Labor Law Must Include All Workers

In January of this year, we published a comprehensive set of recommendations for reforming U.S. labor law.¹Sharon Block and Benjamin Sachs, Clean Slate for Worker Power, Labor and Worklife Program, Harvard Law School (January 23, 2020). Although the recommendations were extensive, the theory that lay behind them was straightforward: our country is facing dual crises of political and economic inequality, and we can help address those crises by giving working people greater collective power in the economy and in politics. Although progressives and conservatives disagree on many things, we all ought to agree that the stark inequalities that now pervade American life constitute grave threats. Politically, the viability of our democracy is threatened by a government that responds to the views of the wealthy but not to those of the poor and middle class. Economically, the viability of our community life is threatened by the fact that that we live in a country where it would take an Amazon worker 3.8 million years, working full time, to earn what Jeff Bezos alone now possesses.

Saving American democracy and American communities will take a wide variety of interventions, but labor law reform must be one of them. In fact, much of the explanation for our current crisis of economic inequality is the decline of the labor movement. Unions redistribute wealth—from capital to workers, from the rich to the poor and middle class— and without unions, we have not had an adequate check on economic concentration. The decline of the labor movement also accounts for much of the current crisis of political inequality. When unions were active and strong, they helped ensure that the government was responsive to the needs and desires of the poor and middle class. Without unions, these poor and middle-class Americans have lost their most effective voice in our democracy. We have seen the consequences of this decline in unionization play out dramatically during the pandemic and recession, which have had devastating consequences for workers trying to navigate their physical and economic survival with so little collective power.

We do have a statute that is supposed to enable working people to form and join unions, but that law – the National Labor Relations Act (NLRA), passed in 1935 – is badly broken and outdated. In fact, in our view, the NLRA is so fundamentally flawed, so fundamentally incapable of providing workers with a viable path to effective unionization, that a complete rewrite of national labor policy is called for. That’s why we titled our reform recommendations Clean Slate and outlined how we would totally redesign American labor law from the ground up. Central to these recommendations is the idea that American workers need a new system of labor law, one that equips them to build power at the level of the workplace and across industries, in corporate boardrooms, and in our democracy.

We reject the idea that the kind of change we need can be accomplished through individual tweaks of the existing structure. We resist the temptation to choose which of our reform recommendations are the most important, believing as we do that the proposals build on and support each other in the creation of a comprehensive, workable system of labor law.

We do not, however, have room here to review this comprehensive call for reform. Instead, we will outline where we grounded all the other Clean Slate proposals: with the effort to expand the inclusiveness of labor law’s reach.

Indeed, one of the primary failings of the NLRA is its lack of inclusivity. Put bluntly, from the outset, the labor statute intentionally excluded large segments of the labor market populated predominantly by workers of color and women. As Ira Katznelson has shown, in order to secure passage of the NLRA (and companion legislation including the Fair Labor Standards Act), President Roosevelt needed the votes of the Southern wing of the Democratic Party.²Ira Katznelson, When Affirmative Action Was White (New York: W.W. Norton & Company, 2005) 53-71. Southern Democrats, however, were unwilling to support the New Deal labor laws without assurances that the enactment of those laws would leave undisturbed the system of Jim Crow segregation that defined Southern labor markets. Roosevelt, and Northern liberals, accepted the deal, and the statute was passed with explicit carveouts for agriculture and domestic work, “the most widespread black categories of employment.”³Ibid., 55. Indeed, at the time of the NLRA’s enactment, nearly half of Black men and fully ninety percent of Black women worked in either the agricultural or domestic sector.⁴Claudia Goldin, Understanding the Gender Gap: An Economic History of American Women (New York: Oxford University Press, 1990), 74–75.

Eighty years later, these discriminatory exclusions – baked into the law at its passage – continue to have profound effects on workers of color and women. In fact, women and people of color constitute nearly one-hundred percent of the agricultural and domestic labor forces today, and so the exclusion of these categories of employment means that union protections are simply unavailable to huge numbers of workers of color and women. Moreover, other legal developments have reinforced and deepened these exclusions. The statute, for example, excludes from coverage anyone working as an independent contractor. And, for many parts of the labor market, this exclusion makes sense: we probably don’t want doctors and plumbers unionizing and fixing their prices. But with the explosion of the mis-use of the independent contractor designation – by massive employers like Uber, Amazon, and FedEx – this exclusion has meant the denial of union rights to millions of workers who ought to enjoy them. And, again, these sectors are predominantly made up of workers of color. Finally, and although the labor statute is itself silent on this point, the courts have carved out from labor law’s full protection immigrants who are working in the country without immigration authorization.⁵Hoffman Plastic Compounds, Inc. v. N.L.R.B., 535 U.S. 137 (2002). This exclusion undermines union access not only for the millions of workers directly targeted by the carveout, but also for the many millions more who count such immigrants as co-workers and whose ability to organize depends on joining together with them.

