(Labor Watch, April 2009 PDF here)
President Barack Obama says that his administration is committed to reviving the nation’s economic prospects and securing the passage of legislation that satisfies important Democratic priorities— including the Employee Free Choice Act. But what if those goals are at cross purposes? What do labor economists have to say about the effects of increased unionization on the whole American economy?
In his second book The Audacity of Hope (2006) Barack Obama credited his successful start in politics to labor union endorsements. As a result, he explained, “I owe those unions.”
The soon-to-be presidential candidate spelled out what that meant: “When their leaders call, I do my best to call them back right away.” There was nothing wrong in doing this, Obama insisted: “I do not consider this corrupting in any way; I do not mind feeling obligated toward home health-care workers or toward teachers. I got into politics to fight for those folks, and I am glad a union is around to remind me of their struggles.”
Candidate Obama announced his support for labor-backed policies such as the Employee Free Choice Act, which will virtually eliminate secret-ballot elections for union representation. Today both the supporters and the opponents of increased unionization expect it to happen should the “card check” provisions of EFCA become law. But this raises a question that is very important in light of America’s recent economic downturn: As workers become more and more unionized, how will it affect the economy? Do Americans overall get more jobs and higher wages when unions organize more workplaces?
Economists have looked closely at this subject. There is a vast literature on the effects of unionization on union and non-union wages, on unemployment and corporate profits, on wage differences, and rates of inflation. Not all these questions are settled. But as lawmakers grapple with a sinking Dow and the shrinking 401(k) balances of their constituents, they ought to pay attention to this work. They need to listen to voices other than union organizers’ voices if they really care about the home health-care workers and the teachers.
Andrew Stern, the politically astute president of the two million member Service Employees International Union (SEIU), has predicted that once EFCA becomes law, organized labor will grow by 1.5 million members every year for the next 10 to 15 years.
In 2007, 15.7 million American workers were union members. Compared to 2006, organized labor enjoyed a rare increase of 311,000 new members. But even with that increase, union membership rolls have dropped an average of 95,000 per year since 2000. In fact, union membership has steadily declined over the last half-century. In 1953, more than 35 percent of workers belonged to unions. In 1983, that number was 20 percent. In 2007, 12.1 percent of workers were in unions.
Many in the labor movement read a lot into the 2006-2007 bump up, which produced an increase in union membership from 12 percent to 12.1 percent. Economists tend to think the difference is a statistical fluke, but union organizers like Stern see it as a sign of things to come. After years of decline, they claim unionization finally hit bottom and is about to increase. To them, EFCA will only accelerate the rate of unionization.
Existing trends aside, Raymond Keating, chief economist of the Small Business and Entrepreneurship Council, agrees that card check could significantly increase union membership. “When you go from a secret ballot to a situation where 50 percent plus one can check off, and your workplace is unionized, that changes the dynamic completely,” Keating told Labor Watch. “I doubt the unions will get back to where they were in the ’50s, ’60s, and ’70s, but this will put a lot more businesses at risk.”
There are others provisions in EFCA that strengthen organized labor’s hand. One would force management into binding arbitration with the union if the parties failed to agree on a contract within a specified period of time. Currently, about one-third of newly unionized shops don’t succeed in negotiating a contract within a year. That possibility would disappear under EFCA, thus further incentivizing union organizing.
Then there is the little-known bill before Congress that would require states to allow collective bargaining for public employees like police officers, firefighters and emergency medical technicians. The so-called “Public Safety Employer–Employee Cooperation Act” is purportedly a measure supporting the rights of homeland security personnel. But the legislation overrules the minority of states and cities that do not recognize public sector unions as the sole bargaining agent for their employees.
The Act makes union collective bargaining a fundamental right. It also forces states and cities to negotiate with the unions on whether to replace merit-based pay systems for public safety employees with systems based on seniority. In the last Congress, the bill, introduced by Sen. Judd Gregg (R-NH) and Rep. Dale Kildee (D-MI), had 36 Senate and 280 House co-sponsors–and it will be reintroduced in Congress this year. (See Labor Watch, “Will Congress Mandate State and Local Unions: ‘Public Safety’ Act Will Turn Federalism Upside Down” by Stan Greer, August 2008.)
If President Obama meets his stated goal of creating (or saving!) 3 million jobs, with up to 20 percent of them in the public sector, he could add 600,000 government employees. Assuming the proportion of public-sector workers in unions stays at 35 percent, this adds up to 210,000 more union members.
