The jury is still out on whether the traditional union is necessary for the new workplace.

—Robert Reich, 1993

Go back about 50 years, when America's middle class was expanding and the economy was soaring.

Paychecks were big enough to allow us to buy all the goods and services we produced. It was a virtuous circle. Good pay meant more purchases, and more purchases meant more jobs. At the center of this virtuous circle were unions.

—Robert Reich, 2009

Apparently, the jury came back. When Robert Reich was Secretary of Labor he was surprisingly equivocal about the value of labor unions—and his misgivings were shared within the Clinton Administration. Ron Brown, Secretary of Commerce from 1993 until his death in 1996, gave unions equally faint praise: "Unions are O.K. where they are. And where they are not, it is not clear yet what sort of organization should represent workers."

Fast-forward 16 years, and such doubts about unions have all vanished, or at least been silenced. Reich's present position is the new (and old) liberal consensus—labor unions are indispensable. As Washington Post columnist Harold Meyerson says, "The one great period of broadly shared prosperity in U.S. history remains the three decades following World War II, which, anything but coincidentally, is the one period in which America had high levels of unionization."

If it's obvious to Reich and Meyerson in 2009 that labor unions are essential, why was their value so doubtful to Reich and Brown in 1993? It's an open question because the argument put forward by Reich and Meyerson on behalf of unions is notably thin. Meyerson assures us that strong unions did not just coincide with shared prosperity. Even if that's so, however, the cause-and-effect relation could just as plausibly operate in the other direction: perhaps organized labor is good at exploiting rather than generating prosperity.

Reich makes the formula for prosperity sound suspiciously simple—unions with the leverage to secure big paychecks will activate the virtuous circle of consuming and hiring. If that's all there is to it, it's hard to imagine why a secretary of labor appointed by a Democratic president could ever have doubted the necessity of unions. The basic facts about the postwar economic boom were as comprehensible in 1993 as they are today.

Cartelizing Labor

Perhaps, then, that's not all there is to it. Federal judge and ubiquitous public affairs commentator Richard Posner expresses the view of many economists when he calls labor unions "cartels" that exist to redistribute wealth from companies' owners and managers to the unions' members and officers. Employee cartels will fold up if employers can obtain the services they sell on more favorable terms outside the cartel. The leverage unions use to redistribute wealth is based on their ability to make this more trouble than it's worth. Thus, unions derive their power from making life unpleasant, and sometimes dangerous, for "scabs" who cross picket lines, and firms that hire from outside the cartel or balk at its terms.

One corollary of Reich's (current) praise for the leverage unions exert is the proposition that economic growth is best served by minimizing profits. The assumption is that workers' paychecks are too small because shareholders' dividends and executives' compensation packages are too big. The AFL-CIO website, for example, maintains an "Executive PayWatch" for those wondering, "Is your CEO raking in the big bucks while running the company into the ground?"

In What Do Unions Do? (1984), an influential treatise on labor economics, the economists Richard Freeman and James Medoff argue that unions are indeed "harmful to the bottom line of company balance sheets." Yet, they say, the "paradox of American unionism" is that a strong union movement promotes a "thriving market economy" even while reducing corporate profits. The beneficial aspect of unions is that they can and often do enhance productivity. Unionized workplaces will be more productive than non-unionized ones to the extent that labor unions increase the likelihood of good industrial relations. Freeman and Medoff believe the quality of industrial relations in unionized firms varies greatly, and is often dysfunctional—but is also good enough, often enough, for unionization to be, on balance, beneficial.

The idea of private firms that, by virtue of being unionized, are simultaneously less profitable and more productive may seem like a contradiction, not a paradox. The best thing to be said for Freeman and Medoff's thesis is that it's not as crazy as it sounds. A unionized firm could be more productive than a non-unionized one if, by virtue of more efficiently utilizing employees with higher morale, it generates quantitatively or qualitatively superior outputs from any given basket of inputs—raw materials, capital, and labor. Furthermore, if the more productive unionized firm is forced to yield a larger share of its revenues to employees who are union members than it otherwise would, then it will also be less profitable.

The problem is that although unions' ability to reduce profits is easy to understand, and their record of doing so unassailable, the argument and evidence that unions enhance productivity is more elusive. When a union uses its leverage against an employer to direct a larger portion of the revenue stream to the union members than they would receive in the absence of the union, it is acting, in Freeman and Medoff's analysis, as a classic monopolist. The consequence of cartelizing the labor market an employer faces is to force that employer to pay more than it would if individual workers were bidding against one another to sell their services.

