Microsoft Word - Board Decision.docx

364 NLRB No. 90

NOTICE: This opinion is subject to formal revision before publication in the

bound volumes of NLRB decisions. Readers are requested to notify the Executive

Secretary, National Labor Relations Board, Washington, D.C. 20570, of any

typographical or other formal errors so that corrections can be included in the

bound volumes.

The Trustees of Columbia University in the City of

New York and Graduate Workers of Columbia–

GWC, UAW. Case 02–RC–143012

August 23, 2016

DECISION ON REVIEW AND ORDER

Y CHAIRMAN PEARCE AND MEMBERS MISCIMARRA,

HIROZAWA, AND MCFERRAN

The threshold question before us is whether students

who perform services at a university in connection with

their studies are statutory employees within the meaning

of Section 2(3) of the National Labor Relations Act.

Here, after a hearing directed by the Board, the Regional

Director applied Brown University, 342 NLRB 483

(2004), where the Board found that graduate student as-

sistants were not employees within the meaning of Sec-

tion 2(3), and dismissed a petition filed by the Graduate

Workers of Columbia-GWC, UAW, which seeks to rep-

resent both graduate and undergraduate teaching assis-

tants, as well as graduate research assistants.

The Board

granted review in this case on December 23, 2015, and

then issued a notice and invitation to file briefs, identify-

ing the primary issue presented, as well as subsidiary

issues that would follow if Brown University were over-

ruled.

We have carefully considered the record, the

The petition defined the bargaining unit sought as follows:

Included: All student employees who provide instructional services,

including graduate and undergraduate Teaching Assistants (Teaching

Assistants, Teaching Fellows, Preceptors, Course Assistants, Readers

and Graders): All Graduate Research Assistants (including those

compensated through Training Grants) and All Departmental Re-

search Assistants employed by the Employer at all of its facilities, in-

cluding Morningside Heights, Health Sciences, Lamont-Doherty and

Nevis facilities.

Excluded: All other employees, guards and supervisors as defined in

the Act.

On January 16, 2016, the Board invited the parties and interested

amici to file briefs addressing the following four issues:

1. Should the Board modify or overrule Brown University, 342

NLRB 483 (2004), which held that graduate student assistants who

perform services at a university in connection with their studies are not

statutory employees within the meaning of Section 2(3) of the Nation-

al Labor Relations Act?

2. If the Board modifies or overrules Brown University, supra, what

should be the standard for determining whether graduate student assis-

tants engaged in research are statutory employees, including graduate

student assistants engaged in research funded by external grants? See

New York University, 332 NLRB 1205, 1209 fn. 10 (2000) (relying on

Leland Stanford Junior University, 214 NLRB 621 (1974)).

positions of the parties and the amici,

the reasoning of

the Brown University Board, and the views of our dis-

senting colleague, who endorses Brown University (as

well as advancing arguments of his own).

For the reasons that follow, we have decided to over-

rule Brown University, a sharply-divided decision, which

itself overruled an earlier decision, New York University,

332 NLRB 1205 (2000) (NYU). We revisit the Brown

University decision not only because, in our view, the

Board erred as to a matter of statutory interpretation, but

also because of the nature and consequences of that error.

The Brown University Board failed to acknowledge that

the Act does not speak directly to the issue posed here,

which calls on the Board to interpret the language of the

statute in light of its policies. The Brown University

Board’s decision, in turn, deprived an entire category of

workers of the protections of the Act, without a convinc-

ing justification in either the statutory language or the

policies of the Act.

As we will explain, our starting point in determining

whether student assistants are covered by the Act is the

broad language of Section 2(3), which provides in rele-

vant part that “[t]he term ‘employee’ shall include any

employee,” subject to certain exceptions

—none of

which address students employed by their universities.

The Brown University Board held that graduate assistants

cannot be statutory employees because they “are primari-

ly students and have a primarily educational, not eco-

nomic, relationship with their university.”

We disagree.

The Board has the statutory authority to treat student

assistants as statutory employees, where they perform

work, at the direction of the university, for which they

3. If the Board concludes that graduate student assistants, terminal

masters degree students and undergraduate students are statutory em-

ployees, would a unit composed of all these classifications be appro-

priate?

4. If the Board concludes that graduate student assistants, terminal

masters degree students and undergraduate students are statutory em-

ployees, what standard should the Board apply to determine whether

they constitute temporary employees?

Briefs were filed in support of the Petitioner by: American Asso-

ciation of University Professors (AAUP); American Federation of

Labor and Congress of Industrial Organizations (AFL-CIO); American

Federation of Teachers (AFT); Ellen Dannin, Attorney; The General

Counsel of the NLRB; Individual Academic Professors of Social Sci-

ence and Labor Studies (IAP); National Association of Graduate-

Professional Students (NAGPS); Service Employees International

Union and Committee of Interns and Resident, SEIU Healthcare

(SEIU-CIR); and United Steelworkers (USW). Filing in support of

Columbia were: American Council on Education (ACE), et al.; Brown

University et al.; Higher Education Council of the Employment Law

Alliance (HEC); and National Right to Work Legal Defense and Educa-

tion Foundation (NRW).

29 U.S.C. §152(3).

Brown University, 342 NLRB at 487.

DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD

are compensated. Statutory coverage is permitted by

virtue of an employment relationship; it is not foreclosed

by the existence of some other, additional relationship

that the Act does not reach.

The unequivocal policy of the Act, in turn, is to “en-

courag[e] the practice and procedure of collective bar-

gaining” and to “protect[ ] the exercise by workers of full

freedom of association, self-organization, and designa-

tion of representatives of their own choosing.”

Given

this policy, coupled with the very broad statutory defini-

tions of both “employee” and “employer,” it is appropri-

ate to extend statutory coverage to students working for

universities covered by the Act unless there are strong

reasons not to do so.

We are not persuaded by the

Brown University Board’s self-described “fundamental

belief that the imposition [sic] of collective bargaining on

graduate students would improperly intrude into the edu-

cational process and would be inconsistent with the pur-

poses and policies of the Act.”

This “fundamental be-

lief” is unsupported by legal authority, by empirical evi-

dence, or by the Board’s actual experience.

Thus, we hold today that student assistants who have a

common-law employment relationship with their univer-

sity are statutory employees under the Act. We will ap-

ply that standard to student assistants, including assis-

tants engaged in research funded by external grants. Ap-

plying the new standard to the facts here, consistent with

the Board’s established approach in representation cases,

we conclude (1) that all of the petitioned-for student-

assistant classifications consist of statutory employees;

(2) that the petitioned-for bargaining unit (comprising

graduate students, terminal Master’s degree students, and

undergraduate students) is an appropriate unit; and (3)

that none of the petitioned-for classifications consists of

temporary employees who may not be included in the

unit. Accordingly, we reverse the decision of the Re-

gional Director and remand the proceedings to the Re-

gional Director for further appropriate action.

29 U.S.C. §151.

Cf. NLRB v. Kentucky River Community Care, Inc., 532 U.S. 706,

711–712 (2001) (upholding Board’s rule allocating burden of proof to

party asserting supervisory exception to Sec. 2(3), citing broad defini-

tion of “employee”).

342 NLRB at 493. Under the Act, collective bargaining can never

be “imposed” on employees by the Board; rather, the Act guarantees

employees full freedom of choice in deciding whether or not to seek

union representation, based on majority support. See National Labor

Relations Act, §§1, 7, & 9, 29 U.S.C. §§151, 157, 159.

. OVERVIEW OF PRECEDENT

A. Board precedent prior to Brown University

The Board has exercised jurisdiction over private,

nonprofit universities for more than 45 years.

During

that time, the Board has permitted collective bargaining

by faculty members at private universities and has had

frequent occasion to apply the Act in the university set-

ting.

The Board first considered the status of graduate

student assistants in Adelphi University, 195 NLRB 639

(1972). There, the Board held that graduate assistants

should be excluded from a bargaining unit of university

faculty members because they did not share a community

of interest with the faculty. However, the Adelphi Board

did not address whether the student assistants were statu-

tory employees. Two years later, the Board held that

certain university research assistants were “primarily

students” and thus not statutory employees, observing

that the relationship between the research assistants and

the university was “not grounded on the performance of

a given task where both the task and the time of its per-

formance is designated and controlled by an employer.”

The Leland Stanford Junior University, 214 NLRB 621,

623 (1974). For similar reasons, the Board dismissed

representation petitions for house staff at teaching hospi-

tals in Cedars-Sinai Medical Center, 223 NLRB 251

(1976) and St. Clare’s Hospital, 229 NLRB 1000

(1977).

In Boston Medical Center, 330 NLRB 152 (1999), the

Board overruled Cedars-Sinai and St. Clare’s Hospital

and held that interns, residents, and clinical fellows

(house staff) at a teaching hospital were statutory em-

ployees entitled to engage in collective bargaining with

the hospital over the terms and conditions of their em-

ployment. In so holding, the Board emphasized the

broad scope of Section 2(3) and noted the absence of any

statutory exclusion for students or house staff. And, con-

trary to St. Clare’s Hospital, in Boston Medical Center

the Board found that the policies of the Act would be

advanced by extending full statutory protection to house

staff.

The Board first held that certain university graduate

assistants were statutory employees in its 2000 decision

in NYU, supra. In NYU, the Board examined the statuto-

ry language of Section 2(3) and the common law agency

See Cornell University, 183 NLRB 329 (1970).

See, e.g., Bradford College, 261 NLRB 565 (1982).

In St. Clare’s Hospital, the Board clarified Cedars-Sinai, observ-

ing that “national labor policy . . . preclude[d] the extension of collec-

tive-bargaining rights and obligations to situations such as the one”

presented, which implicated predominantly academic, not economic,

interests. 229 NLRB at 1002.

330 NLRB at 160.

COLUMBIA UNIVERSITY

doctrine of the conventional master-servant relationship,

which establishes that such a “relationship exists when a

servant performs services for another, under the other’s

control or right of control, and in return for payment.”

In so doing, the Board determined that “ample evidence

exists to find that graduate assistants plainly and literally

fall within the meaning of ‘employee’ as defined in Sec-

tion 2(3)” and by the common law.

The Board’s inter-

pretation was based on the breadth of the statutory lan-

guage, the lack of any statutory exclusion for graduate

assistants, and the undisputed facts establishing that the

assistants in that case performed services under the con-

trol and direction of the university for which they were

compensated.

The NYU Board also relied on Boston Medical Center

to support its policy determination that collective bar-

gaining was feasible in the university context.

In Bos-

ton Medical Center, the Board held that interns, residents

and clinical fellows (collectively, house staff) at a teach-

ing hospital were statutory employees entitled to engage

in collective bargaining with the hospital over the terms

and conditions of their employment.

After 16 years,

Boston Medical Center remains good law today—with

no evidence of the harm to medical education predicted

by the dissenters there—but NYU was overruled only a

few years after it was decided, by a sharply divided

Board’s 2004 decision in Brown University.

B. Brown University

In Brown University, the majority described NYU as

“wrongly decided,” and invoked what it called the “un-

derlying fundamental premise of the Act,” i.e. that the

Act is “designed to cover economic relationships.”

The

Board further relied on its “longstanding rule” that the

Board will decline to exercise its jurisdiction “over rela-

tionships that are ‘primarily educational.’”

In so decid-

ing, the Brown University majority rejected NYU’s reli-

ance on the existence of a common-law employment

relationship between the graduate students and the uni-

versity, stating that “[e]ven assuming arguendo” such a

relationship existed, “it does not follow that [the graduate

assistants] are employees within the meaning of the

Act.”

That issue was “not to be decided purely on the

basis of older common-law concepts,” but rather by de-

termining “whether Congress intended to cover the indi-

332 NLRB at 1206.

Id.

330 NLRB at 164-65.

342 NLRB at 483, 488.

Id. at 488.

Id. at 491.

vidual in question.”

Disavowing the need for empirical

analysis, the Brown University majority instead relied on

what it perceived to be a fundamental tenet of the Act

and a prerequisite to statutory coverage: a relationship

that is primarily economic in character, regardless of

whether it constitutes common-law employment.

In addition to its declaration that graduate assistants, as

primarily students, were necessarily excluded from statu-

tory coverage, the Brown University Board also articulat-

ed a policy rationale based almost exclusively on the

overruled decision in St. Clare’s Hospital, supra, finding

that the St. Clare’s Board had correctly “determined that

collective bargaining is not particularly well suited to

educational decisionmaking and that any change in em-

phasis from quality education to economic concerns will

‘prove detrimental to both labor and educational poli-

cies.’”

That determination ostensibly was supported by

several factors: (1) that the student-teacher relationship is

based on mutual academic interests, in contrast to the

conflicting economic interests that inform the employer-

employee relationship; (2) that the educational process is

a personal one, in contrast to the group character of col-

lective bargaining; (3) that the goal of collective bargain-

ing, promoting equality of bargaining power, is “largely

foreign to higher education”; and (4) that collective bar-

gaining would “unduly infringe upon traditional academ-

ic freedoms.”

The Brown University dissenters, in stark contrast,

noted that “[c]ollective bargaining by graduate student

employees” was “increasingly a fact of American univer-

sity life” and described the majority’s decision as “woe-

fully out of touch with contemporary academic reality.”

According to the dissenters, the majority had misap-

plied the appropriate statutory principles and erred “in

seeing the academic world as somehow removed from

the economic realm that labor law addresses.”

The

dissenters emphasized that the majority’s decision im-

properly disregarded “the plain language of the statute—

which defines ‘employees’ so broadly that graduate stu-

dents who perform services for, and under the control of,

their universities are easily covered” and instead chose to

exclude student assistants.

This decision was based on

“policy concerns . . . not derived from the Act at all,”

reflecting “an abstract view of what is best for American

higher education—a subject far removed from the

Id., citing NLRB v. Bell Aerospace Co., 416 U.S. 267 (1974).

Id. at 489, citing 229 NLRB at 1002.

Id. at 489–490.

Id. at 493 (dissent of Member Liebman and Member Walsh).

Id. at 494.

Id. at 493.

DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD

Board’s expertise.”

Contrary to the majority, the dis-

senters concluded, in line with the Board’s decision in

NYU, that the terms and conditions of graduate-student

employment were adaptable to collective bargaining (as

illustrated by experience at public-sector universities and

at New York University itself) and that empirical evi-

dence contradicted claims that “academic freedom” and

educational quality were harmed by permitting collective

bargaining.

We believe that the NYU Board and the Brown Univer-

sity dissenters were correct in concluding that student

assistants who perform work at the direction of their uni-

versity for which they are compensated are statutory em-

ployees. That view better comports with the language of

Section 2(3) of the Act and common-law agency princi-

ples, the clear policy of the Act, and the relevant empiri-

cal evidence.

II. DISCUSSION

A. The Brown University Board Erred by Determining

that, as a Matter of Statutory Interpretation, Student As-

sistants Could Not Be Treated as Statutory Employees

For reasons already suggested, the NYU Board was on

very firm legal ground in concluding that student assis-

tants could be employees of the university within the

meaning of Section 2(3) of the Act, while also being stu-

dents—and thus permitting collective bargaining when

student assistants freely choose union representation.

We now reaffirm that approach. Where student assis-

tants have an employment relationship with their univer-

sity under the common law test—which they do here—

this relationship is sufficient to establish that the student

assistant is a Section 2(3) employee for all statutory pur-

poses. We do not hold that the Board is required to find

workers to be statutory employees whenever they are

common-law employees, but only that the Board may

and should find here that student assistants are statutory

employees.

Id. at 497.

Id. at 499–500.

Leading scholars of labor law have long agreed with this view.

See, e.g., Michael C. Harper, Judicial Control of the National Labor

Relations Board’s Lawmaking in the Age of Chevron and Brand X, 89

B.U. L. Rev. 189, 222 (2009) (“In Brown, the Board majority departed

from the most relevant precedent, effectively refused to engage any

available evidence, and disagreed with the dissenters and with the New

York University decision in a way that vitiated any claim for special

deference to its expertise in labor relations.”). See also Catherine L.

Fisk & Deborah C. Malamud, The NLRB in Administrative Law Exile:

Problems with Its Structure and Functions and Suggestions for Reform,

58 Duke L. J. 2013, 2076–2077 (2009).

New York University, supra, 332 NLRB at 1206.

Cf. NLRB v. Town & Country Electric, 516 U.S. 85, 94 (1995)

(observing that “[i]n some cases, there may be a question about whether

the Board’s departure from the common law of agency with respect to

1. Section 2(3)

Section 2(3) of the Act defines “employee” to “include

any employee,” subject to certain specified exceptions.

