On July 2^nd, Harvard announced that it had decided to remove a significant number of graduate students from the bargaining unit represented by HGSU. The University said that it would no longer “include any students in the bargaining unit receiving stipends to pursue research toward their degrees.” The explanation for this decision was brief. The email announcement stated only that “[u]nder the law, only employees can unionize. These stipendee students are not employees because they do not perform services for the university in exchange for compensation, a determination that has been recently clarified by multiple decisions interpreting the NLRA.” 

This conclusion runs counter to recent history in the law and at Harvard. First, the law. Under current NLRB precedent, graduate student TAs and RAs are employees of the universities where they work so long as they meet the common law definition of employee and irrespective of whether their relationship with the University could be described as primarily academic or educational. Under the rule of Columbia University, graduate students are employees within the meaning of federal labor law and thus entitled to be included in a bargaining unit where “they perform work, at the direction of the university, for which they are compensated.” Applying this test in Columbia, the Board found that an extremely wide range of graduate student research and teaching assistants met the definition of employee and were appropriately included in a bargaining unit.

Next, recent history at Harvard.  Up until the University’s announcement, all stipendee research assistants had been included in the HGSU bargaining unit and had been covered by the parties’ collective bargaining agreement.

So, what changed?  Since Columbia was decided, a single regional director of the NLRB has decided two cases in which she upheld challenges to the inclusion of certain graduate students in bargaining units at MIT and Brown. In the MIT case, the regional director’s decision was approved by the NLRB on the ground that the graduate students at issue:

do not perform work controlled by the Employer in exchange for compensation. Rather, they perform research (or, occasionally, teach) to further their own academic purposes and are provided with funding to do so regardless of whether their activities also benefit the Employer. 

The NLRB’s approval of the regional director’s decision in MIT explicitly did not rely on her finding “that the fellows are not statutory employees because ‘the compensation received by the fellows is not directly tied to completing particular tasks.’” In other words, the rule of the MIT case is the straightforward proposition that graduate students who do not “perform work controlled by the Employer in exchange for compensation” are not employees. 

It is important to emphasize that neither of the regional director’s decisions alters the prevailing law regarding graduate student employee status that the NLRB established in Columbia. That case is binding precedent and will remain so unless and until it is reversed by another decision of the NLRB. 

Next, what changed at Harvard? One salient change is that the collective bargaining agreement that had included the stipendee students expired.  Under longstanding Board doctrine, the expiration of the collective bargaining agreement gave Harvard the first opportunity to make its argument about the status of the stipendee students.  Of course, much else has changed since Harvard first agreed to include the stipendee students in the bargaining unit.  We have seen the return of President Trump, who previously appointed members of the NLRB who were hostile to collective bargaining generally and at higher education institutions specifically.  

The key question is whether any of these changes — in the law or at Harvard — mean that Harvard is justified in excluding the stipendee students from the bargaining unit.  Many subsidiary questions inform this larger one: Is the work of these graduate students controlled by Harvard in exchange for compensation? Do these graduate students only teach occasionally? Do they pursue research only to advance their own, and not Harvard’s, academic purposes?  Are their stipends not linked to any benefit Harvard may derive from their research? And, finally, do they perform any other tasks for the benefit of Harvard for which they receive their stipends, such as guiding or supporting undergraduates?

At bottom, the relevant question is how many of the stipendee students satisfy the test of performing no work, controlled by Harvard University, in exchange for compensation. While this is a highly fact-specific test that will require lots of line drawing, it would be surprising if all graduate students “receiving stipends to pursue research toward their degrees” meet the test. It would be surprising, in other words, if no Harvard graduate student who receives a stipend does any work that is controlled by Harvard. Whatever the facts, we trust that Harvard will respect the law it cites and exclude only those graduate students who truly perform no work, controlled by the University, in exchange for compensation. And that when the call is a close one, we hope that the University will err on the side of according collective bargaining rights to these critical members of our community.