How much more a particular union can extract from a particular employer will depend on factors over which each will ordinarily have little control. The biggest is that a union's monopoly power vis-à-vis an employer depends heavily on the employer's monopoly power vis-à-vis the consumers of the products and services it provides. The golden age of the United Auto Workers (UAW), for example, was also the heyday of Detroit's Big Three. The union had the most success in pressing contract demands against the automakers when the companies had the greatest ability to pass those costs along to the consumer. The path of least resistance for auto executives under those circumstances was to accede to the union's demands rather than allow profitable factories to be idled by strikes. The UAW declined as the growing popularity of foreign cars eroded Detroit's ability to set prices and maintain profit margins.

Self-Interest Wrongly Understood

In 2008, when a mere 7.6% of all private-sector workers belonged to labor unions, 26.9% of utility companies' employees did so—utilities being, in most cases, regulated monopolies retaining considerable power to set their own rates. And, of course, the only "industry" where unions have flourished in the past 40 years has been government, the ultimate monopoly. The same Bureau of Labor Statistics report shows that 36.8% of public-sector employees were union members last year. (To put the point another way, while private-sector workers were more than five times as numerous as public-sector ones in 2008-108 million compared to 21 million—the number of private-sector unionized employees was only 6% larger than the number of public-sector ones, 8.3 million versus 7.8 million. Unless the trends that have held for decades are reversed, the majority of American union members will soon be government employees.)

Apart from the employer's power to dictate terms to its customers, the leverage any particular union can exert will often depend on essentially arbitrary factors. As Mickey Kaus wrote in 1983:

Some industries are extremely vulnerable to strikes—industries that deal in perishable goods, for example, or industries (e.g., Broadway theaters) where you can set up a picket line that will intercept a lot of customers. In other industries, advances in technology have weakened the power of strikes, as petroleum and chemical workers discovered when they walked out and found that skeleton crews of supervisors could run computer-controlled refineries for a long time. Did the chemical workers deserve to be paid less simply because their industries had become more strike-proof?

The "bargaining leverage" that Robert Reich praises will vary widely among different industries, in ways weakly correlated to any coherent theory of distributive justice. The most successful union leaders will be those who shrewdly assess just how much leverage they do have, and bargain for their members without either overplaying or underplaying their hand.

The death spiral of America's auto industry, however, reveals America's most important industrial union to be a matchless practitioner of self-interest wrongly understood. The UAW's posture toward the auto industry has consistently violated a fundamental Darwinian precept: healthy parasites figure out ways to avoid killing the host organism. Last year BusinessWeek took note of the concessions the UAW, led by its president Ron Gettelfinger, had made to the Big Three when the most recent contract was negotiated. It also noted that, from the industry's perspective, the problem is,

Gettelfinger made those key concessions starting in 2005, but not until Ford and GM were reeling toward massive losses. The union has never given enough to get the companies ahead of the curve. "It's always a day late and a dollar short," says one former GM executive.

"Gettelfinger isn't the problem," according to Kaus. Rather, "[t]he problem isthe system, the American adversarial labor-management negotiating system, in which reasonable people doing what the system tells them they should do wind upproducing undesirable results." In that system, "negotiating ponderous 3-year contracts (in which Gettelfinger must extract every possible concession to please the members who elected him) means contracts adjust too slowly to save the companies from failure if market conditions change."

The 2005 contract, for example, stipulated that the wage and benefit package for new hires would be considerably leaner than ever before—less costly, even, than for starting employees at non-union auto plants operated in the U.S. by foreign companies. The problem, however, is that there haven't been any new hires at the ever-smaller Big Three since 2005. The sort of concessions that might actually help render the U.S. auto companies viable would be actual pay cuts for actual workers, as opposed to hypothetical cuts for prospective ones. But union leaders hold their positions based on what they deliver for today's workers, not for those who might be hired in the future. As BusinessWeek recounts:

How did [Gettelfinger] sell the [2005] concessions to his members? None of the current workers lost much of anything. They kept their pay, and their health-care benefits are still first-rate. Anyone losing a job got buyouts averaging more than $100,000, and they typically head into the pension rolls.