The Supreme Court has observed that the “breadth of

[Section] 2(3)’s definition is striking: the Act squarely

applies to ‘any employee.’”

The “phrasing of the Act,”

the Court has pointed out, “seems to reiterate the breadth

of the ordinary dictionary definition” of the term, a defi-

nition that “includes any ‘person who works for another

in return for financial or other compensation.’”

The Court has made clear, in turn, that the “task of de-

fining the term ‘employee’ is one that ‘has been assigned

primarily to the agency created by Congress to adminis-

ter the Act,’” the Board.

None of the exceptions enumerated in Section 2(3) ad-

dresses students generally, student assistants in particu-

lar, or private university employees of any sort.

The

absence of student assistants from the Act’s enumeration

of categories excluded from the definition of employee is

itself strong evidence of statutory coverage.

Although

Section 2(3) excludes “individuals employed . . . by any

. . . person who is not an employer . . . as defined” in

Section 2(2) of the Act, private universities do not fall

within any of the specified exceptions, and, indeed, as

previously noted, the Board has chosen to exercise juris-

diction over private, nonprofit universities for more than

45 years.

The Act does not offer a definition of the term “em-

ployee” itself. But it is well established that “when Con-

gress uses the term ‘employee’ in a statute that does not

define the term, courts interpreting the statute ‘must in-

fer, unless the statute otherwise dictates, that Congress

means to incorporate the established meaning’” of the

particular questions and in a particular statutory context, renders its

interpretation unreasonable,” but finding no such issue presented be-

cause the “Board’s interpretation of the term ‘employee’ [was] con-

sistent with the common law”). See also Office Employees Int’l Union,

Local No. 11 v. NLRB, 353 U.S. 313 (1957) (Board lacked discretion to

refuse to assert jurisdiction over labor unions as employers, in face of

clear Congressional expression in Sec. 2(2) of Act, defining “employ-

er” to exclude “any labor organization (other than when acting as an

employer)”).

29 U.S.C. § 152(3).

Sure-Tan, Inc. v. NLRB, 467 U.S. 883, 891 (1984).

Town & Country Electric, supra, 516 U.S. at 90, quoting Ameri-

can Heritage Dictionary 604 (3d ed. 1992).

Sure-Tan, Inc., supra, 467 U.S. at 891, quoting NLRB v. Hearst

Publications, Inc., 322 U.S. 111, 130 (1944).

Nor is the employee status of students mentioned anywhere else in

the Act. Thus, “Congress has not directly addressed the precise ques-

tion at issue” in this case. Chevron U.S.A., Inc. v. Natural Resources

Defense Council, Inc., 467 U.S. 837, 843 (1984).

See Sure-Tan, Inc., supra, at 891–892.

See Cornell University, supra, 183 NLRB at 331–333, overruling

Trustees of Columbia University, 97 NLRB 424 (1951).

COLUMBIA UNIVERSITY

term, with reference to “‘common-law agency doc-

trine.’”

Not surprisingly, then, the Supreme Court has

endorsed the Board’s determination that certain workers

were statutory employees where that determination

aligned with the common law of agency.

Other federal

courts have done so as well.

In accordance with the

statute’s broad definition and with the Supreme Court’s

approval, the Board has interpreted the expansive lan-

guage of Section 2(3) to cover, for example, paid union

organizers (salts) employed by a company,

undocu-

mented aliens,

and “confidential” employees,

among

other categories of workers.

The most notable instance in which apparent common-

law employees were found not to be employees under the

Act, in spite of the absence of an explicit statutory exclu-

sion, is the exception that proves the rule. In Bell Aero-

space, cited by the Brown University Board, the Supreme

Court held that “managerial employees” were not cov-

ered by the Act because Congress had clearly implied

their exclusion by the Act’s design and purpose to facili-

tate fairness in collective bargaining.

As the Court

concluded, giving employee status to managers would be

contrary to this purpose: it would place managers, who

would be expected to be on the side of the employer in

bargaining, and non-managerial employees in the same

bargaining “camp,” “eviscerat[ing] the traditional dis-

tinction between labor and management.”

The exclu-

sion of managers rested on legislative history, along with

Town & Country Electric, supra, 516 U.S. at 94, quoting Nation-

wide Mutual Ins. Co. v. Darden, 503 U.S. 318, 322–323 (1992).

Id. at 94–95 (rejecting employer’s argument that common law

principles precluded Board’s determination that paid union organizers,

salts, were statutory employees, and holding that salts fell within rea-

sonable construction of common law definition).

See, e.g., Lancaster Symphony Orchestra v. NLRB, 822 F.3d 563

(D.C. Cir. 2016), enfg. 357 NLRB 1761 (2011) (musicians in a regional

orchestra are statutory employees); Seattle Opera v. NLRB, 292 F.3d

757, 761–762 (D.C. Cir. 2002), enfg. 331 NLRB 1072 (2000) (opera

company’s auxiliary choristers are statutory employees). The Board

has consistently applied common-law principles in its application of

other concepts under the Act, including the Act’s broad definition of an

employer. See, e.g., Browning-Ferris Industries, 362 NLRB No. 186

(2016) (test for joint-employer status).

Town & Country Electric, supra, 516 U.S. at 94 & 97–98 (com-

mon-law principles supported Board’s construction of the term “em-

ployee” to include salts).

Sure-Tan, supra, 467 U.S. at 892 (observing that undocumented

aliens are “not among the few groups of workers expressly exempted

by Congress” from the definition of “employee” and that “extending

the coverage of the Act to [them] is consistent with the Act’s avowed

purpose of encouraging and protecting the collective-bargaining pro-

cess”).

NLRB v. Hendricks County Rural Electric Membership Corp., 454

U.S. 170, 189–190 (1981).

NLRB v. Bell Aerospace Co., 416 U.S. 267, 275 (1974).

Id. at 284 fn. 13.

the intrinsic purpose and structure of the Act. No legisla-

tive history supports excluding student assistants from

statutory coverage, nor does the design of the Act itself.

2. The Brown Board Did Not Adequately Consider the

Text of Section 2(3)

The Brown University Board insisted that Section 2(3)

of the Act must not be examined in isolation; rather, the

Board must “look to the underlying fundamental premise

of the Act, viz. the Act is designed to cover economic

relationships.”

Certainly, the Supreme Court has sug-

gested that, despite the centrality of common-law agency

principles to employee status under the Act, “[i]n doubt-

ful cases resort must still be had to economic and policy

considerations to infuse [Section] 2(3) with meaning.”

But we reject the Brown University Board’s claim that

finding student assistants to be statutory employees,

where they have a common-law employment relationship

with their university, is somehow incompatible with the

“underlying fundamental premise of the Act.” The Act is

designed to cover a particular type of “economic rela-

tionship” (in the Brown University Board’s phrase)—an

employment relationship—and where that relationship

exists, there should be compelling reasons before the

Board excludes a category of workers from the Act’s

coverage.

The fundamental error of the Brown University Board

was to frame the issue of statutory coverage not in terms

of the existence of an employment relationship, but ra-

ther on whether some other relationship between the em-

ployee and the employer is the primary one

—a standard

neither derived from the statutory text of Section 2(3) nor

from the fundamental policy of the Act.

Indeed, in

Contrary to Columbia’s assertion, the fact that Congress has not

enacted legislation to countermand the Board’s Brown decision carries

little weight. One is not to infer legislative intent based on Congress’s

seeming acquiescence to an agency decision unless there is evidence

that Congress actually considered the precise agency action at issue.

Solid Waste Agency of Northern Cook Cty. v. Army Corps of Engineers,

531 U.S. 159, 169–170, fn. 5 (2001). For this reason, we do not rely

today on the fact that Congress took no action to overrule the Board’s

earlier decision in NYU.

342 NLRB at 488.

Allied Chemical & Alkali Workers v. Pittsburgh Plate Glass, 404

U.S. 157, 168 (1971).

Columbia also argues for our adoption of another, similar non-

common-law standard: the “primary beneficiary” analysis used by the

courts in some Fair Labor Standards Act (FLSA) cases, including cases

involving the employee status of student interns. Because the FLSA

definition of a statutory employee is not tethered to the common law (as

the Act’s definition is), and because the FLSA reflects policy goals

distinct from those of the Act, we are not persuaded that the “primary

beneficiary” analysis should govern this case. For the same reason, we

are not persuaded by Columbia’s contention that the Department of

Labor’s recent guidance regarding whether graduate research assistants

are employees within the meaning of the FLSA bears on the separate

DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD

spite of the Brown University Board’s professed adher-

ence to “Congressional policies,” we can discern no such

policies that speak to whether a common-law employee

should be excluded from the Act because his or her em-

ployment relationship co-exists with an educational or

other non-economic relationship.

The Board and the

courts have repeatedly made clear that the extent of any

required “economic” dimension to an employment rela-

tionship is the payment of tangible compensation. Even

when such an economic component may seem compara-

tively slight, relative to other aspects of the relationship

between worker and employer, the payment of compen-

sation, in conjunction with the employer’s control, suf-

fices to establish an employment relationship for purpos-

es of the Act.

Indeed, the principle that student assis-

tants may have a common-law employment relationship

with their universities—and should be treated according-

ly—is recognized in other areas of employment law as

well.

question of whether student assistants who have a common-law em-

ployment relationship with their universities should be regarded as

employees under the NLRA.

Our dissenting colleague observes that an “array of federal statutes

and regulations apply to colleges and universities,” but he does not

identify any statute or regulation that speaks directly (or even indirect-

ly) to the key question here. That Congress is interested in supporting

and regulating postsecondary education, as it surely is, does not demon-

strate a Congressional view on whether or how the NLRA should be

applied to student assistants.

Nor does our colleague identify any potential for conflict between

the Act’s specific requirements and those of federal education law—

with one possible exception, related to educational records, which we

address below. See fn. 93, infra. That application of the Act in some

specific respect might require accommodation to another federal law

cannot mean that the Board must refrain from applying the Act, at all,

to an entire class of statutory employers or statutory employees. Cf.

Sure-Tan, Inc., supra, 467 U.S. at 89–893, 903–904 (affirming Board’s

holding that undocumented workers were statutory employees under

NLRA, but concluding that federal immigration law precluded award-

ing certain remedies for periods when workers were not legally entitled

to be present and employed in United States). See generally Vimar

Seguros y Reaseguros, S.A. v. M/V Sky Reefer, 515 U.S. 528, 533

(1995) (“[W]hen two statutes are capable of co-existence … it is the

duty of the courts, absent a clearly expressed congressional intention to

the contrary, to regard each as effective.”) (quotations omitted).

See Town & Country Electric, supra, 516 U.S. at 88, 95 (although

chief purpose of union salts seeking employment was to organize and

form a union, not to benefit economically, they were nonetheless em-

ployees as they were both paid and controlled by the company with

respect to ordinary workplace duties); Seattle Opera Assn., 331 NLRB

1072, 1073 (2000) (observing that while auxiliary choristers received

some nonmonetary benefit in the form of personal satisfaction at their

involvement in the opera, which is characteristic of a volunteer rela-

tionship, they also received monetary compensation for their effort, and

this fact, along with employer control, made them employees under the

Act), enfd. 292 F.3d 757 (D.C. Cir. 2002).

For purposes of employment law, student assistants cannot be fair-

ly categorized as “volunteers,” rather than employees. See Restatement

of Employment Law §1.02 (“An individual is a volunteer and not an

In sum, we reject the Brown Board’s focus on whether

student assistants have a “primarily educational” em-

ployment relationship with their universities.

The Su-

preme Court has cautioned that “vague notions of a stat-

ute’s ‘basic purpose’ are . . . inadequate to overcome the

words of its text regarding the specific issue under con-

sideration.”

The crucial statutory text here, of course,

is the broad language of Section 2(3) defining “employ-

ee” and the language of Section 8(d) defining the duty to

bargain collectively. It seems clear to us, then, that the

Act’s text supports the conclusion that student assistants

who are common-law employees are covered by the Act,

unless compelling statutory and policy considerations

require an exception. As we explain next, the relevant

considerations strongly favor statutory coverage.

B. Asserting Jurisdiction over Student Assistants Pro-

motes the Goals of Federal Labor Policy

1. Overview of Federal Labor Policy

Federal labor policy, in the words of Section 1 of the

Act, is to “encourag[e] the practice and procedure of col-

lective bargaining,” and to protect workers’ “full free-

dom” to express a choice for or against collective-

employee if the individual renders uncoerced services to a principal

without being offered a material inducement”). As the Restatement

explains, “[w]here an educational institution compensates student assis-

tants for performing services that benefit the institution, . . . such com-

pensation encourages the students to do the work for more than educa-

tional benefits and thereby establishes an employment as well as an

educational relationship.” Id., comment g. The Restatement illustrates

this principle with the following example:

A is a graduate student in biochemistry at university P. In order to

complete the degree requirements, A must work in a laboratory under

P’s auspices, either for pay or as a volunteer. A works in the laborato-

ry of a professor, for which A is paid a yearly stipend and given full

tuition remission. The professor has secured grants to support the re-

search that A is assisting. A is an employee of P. P is providing A with

significant benefits both in order to further A’s education and also to

obtain A’s services on P’s funded research.

Id., illustration 10 (emphasis added).

The Brown University Board insisted that “there is a significant

risk, and indeed a strong likelihood, that the collective-bargaining pro-

cess will be detrimental to the educational process” and announced that

the Board would “decline to take these risks with our nation’s excellent

private educational system.” 342 NLRB at 493. The Board’s state-

ment—coupled not only with the absence of any experiential or empiri-

cal basis for it, but also with the remarkable assertion that no such basis

was required—strongly suggests that the Board acted based on little

more than its own view of what was best for private universities. “No

one in Congress,” an academic critic of Brown University has written,

“would have wanted the Board to determine which workers may be

protected by the Act on the basis of mere suppositions without consid-

eration of how statutory or other goals would be served in practice by

exclusion or coverage.” Michael C. Harper, Judicial Control of the

National Labor Relations Board’s Lawmaking in the Age of Chevron

and Brand X, 89 B. U. L. Rev. 189, 220 (2009).

Mertens v. Hewitt Associates, 508 U.S. 248, 261 (1993) (emphasis

in original).

COLUMBIA UNIVERSITY

bargaining representation. Permitting student assistants

to choose whether they wish to engage in collective bar-

gaining—not prohibiting it—would further the Act’s

policies.

Although the Brown University Board held that student

assistants were not statutory employees, it also observed

that, even assuming they were, the Board would have

“discretion to determine whether it would effectuate na-

tional labor policy to extend collective bargaining rights”

to student assistants and that, in fact, it would “not effec-

tuate the purposes and policies of the Act to do so.”

We disagree not with the claim that the Board has some

discretion in this area,

but with the conclusion reached

by the Brown University Board, including its view that

“empirical evidence” is irrelevant to the inquiry.

have carefully considered the arguments marshaled by

the Board majority in Brown University (as well as the

arguments advanced here by Columbia and supporting

amici, as well as our dissenting colleague), but find that

they do not outweigh the considerations that favor ex-

tending statutory coverage to student assistants.

The claims of the Brown majority are almost entirely

theoretical. The Brown University Board failed to

demonstrate that collective bargaining between a univer-

sity and its employed graduate students cannot coexist

successfully with student-teacher relationships, with the

educational process, and with the traditional goals of

higher education. Labor law scholars have aptly criti-

cized the Brown University decision as offering “no em-

pirical support” for its claims, even though “those asser-

tions are empirically testable.”

The National Labor Relations Act, as we have repeat-

edly emphasized, governs only the employee-employer

relationship. For deciding the legal and policy issues in

this case, then, it is not dispositive that student-teacher

relationship involves different interests than the employ-

ee-employer relationship; that the educational process is

individual, while collective bargaining is focused on the

group; and that promoting equality of bargaining power

is not an aim of higher education. Even conceded, all

342 NLRB at 492 (emphasis added).

However, in exercising this discretion, we tread carefully and with

an eye toward the Act’s purposes. In Northwestern University, 362

NLRB No. 167 (2015), we denied the protections of the Act to certain

college athletes—without ruling on their employee status—because,

due to their situation within and governance by an athletic consortium

dominated by public universities, we found that our extending coverage

to them would not advance the purposes of the Act. Here, conversely,

we have no reason to believe that extending bargaining rights will not

meaningfully advance the goals of the Act.

