economy, wrote economist John Kenneth Galbraith. It was, truth be told, a yawn.

The workers had power and prosperity—or so middle-class liberals believed. But millions of Americans didn’t. Indeed, they still lacked fundamental rights. Young African Americans began sitting in at Southern lunch counters and rode integrated buses into segregated cities. Liberals turned their attention and moral energy to the battles for equal rights. So, for that matter, did the more progressive unions, which were among the few racially diverse large American institutions.

This, in fact, is liberalism’s enduring achievement of the past half-century—the revolution in rights for the previously excluded: blacks, women, Latinos, and, increasingly, gays and lesbians. Title VII of the Civil Rights Act permits workers fired or discriminated against due to race, religion, gender, or age to go to court and collect damages far in excess of those a worker fired for joining a union can collect through the NLRA. As one labor activist puts it, “They can’t fire you for your skin color or your sex or your age. They can just fire you if they want to.”

Liberalism eliminated an entire edifice of privilege. But by thinking the labor question had been settled during the decades of postwar prosperity, and then by not thinking about the labor question at all, liberals ceased to address the issue of power: the economic and political power that capitalism concentrates at the top when unions are weak and regulations watered down.

Liberals and leftists had their reasons for dismissing labor. Throughout the Cold War, the AFL-CIO looked like, and to some degree was, an adjunct of America’s imperial power, an aging politburo that supported every administration’s overseas adventures. It reviled the New Left and the liberals who supported the anti–Vietnam War presidential bids of Eugene McCarthy and George McGovern. It was hostile to feminist and environmental causes. Of course, any number of unions—not just Reuther’s—worked to advance those causes, but they didn’t capture the imagination of many mainstream liberals.

 For decades, labor was complicit in its own demise. The same union leaders who helped create the rift in the 1960s and 1970s between labor and liberals were also largely indifferent to organizing workers in the growing service sector—indeed, to organizing workers at all.

Unions had a problem even closer to home than the indifference of liberals. For decades, labor was complicit in its own demise. The same union leaders who helped create the rift in the 1960s and 1970s between labor and liberals were also largely indifferent to organizing workers in the growing service sector—indeed, to organizing workers at all. As convinced as Daniel Bell that the balance of class power in America had reached a permanent equipoise, George Meany, who presided over the AFL-CIO from 1955 until 1979, told an interviewer in 1972, “Frankly, I used to worry about … the size of the membership. But quite a few years ago, I just stopped worrying about it.”

So, alas, did many of Meany’s peers. Lulled during the years of labor’s power into thinking they’d attained sufficient numbers to keep on winning better contracts, they largely stopped organizing, devoting only 4 percent of their budgets to recruiting new members. As the economy began to grow in regions and sectors where unions had few if any members, and as employer opposition to labor increased, unions were caught flat-footed. For some years, the fact that the raw number of union members didn’t drop masked the decline in their share of the workforce, as did labor’s success in organizing public-sector workers in states and cities under Democratic control. During Ronald Reagan’s presidency, however, the actual number started dropping, too. SEIU’s decades-long drive to organize the janitors who cleaned the office buildings in big cities outside the South and the Hotel and Restaurant Workers’ unionization of the mega-hotels on the Las Vegas Strip were brilliant successes, but such campaigns were rare.

In time, unions awakened to the growing impediment that labor law placed in their path. Beginning in 1965, every time there was a Democrat in the White House and a heavily Democratic Congress on Capitol Hill, labor endeavored to have the law amended so workers could join unions without fear of being fired. They could never convince quite enough Democrats, however, to overcome the Senate’s requirement for a supermajority. Efforts to change the law during the Johnson, Carter, and Clinton presidencies all came up short.In 2010, with Barack Obama in the White House and Democrats in control of the Hill—victories to which labor had contributed about $400 million and the efforts of hundreds of thousands of volunteers—unions tried again. This time, they pushed legislation that would authorize the NLRB to recognize a union if a majority of workers simply signed affiliation cards—a process called “card check”—rather than go through an election that management had learned how to game. Once more, the effort died in the Senate.

It didn’t matter that without a change in labor law, private-sector unions might fade to oblivion. “We didn’t move a single Democrat who wasn’t already with labor to move on our behalf,” says Andy Stern, SEIU president during the 2010 battle. Though the rifts between labor and progressive movements had long since healed, unions couldn’t convince mainstream Democrats that their survival was a matter of existential importance, not just for labor but, given labor’s ability to turn out Democratic voters, for the party as well.

What explains this indifference to labor’s fate? Some leading Democrats believe their party can build an enduring majority that doesn’t have much of a place for unions. Their strategy is based on demography—that through the growing numerical strength of Latinos and Asians and the steady support of blacks and highly educated white professionals, Democrats can put together an electoral coalition that will sweep them into power for many decades. Single black women, the theory goes, vote at a 98 percent rate for the Democrats whether they’re in unions or not. As for unions’ ability to deliver white working-class votes to the Democrats—it’s not going to matter. The white working class is shrinking as a share of the electorate, and the union share of the white working class is shrinking alongside it. Who needs ’em?