Bitterly Adversarial

Making Freeman and Medoff's paradox work is so difficult because profitability and productivity are closely related. More profitable firms aren't distinguished solely by the better golf courses on which their executives and shareholders play. They also have a greater ability to invest in research, to generate new and better products and services, and to develop and utilize better production facilities and technologies. To be more productive, in other words. Capitalism requires capital, after all, and profits create it.

For labor unions to enhance productivity while reducing profits, they would have to play an indispensable role in establishing a workplace environment that is more collegial and goal-oriented than in non-unionized firms. That is, the consequences of good industrial relations would have to be so beneficial that unionized firms could thrive by virtue of receiving a smaller share of a larger revenue stream. Non-unionized firms, by the same measure, would suffer because even though they retained a larger portion of total revenues, those revenues would be significantly smaller than they would have been if labor unions were organizing, guiding, and motivating employees.

To make their case, Freeman and Medoff borrow their analytical framework from Albert Hirschman's influential work in political theory, Exit, Voice, and Loyalty (1970). The "monopoly face" of labor unions is "socially harmful," Freeman and Medoff argue. In that respect, unions' power is based on denying the possibility of exit—either of workers from unions, or firms from labor markets controlled by unions. Nonetheless, the "voice/response face" of organized labor will be beneficial, according to Freeman and Medoff, if management can "adjust to the union and turn unionism into a positive force at the workplace." The efficiency improvements will include lowered costs of training and recruitment, apprenticeship programs that upgrade and certify workers' skills, and "more rational personnel policies" that "raise productivity by reducing organizational slack."

The relationship between unions and employers is inherently and often bitterly adversarial with respect to the fundamental question of compensation. As a result, turning unionism into a positive force is going to be miraculous more often than it will be merely paradoxical. The idea that unions, in their capacity as monopoly cartels, are going to extract every concession they can from their adversaries across the bargaining table, and then turn around to collaborate with management on fashioning workplace reforms that are mutually beneficial is hopeful, to say the least.

Being Naïve

This is especially true because most labor contracts go far beyond establishing wages and benefits, and also stipulate detailed work rules. In the 1980s, for example, the autoworkers relied on a "cumbersome standard contract," according to the journalist Paul Ingrassia, one "featuring nearly 200 job classifications at some GM factories, with rules prohibiting members of one group to perform work reserved for another." Even a sympathetic New Republic article on the "tragic nobility" of America's unionized auto industry concedes, "Work rules and grievance procedures designed to protect diligent workers from unfair managers sometimes ended up protecting less-than-diligent workers from appropriate oversight."

Keeping hookers out of the workplace seems like a good example of appropriate oversight. The blogger Lori Roman recently recounted her experience as a G.M. supervisor who learned that some workers had "brought an RV into the loading yard with a female ‘entertainer' who danced for them and then ‘entertained' them in the RV." Because the otherwise hyper-elaborate contract was silent on the question of lunch-hour prostitution, "Not one person suffered any consequence" for the visiting bordellomobile.

On another occasion, Roman was in charge of "a loading dock and 21 UAW workers who worked approximately five hours per day for eight hours' pay."

They could easily load one-third more rail cars and still maintain their union-negotiated break times, but when I tried to make them increase production ever so slightly they sabotaged my ability to make even the current production levels by hiding stock, calling in sick, feigning equipment problems, and even once, as a show of force, used a fork lift truck and pallets and racks to create a car part prison where they trapped me while I was conducting inventory. The reaction of upper management to my request to boost production was that I should "not be naïve."

Little wonder that Roman's assessment of enlightened management's ability to turn unionization into a "positive force" is less optimistic than Freeman and Medoff's:

To put it bluntly, the UAW takes the hard-earned money of the best workers and spends it defending the very worst workers while tying up the industry with thousands of pages of work rules that make it impossible to be competitive. And the spineless management often makes short-sighted decisions to satisfy the union and maximize immediate benefits over long-term sustainability.

Even now, with the domestic auto industry on taxpayer-funded life support, unionism dictates an implacably adversarial posture. As one autoworker wrote to the Atlantic Monthly's Megan McArdle, "Up until recently one of the popular views was that the big 3 weren't actually in trouble and that management was cooking the books to show a loss in order to demand concessions from the workers and break the union. It may seem silly but I have heard this from several workers at several plants."