342 NLRB at 492–493.

Catherine L. Fisk & Deborah C. Malamud, The NLRB in Adminis-

trative Exile: Problems with Its Structure and Function and Sugges-

tions for Reform, 58 Duke L. J. 2013, 2076–2077 (2009).

these points simply confirm that collective bargaining

and education occupy different institutional spheres. In

other words, a graduate student may be both a student

and an employee; a university may be both the student’s

educator and employer. By permitting the Board to de-

fine the scope of mandatory bargaining over “wages,

hours, and other terms and conditions of employment,”

the Act makes it entirely possible for these different roles

to coexist—and for genuine academic freedom to be pre-

served. It is no answer to suggest, as the Brown Univer-

sity Board did, that permitting student assistants to bar-

gain over their terms and conditions of employment (no

more and no less) somehow poses a greater threat to aca-

demic freedom than permitting collective bargaining by

non-managerial faculty members, “[b]ecause graduate

student assistants are students.”

That the academic-

employment setting poses special issues of its own—as

the Board and the Supreme Court have both recog-

nized

—does not somehow mean that the Act cannot

properly be applied there at all.

2. Applying the Act to Student Assistants Would Not

Infringe upon First Amendment Academic Freedom

The Brown University Board endorsed the view that

“collective bargaining would unduly infringe upon tradi-

tional academic freedoms,” citing the “right to speak

freely in the classroom” and a list of “traditional academ-

ic decisions” including “course length and content,

standards for advancement and graduation, [and] admin-

istration of exams.”

Insofar as the concept of academic

freedom implicates the First Amendment, the Board cer-

tainly must take any such infringement into account.

But there is little, if any, basis here to conclude that treat-

ing employed graduate students as employees under the

Act would raise serious constitutional questions, much

less violate the First Amendment.

The Supreme Court has made clear that academic

freedom, in the constitutional sense, involves freedom

from government efforts “to control or direct the content

342 NLRB at 490 fn. 26.

See NLRB v. Yeshiva University, 444 U.S. 672, 680-681 (1980),

citing Syracuse University, 204 NLRB 641, 643 (1973) (permitting law

school faculty to vote separately from other university faculty members

on questions of representation, based on divergent professional inter-

ests). In Syracuse University, the Board observed that in the “academic

world,” the “basic interests recognized by the Act remain the same, but

their interrelationship, the employer-employee relationship, and even

the employee-employee relationship, does not squarely fit the industrial

model.” 204 NLRB at 643.

342 NLRB at 490, citing St. Clare’s Hospital, supra, 229 NLRB at

1003.

Cf. NLRB v. Catholic Bishop of Chicago, 440 U.S. 490 (1979)

(declining to construe Act as authorizing Board to exercise jurisdiction

over lay faculty members at church-related schools, given serious First

Amendment questions potentially raised).

DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD

of the speech engaged in by the university or those affili-

ated with it.”

No such effort is involved here. Neither

the Brown University majority, nor the parties or amici in

this case, have explained how the “right to speak freely

in the classroom” (in the Brown University Board’s

phrase) would be infringed by collective bargaining over

“terms and conditions of employment” for employed

graduate students, as the Act envisions.

Further, the Supreme Court has explained that

“[a]lthough parties are free to bargain about any legal

subject, Congress has limited the mandate or duty to bar-

gain to matters of ‘wages, hours, and other terms and

conditions of employment.’”

Defining the precise con-

tours of what is a mandatory subject of bargaining for

student assistants is a task that the Board can and should

address case by case.

That approach will permit the

Board to consider any genuine First Amendment issues

that might actually arise—in a concrete, not speculative,

context.

In upholding that Board’s authority to exercise juris-

diction over faculty members at private universities—

provided that they are statutory employees—the Supreme

Court has implicitly rejected the view that some unde-

fined need to preserve academic freedom overrides that

policies of the Act. In Yeshiva University, supra, the

Court found that the full-time university faculty members

there—whose “authority in academic matters [was] abso-

University of Pennsylvania v. EEOC, 493 U.S. 182, 197 (1990)

(emphasis in original) (rejecting university’s First Amendment chal-

lenge to EEOC investigative subpoena under Title VII, seeking materi-

als related to faculty-member tenure review process alleged to be dis-

criminatory).

National Labor Relations Act, §8(d), 29 U.S.C. §158(d) (“[T]o

bargain collectively is the performance of the mutual obligation of the

employer and the representative of the employees to meet at reasonable

times and confer in good faith with respect to wages, hours, and other

terms and conditions of employment . . . but such obligation does not

compel either party to agree to a proposal or require the making of a

concession.”).

First National Maintenance Corp. v. NLRB, 452 U.S. 666, 674–

675 (1981) (footnotes omitted).

In this situation, as with other aspects of labor law, the “‘nature of

the problem, as revealed by unfolding variant situations,’ requires ‘an

evolutionary process for its rational response, not a quick, definitive

formula as a comprehensive answer.”’’ Eastex, Inc. v. NLRB, 437 U.S.

556, 575 (1978), quoting Electrical Workers v. NLRB, 366 U.S. 667,

674 (1961).

In Associated Press v. NLRB, 301 U.S. 103, 132-133 (1937),

which involved the discriminatory discharge of an editorial employee,

the Supreme Court upheld Board jurisdiction over a news-gathering

organization, despite arguments that it would violate the First Amend-

ment freedom of the press. The Court found that Board’s reinstatement

order “in nowise circumscribe[d]” the First Amendment rights of the

Associated Press, observed that the “publisher of a newspaper has no

special immunity from the application of general laws,” and rejected

the contention that because “regulation in a situation not presented

would be invalid,” the Board could not exercise jurisdiction at all.

lute”

—were excluded from the Act’s coverage, as

managerial employees, on that basis. But the Court also

observed that not all university faculty members will be

managerial employees and that “professors may not be

excluded [from statutory coverage] merely because they

determine the content of their own courses, evaluate their

own students, and supervise their own research.”

If the

Brown University Board’s broad view of academic free-

dom were correct, then it seems highly unlikely that the

Yeshiva Court could have contemplated collective bar-

gaining by university professors (as, indeed, the Board

has permitted for many years

) in the face of such an

obvious constitutional obstacle.

3. Empirical Evidence and the Board’s Experience Indi-

cate That It Is Appropriate for the Board to Exercise Ju-

risdiction over Student Assistants

If, historically, the Board had permitted student assis-

tants to engage in collective bargaining, and if actual

experience over the years had demonstrated both that

collective bargaining rarely proved beneficial to students

and that it seriously harmed the ability of private univer-

sities to function effectively, then perhaps the Board

would have had grounds for deciding that the Act cannot

productively be applied in the university setting. But that

is not the case, because collective bargaining by student

assistants at private universities is historically uncom-

mon. Neither administrative experience nor empirical

evidence supported the Brown University Board’s deter-

mination that extending statutory protection to student

assistants would be detrimental to the educational pro-

cess.

It is telling, moreover, that the Brown University

Board gave no weight at all to the analogous experience

of public universities with collective bargaining by stu-

dent assistants or to private universities’ experience with

faculty bargaining, subjects we turn to below. The expe-

rience of student assistant collective bargaining at public

universities provides no support for the fearful predic-

tions of the Brown University Board. In the words of one

scholar, “[t]here appear to be no major disasters that have

arisen because of [graduate-student] unions,” and exam-

ples of collective bargaining in practice “appear to

demonstrate that economic and academic issues on cam-

pus can indeed be separated.”

444 U.S. at 686.

Id. at 690, fn. 31.

See, e.g., Bradford College, 261 NLRB 565 (1982). See also Pa-

cific Lutheran University, 361 NLRB No. 157 (2014).

Judith Wagner DeCew, Unionization in the Academy: Visions and

Realities 98 (2003). See also Josh Rinschler, Students or Employees?

The Struggle over Graduate Student Unions in America’s Private Col-

leges and Universities, 36 J. College & University L. 615, 639–640

COLUMBIA UNIVERSITY

Here, Columbia, its supporting amici, and our dissent-

ing colleague defend the Brown University decision,

echoing the claim that permitting collective bargaining

by student assistants will harm the educational process.

These arguments are dubious on their own terms. Our

skepticism is based on the historic flexibility of collec-

tive bargaining as a practice and its viability at public

universities where graduate student assistants are repre-

sented by labor unions and among faculty members at

private universities.

As the Brown University dissenters observed,

“[c]ollective bargaining by graduate student employees is

increasingly a fact of American university life.”

Recent

data show that more than 64,000 graduate student em-

ployees are organized at 28 institutions of higher educa-

tion, a development that began at the University of Wis-

consin at Madison in 1969 and that now encompasses

universities in California, Florida, Illinois, Iowa, Massa-

chusetts, Michigan, Oregon, Pennsylvania, and Washing-

ton.

At these universities, to be sure, collective bar-

gaining is governed by state law, not by the National

Labor Relations Act.

Even so, the experience with

graduate-student collective bargaining in public universi-

ties is of relevance in applying the Act, as the closest

proxy for experience under the Act.

By way of example, as AFT notes in its amicus brief,

the University of Illinois, Michigan State University, and

Wayne State University include language in their gradu-

ate-assistant collective-bargaining agreements giving

(2010) (“[E]vidence from public institutions as well as from NYU

during the period it had a graduate student union, suggests that unioni-

zation does not result in the sky falling.”).

342 NLRB at 493 (dissent of Members Liebman and Walsh). The

dissenters cited a body of scholarly literature examining this develop-

ment. Id. at fn. 1. See Neal H. Hutchens & Melisa B. Hutchens,

Catching the Union Bug: Graduate Student Employees and Unioniza-

tion, 39 Gonzaga L. Rev. 105, 106–107 (2004); Daniel J. Julius &

Patricia J. Gumport, Graduate Student Unionization: Catalysts and

Consequences, 26 Review of Higher Education 187, 191–196 (2002);

Grant M. Hayden, “The University Works Because We Do”: Collective

Bargaining Rights for Graduate Assistants, 69 Fordham L. Rev. 1233,

1236–1243 (2001). See also DeCew, supra, Unionization in the Acad-

emy at 89-110.

J. Berry & M. Savarese, Directory of U.S. Faculty Contracts and

Bargaining Agents in Institutions of Higher Education (2012). See also

Daniel J. Julius & Nicholas DiGiovanni, Jr., Academic Collective Bar-

gaining: On Campus Fifty Years, Center for Studies in Higher Educa-

tion, Research & Occasional Paper Series: CSHE 4.13 at 5 (April

2013), available at www.cshe.berkeley.edu/publications.

Sec. 2(2) of the Act excludes “any State or political subdivision

thereof” from the definition of “employer.” 29 U.S.C. §152(2).

Cf. Management Training Corp., 317 NLRB 1355, 1357 (1995)

(holding that Board will exercise jurisdiction over employers without

considering extent of their control over purely economic terms and

conditions of employment and citing “successful and effective bargain-

ing” in public sector “where economic benefits play a small role”).

management defined rights concerning courses, course

content, course assignments, exams, class size, grading

policies and methods of instruction, as well as graduate

students’ progress on their own degrees.

This is not to

suggest a prescription for how individual collective-

bargaining agreements should resolve matters related to

the protection of academic freedom and educational pre-

rogatives. Rather, these agreements show that parties

can and successfully have navigated delicate topics near

the intersection of the university’s dual role as educator

and employer.

Other scholars, whose studies were cited in the Brown

University dissent,

confirm that view. Based on their

survey-based research of public universities, they reject

the claim, for example, that collective bargaining will

harm mentoring relationships between faculty members

and graduate students.

More recent survey-based re-

search found “no support” for the contentions that gradu-

ate student unionization “would harm the faculty-student

relationship” or “would diminish academic freedom,”

and observed that “[d]espite the NLRB’s focus on the

potential negative effects on academic outcomes, gradu-

ate students themselves have likely been more concerned

with the basic terms and conditions of employment.”

Although Columbia presented the testimony of an aca-

demic economist to address this study, its expert simply

maintained that the study could not “rule out harm or

benefit” to the faculty-student relationship from collec-

tive bargaining. When the best analytical evidence of-

fered by Columbia suggests merely that neither harm nor

benefit from collective bargaining can be ruled out, the

dire predictions of the Brown University Board are un-

dercut.

Agreement by and between the Board of Trustees of the Universi-

ty of Illinois and Graduate Employees’ Organization IFT/AFT 6300

(August 16, 2012—August 15, 2017), available at

http://www.ahr.illinois.edu/geo.pdf; Collective Bargaining Agreement

between Michigan State University and The Graduate Employees Un-

ion, Local 6196, AFT-Michigan/AFL–CIO (May 2015—May 2019),

available at https://www.hr.msu.edu/documents/contracts/GEU2015-

2019.pdf; Collective Bargaining Agreement between Wayne State

University and The Graduate Employees Organizing

Committee, AFT (May 2015—February 2018), available at

http://provost.wayne.edu/pdf/geoc_contract_2015–2018_w_toc.pdf.

342 NLRB at 499-500.

Julius & Gumport, supra, Graduate Student Unionization: Cata-

lysts and Consequences, 26 Review of Higher Education at 201, 209;

Gordon J. Hewitt, Graduate Student Employee Collective Bargaining

and the Educational Relationship between Faculty and Graduate Stu-

dents, 29 J. Collective Negotiations in the Public Sector 153, 159–164

(2000).

Sean E. Rogers, Adrienne E. Eaton, and Paula B. Voos, Effects of

Unionization on Graduate Student Employees: Faculty-Student Rela-

tions, Academic Freedom, and Pay, 66 Industrial & Labor Relations

Rev. 487, 507 (2013).

DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD

Columbia and supporting amici point to a few individ-

ual examples arising from the 28 public universities and

64,000 represented public university student assistants,

along with NYU on the private side, in which, they con-

tend, collective bargaining by student assistants has

proven detrimental to the pursuit of the school’s educa-

tional goals. They note the occurrences of strikes and

grievances over teaching workload and tuition waivers.

Similarly, they point to grievances over classroom as-

signments and eligibility criteria for assistantships. But

labor disputes are a fact of economic life—and the Act is

intended to address them.

Columbia and its supporting amici suggest that collec-

tive-bargaining demands would interfere with academic

decisions involving class size, time, length, and location,

as well as decisions concerning the formatting of exams.

They also worry that disputes over whether bargaining is

required for such issues may lead to protracted litigation

over the parties’ rights and obligations as to a given is-

sue, for example, over the propriety of a university’s

change in class or exam format, thus burdening the time-

sensitive educational process. However, to a large ex-

tent, the Board’s demarcation of what is a mandatory

subject of bargaining for student assistants, and what is

not, would ultimately resolve these potential problems.

Moreover, there is no good reason to doubt that unions

and universities will be able to negotiate contract lan-

guage to delineate mutually satisfactory boundaries of

their respective rights and obligations.

Indeed, faculty

members have successfully negotiated collective-

bargaining agreements that address terms and conditions

of employment at private universities while contractually

ensuring academic freedom for decades.

Indeed, decisions concerning management’s right to control its

fundamental operations and to produce a product of its choosing are

issues of concern to manufacturing employers as well. Collective bar-

gaining does not limit such a management right normally, unless the

parties consent to it. See First National Maintenance v. NLRB, 452

U.S. 666, 677–679 (1981). However, in a common feature of collective

bargaining across economic sectors, employers must, and do, routinely

address incidental operational impacts based on their agreements with

unions. For example, an employer chooses the amount and type of

product it produces, but an employer must bargain about employees’

hours of work and must operate within whatever work-hours constraint

it agrees to.

Notably, the NYU graduate assistants union, voluntarily recog-

nized by that university after the Board overruled its NYU decision, has

given the university control over academic matters in the parties’ col-

lective-bargaining agreement.

As AAUP notes in its amicus brief, many of its unionized faculty

chapters’ collective-bargaining agreements expressly refer to and quote

the AAUP’s 1940 Statement of Principles on Academic Freedom and

Tenure, which provides a framework that has proven mutually agreea-

ble to many unions and universities. See 1940 Statement of Principles

The notion that the parties themselves can resolve,

through the bargaining process, many of the latent con-

flicts suggested by Columbia and amici (as well as by

our dissenting colleague)—and hence forge successful

bargaining relationships—is not a theoretical one. The

experience at New York University is a case in point.

Even after Brown University issued, NYU continued—

after a brief interruption—to voluntarily recognize its

graduate assistants union and successfully negotiated

collective-bargaining agreements with that union.

Both the original and successor agreements at NYU

addressed such matters as stipends, pay periods, disci-

pline and discharge, job posting, a grievance-and-

arbitration procedure, and health insurance—nearly all

familiar mandatory subjects of bargaining across the pri-

vate sector, which appear to have been successfully

adapted to a university setting.

The agreements also

incorporate a “management and academic rights” clause,

which would tend to allay fears that collective bargaining

will attempt to dictate academic matters.

