The departure of the AFL-CIO's three largest unions has already reduced the federation's membership by a third and its income by a fourth. Other unions may head for the exit.
All that competition for union dues is bad news for the AFL-CIO hierarchy, but not necessarily for unions in general, much less for workers in general.
This seems an opportune time to reconsider what a trade union is, and what it can and cannot do. Any union is a nonprofit enterprise that markets services for a fee. If many existing and potential members think the services are not worth the dues, then membership should be expected to shrink. A union's offerings might include assistance with training, career counseling, help with work-family scheduling, and financial and other information services.
Leaders of the AFL-CIO claim to represent those who pay union dues. But this introduces an "agency problem." Just as interests of corporate managers can differ from interests of corporate stockholders, the personal interests of union bosses may likewise differ from the interests of union members.
Being in a position to spend other people's money is a great source of power and prestige, including the offer of campaign help and money to politicians. The AFL-CIO uses about 36 percent of union dues to support the leadership's personal political preferences, although some 40 percent of union members voted for President Bush. Even members who agree with the union leaders' political tastes may nonetheless regard political crusades as a huge waste of their money.
Old guys running the "labor movement" tend to depict their efforts in quaint Marxist terms, as a contest between a huge army of wage slaves ("working Americans") against a small managerial-professional elite. In reality, there were 137.7 million Americans working in 2003, but only 72.9 million -- or 53 percent -- were still being paid by the hour, and a fourth of those were part-timers. Compare that with 47.9 million employees -- 34.8 percent -- who worked in managerial and professional positions. AFL-CIO protectionist lobbying, aiming to raise the price of imports, is an unlikely way to appeal to salaried service workers, Teamsters and others exiting the AFL-CIO.
"Change to Win" union bosses dislike the AFL-CIO's emphasis on partisan politics. But they, too, define goals in terms of their own personal interest -- namely, presiding over bigger unions with more power, influence and money. Nobody explains how existing union members might benefit from having their dues spent on trying to recruit more union members -- a frequently futile task estimated to cost as much as $3,000 per new member.
Although average pay is often higher for union workers, some of that gap reflects unions organizing the largest firms in the biggest cities -- those that always paid relatively high wages. Also, as Harvard economist Richard Freeman explained, "unions negotiate compensation packages that will create queues of job applicants and permit employers to cream (select) the best." In that case, we're not comparing comparable workers.
H. Gregg Lewis famously estimated the wage gap between union and nonunion wages at just 15 percent from 1967 to 1979. In 2003, however, David Blanchflower and Alex Bryson found the wage premium was substantially lower than in the '70s.
Labor compensation hovered around 70 percent of national income for decades, regardless of unionization. Whatever gains unions made were at the expense of other workers, consumers and taxpayers, not investors or owners.
Organized labor can widen the gap between union and nonunion pay only by making union labor relatively scarce. Suppose some unions attain sufficient monopoly power to force wages up. Whenever the price of anything goes up, demand goes down. There must then be fewer jobs at unionized firms. And that, in turn, leaves more jobseekers displaced into the nonunion sector -- thus, depressing nonunion wages.
In 1992, Henry Farber and Alan Krueger of Princeton found that "virtually all of the decline in union membership ... is due to a decline in worker demand for union representation." That is likely still true, with the notable exception of government employees -- 35 percent of whom are unionized, compared with 8 percent among private workers.
In "The Economics of Trade Unions," Albert Rees concluds: "The likeliest effect of unions on the distribution of income is to redistribute it among workers. ... First, the money wages of nonunion workers may be held down by the reallocation of labor produced by unionism; second, the nonunion workers may have to pay more for the products produced by union labor." But it is not as easy as it once was to pass on higher labor costs to consumers.
Unions are most likely to push their members' wages above those of other workers when (1) there are no good substitutes for union labor, (2) there are no good substitutes for the employer's product and (3) union labor is a small part of total costs.
A classic example was airline pilots under the Civil Aeronautics Board's regulated cartel. Employers could not replace pilots with labor-saving machinery, consumers were not free to choose a cheaper airline and pilots' salaries were a fraction of airline expenses. For similar reasons, the "Ma Bell" telephone monopoly was another irresistible target for unions. Computers were too primitive and costly to replace many telephone operators, and consumers were not permitted to buy phones or long-distance service from anyone but AT&T.
The only major sector in which competition is still legally banned is public services. Employers in public schools and other tax-financed services have little incentive to economize on costs by substituting nonunion workers or labor-saving technology. No matter how inflated the cost of public services may be, it would be literally criminal to refuse to pay for them. Public service monopolies thus allow unions to gain at the expense of taxpayers.
"Between 1982 and 1993," wrote James Poterba and Kim Rueben, "wages and salaries grew 69.2 percent in the public sector, and 52.2 percent in the private sector."
Breaking the AFL-CIO's stranglehold on union politics and services will be beneficial for the same reason competition is beneficial in economics and politics. Those trying to sell union services to workers may actually offer more and better services, for a change. The economic impact on private employers is unlikely to change much because, in a world of intense competition, excessive labor demands just "kill the goose." When it comes to tax-financed public services, by contrast, union gains are taxpayer losses.