The Saturn Experiment

The "new workplace" that Clinton Democrats like Reich envisioned was not supposed to be so angry, or so inefficient. Rather, it would be founded on the "voice/response face" emphasized by Freeman and Medoff. A 1993 New York Timesarticle, for example, contained a detailed, hopeful account of labor-management "work councils" that set production schedules, judge quality, determine wages, hire new people, and make job assignments.

The most radically new workplace was supposed to be the Saturn division of General Motors, which produced its first car in 1990. Writing in Newsweek this year, Paul Ingrassia laments that Saturn was "a major missed opportunity" for Detroit "to recalibrate the bitter business-labor relationship in ways that could have had far-reaching implications for the entire industry." A UAW apostate, Donald Ephlin, the head of the union's G.M. department when the Saturn division was created in 1985, agreed to a very unorthodox contract for its workers: a mere handful of job classifications, one-fifth of compensation tied to quality and productivity, and every employee to spend at least 5% of work time in skills training. Most important, writes Ingrassia, "Workers had a voice in everything, even hiring decisions."

Vice President Al Gore took time off from reinventing government in 1993 to visit the Saturn factory in Spring Hill, Tennessee, and proclaim that his goal was to "Saturnize" the federal bureaucracy. It was already clear, however, that G.M. couldn't even Saturnize Saturn. In 1989 Ephlin was replaced as the head of UAW's G.M. department by Stephen Yokich, recently described as "downright scary" in the Wall Street Journal, a union leader who "ate [corporate] vice presidents for breakfast." Yokich set out to eliminate everything that was distinctively flexible about industrial relations at Saturn. In 1995 he was elected UAW president. In 2003, Saturn workers voted to rejoin the same master-contract as all other G.M. employees represented by UAW, and the new factory constructed for the sole purpose of building Saturns was renamed "GM Spring Hill Manufacturing." As was universally expected, General Motors threw Saturn overboard this year in its desperate struggle to survive.

The fact that Saturn failed doesn't prove that it had to fail. One suspects, however, that Clinton Democrats searching for a new workplace were attracted to the very quality that proved to be Saturn's gravest problem. "Consensual decision making was valued at Japanese [auto] plants, too," Ingrassia points out, "but management retained the right to run the place." One Toyota executive told him that, by contrast, Saturn's version of workplace democracy "was like having two people share one pair of pants."

The New Old Liberalism

"Which side are you on?" the old labor anthem demanded to know. It's a question, in all its forms, the Clinton Administration spent eight years trying to dodge, down to haggling over what the meaning of "is" is. Taking power after the fall of the Berlin Wall had certified the intellectual bankruptcy of socialism, the Clinton liberals sought a "third way" between relying on markets and relying on the State. In the new workplace the industrial relations conflicts that had enervated American industry would be triangulated. Unlike the old, adversarial firms where owners owned, managers managed, and workers worked, the new one would thrive by having "stakeholders" collaborate. Everybody would be involved and nobody would really be in charge.

The Obama liberals, by contrast, know which side they're on. The collapse of the Soviet Union was long ago, but the collapse of Lehman Brothers is recent, vivid, and important. The position staked out by the Obama Administration "reverses and repudiates the economic philosophy that has dominated America since 1981," Robert Reich wrote in March. It repudiates, that is, both the Reaganites who espoused that philosophy and the Clintonites who felt the need to accommodate it.

As a result, liberalism is back where it started. Conservatives can relinquish their hopes, and paleoliberals their fears, that the Obama Administration will be guided by pragmatists who have learned from Friedrich Hayek to respect markets' usefulness and government's limits. Franklin Roosevelt asserted in 1932 that business leaders "must, where necessary, sacrifice this or that private advantage; and in reciprocal self-denial must seek a general advantage." The government would be standing by, FDR promised, if they were slow to work out the desired arrangements. One administration official told the Senate in 1933, "We have reached a stage in the development of human affairs where it has become intolerable to have our primitive capitalistic system operated by selfish individualists engaged in ruthless competition."

By the time the Supreme Court put Roosevelt's National Recovery Administration out of its misery in 1935, however, even the most committed New Dealers had stopped defending it. This vehicle for the renunciation of private advantage was supposed to be a set of agreements, crafted by each industry, that codified pricing, business practices, and labor relations, and which acquired the force of law upon the president's signature. According to historian David Kennedy, the "lamentably predictable results" were, instead, "codes that amounted to nothing less than the cartelization of huge sectors of American industry under the government's auspices." What's more, "NRA mushroomed into a bureaucratic colossus," under which even running a hardware store required compliance with 19 separate codes.