In the most

recent agreement, in effect from September 1, 2014, to

August 31, 2020, the clause preserved the university’s

right to “determine . . . qualifications . . . and assignment

of graduate employees; to determine the processes and

criteria by which graduate employees’ performance is

evaluated; . . . to schedule hours of work; . . . to deter-

mine how and when and by whom instruction is deliv-

ered; . . . to introduce new methods of instruction; . . .

and to exercise sole authority on all decisions involving

on Academic Freedom and Tenure, available at

https://www.aaup.org/file/1940%20Statement.pdf.

The evidence all seems to suggest that the bread-and-butter eco-

nomic concerns reflected in the NYU collective-bargaining agreement

are what drive American graduate students to seek union representa-

tion. See, e.g., Julius & Gumport, supra, Graduate Student Unioniza-

tion: Catalysts and Consequences, 26 Review of Higher Education at

196 (“[D]ata show that the unionization of these individuals is driven

fundamentally by economic realities.”); Gerrilynn Falasco & William J.

Jackson, The Graduate Assistant Labor Movement, NYU and Its After-

math: A Study of the Attitudes of Graduate Teaching and Research

Assistants at Seven Universities, 21 Hofstra Labor & Employment L. J.

753, 800 (2004) (“Overwhelmingly, the respondents from the seven

universities surveyed indicated that the most important issues to them

were wages and health insurance”).

Collective-Bargaining Agreements between NYU and Internation-

al Union, UAW, AFL-CIO and Local 2110, Technical Office and Pro-

fessional Workers, UAW, Sept. 1, 2001-Aug. 31, 2005, and Sept. 1,

2014-Aug. 31, 2020, Art. XXII. See Brown University, supra, 342

NLRB at 499 (dissent) (discussing agreement). The cited clause would

seem to confirm the view of one scholar that the “historical develop-

ment of professional norms of academic freedom creates a strong po-

tential for faculty and GAs [employed graduate students] to find com-

mon interests in carrying out the core functions of the university.” Risa

L. Lieberwitz, Faculty in the Corporate University: Professional Iden-

tity, Law and Collective Action, 16 Cornell J. L. & Public Policy 263,

326 (2007) (discussing events at Cornell).

COLUMBIA UNIVERSITY

academic matters . . . decisions regarding who is taught,

what is taught, how it is taught and who does the teach-

ing.”

Moreover, to the extent disputes nonetheless do arise,

the process of resolving such disputes over the bounda-

ries of parties’ rights and obligations is common to near-

ly all collective-bargaining contexts in which manage-

ment seeks to act in some way it believes is important to

its business, including critical sectors such as national

security and national defense. Not long after Brown

University was decided, for example, the Board observed

that “for over 60 years, in times of both war and peace,

the Board has asserted jurisdiction over employers and

employees that have been involved in national security

and defense,” and that the Board could “find no case in

which our protection of employees’ Section 7 rights had

an adverse impact on national security or defense.”

Similarly, in the acute care hospital sector, the Boston

Medical Center Board, supra, recognized house staff at

teaching hospitals as statutory employees, and the

Board’s experience since that decision has provided no

support for one dissenting member’s prediction that

“American graduate medical education [would] be irrep-

arably harmed”

if the Board asserted jurisdiction over

house staff.

These critical sectors have proven able to effectively

integrate collective bargaining, with its occasional dis-

putes and attendant delays, into their modes of doing

business. We have no reason to doubt that the higher-

education sector cannot do the same. Indeed, some of

the practical concerns raised by Columbia and amici

seem to be generic complaints about the statutory re-

quirements inherent in a collective-bargaining relation-

ship, rather than education-specific concerns. For exam-

ple, it is posed as problematic by amici that a university

may have to bargain, at least as to effect, over the elimi-

We recognize that part of the ostensible reason NYU decided to

withdraw recognition from its union of student assistants was the rec-

ommendations of two university committees, which cited the union’s

filing of grievances that were perceived as threatening to undercut

NYU’s academic decisionmaking. We need not decide whether this

perception was accurate or whether it would hold true over time and at

other universities where student assistants organized. That employees

may invoke their Sec. 7 rights to a greater or lesser degree has no bear-

ing on whether the Act should be interpreted to grant them those rights.

We note, in any case, that the NYU collective-bargaining relationship

was new, that the collective-bargaining agreement was untested, and

that ultimately the university appears to have prevailed in grievance-

arbitration proceedings in its assertion of its academic prerogatives.

Firstline Transportation Security, 347 NLRB 447, 453 (2006) (re-

jecting argument that Board should decline to assert jurisdiction over

privately-employed airport security screeners, on national-security

grounds).

Boston Medical Center, supra, 330 NLRB at 182.

nation of assistantship positions. However, bargaining

over staffing levels is a core concern of employees, and

standard fare for collective bargaining. Fulfilling one’s

obligation to bargain about job loss or staffing levels, or

the effects thereof, has not proven unduly burdensome to

countless other unionized workplaces.

Similarly, Co-

lumbia and amici, as well as our dissenting colleague,

also raise the specter of strikes (and lockouts), and the

impact they might have on the educational trajectory of

students and on their considerable investment in their

education; but the problems raised by strikes are com-

mon to nearly all industries in which the Board accords

employees bargaining rights.

Moreover, we cannot give credence to the dissent’s

speculation that, among other things, the provisions of

the Act might negatively interfere with university confi-

dentiality practices or standards of decorum, for example

by authorizing abusive language by student assistants

directed against faculty. The Act’s provisions pertaining

to document production and the boundaries of protected

conduct are, and always have been, contextual. The

Board evaluates such claims in light of workplace stand-

ards and other relevant rules and practices.

Moreover, while focusing on a few discrete problems

that may arise in bargaining—without considering the

likelihood that they would both actually occur and not be

amenable to resolution by bargaining partners acting in

good faith—Columbia and amici neglect to weigh the

See, e.g., St. Anthony Hospital Systems, 319 NLRB 46, 47 & 50

(1995) (discussing a hospital’s duty to bargain with nurses’ union over

staffing levels).

Strikes may affect the operations of an employer—especially in in-

dustries where the provision of goods and services may be time-

sensitive. However, these already include many sectors where collec-

tive bargaining has long been a staple of workplace life, such as those

involving caring for hospital patients, maintaining critical infrastruc-

ture, publishing newspapers, unloading billions of dollars in overseas

freight, and teaching students in college or pre-college settings (as in

the case of unionized faculty in private schools). In all of these situa-

tions, strikes pose the possibility of a disruption with significant costs,

given the value and/or time-sensitivity of the goods and services in-

volved. Yet, the Act permits collective bargaining in these sectors; and

indeed, employers have incorporated the risks of economic conflict into

their negotiation strategies and modes of doing business. The Board’s

experience demonstrates that parties grasp the seriousness of recourse

to economic weapons, and do not do so lightly or without the convic-

tion that their position is worth fighting for. There is no reason to set

universities apart, where the Act otherwise points toward coverage.

Our dissenting colleague points to the potential for unique conse-

quences to student assistants in the event of a work stoppage, such as

loss of academic credit and tuition waivers. As we have noted, the Act

permits collective bargaining—and permits parties to decide for them-

selves what risks to take—even where there is a potential for economic

disruption. However, to the extent a work stoppage cannot be avoided,

parties will frequently resolve a strike or lockout on terms that address

questions pertaining to vested rights and other such matters that arose

because of the work stoppage.

DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD

possibility of any benefits that flow from collective bar-

gaining, such as those envisioned by Congress when it

adopted the Act. In this connection, it is worth noting

that student assistants, in the absence of access to the

Act’s representation procedures and in the face of rising

financial pressures, have been said to be “fervently lob-

bying their respective schools for better benefits and in-

creased representation.”

The eagerness of at least some

student assistants to engage in bargaining suggests that

the traditional model of relations between university and

student assistants is insufficiently responsive to student

assistants’ needs. That is not to say collective bargaining

will necessarily be a panacea for such discontent, but it

further favors coverage by the Act, which was designed

to ameliorate labor unrest.

Finally, we disagree with the suggestion of our dissent-

ing colleague that the Act’s procedures are ill suited here,

because student assistants have finite terms and because

the academic world may experience a fast pace of devel-

opment in fields of study, and thus because, in the time it

takes for the Board to resolve a question arising under

David Ludwig, Why Graduate Students of America Are Uniting,

The Atlantic, available at

http://www.theatlantic.com/education/archive/2015/04/graduate-

students-of-the-world-unite/390261/. See also Rachel Bernstein, Ivy

League Graduate Students Push for Unionization, Science, available at

http://www.sciencemag.org/careers/2015/04/ivy-league-graduate-

students-push-unionization (“Graduate students’ concerns include

inadequate health insurance, high prices for dependent coverage on

student health insurance policies, and insufficient child care and family

leave support.”). Indeed, some scholarship suggests that universities

are actively seeking to derive greater profit from instructional and

research activities, and to lower their teaching costs. Such endeavors—

to lower labor costs to increase profit—are hallmarks of the sort of

economic dynamic in which, historically, employees’ bargaining rights

have played an important countervailing role. See generally The Cor-

poratization of Higher Education, 39 Monitor on Psychology 50

(2008), available at http://www.apa.org/monitor/2008/12/higher-

ed.aspx.

Indeed, it is important to note the policy judgment embodied in the

Act: that collective bargaining can help avert workplace unrest that

may occur in the absence of a process for employees to choose repre-

sentation, bargain collectively, and resolve disputes peacefully. See

NLRB v. Insurance Agents’ International Union, 361 U.S. 477, 488

(1960) (good-faith bargaining “may narrow the issues, making the real

demands of the parties clearer to each other, and perhaps to themselves,

and may encourage an attitude of settlement through give and take.”).

The Act is designed to lessen conflict by channeling disputes into struc-

tured negotiations and reflects the judgment of Congress that collective

bargaining, with its occasional attendant workplace conflicts such as

strikes and lockouts, is a right to be accorded broadly and across many

industries.

It is noteworthy that at NYU, graduate assistants struck after the

Board reversed its NYU decision and the school withdrew recognition

from their union. Without the protection of the Act, student assistants

lacked recourse to the orderly channels of bargaining and instead chose

to resort to more a disruptive means of resolving their dispute with the

University.

the Act, there may be significant turnover or other

changes involving affected employees. It goes without

saying that the resolution of cases under the Act, both

representation and unfair labor practice cases, before the

Board and the courts can be time-consuming. However,

this is simply not a basis on which to deny the Act’s pro-

tections to student assistants. The alternative—to deny

coverage because of the effects of procedural delays—

would seem to countenance the denial of the Act’s cov-

erage to large groups of employees whose tenures are

short or industries where there is a rapid pace of

change.

In sum, there is no compelling reason—in theory or in

practice—to conclude that collective bargaining by stu-

dent assistants cannot be viable or that it would seriously

interfere with higher education. We have put supposi-

tions aside today and have instead carefully considered

the text of the Act as interpreted by the Supreme Court,

the Act’s clearly stated policies, the experience of the

Board, and the relevant empirical evidence drawn from

collective bargaining in the university setting. This is not

a case, of course, where the Board must accommodate

the National Labor Relations Act with some other federal

statute related to private universities that might weigh

against permitting student assistants to seek union repre-

sentation and engage in collective bargaining. Finding

student assistants to be statutory employees, and permit-

In cases involving seasonal workplaces or those with significant

turnover, the Board has held elections even though, when there have

been employer challenges, bargaining may not begin until well after the

election. In this connection, the Board generally presumes that new

employees support the union in the same proportion as those who vot-

ed. See generally NLRB v. Curtin Matheson Scientific, Inc., 494 U.S.

775, 779 (1990). Otherwise, it would be difficult for employees in any

workplace with high turnover to ever achieve representation, because

the nature of administrative adjudication, as well as the provision of

due process to an employer’s challenges to certification may delay a

final ruling on certification. Conversely, when an employer's operation

is seasonal in nature or otherwise involves peaks and valleys in em-

ployment, the Board retains the discretion to adjust the election date to

ensure that a representative group of employees will be able to express

their choice concerning representation. See, e.g., Tusculum College,

199 NLRB 28, 33 (1972) (adjusting date of election to the beginning of

the fall semester to ensure that a representative complement of the

petitioned-for faculty would have an opportunity to express their wish-

es).

In unfair labor practice cases, remedies are commonly applied after

affected individuals have ceased working for an employer. But reme-

dies are tailored to address the violation of the Act relative to the con-

text in which they are applied. Backpay or other monetary remedies

may still be appropriate even after an individual is no longer in an

employment relationship with a respondent. Other remedies, such as in

a bad-faith bargaining case, involve the unit as a whole, and thus may

be implemented in the future, notwithstanding the turnover of individu-

al employees. Notice-posting requirements ensure that future comple-

ments of employees are aware of any violations and provide assurances

that the violations will not continue in the future.

COLUMBIA UNIVERSITY

ting them to seek union representation, does not conflict

with any federal statute related to private universities, as

far as we can discern. Certainly the Brown University

Board cited no statutory conflict, nor have the parties and

amici in this case.

Our conclusion is that affording

student assistants the right to engage in collective bar-

gaining will further the policies of the Act, without en-

gendering any cognizable, countervailing harm to private

higher education.

Accordingly, we overrule Brown University and hold

that student assistants who have a common-law employ-

ment relationship with their university are statutory em-

ployees entitled to the protections of the Act.

Where a party does not actually raise a supposed conflict between

the Act and another federal statute, the Board is not required to consid-

er the issue. Can-Am Plumbing, Inc., 350 NLRB 947, 947–948 (2007).

For his part, our dissenting colleague takes us to task for failing to

accommodate the Act with the “broad range of federal statutes and

regulations [that] apply to colleges and universities,” which “govern,

among other things, the accreditation of colleges and universities, the

enhancement of quality, the treatment of student assistance, gradu-

ate/postsecondary improvement programs, and the privacy of student

records.” But our colleague does not explain how any one of these

education statutes and regulations bears on the specific issue posed in

this case or how the Board should accommodate the Act to them—short

of not applying our statute at all to student assistants. That alternative,

of course, is disfavored, unless a conflict between two federal statutes

is truly irreconcilable. See Lewis v. Epic Systems Corp., 823 F.3d

1147, 1157 (7th Cir. 2016) (rejecting asserted conflict between NLRA

and Federal Arbitration Act). There is no such conflict here, as we

have already explained. See fn. 50, supra.

That an industry or economic sector is governed in certain respects

by other federal laws, in addition to being covered by the NLRA, can-

not mean that the Board must determine, in the abstract, whether the

general policies of those other laws might be better accomplished if the

Act did not apply, notwithstanding the absence of any exemption from

coverage in the statutory text or the legislative history. It is far too late

in the day—45 years after the Board’s decision in Cornell University,

supra—to argue that the Act cannot safely be applied to private univer-

sities. The supposed conflicts our colleague conjures would seem to

arise, for example, in cases where the Board applies the Act to universi-

ty faculty members, but there is no suggestion in our case law that an

actual conflict has ever arisen (or even been raised by a university

employer).

Our colleague identifies one specific, potential conflict between the

NLRA and federal education law related to the disclosure of education-

al records, where Sec. 8(a)(5) of the Act, which grants unions the right

to information necessary for carrying out their collective-bargaining

duties, conceivably could require a disclosure that the Family Educa-

tional Rights and Privacy Act (FERPA) might otherwise prohibit. Any

such conflict can and should be addressed in the particular factual set-

ting in which it arises. Suffice it to say that the Act recognizes that a

union’s right to information may, in a particular context, be subordinat-

ed to a legitimate confidentiality interest. See, e.g., Olean General

Hospital, 363 NLRB No. 62, slip op. at 7–8 (2015) (considering state

laws protecting patient confidentiality). See generally Detroit Edison

Co. v. NLRB, 440 U.S. 301 (1979).

III

. APPLICATION TO THIS CASE

We now apply our holding to the facts of this case.

For the reasons that follow, we conclude (1) that all of

the petitioned-for student-assistant classifications consist

of statutory employees; (2) that the petitioned-for bar-

gaining unit (comprising graduate students, terminal

Master’s degree students, and undergraduate students) is

an appropriate unit; and (3) that none of the petitioned-

for classifications consists of temporary employees who

must be excluded from the unit by virtue of the limited

length of their employment.

A. Statement of Facts

Columbia is a nonprofit postsecondary educational in-

stitution located in New York City. Columbia’s major

sources of annual revenue include tuition (net revenue of

$940 million in 2015, nearly a quarter of operating reve-

nue) and government contracts and grants ($750 mil-

lion).