What is the likely impact of increased union membership? There’s no doubt that unions help their members gain significant advantages over non-union workers in wages and benefits (although union pension plans are arguably inferior to their non-union counterparts). However, this creates a problem, because these extra dollars and benefits aren’t the result of free choice and competition in the market. Economists explain that everyone benefits when a worker’s compensation rises through a market process. Wages rise when there is an advance in productivity, or a worker gains new skills, or there is increased demand for a company’s products and thus increased demand for workers to make these products. In these instances, the worker’s extra worth matches his extra cost.
But when legal mandates like minimum wage laws or collective bargaining supplant market processes, the worker’s gain is the employer’s loss. Labor economics becomes a zero-sum game when wage and benefit hikes are compelled by monopoly representation coupled to union agitation and the threat of strikes.
Economists further explain that when one company is unionized, other companies will pay lower wages and unemployment will increase. That sounds counterintuitive, but here’s why it happens:
*When a union successfully demands higher compensation for the same (or less) work, it makes labor more expensive for the company.
*With labor more expensive, the company buys less of it; it employs fewer workers. *Workers who would typically work in the now-unionized company will have to go elsewhere, thereby increasing the supply of labor to non-union companies.
*This greater supply of non-union labor will cause its price (i.e., wages) to fall.
The same factors affect a unionized public-sector workforce. Governments that must employ more highly-paid workers will hire fewer of them. Workers will lose their jobs or will fail to be hired, which will increase the overall supply of labor to other employers causing wages to fall. When governments are under pressure to retain or increase their workforce and pay higher wages, then every extra dollar paid ultimately comes from taxpayers’ pockets.
In their book Free to Choose (1980), Nobel laureate Milton Friedman and his wife Rose described the discriminatory effects of union power over non-union wage rates. They rejected the union claim that high union wages came at the expense of an employer’s high profits—often labeled “obscene.” In fact, they pointed out that labor costs amounted to 80 percent of national income. By contrast, corporate profits were less than 10 percent of national income, and just six percent after taxes. Slight increases in labor costs could easily wipe out profits.
The Friedmans also explored unions’ effects on non-union employees. In the mid-20th century, they estimated, unionization had raised the wages of 10 to 15 percent of all workers by 10 to 15 percent. They calculated that other workers made about 4 percent less than they would have otherwise as a result. The Friedmans’ estimates for that time were confirmed by other studies. In recent years, however, the estimates have varied remarkably, depending on methodology and the sector studied. Some studies indicate the union wage advantage over non-union labor has risen as high as 30 percent.
The Friedmans’ theory of the impact of unionization on wages has drawn two counter-arguments. One (summarized by the Congressional Research Service in its report on the impact of EFCA’s card-check provision) asserts that there is an “imperfect market” for labor in companies that do not have to compete for workers. In a “monopsony” (a sort of inverted monopoly), where many employees compete to work for a single employer, unions can extract more worker benefits without causing non-union unemployment and lower wages.
It’s possible to build a theoretical model for this case, Robert Murphy, a Pacific Research Institute senior fellow and the author of The Politically Incorrect Guide to Capitalism (Regnery, 2007), told Labor Watch. However, Murphy finds little empirical evidence for it. “And even if you concede it happens in the real world, it’s not the case that unions are going to go after sectors where these very special conditions hold.”
The other counter-argument to the Friedmans’ argument is the theory, associated with Keynesian economist John Kenneth Galbraith, that holds that unions are a “countervailing power” to Big Business and therefore can actually increase economic efficiency. Murphy agrees that corporations often do evade the discipline of marketplace competition. He adds, though, that this is “because they’re in league with the government. They can get regulations passed that will have a crippling effect on their smaller competitors.”
“But the solution is the get the government out of the market, not to embolden the unions. In a free market, the only way to get big is to serve customers better than your competitors do,” Murphy argues.
Cars, Trains, and Chalkboards
While it’s false to say that union benefits come from corporate profits, it’s true that unions do affect a company’s well-being. Unionized companies often suffer when they must compete with non-union ones, particularly during times of crisis. And when unions cause employers to become less productive, they hurt the economy in general and workers in particular.