The spirit of the New Deal agriculture policies lives on in farm subsidies that no liberal intellectual defends on the merits, and which no administration or Congress can dismantle. The subsidies are a textbook example of client politics, where the benefits are concentrated on a small segment of the small agricultural sector, while the costs are distributed across the broad public. The public bears the costs twice, in fact—as taxpayers financing the farmers' subsidies, and as consumers buying artificially expensive food.

The organization of the labor market under Wagner Act unionism, according to Posner, is of a piece with the New Deal's efforts to cartelize industry and agriculture. During the brief neo-liberal moment at the dawn of the Reagan era, unionism was as intellectually disreputable on the American Left as farm subsidies. The editor of the Washington Monthly, Charles Peters, wrote in 1983 that he favored "freeing the entrepreneur from economic regulation that discourages desirable competition," but did not support "unions that demand wage increases without regard to productivity increases. That such wage increases have been a substantial factor in this country's economic decline is beyond a reasonable doubt."

Re-unionizing America

Among the ways in which the Obama Administration is not letting the economic crisis go to waste is by making clear that liberalism is rushing back to the pre-Reagan era, when sensible thoughts such as Peters's were banished from acceptable discourse. 2009 is 1933 all over again, when the creation of wealth was morally repugnant and, somehow, economically irrelevant, while the governmentally sanctioned reallocation of wealth from the less to the more deserving was the necessary and sufficient condition for social justice. Walter Reuther was acting as a good unionist and a good New Dealer in 1946 when he led a UAW strike against General Motors over his demand for a 30% wage increase coupled with a freeze in the retail price of all G.M. cars. Barack Obama is doing the same today, when he uses taxpayer funds to buy the Chrysler corporation for his supporters in the autoworkers' union, then publicly berates the "small group of speculators" who had the temerity to complicate his plan "by refusing to sacrifice like everyone else." That is, they wouldn't drop their demand for the return of an inconveniently large fraction of the money they had lent the company.

The same contempt for the neutrality of law—the same "by any means necessary" spirit one expects from a president who got his start in politics as a community organizer—animates the Obama Administration's support for "card check" legislation that would permit unions to organize workplaces without secret-ballot elections. Even a recent, favorable Washington Monthly article on the "Employee Free Choice Act" (EFCA) by T.A. Frank admitted, "Card check is worth fighting for—except for the ‘card check' part." Unfortunately, the card check part isn't even the worst part of EFCA, which would allow government arbitrators to set wages, benefits, and work rules in a two-year contract if newly unionized firms and the union representing its workers do not arrive at a negotiated deal. We've seen how even-handedly the Obama Administration treated labor and management in forging a packaged bankruptcy plan for General Motors. A memo from UAW to its G.M. members on the eve of the bankruptcy filing assured them that the union's "concessions" involved "no loss in your base hourly pay, no reduction in your healthcare and no reduction in pensions."

The question is whether, in their exhilaration at being free at last to ignore markets and their defenders, Obama liberals will forget the excesses caused by the hostility to markets, excesses that brought conservatives to power. If liberals rediscovering their inner unionist want to look at a really vigorous labor movement, they can revisit the Great Britain that turned to Margaret Thatcher 30 years ago, after it had become unclear, in George Will's words, whether Parliament or the labor movement actually ran the country. Long before Obama appointed Steven Rattner to oversee the restructuring of the American automobile industry, Rattner had been a reporter for the New York Times. In that capacity he visited Ford's plant in Halewood, England, in 1981, where he discovered that the factory had been shut down by 20 strikes that year—which sounds bad, but was a big improvement from the 300 it had endured in 1976. The state of industrial relations at the factory was such, according to Rattner, that "one worker greeted a news photographer by exposing himself."

Whatever Halewood may have lacked—productivity, quality, collegiality, decency—it had plenty of the union bargaining leverage that Reich singles out now as the crucial element to make America's economy soar. We may get the chance to learn whether the re-unionization of America will lead to Reich's virtuous circle, or to one of the more hellish ones described by Dante.

* * *

For Correspondence on this essay, click here (Winter 2009/2010) and here (Fall 2009).