Graduate students at Columbia are selected by the fac-

ulty of the academic departments into which they are

accepted on the basis of academic performance, as

demonstrated by educational background and standard-

ized test scores. In general, Ph.D. candidates spend five

to nine years of study within their discipline, during

which they take coursework, as well as prepare a doctor-

al thesis, that the candidates develop with guidance of

faculty or in connection with their laboratory work. Dur-

ing their enrollment, candidates are subject to various

academic requirements, including timely progress toward

a thesis and proficiency in coursework. Most Ph.D. can-

didates are required to take on teaching duties for at least

one semester as part of their academic requirements, alt-

hough many departments require additional semesters of

teaching as a condition for obtaining a degree.

Columbia fully funds most Ph.D. student assistants,

typically providing tuition and a stipend, for at least their

See Consolidated Financial Statements, The Trustees of Columbia

University in the City of New York, available at

http://finance.columbia.edu/files/gateway/content/reports/financials201

5.pdf. Tuition and government grants are the second and third largest

revenue streams. Columbia, which operates a hospital, earned the most

revenue, approximately $1 billion, from patient care.

A substantial portion of the tuition revenue comes from students in

graduate and professional programs, including law and business. How-

ever, the approximately 8500 undergraduates at Columbia, paying

approximately $25,000 in base tuition per semester, generate about

$376 million in tuition revenue, of which only $148 million is offset by

grant aid from Columbia. See Columbia University, Institutional Grant

Aid by School, 2014–2015, available at

http://www.columbia.edu/cu/opir/abstract/opir_institutional_grant_aid_

1.htm.

There is some variation as to the specifics of degree program re-

quirements and the nature of student assistant duties across schools and

programs at Columbia.

DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD

first 5 years of study. In most students’ second through

fourth year, taking on teaching or research duties is a

condition for full receipt of such funding. For most

Ph.D. candidates, the first and fifth years are funded

without a condition of service. In students’ sixth year

and beyond, teaching-based support may be available.

Research-based financial support, unlike teaching sup-

port, frequently comes in whole or in part from sources

outside the University. Grants from government or other

outside entities, generally to support a specified research

task, often cover research assistants’ financial awards.

However, the University will make up any shortfall if

outside grants provide a level of funding that falls below

the standard graduate funding package.

Terminal Master’s degree students (as opposed to

those who earn the degree as an intermediate step toward

earning a Ph.D.) typically earn their degrees in shorter

time periods and do not prepare a thesis. They receive

very little financial aid, although some take on teaching

duties for which they receive compensation.

The nature of teaching duties for a teaching assis-

tantship varies. Columbia’s teaching assistants, known

as Instructional Officers, fall into various subsidiary cat-

egories, which involve varying levels of discretion and

involvement in course design. Undergraduate, Master’s

degree, and Ph.D. student assistants can all serve in

teaching assistant roles, with some similarities in their

duties, although Ph.D. teaching assistants may take on

the most advanced duties. Notably, some Instructional

Officers teach components of the core curriculum, which

is Columbia’s signature course requirement for all un-

dergraduate students regardless of major. Instructional

Officers generally work up to 20 hours per week, and

they are typically appointed for one or two semesters at a

time.

Instructional Officers include the specific classifica-

tions of Teaching Fellows and Teaching Assistants.

Teaching Fellows are doctoral students in the Graduate

School of Arts and Sciences, while Teaching Assistants

may be either doctoral or Master’s degree students and

perform similar functions outside the Graduate School of

Arts and Sciences. Teaching Fellows and Teaching As-

sistants spend 15–20 hours a week undertaking a wide

range of duties with respect to a course. Their duties

may include grading papers and holding office hours,

leading discussion or laboratory sessions, or assuming

most or all the teaching responsibilities for a given

course. Columbia maintains other, specialized teaching-

assistant funding as well. Instructional officers who par-

ticipate in the Teaching Scholars Program, a category

that includes Ph.D. students who are somewhat advanced

in their studies, teach courses that they have designed for

undergraduates in their junior or senior years. Under-

graduates serve in the Teaching Assistant III classifica-

tion. They are responsible for grading homework and

running laboratory or problem sections that are ancillary

to large classes within the School of General Studies and

Columbia College.

The category of Instructional Officers also includes

classifications of Preceptors and Readers (sometimes

referred to as “Graders”). Preceptors are graduate assis-

tants who teach significant undergraduate courses with

high levels of independence. These positions are gener-

ally available only to graduate students far along in their

studies because they require the highest level of teaching

ability. Preceptors hold office hours, design and grade

all exams and assignments in their courses, and assign

final grades to their students. Readers/Graders are Mas-

ter’s degree students who are appointed to grade papers

under the direction of a course instructor. Finally,

Course Assistants, who are not Instructional Officers and

do not receive semester-long appointments, assist faculty

with administering classes by performing clerical tasks

that may include proctoring exams, printing and collect-

ing homework, answering students’ questions, and occa-

sionally grading assignments.

Research Officers generally participate in research

funded by outside entities. The research grants specify

the nature of the research and the duties of the individu-

als working on the grant. The revenue from the grant

beyond the amounts allotted to research assistantships

goes to Columbia’s general operating expenses. Gradu-

ate Research Assistants in Ph.D. programs must both

comply with the duties specified by the grant and simul-

taneously carry out research that they will ultimately

present as part of their thesis.

Departmental Research

Assistants, by contrast, are Master’s degree students and

are appointed and funded by the University and provide

research assistance to a particular department or school

within the University.

Teaching and research occur with the guidance of a

faculty member or under the direction of an academic

department. In the teaching context, poor performance

by an instructional officer is addressed through remedial

training, although in one instance poor performance re-

sulted in the University’s removal of a student’s teaching

duties, and the cancellation of his stipend.

B. Application of the Revised Section 2(3) Analysis

For the reasons that follow, we conclude that the peti-

tioned-for student-assistant classifications in this case

Training grant recipients are subject to slightly different condi-

tions, and are discussed below.

As reflected in its petition, the Union here seeks to represent:

COLUMBIA UNIVERSITY

comprise statutory employees: individuals with a com-

mon-law employment relationship with Columbia Uni-

versity. At the hearing before the Regional Director,

Columbia seemingly conceded

that, if the Board were

to adopt the common-law test, the petitioned-for individ-

uals—with the exception of students operating under

training grants—were employees under the Act. In its

brief to the Board on review, however, Columbia argues

that research assistants are not common-law employees,

citing the Board’s decision in Leland Stanford, supra.

With respect to teaching assistants, Columbia confines

itself to arguing that the common-law test should not be

the standard of statutory employment (a position we have

rejected). Below, we begin by examining the common-

law employment status of Columbia’s student assistants

generally. We then address arguments specific to the

status of Columbia’s research assistants and overrule

Leland Stanford because its reasoning cannot be recon-

ciled with the general approach we adopt today.

1. Instructional Officers

Common-law employment, as noted above, generally

requires that the employer have the right to control the

employee’s work, and that the work be performed in ex-

change for compensation.

100

That is the case here.

All student employees who provide instructional services, including

graduate and undergraduate Teaching Assistants (Teaching Assistants,

Teaching Fellows, Preceptors, Course Assistants, Readers and Grad-

ers): All Graduate Research Assistants (including those compensated

through Training Grants) and All Departmental Research Assistants

employed by the Employer at all of its facilities, including Morning-

side Heights, Health Sciences, Lamont-Doherty and Nevis facilities.

At the hearing, Columbia’s counsel stated, “If the Board finds that

students who provide services to their institutions are employees based

on common law test of employment, if you will, then our position

would be that the graduate research assistants and the teach-

ing assistants would be considered employees and part of an appropri-

ate bargaining unit, but that the students on training grants are simply

not employees because they’re not employed in a University position,

that they’re simply supported by the Government to be students and

they don’t provide a service to the University.”

This category encompasses the Teaching Assistants, Teaching Fel-

lows, Preceptors, and Readers/Graders named in the petition. Course

assistants, a classification named in the petition, do not appear to be

Instructional Officers and are not appointed on a semester-long basis.

The increments of their employment—they may work in less-than-

semester-long intervals—may raise questions about their eligibility.

However, we leave such determinations to the Regional Director in

determining an eligibility formula (as we discuss, infra) in the first

instance.

100

See, e.g., Seattle Opera, supra, 292 F.3d at 762 (“[T]he person

asserting statutory employee status [under the Act] does have such

status if (1) he works for a statutory employer in return for financial or

other compensation … and (2) the statutory employer has the power or

right to control and direct the person in the material details of how such

work is to be performed.”) (emphasis in original). Accord Restatement

of Employment Law §1.02 (2015) (“Where an educational institution

Here, the University directs and oversees student assis-

tants’ teaching activities. Indeed, the University pos-

sesses a significant interest in maintaining such control,

as the student assistants’ work advances a key business

operation of the University: the education of undergrad-

uate students. The record shows that teaching assistants

who do not adequately perform their duties to the Uni-

versity’s satisfaction are subject to corrective counseling

or removal.

Instructional officers receive compensation in ex-

change for providing services to the University. Receipt

of a full financial award is conditioned upon their per-

formance of teaching duties. When they do not perform

their assigned instructional duties, the record indicates

they will not be paid. For instance, after one assistant,

Longxi Zhao, was removed from his teaching assis-

tantship, his stipend was cancelled. In his termination

letter, the University indicated that, “[T]his termination

is effective immediately. As a result, you will no longer

receive a salary for this position.” This letter, in connec-

tion with the explicit conditioning of awards on perfor-

mance of teaching duties, demonstrates that the Universi-

ty offers student assistants stipends as consideration for

fulfilling their duties to perform instructional services on

the University’s behalf.

Although the payments to Ph.D. student assistants may

be standardized to match fellowship or other non-work

based aid, these payments are not merely financial aid.

Students are required to work as a condition of receiving

this tuition assistance during semesters when they take

on instructional duties, and such duties confer a financial

benefit on Columbia to offset its costs of financial aid,

even if it chooses to distribute the benefit in such a way

that equalizes financial aid for both assistants and non-

assistant students. Indeed, in semesters where a student

assistant would normally be required to work as a condi-

tion of funding, he or she may opt not to work only if he

or she finds a source of outside fellowship aid. Also, the

stipend portion of the financial package given to assis-

tants is generally treated as part of university payroll and

is subject to W-2 reporting and I-9 employment verifica-

tion requirements.

Even though it is unnecessary to delve into the ques-

tion of whether the relationship between student assis-

tants and their universities is primarily economic or edu-

cational—we have overruled the Brown University

standard—the facts in this case do not suggest a primari-

ly educational relationship, but rather, simply point to the

compensates student assistants for performing services that benefit the

institution . . . such compensation encourages the student to do the

work for more than educational benefits and thereby establishes an

employment as well as an educational relationship”).

DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD

difficulty of the analytical exercise required by the prior

approach.

While overlooked by the Brown University Board,

there is undoubtedly a significant economic component

to the relationship between universities, like Columbia,

and their student assistants. On average, private nonprof-

it colleges and universities generate a third of their reve-

nue from tuition, and 13 percent from government grants,

contracts and appropriations.

101

Columbia, for example,

generates nearly a billion dollars in annual tuition reve-

nue and over a half-million in government grants and

contracts.

102

Teaching assistants frequently take on a role akin to

that of faculty, the traditional purveyors of a university’s

instructional output. The teaching assistants conduct

lectures, grade exams, and lead discussions. Significant

portions of the overall teaching duties conducted by uni-

versities are conducted by student assistants. The dele-

gation of the task of instructing undergraduates, one of a

university’s most important revenue-producing activities,

certainly suggests that the student assistants’ relationship

to the University has a salient economic character.

While Columbia’s pool of student assistants consists of

enrolled students who were selected based on the Uni-

versity’s academic admissions process, this fact is not

inconsistent with an economic relationship. Students

pre-selected for their academic proficiency would natu-

rally tend to constitute a labor pool geared toward the

endeavor of teaching or researching in a university set-

ting, and their usage as instructors and researchers

achieves the efficiency of avoiding a traditional hiring

process for these jobs.

103

101

See National Center for Education Statistics, Postsecondary Rev-

enues by Source, available at

http://nces.ed.gov/programs/coe/pdf/Indicator_CUD/coe_cud_2015_06.

pdf.

102

See Consolidated Financial Statements, The Trustees of Colum-

bia University in the City of New York, available at

http://finance.columbia.edu/files/gateway/content/reports/ finan-

cials2015.pdf.

103

The claim that universities could more inexpensively hire adjunct

faculty to perform the duties also does not establish that the relationship

is primarily educational. Indeed, it is unclear that using students in

these roles is more costly to a university. As previously noted, a uni-

versity that makes use of an existing pool of student labor garners the

efficiency benefit of avoiding costly labor searches. Moreover, the

financial packages offered to graduate students are dictated in part by

the need to be competitive with other schools also seeking to attract top

graduate students. Although it may pay student assistants more com-

pensation than it would need to pay to attract an employee hired on the

open market, a university also receives the benefit of making itself

more attractive in recruiting graduate students. Compensation to a

student assistant is offset, then, by the benefits of hiring students. Thus,

because it fails to account for all the benefits that accrue to a university

by using its graduate students to fill assistantships, the argument that

And, the fact that teaching may be a degree require-

ment in many academic programs does not diminish the

importance of having students assist in the business of

universities by providing instructional services for which

undergraduate students pay tuition.

104

Indeed, the fact

that teaching assistants are thrust wholesale into many of

the core duties of teaching—planning and giving lec-

tures, writing exams, etc., including for such critical

courses as Columbia’s Core Curriculum—suggests that

the purpose extends beyond the mere desire to help in-

culcate teaching skills.

We have no difficulty, then, in finding that all of the

petitioned-for classifications here comprise statutory

employees—with the possible exception of research as-

sistants. That issue, as we explain next, is more compli-

cated in light of Board precedent.

2. Student Research Assistants

As indicated, Columbia argues that student research

assistants have no common-law employment relationship

with the University. It relies on Leland Stanford’s de-

termination that certain externally-funded research assis-

tants were not employees.

105

That holding was later ap-

plied by the NYU Board in finding that some research

assistants in that case were not statutory employees, even

as it reversed the overall exclusion of student assistants

from the Act’s protections.

106

Applying our holding to-

day regarding the employment status of student assis-

tants, we find that core elements of the reasoning in Le-

land Stanford are no longer tenable. We further find

that, under the common-law test discussed in our deci-

sion today, research assistants at Columbia are employ-

ees under the Act.

In Leland Stanford, the student research assistants re-

ceived external funding to cover their tuition while they

essentially went about pursuing their own individual ac-

student assistants are more costly to a university than an employee

hired in the free market is not self-evidently true.

104

As the American Association of University Professors, an organi-

zation that represents professional faculty—the very careers that many

graduate students aspire to—states in its brief, teaching abilities ac-

quired through teaching assistantships are of relatively slight benefit in

the attainment of a career in higher education. While the evidence does

suggest that graduate research assistantships dovetail more strongly

with the career/educational goals of graduate students than teaching

assistantships, it is by no means clear that education overshadows eco-

nomics in the case of research assistants either.

105

Supra, 214 NLRB at 623.

106

See NYU, supra, 332 NLRB at 1209 fn.10, citing Leland Stan-

ford, supra (“[W]e agree that the Sackler graduate assistants and the

few science department research assistants funded by external grants

are properly excluded from the unit [because] [t]he evidence fails to

establish that the research assistants perform a service for the Employer

and, therefore, they are not employees as defined in Section 2(3) of the

Act.”).

COLUMBIA UNIVERSITY

ademic goals in a manner of their own choosing. They

were not subject to discharge for failure to perform satis-

factory work, but would at worst receive a non-passing

grade for their coursework. Further, the Board conclud-

ed, the award the research assistants received was not

correlated to the nature or quantum of services they ren-

dered. The Board also contrasted the student research

assistants with non-student research associates who were

employees. The student researchers received none of the

fringe benefits that these non-student employee research

associates received. And these non-student research as-

sociates already had their academic degrees, were under

the direction of their department, and were subject to

discharge.

In view of these facts, the Stanford Board found that

the externally-funded research assistants were “primarily

students,” and concluded that their relationship with the

university was not one of employment because it was

“not grounded on the performance of a specific task

where both the task and the time of its performance is

designated and controlled by an employer [but] [r]ather it

is a situation of students within certain academic guide-

lines having chosen particular projects on which to spend

the time necessary.”

107

Leland Stanford thus, in many respects, focused upon

the absence of the common-law features of employment

of the externally-funded research assistants. It contrasted

the research assistants with non-student, employee re-

search associates who worked under the direction of their

department, and it noted that the work performed by the

externally-funded research assistants was largely done at

the students’ own discretion and for their own benefit. It

also observed that these student research assistants could

not be disciplined in a traditional sense. Their researcher

status, and presumably their aid award, was not termina-

ble based on a failure to meet any obligations of the

grant, undermining a claim that the aid was compensa-

tion.