Put together, these race- and gender-salient exclusions – and similar ones in cognate employment laws – have helped construct a set of labor rights not accessible to all on equal terms. To put it mildly, we are seeing today the tragic consequences of this inequitable labor market as workers of color and women suffer the economic consequences of the COVID-19 pandemic in highly disproportionate ways, including higher rates of unemployment,⁶Patricia Cohen and Ben Casselman, “Minority Workers Who Lagged in a Boom Are Hit Hard in a Bust,” The New York Times (June 6, 2020). lower access to unemployment benefits,⁷Ava Kofman and Hannah Fresques, “Black Workers Are More Likely to Be Unemployed but Less Likely to Get Unemployment Benefits,” ProPublica (August 24, 2020). and higher incidence of COVID-related hospitalization and death.⁸Tiffany Ford, Sarah Reber, and Richard Reeves, “Race gaps in COVID-19 deaths are even bigger than they appear,” The Brookings Institution (June 18, 2020).

Any reform of our labor law must therefore begin by ensuring that all workers – including those who unjustly have been excluded since the outset – have access to union rights. Racial and gender equity should certainly be a bipartisan project, and so it should be possible to reach agreement on a viable set of reforms designed to increase labor law’s inclusivity. At a minimum, a reformed statute should dispense with the categorical exclusion of agricultural and domestic workers – exclusions that have been inseparable from Jim Crow since they were drafted.

Given the structure of these industries, ending the exclusions would need to be accompanied by other statutory amendments. For example, the inclusion of agricultural workers would require changes to the law’s constraints on the object of collective action. Agricultural workers are employed by growers, but organizations like the Coalition of Immokalee Workers have shown that applying pressure to growers doesn’t succeed in raising wages. Why? Because downstream players in the industry, like fast food restaurants and food retailers (e.g., Wal-Mart), effectively set wages by keeping growers’ margins so tight. Successful labor organizing in agriculture therefore depends on workers’ ability to influence these restaurants and retailers, which means the law must protect campaigns that apply pressure down the production chain. In particular, the law must offer protection to campaigns designed to persuade consumers to pressure restaurants and retailers to pay the growers more and ultimately to pass those increased commodities prices on to the workers. The NLRA, however, often prohibits these types of campaigns under the ban on so-called secondary boycotts. To make NLRA organizing an effective tool for agricultural workers – and others who work in similarly structured industries – secondary boycott rules would therefore need to be adjusted.

Similarly, the inclusion of domestic workers would require changes to the NLRA’s constraints on which workers can organize together. In our current system, workers may only organize within the bounds of the single firm – or “employer,” to use the statutory term – where they work. Because domestic workers are often the sole employee of their employer – the homeowner or family for which they work – this limitation means that domestic workers have the right to organize unions of one worker; the limitation, that is, renders the right to organize meaningless.  To address this problem, adjustments to the rules of bargaining units would be required. Ideally, sectoral bargaining would be permitted, thus enabling domestic workers – and others – to organize and set conditions across entire industries, a topic that David Rolf and Oren Cass have discussed here.

In our view, ending the agricultural and domestic worker exclusions should be accompanied by two other moves, both of which would also repair some of the racial and gender inequities that have plagued labor law for too long. First, we would amend the statute to make clear that union protections extend to all people working in the United States, irrespective of immigration status. Such a reform would have clear benefits for workers’ ability to organize: no longer would employers be able to defeat organizing drives by dividing workers along immigration lines. And while we understand that such a reform might generate opposition on the other side of the aisle, it actually should find support regardless of one’s view of the immigration issue at stake: as Justice Breyer has explained, if you deny labor law’s protection to unauthorized workers – as current law does – you make it more likely that employers will seek to hire them.

Finally, we would amend the statute’s definition of employee such that only genuine independent contractors were excluded, and actual employees – like Uber drivers and Amazon delivery workers – could not be misclassified and denied union rights. As noted, because so many gig workers – and workers in other industries where misclassification rates are high – are immigrants, workers of color, and women, this move would also help make labor law significantly more inclusive. We would accomplish this reform by replacing the statute’s current definition of employee with the so-called “ABC test” for employment, which is now in place in several state jurisdictions.⁹“Employee or Contractor? The Complete List of Worker Classification Tests By State,” Wrapbook (October 25, 2019). Importantly, the ABC test starts with the presumption that workers are employees, covered by labor protections, and puts the burden on employers to prove the exclusion.  Employers must prove each of the following three factors: (A) that the employer doesn’t exert control over the workers at issue; (B) that the work performed is outside the usual scope of the employer’s business; and (C) that the worker is engaged in an independent trade, occupation or business. The adoption of the ABC test would extend collective bargaining rights to millions of workers who need and deserve labor law’s protection, without sweeping in true entrepreneurs, freelancers, or businesspeople.

As we said at the outset, much more needs to be done. Inclusion is a necessary first step toward fixing America’s broken labor law system, but it is only a first step because workers deserve to be included in a labor law that actually works – a labor law that enables them to build and exercise meaningful economic and political power. Constructing that kind of labor law would mean creating a diverse array of representational structures at the workplace, guaranteeing opportunities for sectoral bargaining, building power in the corporate boardroom, and expanding opportunities for political participation, and doing so in a way that ensures that all these pieces work together in a coherent whole. If we succeeded in building that kind of labor law, we could make real progress toward addressing the dual crises of economic and political inequality that now threaten American communities and American democracy.