But this strategy falls short in two particulars. First, it’s the unions on whom Democrats count to turn out much of the working-class black and Latino vote. Second, and more fundamentally, a coalition based chiefly on demography cannot stand, at least not for long. The New Deal coalition put together by Franklin Roosevelt, which lasted until 1968, may have been initially based in rising minority groups—Catholic and Jewish immigrants and their children—who were often as suppressed and reviled by Republican Protestant voters of that time (and Southern Democratic whites as well) as blacks and Latinos are today. But Roosevelt’s coalition, while incorporating these new groups in its circles of power, also bettered their economic lot in life.***

Can a new Democratic coalition all but devoid of a union presence and subject to the growing influence of corporate America and the financial elite do the same? Can it restore equitable growth? That would be squaring a circle. Already, some Democratic mayors, among them Chicago’s Rahm Emanuel and Newark’s Cory Booker, are building coalitions that array their city’s corporate elites and minority communities against their cities’ unions. The irony here is that their cities’ unions are largely responsible for expanding the middle class within those minority communities. Nonetheless, this municipal version of the Democrats’ top-bottom coalition could prove to be the model for the Democratic Party of the future. By ceding control over taxes, trade, and worker rights to the party’s corporate funders, however, this model omits a plausible vision of how to reconstruct broadly shared prosperity. Absent worker power, you can’t get there from here.

V. The Murky Road Back (Or anyway, to someplace)

Washington teems with liberal think tanks that produce plausible policy papers on how to rebuild the economy, and magazines like the Prospect that publish articles suggesting remedies for our economic woes. What Washington and America, what liberals and Democrats lack are institutions with enough power to rekindle class conflict and challenge class dominance. The ideas in both the labor movement and liberal America on how to build such institutions—be they unions or some new entities that do the work that unions have done—are few and fledgling. “If the next big idea was readily at hand,” says Wilma Liebman, who chaired the National Labor Relations Board between 2009 and 2011, “someone would have thought of it.”

Nobody has yet, but here are some provisional approaches.

One set of ideas derives from successful campaigns waged by workers and their allies in liberal cities and states over the past 15 years to win increased wages, inner-city hiring agreements, union recognition, new parks, and clean-air standards. The laboratory for this experiment has been Los Angeles, once an anti-union town that was transformed into labor-friendly terrain by the mobilization of the city’s vast immigrant community. There were two primary architects of this transformation: First, Miguel Contreras, who headed the Los Angeles County Federation of Labor, an association of more than 350 local unions, from 1996 until his death in 2005. Second, Madeline Janis, an attorney who at the behest of Contreras and his wife, hotel union leader Maria Elena Durazo, established what was first known as Los Angeles’s living-wage coalition, and today is the Los Angeles Alliance for a New Economy (LAANE).

Janis’s thesis was that businesses that operate on government-owned or -assisted properties or that have government contracts should repay the city by bettering the lives of the workers they employ and the communities in which they operate. LAANE’s first major victory was persuading the city council to require cleaning companies with which it had contracts to pay their workers a living wage—a sum several dollars higher than California’s minimum wage (or a bit lower than that if they provided their workers with health coverage). Over the years, the scope of such ordinances was broadened to encompass card-check unionization at hotels and sports arenas that received redevelopment funds; local-hiring requirements for developers of major projects; and clean-air standards for trucks at the Port of Los Angeles. Some of these ordinances have served as models for living-wage and other laws in more than 140 other cities. “A bank that makes an investment wants a return for its money, and so does the public,” Janis says. “The returns to the public should include good jobs, child-care centers, cleaner air, affordable housing.”

Realizing such returns, of course, requires political heft. That’s where Contreras came in. A few months after he took control of the federation, he turned to some immigrant-dominated local unions and asked them to do something that no one had really done in far-flung Los Angeles: walk precincts, lots of them. The federation soon developed an extensive, permanent field operation, particularly in Latino and black neighborhoods. It recruited its own candidates to run for office, authored its own platform for elections (including support for the programs LAANE promoted), and won election after election in a majority of council districts.

The premise behind the Los Angeles model is that liberal cities can enact pro-worker reforms that are more difficult to pass in statehouses and all but impossible in Congress. Which raises a question: What if, the next time we have a Democratic president and Congress, labor didn’t return to the Sisyphean challenge of reforming the National Labor Relations Act but instead advocated scrapping it altogether? The act preempts states and cities from establishing labor-relations statutes of their own (with the exception of Taft-Hartley’s provision allowing states to pass right-to-work laws). Suppose more liberal states were free to regulate labor relations in their jurisdictions. They might let workers win union recognition though card check or require companies that refuse to sign a first contract with their workers to submit to binding arbitration after six or nine months have elapsed. (These were all provisions in the labor-reform bill that failed to clear the Senate in 2010.)

Labor economist Richard Freeman has estimated the gains and losses unions would experience if labor law were left to the states. In industries like manufacturing, where businesses could relocate to anti-labor states, unions would suffer a decline in members, which Freeman calculates at around 500,000. But the gain in industries like construction, hotels, and retail—sectors where businesses can’t relocate—could total four million in union-friendly states.