However, the Leland Stanford decision arguably sug-

gested that the mere fact that the performance of a task

that advanced a student’s personal educational goals

could negate an employment relationship. It described

the status of the research assistants as akin to “the situa-

tion of all students,” who work on academic projects, and

suggested the importance of the fact that they were “sim-

ultaneously students” as well as researchers.

108

Because

Leland Stanford thus relied in part on the existence of a

student relationship in determining employee status, ra-

ther than determining whether a common-law relation-

107

214 NLRB at 623.

108

Id.

ship existed, we now overrule it, alongside Brown Uni-

versity, as inconsistent with the approach adopted today,

which better reflects the language and policies of the

Act.

109

The premise of Columbia’s argument concerning the

status of its research assistants is that because their work

simultaneously serves both their own educational inter-

ests along with the interests of the University, they are

not employees under Leland Stanford. To the extent

Columbia’s characterization of Leland Stanford is cor-

rect, we have now overruled that decision. We have re-

jected an inquiry into whether an employment relation-

ship is secondary to or coextensive with an educational

relationship. For this reason, the fact that a research as-

sistant’s work might advance his own educational inter-

ests as well as the University’s interests is not a barrier to

finding statutory-employee status.

110

Nonetheless, it remains the case that if a student re-

search assistant is not an employee under the common-

law test, we would not normally find the assistant to be

an employee under the Act. But there is nothing about

the nature of student-assistant research that would auto-

matically imply an absence of the requisite control under

the common-law test. It is theoretically possible that

funders may wish to further a student’s education by

effectively giving the student unconditional scholarship

aid, and allowing the student to pursue educational goals

without regard to achieving any of the funder’s own par-

ticular research goals. But where a university exerts the

requisite control over the research assistant’s work, and

specific work is performed as a condition of receiving

the financial award, a research assistant is properly treat-

ed as an employee under the Act.

The research assistants here clearly fall into this latter

category of common-law employees. The research of

Columbia’s student assistants, while advancing the assis-

tants’ doctoral theses, also meets research goals associat-

ed with grants from which the University receives sub-

stantial income.

111

The research assistants here work

under the direction of their departments to ensure that

109

Our dissenting colleague contends that the longstanding nature of

the Board’s Leland Stanford precedent favors leaving its holding—that

research assistants are not statutory employees—untouched. However,

as we have already explained, we believe that the better view is that

student assistants, including research assistants, should be regarded as

employees under the Act.

110

In this respect, our decision is entirely consistent with other em-

ployee-status decisions under the Act involving individuals who work,

at least in part, to advance personal interests. See fn. 39–41, supra.

111

One can conceive of countless employment situations where the

employee gains personally valuable professional experience and skills

while simultaneously performing a valuable service for his or her em-

ployer.

DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD

particular grant specifications are met. Indeed, another

feature of such funding is that the University typically

receives a benefit from the research assistant’s work, as it

receives a share of the grant as revenue, and it is relieved

of any need to find other sources of funding for graduate

students under a research grant; thus it has an incentive

to ensure proper completion of the work in accordance

with the grant. Further, a research assistant’s aid pack-

age requires fulfillment of the duties defined in the grant,

notwithstanding that the duties may also advance the

assistant’s thesis, and thus the award is compensation.

Students, when working as research assistants, are not

permitted to simply pursue their educational goals at

their own discretion, subject only to the general require-

ment that they make academic progress, as they would be

in semesters where they were under some form of finan-

cial aid other than a research grant.

112

The funding here is thus not akin to scholarship aid

merely passed through the University by a grantor with-

out specific expectations of the recipients. Because Co-

lumbia directs the student research assistants’ work and

the performance of defined tasks is a condition of the

grant aid, we conclude that the research assistants in this

case are employees under the Act.

113

Columbia argues that, even if research assistants gen-

erally are common-law employees, the research assis-

tants funded by a specific form of grants known as train-

ing grants present unique circumstances and lack the

characteristics of common-law employment. However,

the record shows that Columbia, which receives revenue

from these training grants, is charged with ensuring that

research assistants thereunder receive appropriate train-

ing within a formalized program (consistent with the

funder’s goal of having a well-trained workforce in bio-

medical and behavioral research), and accordingly it

oversees and directs the research assistants who receive

the grants. Additionally, research assistants often receive

funds from research and training grants simultaneously.

Further, participation in specific training activities is a

requirement for receipt of training grants; thus, notwith-

112

Stanford found that the fact that the university equalized financial

packages for research assistants and other graduate students suggested

that funding for research assistants was financial aid and not compensa-

tion. 214 NLRB at 622. As previously discussed, we do not believe

that Columbia’s practice to distribute the benefits it receives from stu-

dent-assistant labor, in order to equalize aid packages, demonstrates

that funding for an assistantship is not compensation, given that the

research work assigned in a given semester is a requirement for receipt

of aid.

113

Indeed, in NYU, the Board upheld the Regional Director’s deter-

mination that those research assistants who were “assigned specific

tasks and . . . [who] work[ed] under the direction and control of the

faculty member,” were employees eligible for inclusion in the unit. See

NYU, supra, 332 NLRB at 1221 fn.51.

standing the grantor’s statement that the grant aid is not

salary, it is a form of compensation.

C. Student Assistants in the Petitioned-for Unit Share a

Community of Interest

We now turn to the question of whether the petitioned-

for unit is a unit appropriate for collective bargaining.

Columbia argues that the petitioned-for unit is inappro-

priate because it groups undergraduate and Master’s de-

gree student assistants together in the same unit with

Ph.D. assistants. According to Columbia, differences in

pay and benefits, duties, and remunerative interests

demonstrate the absence of a community of interest. We

disagree.

The first and central right that Section 7 of the Act

grants employees is “the right to self-organization, to

form, join, or assist labor organizations, to bargain col-

lectively through representatives of their own choosing

. . . .” Section 9(b) provides that “[t]he Board shall de-

cide in each case whether, in order to assure to employ-

ees the fullest freedom in exercising the rights guaran-

teed by this Act, the unit appropriate for the purposes of

collective bargaining shall be the employer unit, craft

unit, plant unit, or subdivision thereof.” As recognized

by the Supreme Court, Section 9(a) “suggests that em-

ployees may seek to organize ‘a unit’ that is ‘appropri-

ate’—not necessarily the single most appropriate unit.”

114

In other words, “[m]ore than one appropriate bargaining

unit logically can be defined in any particular factual

setting.”

115

In making the determination of whether the proposed

unit is an appropriate unit, the Board's “focus is on

whether the employees share a ‘community of inter-

est.”’

116

In determining whether employees in a pro-

posed unit share a community of interest, the Board ex-

amines:

[W]hether the employees are organized into a separate

department; have distinct skills and training; have dis-

tinct job functions and perform distinct work, including

inquiry into the amount and type of job overlap be-

tween classifications; are functionally integrated with

the Employer's other employees; have frequent contact

with other employees; interchange with other employ-

114

American Hospital Assn., 499 U.S. 606, 610 (1991) (emphasis in

original). See also Specialty Healthcare & Rehabilitation Center of

Mobile, 357 NLRB 934, 942 (2011), enfd. 727 F.3d 552 (6th Cir.

2013).

115

Country Ford Trucks, Inc. v. NLRB, 229 F.3d 1184, 1189 (D.C.

Cir. 2000) (quoting Operating Engineers Local 627 v. NLRB, 595 F.2d

844, 848 (D.C. Cir. 1979)).

116

Specialty Healthcare, supra, 357 NLRB at 942, quoting NLRB v.

Action Automotive, Inc., 469 U.S. 490, 491 (1985).

COLUMBIA UNIVERSITY

ees; have distinct terms and conditions of employment;

and are separately supervised.

117

Further, to honor the statutory command to maximize

employees’ freedom in choosing a representative, the

Board has held that a petitioner's desire concerning the

unit “is always a relevant consideration.”

118

Although

Section 9(c)(5) of the Act provides that “the extent to

which the employees have organized shall not be control-

ling,” the Supreme Court has made clear that the extent

of organization may be “consider[ed] . . . as one factor”

in determining if the proposed unit is an appropriate

unit.

119

We thus consider the unit expressed in the peti-

tion to be a factor, although not a determinative one.

In Specialty Healthcare and Rehabilitation Center of

Mobile,

120

the Board held that a unit is appropriate if the

employees in the proposed unit constitute a readily iden-

tifiable grouping and share a community of interest.

Here, the proposed unit consists of a readily identifiable

grouping of employees: all student employees who pro-

vide instructional services and all research assistants at

Columbia University’s campuses. We further find that

the employees in the unit share a community of interest

and agree with the Regional Director who, in concluding

the unit was appropriate, found substantial similarities

among the types of work of all the student assistants in

the proposed unit. She noted that they work in similar

settings (in labs and classrooms at the same university)

and serve similar functions with respect to the Universi-

ty’s fulfillment of its teaching and research mission.

Thus, the petitioned-for unit of student assistants per-

forming instructional services and research services pro-

vide supplemental educational services to the faculty and

the University and therefore constitute a readily identifi-

able grouping of employees within the University’s op-

erations that share a community of interest.

While Columbia argues that there are some dissimilari-

ties, such as differences in the difficulty and independ-

ence of work assignments, as well as in pay and benefits,

among the categories of student assistants, we find that—

although there might potentially be other appropriate unit

groupings among these student assistants—the peti-

tioned-for classifications share a sufficient community of

interest to form an appropriate unit. We note that all of

the student assistants here are performing a supplemental

educational service. That is, their duties are functionally

integrated into a system designed to meet the university’s

teaching and research missions in non-faculty roles. Alt-

117

Specialty Healthcare, supra, 357 NLRB at 942 (quoting United

Operations, Inc., 338 NLRB 123, 123 (2002)).

118

Marks Oxygen Co., 147 NLRB 228, 230 (1964).

119

NLRB v. Metropolitan Life Ins. Co., 380 U.S. 438, 442 (1965).

120

357 NLRB at 946.

hough some of the assistantships undertaken by Ph.D.

students may involve advanced duties, in many cases

their roles are similar to those of Master’s and under-

graduate assistants who fill related positions. And even

when the Ph.D. assistants take on more advanced roles,

there is often still an overlap of job duties with Master’s

and undergraduate student assistants.

121

Further, all student assistants work under the direction

of the University. Most are appointed on a semester-long

basis and are paid in part through a tuition remission and

in part via a bimonthly stipend. Although it is the con-

tention of our dissenting colleague that the “broad array”

of employees within the unit militates against its appro-

priateness, we note that the Act countenances broad units

where there are factors establishing a community of in-

terest. For example, the Board has held, consistent with

the Act’s text, that similarly situated employees can form

an appropriate employer-wide unit.

122

We find under

these circumstances that differences in level and type of

compensation

123

and some differences in the nature of

work assignments, do not negate the shared community

of interest of employees in the petitioned-for unit, given

the many other relevant similarities.

Columbia also argues that, because they are in shorter-

term degree programs geared toward rapid graduation

and job-market entry, Master’s and undergraduate stu-

dent assistants are less likely to be concerned with issues

of housing costs, quality of health care, and availability

of dependent health coverage. Assuming the veracity of

Columbia’s speculation regarding Master’s and under-

graduate students’ likely priorities, it is nonetheless the

case that classifications in a unit need not have complete

identity of interests for the unit to be appropriate. While

Master’s and undergraduate assistants may, arguably,

121

Notably, a Columbia Vice-Provost testified to “considerable simi-

larity” between the teaching duties of a Master’s degree teaching assis-

tant and a Ph.D. appointed as a teaching fellow, ostensibly a higher-

difficulty position. The record also indicates that, while Ph.D. assis-

tants take on the most advanced and independent teaching assignments,

there are various types of discrete teaching duties, such as grading and

leading discussion sections, that are performed by both undergradu-

ate/Master’s student assistants and Ph.D. assistants.

122

Sec. 9(b) provides, inter alia, that: “The Board shall decide in

each case whether . . . the unit appropriate for the purposes of collective

bargaining shall be the employer unit, craft unit, plant unit, or subdivi-

sion thereof.” 29 U.S.C. 159(b). See also Hazard Express, Inc., 324

NLRB 989 (1997) (finding appropriate a unit including “drivers, dock-

workers, and helpers and excluding clericals); Jackson’s Liquors, 208

NLRB 807, 808 (1974) (“not[ing] that the employerwide unit, being

one of the units listed in the Act as appropriate for bargaining, is pre-

sumptively appropriate”).

123

See Los Angeles Water & Power Employees’ Assn., 340 NLRB

1232, 1236 (2003) (difference in pay not controlling as to unit determi-

nation in light of other similarities); Four Winds Services, 325 NLRB

632, 632 (1998) (same).

DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD

have some different priorities from those of Ph.D. assis-

tants, there are also overarching common interests. For

most student assistants, there will be a shared desire to

successfully balance coursework with job responsibili-

ties, as well as a shared desire to mitigate the tuition and

opportunity costs of being a student. Additionally, all

student assistants are likely to share a desire to address

policies affecting job postings, pay periods, stipend dis-

bursement, and personal health insurance coverage. Stu-

dent assistants also have common interests in developing

guidelines for discipline and discharge and establishing a

grievance-and-arbitration procedure. While Ph.D. assis-

tants, as longer-term students, may be somewhat more

concerned with certain types of remuneration, such as

housing subsidies, their interests are certainly not at odds

with those of the shorter-term employees. Indeed the

unit’s overarching interest in addressing issues pertaining

to one’s simultaneous employment and enrollment as a

student provides ample basis on which to pursue a com-

mon bargaining agenda.

Therefore, applying traditional community of interest

factors to these facts, we conclude that the petitioned-for

unit is an appropriate unit.

124

D. None of the Petitioned-for Classifications Contain

Temporary Employees Who Must Be Excluded From the

Unit

Columbia argues that certain classifications must be

excluded from the unit because they comprise temporary

employees, who may not be included in the unit. We

reject this argument.

In its analysis of whether an employee should be ex-

cluded from a unit as a “temporary employee,” the Board

focuses on “the critical nexus between an employee's

temporary tenure and the determination whether he

shares a community of interest with the unit employ-

ees.”

125

To determine whether an alleged temporary em-

ployee shares a community of interest, the Board exam-

ines various factors, including “whether or not the em-

ployee’s tenure is finite and its end is reasonably ascer-

tainable, either by reference to a calendar date, or the

completion of a specific job or event, or the satisfaction

of the condition or contingency by which the temporary

employment was created.”

126

124

We stress that the bargaining relationship here pertains only to

undergraduates’ employment relationship and does not interfere with

any other role the university may play with respect to students’ aca-

demic or personal development. Since undergraduate student assistants

share a community of interest with the other student assistants, they are

appropriately included in the same unit.

125

Marian Medical Center, 339 NLRB 127, 128 (2003).

126

Id.

However, the determination is not based on the nature

of an employee’s tenure in a vacuum; rather, the nature

of the alleged temporary employees’ employment must

be considered relative to the interests of the unit as a

whole. The practice of excluding temporary employees

from a unit merely recognizes that, “as a general rule,”

employees of a defined, short tenure are “likely” to have

divergent interests from the rest of the unit.

127

Here, Columbia argues that undergraduate and termi-

nal Master’s assistants in the petitioned-for unit are

“temporary” in the sense that they are employed for rela-

tively short, finite periods of time, averaging only about

two (not necessarily continuous) semesters of work.

However, all the employees in this unit, which we find to

be an appropriate, serve finite terms. Although the Ph.D.

student assistants typically serve for the longest periods,

all the classifications perform similar duties in (not nec-

essarily continuous) semester increments.

128

Thus, in

some sense, one could argue that all the student assistants

here are temporary. Yet the Board has made clear that

finite tenure alone cannot be a basis on which to deny

bargaining rights, because “[i]n many employment rela-

tionships, an employee may have a set tenure and, in that

sense, may not have an indefinite departure date . . . . To

extend the definition of ‘temporary employee’ to [all]

such situations, however, would be to make what was

intended to be a limited exception swallow the whole.”

129

Therefore we must look beyond the finite tenure of the

student assistants at issue, and consider whether and to

what extent their tenure affects their community of inter-

est with the unit or their ability to engage in meaningful

bargaining.

130

Further, we find that Master’s and undergraduate stu-

dent assistants’ relatively short tenure, within the context

of this unit, does not suggest a divergence of interests

that would frustrate collective bargaining.

131

In Manhat-

127

Id.

