If the PRO Act Were Law, The Amazon Union Election Would Have Looked Very Different

Adult and boy with signs in front of Whole Foods supporting Alabama Amazon workers attempting to unionize.

Alabama Amazon workers voting whether or not to unionize were supported by a Detroit-area rally on March 6. Under the proposed PRO Act, it would have been far more difficult for Amazon to interfere with the election. Photo: Jim West / jimwestphoto.com

Voting in the National Labor Relations Board-conducted election at Amazon’s Bessemer, Alabama facility came to a close on Monday. In many ways the election was unusual, marked by the sheer size of the agreed bargaining unit (5,800 warehouse workers) and the length of voting (nearly two months of mail-balloting necessitated by the pandemic). But in many more ways, the election was perfectly ordinary: the employer took advantage of its numerous lawful (and likely unlawful) advantages built into the National Labor Relations Act to wage a ferocious ideological campaign against the petitioning Retail, Wholesale and Department Store Union (RWDSU), utilizing many of the oldest weapons in management’s impressive arsenal.

This is the way things have been under the NLRA since the 1960s, the decade when scholars and activists first began criticizing and proposing major changes to the NLRB’s election machinery. Rather than dwell on the latest episode in labor law’s broken system of certifying bargaining representatives, I thought it would be helpful to analyze how the election would have been different if the Protecting the Right to Organize Act—currently pending debate and a vote in the Senate—were already law. I’ve broken down the changes for simplification.

  • Amazon would be prohibited from conducting captive audience meetings. Research shows that employers deploy these meetings in almost 90 percent of NLRB elections, in no small part because they’re so effective in channeling the employer’s message and coercing the captive employees. Amazon was no exception to this phenomenon. Under the PRO Act, the very use of these meetings would constitute an Unfair Labor Practice as a violation of Section 8(a)(1) of the NLRA.

  • Amazon would not have had legal standing to participate in representation proceedings. This is important for pre-election conduct, as Amazon used its time at the December hearings to inflate the proposed bargaining unit to a wall-to-wall unit and attempt to stall for time to begin campaigning against the union. These sort of dual stalling-and-gerrymandering tactics would be impossible under the PRO Act, as the pre-election hearing would not have included Amazon at all.

  • The Union’s proposed bargaining unit would have likely been accepted. Through the PRO Act’s codification of the Obama Board’s Specialty Healthcare decision, a union’s proposed bargaining unit is to be accepted by the Labor Board so long as the unit shares a sufficient “community of interest.” Combined with the removal of employer standing, these changes would have all but guaranteed the vote covered only the 1,500 or so employees the union originally petitioned for instead of the wall-to-wall compromise that was agreed upon by the Union as a means of speeding up the process.

  • If the union won, it would be guaranteed to secure a first contract. The PRO Act installs a system of interest arbitration to take hold if the parties do not reach an initial collective bargaining agreement within 90 days of a union’s certification following an election victory. This would have taken away a key campaign talking point from Amazon, which was able to truthfully tell its employees that the Union was not assured to provide its members a contract even if it won the election.

  • The election would have been over months ago. The PRO Act provides for faster pre-election proceedings (through codification of the Obama Board’s 2014 “quickie election” rules) and the option for electronic voting at the union’s preference. While the mail-ballot election in many ways has likely benefitted the union, the sheer size of the unit and USPS’s pandemic-era issues likely convinced the NLRB that voting needed to last almost two months to give everyone a chance to vote. Amazon has been allowed to continue campaigning against the Union during this time. Electronic voting could logically be done faster and circumvent any post-office problems, in addition to undercutting the employer’s built-in influences from in-person voting on company property.


    Give $10 a month or more and get our "Fight the Boss, Build the Union" T-shirt.

  • Amazon would face far greater consequences for unlawfully interfering with the election. As I’ve discussed, current NLRB case law makes bargaining orders a very rare occurrence. Combined with the agency’s lack of punitive remedies, employers are greatly incentivized to commit unfair labor practices or otherwise interfere with the “laboratory conditions” of the vote, as the worst that will usually happen is a re-run of the election. Under the PRO Act, a union that enjoyed majority support before filing its petition would be certified through card check in elections that have been set aside because the employer committed an unfair labor practice or otherwise interfered with a fair election, and where the employer did not demonstrate that the violation or other form of interference was unlikely to have affected the outcome of the election. A simple 8(a)(1) violation concerning supervisor threats or promises could conceivably be enough to certify a losing union under the PRO Act’s strict language, so Amazon would have been forced to strictly regulate both its and its agents’ conduct.


    While Amazon’s vast wealth and resources would make the PRO Act’s provision for monetary fines for ULPs less powerful than at smaller companies, these fines can pile up for a national employer that engages in similar unlawful conduct at multiple facilities (as Amazon is now accused of doing). Moreover, the Act’s installation of personal liability for bad individual actors would allow the NLRB to hold Jeff Bezos, Dave Clark, or any other member of Amazon management personally responsible for any ULPs that they directed themselves, which could be a powerful worker rallying cry.

  • The union would have far more economic weapons at its disposal in exacting leverage and building solidarity. The PRO Act expressly legalizes the use of intermittent strikes, partial strikes, and production slowdowns. While the Union has seemed content to rely on peaceful tactics and the NLRB’s election machinery for certification, unions would not be confined to such tactics in the face of employer hostility. These sorts of job actions could be immensely effective against Amazon given its just-in-time methods underlying its global supply chain. If nothing else, it’s more tools in the workers’ tool box that the employer has to account for.

  • Alabama’s “Right-to-Work” law would be repealed. While RTW laws are often exaggerated in their legal effect, as they only affect the specific issue of union security clauses and unions’ right to collect dues, most people understand these laws to have powerful signaling effects on politicians, businesses, communities, and the workers themselves. With all RTW laws repealed under the PRO Act, Alabama employers like Amazon would not have the head start of overseeing a workforce that has been conditioned for years on the idea that unions don’t belong in the Deep South.

Of course, every provision in the PRO Act matters for fixing our broken labor laws and revitalizing the United States labor movement, creating a new ecosystem for worker power and organizing. But these changes I’ve highlighted would have had immediate impact on the legal structures at play and the parties’ strategies before the Labor Board. Whatever the result of the vote, the PRO Act is vital to restoring the NLRA’s promise of bringing industrial democracy to American workplaces.

Brandon Magner is a labor lawyer. He writes about unions and the law at his Substack, Labor Law Lite. This story originally appeared there.