A nonsense fact in a Super Bowl ad
“Only ten percent of people in unions today actually voted to join the union.”
— Voiceover from an ad sponsored by the Center for Union Facts which aired during the Super Bowl
A group that supports a bill in Congress that would require every unionized workplace to recertify their union every three years made this interesting claim in a TV ad that ran during the Super Bowl. The Center for Union Facts also asserted this fact in an advertisement that ran in The New York Times, featuring the dictatorial leadership of North Korea as apparent stand-ins for union leaders.
The Center for Union Facts is part of a web of pro-corporate organizations run by Rick Berman, who has also battled Mothers Against Drunk Driving, disputed evidence regarding mercury levels in fish and countered a perceived link between high-fructose corn syrup and obesity. His Web Site features a 60 Minutes profile in which he says, “I do get paid for educating people; if that’s my biggest crime, I stand accused.” (A more negative take on the Berman enterprise can be found here.)
A key feature of Berman’s ads are wicked, often sarcastic humor against “union bosses,” “food police” and the like. In this particular ad, auto mechanics bemoan their lost wages to union dues and ask each other who voted for the union. They ultimately conclude it must be the oldest guy in the shop. (One report on Monday said Berman actually portrays one of the “mechanics” in the ad.)
We take no position on the Employee Rights Act, but wondered whether this statistic was valid, especially since Sen. Orrin Hatch (R-Utah) has also cited it on the Senate floor in support of the legislation.
J. Justin Wilson, managing director of the Center for Union Facts (and a player in other Berman-run groups), said he personally calculated this statistic by examining National Labor Relations Board annual reports from 1964 (specifically Table 14) and job tenure data for unionized employees from the Current Population Survey, which is jointly sponsored by the Census Bureau and the Bureau of Labor Statistics.
Essentially, Wilson tried to estimate the proportion of employees who both would have voted for the establishment of a union at their companies and were still in their jobs. Because of limitations in the data, he said some guesswork was involved but he came up with a statistic of 9.25 percent. [Update: Wilson objects to the word “guesswork,” which is our word, not his, based on our conversation. He says the data has a bias in favor of the unions.]
Wilson suggests this statistic shows how entrenched unions have become, because workers at companies with established unions have not had a chance to decide whether or not to have a union.
However, there is a process for changing a union—or even decertifying a union, which the Center for Union Facts says on its Web Site is “not rare—several hundred take place in a typical year.” Wilson did not count “no” votes or decertification votes when doing his calculations.
The more we dug into the NLRB reports, the more dubious this statistic became. The reports show consistent support for unions when the matter has been put to a vote through the NLRB process. Wilson also acknowledged that the NLRB elections are an imperfect guide because unions can pursue an alternative process known as “card check,” in which workers sign authorization forms asking for union representation. If more than 50 percent of workers turn in cards, the vote can be canceled.
Richard Freeman, a professor at Harvard University and one of the leading labor economists in the country, in 1994-95 surveyed workers and found that “unionized workers strongly support their unions, and that the vast majority would vote to retain them in an election for union representation.” The survey also found that “nonunion workers have a less positive view of unions, but about one-third would vote for a union if given an opportunity.” (The results were published in the 1999 book “What Workers Want,” with Joel Rogers.)
Freeman and Rogers released an updated version of the book in 2006 and said that more recent surveys indicated that worker attitudes had not changed.
Wilson counters that attitudes have changed, and says his group sponsored a survey in 2009 in which 82 percent of nonunionized workers said they did not want their job to be unionized.
Union membership certainly has continued a long slide over the years, though may have begun to stabilize. Union workers made up more than 14 percent of the wage and salary workers in 1994-95; it is now 11.8 percent, virtually changed from 2010, according to the Labor Department.
Most of these unionized workers are in the public sector, such as teachers. The workers depicted in the advertisement actually are relatively unusual; only 6.9 percent of private-sector workers are unionized.
Ultimately, the Center’s statistic tells us little about worker attitudes about unions. The NLRB reports show that when there is a vote, workers usually vote to form a union. They also have the option of decertifying a union or changing their union leadership. The only thing that this statistic tells us is how many workers are still in their jobs from that original union vote.
“I don't see what the point is,” Freeman said in an e-mail. “Most people who are union today JOIN a company that is union. Very few people vote in NLRB elections because there are very few elections. So no surprise that only 10 percent or so of those who are members voted for the union….This is really about what the status quo is. It says nothing about people wanting or not wanting unions.”
Or as Jared Bernstein, a former economist for the Labor Department now with the Center for Budget and Policy Priorities, put it: “It is a bit like saying Virginia isn’t a state because none of its current residents voted for statehood.”
The Pinocchio Test
In the end, this is a nonsense fact. On its face, it may be technically correct, though involves calculations from two different data sets, which usually leads to trouble. The number of people who cast an original vote for unionization says little about current support for the union. Indeed, workers now working in union shops may have actually sought out that job specifically because it was unionized. So the dynamic in the union may actually be the opposite of what is depicted in the ad.
(UPDATE: For a counterargument, Mark Hemingway of the Weekly Standard strongly disagreed with our findings and attacked our methods.)
Check out our candidate Pinocchio Tracker
Track each presidential candidate's campaign ads .