Union “featherbedding” (or “make-work”) is one way to lower employer productivity. Probably the most extreme recent example is the “jobs bank” that the United Auto Workers contract imposed on the auto companies. Instead of laying off workers, auto companies were required to put them on furlough and to give them 85 percent of their pay. The jobs bank was finally suspended in late 2008, when the companies faced insolvency.
Then there’s the “100-mile rule,” awarding overtime to railroad workers for traveling more than 100 miles on a train. In the days of steam locomotives a railroad worker probably traveled 100 miles while doing a full day’s work and deserved the overtime. But the railroad unions fought to keep the rule long after it became nonsensical.
Unions also hurt productivity whenever they resist company efforts to adapt to changing environments. For instance, the UAW has opposed the introduction of robot technology because robots would replace union members. Similarly, teachers’ unions oppose technological improvements that may cost teacher jobs.
Hoover Institution fellows John Chubb and Terry Moe observe that the mission of teachers unions “is to protect the jobs of teachers in the regular public schools, and real technological change — which outsources work to distant locations, allows students and money to leave, substitutes capital for labor, and in other ways disrupts the existing job structure — is a threat to the security and stability that the unions seek.”
Unionized companies pay higher wages while driving down wages elsewhere. This pay discrepancy affects not only workers but unionized companies, because non-union companies can offer consumers a similar product at a lower cost. The unionized Big Three automakers in Detroit have learned this lesson as they compete with non-union auto companies based in Southern right-to-work states.
In a classic article in The Atlantic, the journalist Gregg Easterbrook detailed union responses to the recession economy of 1983. The article, called “Voting for Unemployment,” described how union members deliberately voted to let the factories that employed them close rather than accept less compensation than they’d been promised. Sometimes the unions counseled them that the employers were bluffing, that the companies kept “secret books” showing mountains of profit. But on other occasions, union workers with seniority actually stood to make more from their benefits packages if the factory closed.
Because they realize that their contract negotiations with a single employer can produce unanticipated consequences, Big Labor has resorted to what is called “pattern bargaining” in which it negotiates a “master contract” with all the major employers in an industry and then “patterns” it in bargaining with smaller companies. The union strategy has been that if all employers have to pay their unionized workers the same wage and pass the costs on to consumers, no one company can exploit a wage difference.
In late 2007, the UAW negotiated a four-year contract with the Big Three automakers, essentially freezing the average pre-benefits wage at $28 per hour. When the contracts expire, the UAW will pick the company most likely to give it the best deal and use the results to secure comparable agreements with the other two. Observers claimed the 2007 contracts (Ford’s was 2,215 pages long) would close the wage gap between Detroit and the South, and return the Big Three companies to competitiveness. Needless to say, Detroit’s pattern-bargained agreements have done nothing to overcome its financial problems.
Equal, or Else!
Of course, leftwing activists and pundits will never be persuaded by these examples of how forced unionism lowers worker wages overall. That’s because they have an ideological commitment to an equality of results; equality is a goal in itself for the Left. They want government to impose it on all workers. They have an aversion to letting employers and employees reach voluntary agreements that result in paying people different wage rates.
Leftists believe the decline in union membership is a principal reason why U.S. income distribution has become more unequal in recent years. “What are the mechanisms [of the rise in inequality?” wrote economist and New York Times columnist Paul Krugman in an e-mail to Mark Thoma of the Economist’s View blog. “Unions are probably the top of the list; I believe that there’s a qualitative difference between wage bargaining in an economy with 11 percent of workers unionized, which is what we had in the early 30s, and one with 35 percent unionization, which is what emerged from World War II.”
The idea isn’t far-fetched. “Within the union sector, wages are more compressed than outside the union sector,” observes U.C. Berkeley economist David Card, who has studied the issue. He told Thoma, “Unions try to equalize wages for people doing similar jobs.” But, as Friedman noted, unions also create inequality by raising wages for their own members while lowering wages for non-union employees.
The real question is one of effect. People who believe equality is a paramount issue need to ask: Do unions reduce inequality? Card examined the effect of unionization on inequality between the mid-1970s and the early-1990s, and his results indicate the answers to be “yes” and “no.” Unionization decreased inequality among men, but not among women (mainly because women’s total union membership hadn’t changed much). Even among men, unionism’s decline explained only 10 to 20 percent of wage inequality’s rise—over a 25-year period in which the union-membership rate fell from one in four workers to about one in six.
Even if the predictions of SEIU’s Andy Stern come true and the proportion of unionized workers doubles to 25 percent, no small feat, the U.S. is still likely to have about as much income “inequality” as it has now.
