The Joy Silk Doctrine and Cemex

The new rule, the board emphasized in its opinion, is meant to expedite the union recognition process. Employees may “designate” a union as their bargaining representative when a majority sign cards that authorize the union to do so. Under prior law, an employer was free to do nothing when presented with such a situation. This response placed the onus for election-petition filling on the union, making a board-supervised election nearly the only path toward collective bargaining — a long, drawn-out affair with a highly uncertain outcome.

And the outcomes were frequently tragic: in many cases, when presented with a demand to bargain, the employer’s answer has been to terminate union supporters to eliminate pro-union workers and intimidate others. Such interference would often result in the board “setting aside” the election, ordering the employer to cease and desist, and scheduling another election in the future. The employer could rinse and repeat until it had crushed the union drive. The Cemex decision attempts to deter this kind of union busting.

The new opinion is, in some respects, less an innovation and more a reversion to an older state of the law. Under the board’s 1949 Joy Silk decision, an employer was legally obligated to bargain with a card-majority union, unless the employer had “good-faith” reasons — honest reasons, not made-up ones — to doubt the union’s majority support. As the board understood at that time, an NLRB-supervised election was not the only way for workers to choose their union. The National Labor Relations Act refers only to representatives “designated or selected” by employees; it does not prescribe any one method by which workers can choose a union. So there is nothing radical or unprecedented about the board’s decision.

In fact, the board’s general counsel, Jennifer Abruzzo — in effect the board’s prosecutor — asked it to reinstate the Joy Silk standard in the Cemex proceedings. The board declined to do so, acknowledging federal-court criticisms of Joy Silk made before the NLRB abandoned it. According to the courts, attempting to divine the honesty of the employer’s refusal to recognize the union was too difficult and too subjective.

In that respect, the board’s Cemex decision is an innovation. Rather than attempting to read the employer’s mind, the ruling puts a choice to the employer: allow a fair election now or bargain with the union later. In other words, the employer is not obligated to bargain merely on a showing of majority support for the union. The employer may still petition for an election if it has questions about majority status. But if it does so, it must respect the process; if it does not, the employer will be ordered to bargain with the union.

Union advocates have unsurprisingly extolled the Cemex decision. Last week, for instance, the American Prospect ran a piece by Harold Meyerson titled “Biden’s NLRB Brings Workers’ Rights Back From the Dead.” The article’s subheadline declared: “A decision last Friday makes union organizing possible again.”

The decision certainly makes union recognition easier. But I would also laud the ruling purely for reducing the role of the board in supervising the union recognition process. Prior to Cemex, the only path to recognition in the face of an unwilling employer was a board-supervised election. And securing a fair election depended on the NLRB’s ability — and willingness — to police the election so that it is conducted in, in the board’s words, “laboratory conditions.”

By moving the process closer to the more informal method of card-check recognition, workers and unions can avoid the delays, costs, and tedious legal minutiae of paper battles before the board. This is a good thing, independent of the immediate consequences of the decision.

The Effects of the Ruling

Confidence in Cemex’s positive effects runs high, and no one would be more thrilled than me to see union recognition victories soar in the wake of the board’s decision. But a dose of skepticism is warranted here.

At the OnLabor blog, Tascha Shahriari-Parsa wrote after the ruling that Cemex will “significantly dissuade flagrant transgressions of our labor laws — possibly more than any Board decision of the last half-century.” Last year in the American Prospect, Meyerson wrote, “In the five years before Joy Silk was struck down, charges of employer intimidation totaled about 1,000 cases a year.” After the board rejected Joy Silk, “charges exploded to a peak of 6,493 in 1981, after which they fell along with unionization efforts generally.” Under Joy Silk, when the requirements for securing union recognition were less onerous, employers found little use in resisting through threats, firings, or other unfair labor practices. After Joy Silk, employers found it much more attractive to undermine the union’s election drive when the standard recourse for the employer’s recognition refusal was to run, and often rerun, a board election.

But it is hard to know whether Meyerson’s figures reflect the change in the law, as he implies, or rather other, more complicated changes in culture, society, and the economy. Joy Silk was effectively abandoned in NLRB v. Gissel Packing Co., Inc. in 1969 and formally overturned by Linden Lumber Division, Summer & Co. v. NLRB in 1971 just as the postwar capital-labor “accord” — or, perhaps more accurate, détente — was coming apart. Employers were ramping up their all-out assault on unions at this time, and it would be hard to credit that sea change to a single NLRB ruling.

Joy Silk had also been substantially weakened prior to its abandonment. At the time Joy Silk was decided, an employer’s substantial unfair labor practices were sufficient evidence of the employer’s bad faith in refusing recognition. Under later decisions, substantial unfair labor practices became virtually necessary to demonstrate bad faith. So it is unclear how effective Joy Silk was in the years before its abandonment.

No Substitute for Worker Solidarity

The extent of employer intimidation may also not be the best measure of success of board rulings in aiding unionization. The effectiveness of the Cemex rule ultimately rests on the board’s bargaining order remedy in cases where the employer attempts to undermine the election. But bargaining orders themselves have not been terribly effective remedies, especially against employers who are intent on resisting unionization.

In a 1997 paper, professors Terry A. Bethel and Catherine Melfi examined the outcomes of a small sample of NLRB bargaining orders. Only about 20 percent of bargaining orders resulted in at least one collective bargaining agreement. Most significant, in around 46 percent of bargaining orders, the union abandoned bargaining — certainly as a result of employer intransigence. In 25 percent of cases, the employer’s business closed, and in 8 percent of cases the union ended up being decertified as the employees’ bargaining representative.

To be clear, these numbers represent the results of what are called Gissel bargaining orders, which are issued when employer intimidation is particularly severe and pervasive. Nevertheless, the numbers are not reassuring. They demonstrate that employers who adamantly resist bargaining can eventually prevail, despite the illegality of their conduct.

All of this underscores an important lesson: ultimately, there is no substitute for solidarity in the workplace. Workers are their own enforcement policy, and when a culture of solidarity prevails workers’ ability to collectively withhold their labor is sufficient deterrence against a recalcitrant employer.

Without a doubt, building such a culture is hard to do and difficult to maintain. Legal enforcement presents a supposedly faster and less painful “easy way out,” and many union advocates understandably decry the board’s weak enforcement powers. But stronger government enforcement of collective bargaining rights and other labor protections have their limits. If you believe government intervention is necessary in order to help unions improves wages, hours, safety, and other working conditions, there is no reason why the government cannot simply dispense with unions and regulate the workplace directly.

In fact, the role of unions goes beyond these narrowly economic functions. Besides enforcing the law, unions are also “schools of democracy” and fulfill other civil-society functions, like building solidarity across racial divides. But the problem in this case is that state-dependent unions cannot satisfy these state-independent roles. Under a regime of labor relations where unions are heavily dependent on legal protection, unions are more likely to become the unwitting creatures of the state or political parties, as demonstrated by the experience of other countries.

This is why Cemex’s small but significant deregulatory step is a virtue itself. Making workers the authors of their own organizing is an essential step to rebuilding a culture of solidarity. A union culture or labor law that demands nothing from workers cannot inspire a movement to transform society.