128

Indeed, to exclude Master’s and undergraduate student assistants

here who share a community of interest with the unit as a whole might

undercut the integrity of the overall bargaining unit, because these

employees perform not-readily differentiable work compared to Ph.D.

student assistants, and thus could easily be utilized as substitutes for

bargaining unit employees. See generally Outokumpu Copper Frank-

lin, Inc., 334 NLRB 263, 263 (2001) (temporary employees who

worked work side-by-side at same jobs under same supervision as other

employees were properly included in unit).

129

Boston Medical, supra, 330 NLRB at 166.

130

To the extent that cases like San Francisco Art Institute, 226

NLRB 1251 (1976), suggest that the mere fact of being a student in

short-term employment with one’s school renders one’s interests in the

employment relationship too “tenuous,” such cases are incompatible

with our holding here today and are overruled.

131

This case is distinguishable from Goddard College, 216 NLRB

457, 458 (1975), cited by the dissent. In that case there was a signifi-

cant difference in employment expectations between the visiting pro-

COLUMBIA UNIVERSITY

tan College, 195 NLRB 65, 65–66 (1972), the Board

found that faculty members on terminal contracts shared

a community of interest with their colleagues for the du-

ration of their employment and were therefore properly

included in a faculty bargaining unit.

132

The Board has

never held that, regardless of community of interest with

the broader unit, arguably temporary employees should

be denied bargaining rights altogether.

133

If, under the

specific circumstances of a case, alleged temporary em-

ployees possess a sufficient interest in bargaining over

terms and conditions of employment to allow for suc-

cessful and stable collective bargaining on their behalf,

they are permitted to bargain collectively within an ap-

propriate unit.

134

Here, even the Master’s and undergraduate student as-

sistants typically serve more than one semester—and

thus their tenure is not so ephemeral as to vitiate their

interest in bargaining over terms and conditions of em-

ployment. Indeed, a semester at Columbia is not some

insignificant or arbitrary period of time spent performing

a task, but rather it constitutes a recurring, fundamental

unit of the instructional and research operations of the

University. And notwithstanding the length of any indi-

vidual assistant’s tenure, the University will continuously

employ groups of Master’s and undergraduate student

assistants to perform research and instructional duties

across semesters (and, although the precise composition

of these groups will differ from semester to semester,

there will typically be some individual student assistants

fessors, who typically worked at the university less than a year, and

tenured faculty. The question there was whether it would be appropri-

ate to combine individuals with a long-term relationship to an employer

with those whose long-term interests lay elsewhere. Here, as we have

noted, all the employees—both the allegedly temporary Master’s and

undergraduate assistants, as well as the Ph.D. assistants—serve finite

periods in semester increments, during which they perform similar

services to the University. Further, as student assistants working during

academic semesters, they will share common interests in addressing

issues unique to their status, such as those involving the balancing of

work and academic obligations. See Goucher College, 364 NLRB No.

71, slip op. at 2 & fn. 7 (2016) (finding that terminal-appointment fac-

ulty should be included in unit because they share a community of

interest with nontenure faculty “that is more significant than whether or

not they have a reasonable expectation of reappointment”). Also, un-

like the visiting faculty in Goddard, most student assistants possess a

long-term goal of achieving employment elsewhere, as opposed to

cultivating a permanent relationship with the University.

132

See also Rensselaer Polytechnic Institute, 218 NLRB 1435, 1437

(1975).

133

See Kansas City Repertory Theatre, 356 NLRB 147, 147 (2010)

(“The logical consequence of the Employer's argument is that tempo-

rary or intermittent employees cannot exercise the rights vested in

employees by Section 9 of the Act. However, no such exclusion ap-

pears in the definition of employees or elsewhere in the Act.”).

134

See Manhattan College, supra, at 65–66.

who are carried over from one semester to another).

135

Because the University’s employment of Master’s and

undergraduate student assistants is regularly recurring,

with some carryover between semesters, and their indi-

vidual tenures are neither negligible nor ad hoc, we be-

lieve that as a group, they, together with the Ph.D. assis-

tants, form a stable unit capable of engaging in meaning-

ful collective bargaining.

136

Accordingly, we find that none of the petitioned-for

classifications consists of temporary employees who

should be excluded from the unit by virtue of their finite

tenure of employment.

E. Voting Eligibility Formula

There remains the issue of which of the employees in

the petitioned-for unit—some of whom, on account of

intermittent semester appointments, may not be eligible

to vote under the Board’s traditional eligibility date ap-

proach—should nonetheless be permitted to vote because

of their continuing interest in the unit. In this connec-

tion, although it does not fully address the eligibility

question, the Petitioner has suggested in its brief that

student assistants who have been appointed for at least

one semester should be deemed eligible.

We observe that the unique circumstances of student

assistants’ employment manifestly raise potential voter

eligibility issues. The student assistants here tend to

work for a substantial portion of their academic career,

but not necessarily in consecutive semesters; thus, during

any given semester, individuals with a continuing interest

in the terms and conditions of employment of the unit

may not be working. The Board has long recognized that

certain industries and types of employment, particularly

those with patterns of recurring employment, may neces-

sitate rules governing employee eligibility. The Board

135

We are not, as the dissent suggests, establishing a special rule for

student assistants. Rather, we are applying relevant principles concern-

ing the establishment of units of employees, including some with rela-

tively short, finite tenures, to the particular circumstances of student

assistants. The question we must ask before denying a category of

employees the right to bargain collectively is whether their tenure pre-

cludes meaningful bargaining. Otherwise, to deny bargaining rights

merely because one has a short tenure, would be antithetical to the Act.

The evidence here indicates that meaningful bargaining is possible

within such a unit. Notably, student assistant collective bargaining at

public universities sometimes involves units of students without exclu-

sions based on expected duration of employment; yet there is no evi-

dence that this has proven an impediment to effective bargaining. See,

e.g., Collective Bargaining Agreement between Michigan State Univer-

sity and The Graduate Employees Union, Local 6196, AFT-

Michigan/AFL–CIO (May 2015—May 2019), supra.

136

Cf. Kelly Bros. Nurseries, 140 NLRB 82, 85 (1962) (category of

seasonal employees properly included in a bargaining unit where the

employer relied on these employees to serve over recurring production

seasons, and where there was some employee holdover from season to

season).

DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD

attempts to strike a balance between the need for an on-

going connection with a unit and concern over disenfran-

chising voters who have a continuing interest notwith-

standing their short-term, sporadic, or intermittent em-

ployment.

137

Setting such rules on a pre-election basis

by use of eligibility formulas also serves the efficiency

goal of avoiding protracted post-election litigation over

challenges to individual voters.

Such eligibility formulas attempt to include employees

who, despite not being on the payroll at the time of the

election, have a past history of employment that would

tend to signify a reasonable prospect of future employ-

ment.

138

We have traditionally devised these formulae

by examining the patterns of employment within a job or

industry, and determining what amount of past employ-

ment serves as an approximate predictor of the likelihood

of future employment.

For example, in a case involving adjunct faculty, the

Board noted the importance of “prevent[ing] an arbitrary

distinction” which disenfranchises employees with a con-

tinuing interest in their employment within the unit but

who happen not to be working at the time of the elec-

tion.

139

In the particular circumstances of that case, the

Board looked at factors including whether adjuncts had

signed teaching contracts and the extent to which they

had actually taught over previous semesters.

140

Here, the record contains data concerning the average

number of semesters worked relative to a student assis-

tant’s time enrolled at the University, as well as data

concerning typical patterns of work over the academic

career of a Ph.D. student assistant. But neither the Re-

gional Director nor the parties have specifically ad-

dressed what an appropriate formula would be under

these circumstances. Having determined the appropriate

unit, we therefore remand this case and instruct the Re-

gional Director to take appropriate measures, including

reopening the record, if necessary, to establish an appro-

priate voting eligibility formula.

ORDER

The Regional Director’s Decision is reversed. The

proceeding is remanded to the Regional Director for fur-

137

See Steiny & Co., 308 NLRB 1323, 1325 (1992); Trump Taj Ma-

hal Casino, 306 NLRB 294 (1992).

138

See Trump Taj Mahal Casino, supra, 306 NLRB at 296 (reiterat-

ing Board’s obligation to be “flexible in . . . devis[ing] formulas . . . to

afford employees with a continuing interest in employment the opti-

mum opportunity for meaningful representation”).

139

C.W. Post Center, 198 NLRB 453, 454 (1972).

140

Id. See also Steiny, supra, 308 NLRB at 1326 (articulating for-

mula for construction work that accounts for non-continuous work

patterns).

ther appropriate action consistent with this Decision and

Order.

Dated, Washington, D.C. August 23, 2016

______________________________________

Mark Gaston Pearce, Chairman

______________________________________

Kent Y. Hirozawa, Member

______________________________________

Lauren McFerran, Member

(

SEAL) NATIONAL LABOR RELATIONS BOARD

EMBER MISCIMARRA, dissenting.

In this case, my colleagues decide that college and

university students are “employees” for purposes of col-

lective bargaining under the National Labor Relations

Act (NLRA or Act) when serving in a variety of academ-

ic assistant positions. An assortment of student positions

are involved here: the petitioned-for bargaining unit in-

cludes all “student employees” who engage in “instruc-

tional services,” including “graduate and undergraduate

Teaching Assistants,” “Teaching Fellows,” “Preceptors,”

“Course Assistants,” “Readers,” and “Graders,” plus

“Graduate Research Assistants” and “Departmental Re-

search Assistants.” No distinctions are drawn based on

subject, department, whether the student must already

possess a bachelor’s or master’s degree, whether a par-

ticular position has other minimum qualifications,

whether graduation is conditioned on successful perfor-

mance in the position, or whether different positions are

differently remunerated. As a result of today’s decision,

all of these university student assistant positions

are

made part of a single, expansive, multi-faceted bargain-

ing unit.

I believe the issues raised by the instant petition re-

quire more thoughtful consideration than the Board ma-

jority’s decision gives them. In particular, my colleagues

For ease of reference, I use the terms college and university inter-

changeably. For the same reason, I use the term student assistants to

refer to all types of students encompassed within the petitioned-for

bargaining unit—i.e., all “student employees” who engage in “instruc-

tional services,” including “graduate and undergraduate Teaching As-

sistants,” “Teaching Assistants,” “Teaching Fellows,” “Preceptors,”

“Course Assistants,” “Readers,” and “Graders,” as well as “Graduate

Research Assistants” and “Departmental Research Assistants.”

COLUMBIA UNIVERSITY

disregard a fundamental fact that should be the starting

point when considering whether to apply the NLRA to

university students. Full-time enrollment in a university

usually involves one of the largest expenditures a student

will make in his or her lifetime, and this expenditure is

almost certainly the most important financial investment

the student will ever make. In the majority of cases, at-

tending college imposes enormous financial burdens on

students and their families, requiring years of preparation

beforehand and, increasingly, years of indebtedness

thereafter. Many variables affect whether a student will

reap any return on such a significant financial invest-

ment, but three things are certain: (i) there is no guaran-

tee that a student will graduate, and roughly 40 percent

do not;

(ii) college-related costs increase substantially

the longer it takes a student to graduate, and roughly 60

percent of undergraduate students do not complete de-

gree requirements within four years after they commence

college;

and (iii) when students do not graduate at all,

there is likely to be no return on their investment in a

college education.

I respect the views presented by my colleagues and by

advocates on all sides regarding the issues in this case.

However, Congress never intended that the NLRA and

collective bargaining would be the means by which stu-

dents and their families might attempt to exercise control

over such an extraordinary expense. This is not a com-

mentary on the potential benefits associated with collec-

U.S. Department of Education, National Center for Education Sta-

tistics (NCES), The Condition of Education 2016, Undergraduate Re-

tention and Graduation Rates (excerpted at

https://nces.ed.gov/fastfacts/display.asp?id=40) (last visited Aug. 5,

2016). The NCES reports that the “6-year graduation rate for first-

time, full-time undergraduate students who began their pursuit of a

bachelor's degree at a 4-year degree-granting institution in fall 2008

was 60 percent. That is, 60 percent of first-time, full-time students who

began seeking a bachelor's degree at a 4-year institution in fall 2008

completed the degree at that institution by 2014.” Id.

Dr. Peter Cappelli, Will College Pay Off?, p. 18 (Public Affairs

2015) (hereinafter “Cappelli”). See also Complete College America,

Four-Year Myth: Make College More Affordable—Restore the Promise

of Graduating on Time, pp. 4, 6 (2014) (http://completecollege.org/wp-

content/uploads/2014/11/4-Year-Myth.pdf) (reporting that only 36

percent of students attending flagship 4-year college bachelor degree

programs graduate on time); Danielle Douglas-Gabriel, Why So Many

Students Are Spending Six Years Getting a College Degree, Washing-

ton Post, Dec. 2, 2014 (https://www.washingtonpost.

com/news/wonk/wp/2014/12/02/why-so-many-students-are-spending-

six-years-getting-a-college-degree/) (last visited Aug. 5, 2016).

University coursework may result in various personal benefits even

if students fail to earn a degree. However, there is little doubt that the

financial return on the investment required to attend college requires

graduation. See, e.g., Cappelli, p. 48 (“The biggest cost associated with

going to college, though, is likely to be the risk that a student does not

graduate on time or, worse, drops out altogether. There is virtually no

payoff from college if you don’t graduate.”).

tive bargaining in the workplace. Rather, it is a recogni-

tion that for students enrolled in a college or university,

their instruction-related positions do not turn the academ-

ic institution they attend into something that can fairly be

characterized as a “workplace.” For students, the least

important consideration is whether they engage in collec-

tive bargaining regarding their service as research assis-

tants, graduate assistants, preceptors, or fellows, which is

an incidental aspect of their education. If one regards

college as a competition, this is one area where “winning

isn’t everything, it is the only thing,” and I believe win-

ning in this context means fulfilling degree requirements,

hopefully on time.

The Board has no jurisdiction over efforts to ensure

that college and university students satisfy their postsec-

ondary education requirements. However, Congress has

certainly weighed in on the subject: an array of federal

statutes and regulations apply to colleges and universi-

ties, administered by the U.S. Department of Education,

led by the Secretary of Education. My colleagues disre-

gard the Board’s responsibility to accommodate this ex-

tensive regulatory framework. In addition, I believe col-

lective bargaining and, especially, the potential resort to

economic weapons protected by our statute fundamental-

ly change the relationship between university students,

including student assistants, and their professors and ac-

ademic institutions. Collective bargaining often produc-

es short-term winners and losers, and a student assistant

in some cases may receive some type of transient benefit

as a result of collective bargaining pursuant to today’s

decision. Yet there are no guarantees, and they might

end up worse off. Moreover, I believe collective bar-

gaining is likely to detract from the far more important

goal of completing degree requirements in the allotted

time, especially when one considers the potential conse-

quences if students and/or universities resort to economic

weapons against one another. I also believe that the

Board’s processes and procedures are poorly suited to

deal with representation and unfair labor practice cases

involving students. Add these up, and the sum total is

uncertainty instead of clarity, and complexity instead of

The expression “winning isn’t everything, it’s the only thing” is

commonly attributed to legendary football coach Vince Lombardi, who

was the head coach for the Green Bay Packers from 1959 to 1967.

However, it appears to have been originated by Henry Russell (Red)

Sanders, who was the head coach of the University of California, Los

Angeles (UCLA) Bruins football team from 1949 to 1957. See Wik-

ipedia, Winning isn’t everything; it’s the only thing

(https://en.wikipedia.org/wiki/Winning_isn%27t_everything;_it%27s_t

he_only_thing) (last visited Aug. 5, 2016). Red Sanders also referred

to the rivalry between UCLA and the University of Southern California

(USC) and famously stated: “Beating [USC] is not a matter of life or

death, it’s more important than that.” Id.

DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD

simplicity, with the risks and uncertainties associated

with collective bargaining—including the risk of break-

down and resort to economic weapons—governing the

single most important financial decision that students and

their families will ever make.

For these reasons, I agree with former Member Brame,

who stated that the Board resembles the “foolish repair-

man with one tool—a hammer—to whom every problem

looks like a nail; we have one tool—collective bargain-

ing—and thus every petitioning individual looks like

someone’s ‘employee.’” Boston Medical Center Corp.,

330 NLRB 152, 182 (1999) (Member Brame, dissent-

ing). Accordingly, as explained more fully below, I re-

spectfully dissent.

ISCUSSION

The Board here changes the treatment that has been af-

forded student assistants throughout the Act’s history of

80 years, with the exception of a four-year period that

was governed by the Board’s divided opinion in New

York University (NYU).

Prior to NYU, the Board in

Adelphi University

and The Leland Stanford Junior Uni-

versity

held that various student assistants could not be

included in petitioned-for units. After NYU, the Board

similarly held that various student assistants were not

employees in Brown University.