Freeman’s suppositions about union--friendly states, which he defined as the 27 states that haven’t passed right-to-work laws, are probably too optimistic, however. Many of those states don’t have liberals or Democrats in power. Even in such progressive bastions as California or New York, there’d be no guarantee that Governor Jerry Brown, who vetoed a card-check bill for California farmworkers last year, or Governor Andrew Cuomo, who has cultivated an adversarial relationship with many of New York’s unions, would be inclined to make it easier for workers to organize in the private sector. Few states, far fewer than 27, are home to labor movements strong enough to persuade state government to pass legislation, in the face of the almost-certain outrage from business, that would make organizing easier.

Another approach to reversing labor’s decline is to broaden the scope of unions’ demands and, with that, their potential to connect with a larger public. One leading proponent of this idea is Stephen Lerner, the veteran union organizer and strategist who devised and ran SEIU’s Justice for Janitors campaign. Of late, Lerner has been working with community organizations and citizens’ action groups, advising them in a set of campaigns directed against the country’s leading banks. He argues that worker and community-organizing campaigns should target the public and private banks that control an increasing share of the economy as well as corporations like Wal-Mart rather than the multitudes of small firms that make up its supply chain. Going after the 1 percent, Lerner says, is a way unions can take the battle to the economy’s commanding heights and expand the coalition of engaged activists beyond those involved in traditional union campaigns. (If Lerner’s rhetoric sounds like Occupy Wall Street’s, that’s because he laid out a similar battle plan some months before the protests at Zuccotti Park.)

Lerner envisions four sets of protagonists: beleaguered homeowners waging a campaign to renegotiate mortgages; student organizations demanding a reduction in college debt; municipalities suing banks that devised dubious deals for them; and bank tellers seeking unionization. It’s a strategy reminiscent of the 1930s, when tenant rent strikes and armed farmers stopping foreclosure sales were common occurrences. Such campaigns would require acts of civil disobedience. In fact, groups with which Lerner is working have already disrupted shareholder meetings and occupied banks and homes facing foreclosure.

Could such campaigns be scaled up to a level that compels banks to accede to their demands? Lerner argues that it’s possible: “The movements of the ’30s and ’60s were successful because they were disruptive.” Today’s unions may not be able to practice large-scale civil disobedience, he says, but they can certainly encourage the activities of other groups.

Lerner is far from alone in believing that unions can no longer confine their organizing to workers they hope to represent in collective bargaining. For the past half-decade, AFL-CIO canvassers have gone door to door in white working-class neighborhoods in a notably successful campaign called Working America, enrolling more than three million nonunion swing voters to persuade them to support union-backed candidates at the polls. In the past few months, the group has also begun to talk to its members about their workplace concerns and how they might address them. In an even more radical move away from organizing workers, SEIU last year suspended its unionization drives and instead invested tens of millions of dollars in community--organizing projects in 17 cities. The union hoped to create a massive network of minority voters that could move American politics leftward, but the campaign proved largely unsuccessful.

A problem of labor’s own making afflicts such projects: the lack of a unified labor movement. In 2005, SEIU and several other major unions left the AFL-CIO to form a mini--federation of their own, Change to Win, that focused on unionizing workers in industries that could not be offshored. Seven years later, Change to Win has yet to claim any victories. The fragmentation of the movement has kept initiatives like the AFL-CIO’s and SEIU’s from reaching critical mass. Such efforts require funding from the whole of organized labor, since they don’t produce dues-based revenue even when they succeed. Labor’s fissures remain an impediment to its prospects for fighting bigger battles with potentially bigger rewards.

Larry Cohen, the president of the Communications Workers of America (CWA), has been a leading proponent of labor’s reunification. He also believes American unions can learn from the successes that labor has experienced in other nations when they helped form broader social movements. In recent decades, he points out, unions in Brazil, South Korea, and South Africa have flourished by becoming linchpins in campaigns for democracy, much as European unions did 100 years ago.

Having played a major role in the unsuccessful 2010 battle to persuade Congress to reform labor law, Cohen is convinced labor needs “to build a movement of 50 million Americans for democracy.” If labor wants to win more rights for workers, he argues, it must first break the stranglehold that big money and conservatives have on our politics by winning campaign--finance reform, abolishing the Senate’s supermajority requirements, outlawing voter suppression, and legalizing undocumented immigrants.

Beyond some visionary officials and activists, does labor recognize how radically it has to change? Too many leaders, says one union veteran, still think that “labor is selling toothpaste, something that’s always needed, that’s not going to go out of business. But we’re not selling toothpaste. We’re selling landline telephones. There won’t be a market for landline phones ten years from now.”

Can unions move beyond landlines? Even if they can, will our corporate-dominated political system still indulge their existence, let alone permit their rebirth? Labor can’t win this fight by itself. It not only needs to make common cause with the broader liberal community; it needs the liberal community to understand and embrace Schlesinger’s dictum that the only alternative to class dominance is class conflict. Heightening that conflict and strengthening unions, or whatever new institutions emerge to fight for working people, must now become liberalism’s primary task.