Whip Inflation Now
Unionization may also indirectly cause, or interact poorly with, inflation. The simple explanation for the impact of unionization on inflation is that unions increase the price of labor, which raises the cost of a company’s products. What is inflation if not prices going up?
That’s not quite right, says the Pacific Research Institute’s Robert Murphy: “Unions might cause the prices of union-made products to go up, but they don’t give consumers more money to spend. Say the auto industry unionizes. Consumers might spend more on cars, but they have to spend less on something else. It just shifts money from one sector to another, unless the government is printing more money.”
Murphy also has a more political analysis. He notes that vote-seeking politicians are on the horns of a dilemma: While they have every incentive to prevent the kind of overall job losses and wage cuts that unionization inevitably produces, many politicians curry favor with unions. One solution is to take actions that will increase the rate of inflation. Ordinarily, employment increases during inflationary periods because employers can pay workers with cheaper dollars.
However, if unions grow stronger, they will demand that wages keep pace with inflation. This is likely to stall gains in employment, which is a recipe for “stagflation,” a period of when price inflation co-exists with economic stagnation. Murphy predicts high inflation as trillions of dollars circulate through the financial system courtesy of the Obama administration and Federal Reserve chairman Ben Bernanke.
Unions in the New Millennium
Three factors determine the rate of unionization: 1) The percentage of workers employed in high-union versus low-union sectors in the economy, 2) how hard unions try to organize workplaces, and 3) their rate of success. Anything that makes the job of union organizers easier will increase the rate of unionization.
Under current law, in the private sector, union organizers must get 30 percent of workers to sign authorization cards, at which point the National Labor Relations Board (NLRB) oversees a secret-ballot election. (Public-sector labor law varies state by state.) If the organizers get a majority of the votes, a union is formed.
Alternately, if the organizers can get 50 percent of workers to sign cards, they can form a union right away. This latter method is known as “card check.” The catch is that when organizers use card check, the company can refuse to recognize the results, and demand an NLRB-supervised secret-ballot election.
An election places workers in the privacy of a voting booth, where it’s easier to ignore pressure from organizers and co-workers. But the Employee Free Choice Act would take away the employer’s right to demand a secret-ballot election, thereby increasing unions’ chances of organizing workplaces.
There is a good chance mandatory card check will become law. President Obama says he supports it, and so do most of the Democrats who control Congress. “Even if it gets pushed back in the debate over the economy, I don’t think it will go away,” says Raymond Keating of the Small Business and Entrepreneurship Council.
Keating noted the role unions played in the election and the importance organized labor places on card check. Shortly after Labor Watch spoke to Keating, the Service Employees International Union announced an $85 million drive to promote the legislation. It is hard to tell how dramatically card check would help unions, and whether it could stall or reverse the trend of declining unionization. In 2000, Princeton University’s Henry Farber and Bruce Western published a notable paper arguing that the decline of American unionization stemmed from changes in the economy, not from anti-union measures. The truth was that fewer Americans were working in high-union fields such as manufacturing.
“The prospects are dim for the reversal of the downward spiral of labor unions based on increased organizing activity,” Faber and Western wrote. In the long term, they’re right, as long as workers continue to leave high-union fields. But what if union organizing occurs under a radically different legal regime? There’s every reason to think EFCA could spur increased unionization of workplaces.
In a report last updated in 2007, the Congressional Research Service noted that under card-check regimes, union organizers make more attempts to unionize employees, and have a higher rate of success in doing so. The report cited a study of Canadian labor law that found the union “win” rate was nine percentage points higher when card-check was in place.
As W. James Antle III noted in the December Labor Watch, when the AFL-CIO gets companies to allow card check, it wins 75 percent of the time. In 2004, even without the EFCA, more American workers unionized by using the card check mechanism than by relying on an NLRB-supervised secret ballot election. The AFL-CIO organized five times more union representation drives with card check than with a secret ballot.
The bottom line is that while unions doubtless raise compensation for their members, there is a ripple effect on the economy. Non-union workers lose jobs, aren’t hired, or are hired at lower wages; companies have a harder time competing, staying productive, and remaining in business; and inflation or, worse, stagflation can result. Decision makers in government should bear these things in mind when considering the latest brainstorm from Big Labor.
Robert VerBruggen is an associate editor at National Review.