I disagree with my colleagues’ decision to apply the

Act to college and university student assistants. In my

view, this change is unsupported by our statute, and it is

ill-advised based on substantial considerations, including

those that far outweigh whether students can engage in

collective bargaining over the terms and conditions of

education-related positions while attempting to earn an

undergraduate or graduate degree.

The Supreme Court has stated that “the authority struc-

ture of a university does not fit neatly within the statutory

scheme” set forth in the NLRA. NLRB v. Yeshiva Uni-

versity, 444 U.S. 672, 680 (1980). Likewise, the Board

has recognized that a university, which relies so heavily

on collegiality, “does not square with the traditional au-

thority structures with which this Act was designed to

cope in the typical organizations of the commercial

world.” Adelphi University, 195 NLRB at 648. The ob-

vious distinction here has been recognized by the Su-

preme Court and the Board: the lecture hall is not the

factory floor, and the “industrial model cannot be im-

posed blindly on the academic world.” Syracuse Univer-

332 NLRB 1205 (2000).

195 NLRB 639, 640 (1972).

214 NLRB 621(1974).

342 NLRB 483 (2004).

sity, 204 NLRB 641, 643 (1973); see also Yeshiva, 444

U.S. at 680.

The Board has an uneven track record in its efforts to

apply the NLRA to colleges, universities and other edu-

cational institutions. In Yeshiva, the Board summarily

rejected the university’s position that its faculty members

were managerial employees who were exempt from the

Act. The Supreme Court reversed, finding that the facul-

ty members constituted managerial employees and that

the Board’s conclusions were neither consistent with the

Act nor rationally based on articulated facts. 444 U.S. at

686–691.

In NLRB v. Catholic Bishop of Chicago, 440 U.S. 490

(1979), the Supreme Court rejected the Board’s exercise

of jurisdiction over lay faculty members at two groups of

Catholic high schools, concluding that to do so would

give rise to “serious First Amendment questions” involv-

ing church/state entanglement and that there was insuffi-

cient evidence Congress intended that “teachers in

church-operated schools should be covered by the Act.”

Id. at 504–507; see also Pacific Lutheran University, 361

NLRB No. 157 (2014).

In Boston Medical Center, a divided Board found that

interns, residents and fellows at a teaching hospital were

employees under the Act.

However, the majority did

not change the status of university student assistants,

whom the Board had previously determined not to be

employees.

And as noted previously, except for the

four-year period governed by NYU,

the Board has con-

sistently held that university student assistants are not

employees,

most recently in Brown University,

where

the Board reaffirmed that a student assistant’s relation-

ship with a university is “primarily educational.” Brown,

342 NLRB at 487.

Cf. Northwestern University, 362 NLRB No. 167 (2015) (declin-

ing to exercise jurisdiction over grant-in-aid scholarship football play-

ers without reaching the question of “employee” status under the

NLRA).

330 NLRB at 159–165. Members Hurtgen and Brame dissented.

Id. at 168–170 (Member Hurtgen, dissenting); id. at 170–182 (Member

Brame, dissenting).

The majority in Boston Medical explained that hospital interns,

residents and fellows—referred to as “house staff”—were materially

different from students, including student assistants:

[W]hile house staff possess certain attributes of student status, they are

unlike many others in the traditional academic setting. Interns, resi-

dents, and fellows do not pay tuition or student fees. They do not take

typical examinations in a classroom setting, nor do they receive grades

as such. They do not register in a traditional fashion. Their education

and student status is geared to gaining sufficient experience and

knowledge to become Board-certified in a specialty.

Id. at 161 (footnote omitted).

332 NLRB at 1205.

See fns. 7-9, supra.

342 NLRB at 483.

COLUMBIA UNIVERSITY

I agree with the Board majority’s reasoning in Brown.

There, the Board considered whether “graduate student

assistants who are admitted into, not hired by, a universi-

ty, and for whom supervised teaching or research is an

integral component of their academic development”

should be deemed employees under the Act. Brown, 342

NLRB at 483. The Board majority held that these indi-

viduals were not “employees,” based on the conclusion

that “graduate student assistants, who perform services at

a university in connection with their studies, have a pre-

dominately academic, rather than economic, relationship

with their school.” Id. The Board majority stated that

the “fundamental premise of the Act” was “to cover eco-

nomic relationships,” and the majority recognized “the

simple, undisputed fact that all the petitioned-for indi-

viduals [were] students and must first be enrolled at

Brown” before they could be graduate assistants. Id. at

488. The majority emphasized that the work done by

graduate assistants was “part and parcel of the core ele-

ments of the Ph.D. degree.” Id. In the case of most doc-

toral students who provided instruction, for example, the

majority observed that “teaching is so integral to their

education that they will not get the degree until they sat-

isfy that requirement.” Id.; see also Leland Stanford, 214

NLRB at 621, 622 (student research assistants who re-

ceived stipends to perform research projects were not

employees, since the research was “part of the learning

process” and a step leading to the “thesis and . . . to-

ward[s] the goal of obtaining the Ph.D. degree”). The

Board majority in Brown concluded it was likely that

collective bargaining would impermissibly interfere with

academic freedom and be “detrimental to the educational

process.”

The majority explained:

Imposing collective bargaining would have a deleteri-

ous impact on overall educational decisions by the . . .

faculty and administration. These decisions would in-

clude broad academic issues involving class size, time,

length, and location, as well as issues over graduate as-

sistants’ duties, hours, and stipends. In addition, collec-

tive bargaining would intrude upon decisions over who,

what, and where to teach or research—the principal

prerogatives of an educational institution . . . .

Apart from my belief that the Board correctly ad-

dressed these issues in Brown, I especially disagree with

several aspects of my colleagues’ opinion to the contrary.

Brown, 342 NLRB at 493.

Brown, 342 NLRB at 490. Cf. Yeshiva, 444 U.S. at 686 (recog-

nizing that academic freedom applies not only to the clash of ideas

among faculty but also to debate concerning “which students will be

admitted, retained, and graduated”).

1. The Financial Investment Associated With a Uni-

versity Education, and the Mistake of Making Academic

Success Subservient to the Risks and Uncertainties of

Collective Bargaining and the Potential Resort to Eco-

nomic Weapons. Given the critical importance of higher

education, I believe the time is long past when the ques-

tion of whether to apply the NLRA to students can ap-

propriately be decided based on the standard lines of

division that are commonplace in matters that come be-

fore the Board. Many parties tend to favor union repre-

sentation and collective bargaining generally, and one

can reasonably expect many of these parties to support

union representation and collective bargaining for uni-

versity student assistants. Likewise, when some parties

tend to oppose union representation or collective bar-

gaining, it is unsurprising when they oppose these things

for student assistants as well. The Board’s role should be

different. We administer a statute enacted by Congress

that was adopted with a focus on conventional workplac-

es, not universities. For this reason, as noted above, the

Board and the courts have recognized that unique issues

arise in applying the NLRA to academic work settings,

even when dealing with college and university faculty.

Moreover, the NLRB has no regulatory authority over

efforts to ensure that undergraduates and graduate stu-

dents at colleges and universities satisfy their degree re-

quirements. And the Board should not ignore the fact

that, for the vast majority of students, attendance at a

college or university has a paramount goal—to obtain a

degree—and this goal, if attained, is usually achieved at

enormous expense. Neither should the Board disregard

the unfortunate reality in the United States that many

students never receive their degree.

I believe my colleagues—though armed with good in-

tentions—engage in analysis that is too narrow, exclud-

ing everything that is unique about the situation of col-

lege and university students. In particular, my colleagues

disregard what hangs in the balance when a student’s

efforts to attain an undergraduate or graduate degree are

governed by the risks and uncertainties of collective bar-

gaining and the potential resort to economic weapons by

students and universities. What hangs in the balance has

College graduation rates in a significant number of other countries

are higher than in the United States (although it appears difficult to

obtain recent data that permit a reliable comparison). See, e.g., Institute

of Education Sciences, National Center for Education Statistics, Youth

Indicators 2011 (identifying 12 countries having higher college gradua-

tion rates than the United States among first-time college students, and

11 countries having lower rates) (https://nces.ed.gov/pubs2012/

2012026/chapter2_23.asp and https://nces.ed.gov/pubs2012/

2012026/figures/figure_23.asp) (last viewed Aug. 1, 2016); Cappelli,

p. 29 (“[T]he United States has among the worst [college] graduation

rates of any country.”).

DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD

immense importance, and it does not come cheap for the

great majority of undergraduate and graduate students

and their families. As one commentator has explained,

“college is for many people the biggest financial decision

they will ever make,” it “makes more demands on our

cognitive abilities than most of us will ever see again in

our lives,” and the “biggest cost associated with going to

college . . . is likely to be the risk that a student does not

graduate on time or, worse, drops out altogether. There

is virtually no payoff from college if you don’t gradu-

ate.”

My colleagues ignore these considerations, and they

disclaim any responsibility to address anything other

than the need to promote collective bargaining. In their

words:

We have put suppositions aside today and have instead

carefully considered the text of the Act as interpreted

by the Supreme Court, the Act’s clearly stated policies,

the experience of the Board, and the relevant empirical

evidence drawn from collective bargaining in the uni-

versity setting. This is not a case . . . where the Board

must accommodate the National Labor Relations Act

with some other federal statute related to private uni-

versities that might weigh against permitting student

assistants to seek union representation and engage in

collective bargaining.

Regarding examples where bargaining involving student

assistants (according to Columbia University and other par-

ties) “has proven detrimental to the pursuit of the school’s

educational goals,” my colleagues state that “labor disputes

are a fact of economic life,” and “the Act is intended to ad-

dress them.”

They conclude:

The National Labor Relations Act . . . governs only the

employee-employer relationship. For deciding the le-

gal and policy issues in this case, then, it is not disposi-

tive that the student-teacher relationship involves dif-

ferent interests than the employee-employer relation-

ship; that the educational process is individual, while

collective bargaining is focused on the group; and that

promoting equality of bargaining power is not an aim

of higher education. Even conceded, all these points

simply confirm that collective bargaining and educa-

tion occupy different institutional spheres.

I disagree with this analysis because it is contrary to

what the Supreme Court has stated repeatedly is the

“‘primary function and responsibility of the Board,’”

Cappelli, pp. 8, 26, 48 (emphasis added). See also fns. 2-3, supra

and accompanying text.

Majority opinion, supra, slip op. at 12 (emphasis added).

Majority opinion, supra, slip op. at 10 (emphasis added).

Majority opinion, supra, slip op. at 7 (emphasis added).

which is “‘applying the general provisions of the Act to

the complexities of industrial life.’” Ford Motor Co. v.

NLRB, 441 U.S. 488, 496 (1979) (quoting NLRB v. In-

surance Agents, 361 U.S. 477, 499 (1960); NLRB v. Erie

Resistor Corp., 373 U.S. 221, 236 (1963)); see also

NLRB v. J. Weingarten, Inc., 420 U.S. 251, 266-267

(1975) (“The responsibility to adapt the Act to changing

patterns of industrial life is entrusted to the Board.”).

The instant case does not involve “industrial life.”

Yet this only serves to reinforce the inappropriateness of

“blindly” imposing collective bargaining and the rest of

the NLRA on students in “the academic world.” Syra-

cuse University, 204 NLRB at 643; see also Yeshiva, 444

U.S. at 680; Adelphi University, 195 NLRB at 648. The

Board has applied the NLRA to college and university

faculty members, which has presented its own challeng-

es, as noted previously. The best interests of students,

however, necessarily revolves around whether they ob-

tain the education that costs so much in time and money

and means so much to their future. The Board has no

expertise regarding these issues, and Congress did not

adopt our statute to advance the best interests of college

and university students. This makes it inappropriate to

summarily dismiss concerns in this area as being “not

dispositive.”

Even more objectionable is my colleagues’ statement

that the instant case involves no need to “accommodate

the National Labor Relations Act with some other federal

statute related to private universities that might weigh

against permitting student assistants to seek union repre-

sentation and engage in collective bargaining.”

This is

contrary to Southern Steamship Co. v. NLRB, 316 U.S.

31, 47 (1942), where the Supreme Court stated that “the

Board has not been commissioned to effectuate the poli-

cies of the [Act] so single-mindedly that it may wholly

ignore other and equally important Congressional objec-

tives.”

Regarding the need to accommodate other “Congres-

sional objectives,” id., there is no shortage of federal

mandates applicable to colleges and universities that, to

borrow my colleagues’ words, “might weigh against

permitting student assistants to seek union representation

and engage in collective bargaining.” Again, a broad

range of federal statutes and regulations apply to colleges

When the NLRA was adopted, Congress contemplated that the

Act would primarily apply to industrial plants and manufacturing facili-

ties. Sec. 1 of the Act refers to “industrial strife or unrest” and sets

forth a policy to encourage “practices fundamental to the adjustment of

industrial disputes,” and the Supreme Court has acknowledged that the

“Act was intended to accommodate the type of management-employee

relations that prevail in the pyramidal hierarchies of private industry.”

Yeshiva, 444 U.S. at 680.

Majority opinion, supra, slip op. at 12.

COLUMBIA UNIVERSITY

and universities, with significant involvement by the U.S.

Department of Education, led by the Secretary of Educa-

tion. Relevant laws include, among many others, the

Higher Education Opportunity Act, enacted in 2008,

which reauthorized the Higher Education Act of 1965,

and the Family Educational Rights and Privacy Act

(FERPA), enacted in 1974.

These statutes govern,

among other things, the accreditation of colleges and

universities, the enhancement of quality, the treatment of

student assistance, graduate/postsecondary improvement

programs, and the privacy of student records. In 2015, a

task force created by a bipartisan group of U.S. Senators

reviewed the Department of Education’s regulation of

colleges and universities and recommended, among other

things, that the Department’s regulations “be related to

education, student safety, and stewardship of federal

funds” and “not stray from clearly stated legislative in-

tent.”

The extensive federal regulation of colleges and

universities focuses on access, availability, affordability

and effectiveness, all of which relate to the ability of

students to satisfy educational objectives. This supports

my view that collective bargaining—and especially the

resort to economic weapons between and among student

assistants, faculty members, and administrators—is like-

ly to substantially affect the educational process, separate

from any impact on the economic interests of student

assistants.

Furthermore, it is already clear that current Board law,

if applied to university student assistants, may contradict

federal education requirements. For example, FERPA

broadly restricts the disclosure of educational records,

including student disciplinary records.

However, cur-

Pub. L. 110–315 (2008).

Pub. L. 89–329 (1965). The Higher Education Act of 1965 was

reauthorized in 1968, 1972, 1976, 1980, 1986, 1992, 1998, and 2008. It

was extended through 2015, and reauthorization bills remain pending in

Congress. See U.S. News University, House Approves Five Bipartisan

Bills to Improve Higher Education Act (July 15, 2016) (https://www.

usnewsuniversitydirectory.com/news/paying-for-school/house-

approves-five-bipartisan-bills-to-improve-higher-education-act/) (last

viewed July 28, 2016).

Pub. L. 93–380 (1974). FERPA has been amended multiple times

since its initial passage and may be found in its current form at 20

U.S.C. § 1232g.

Task Force on Federal Regulation of Higher Education, Recali-

brating Regulation of Colleges and Universities, pp. 2, 4 (2015)

(http://www.help.senate.gov/imo/media/

Regulations_Task_Force_Report_2015_FINAL.pdf) (last visited Aug.

5, 2016). The Senate Committee on Health, Education, Labor, and

Pensions also held a hearing on February 24, 2015 devoted to the Task

Force’s report. See http://www.help.senate.gov/hearings/recalibrating-

regulation-of-colleges-and-universities-a-report-from-the-task-force-

on-government-regulation-of-higher-education (Feb. 24, 2015) (last

viewed July 28, 2016).

See http://www2.ed.gov/policy/gen/guid/fpco/brochures/postsec.

html (last viewed Aug. 3, 2016) (indicating that “student disciplinary

rent Board law, if applied to university students, would

require the disclosure of confidential witness statements

(absent proof that the witnesses required protection from

retaliation in the particular circumstances presented),

and Board law would prevent university officials from

routinely requesting nondisclosure of matters discussed

in investigatory interviews involving student assistants.

records are protected as education records under FERPA,” although

disclosure without the student’s consent is permitted in certain circum-

stances). FERPA regulations indicate that “education records” do not

include records relating to employees of an educational institution, but

this exclusion applies only if employment-related records “[r]elate

exclusively to the individual in that individual’s capacity as an employ-

ee.” 34 CFR § 99.3 (defining “education records”). According to the

U.S. Department of Education, the employment-related records of

“graduate student teaching fellows/assistants,” whose appointments are

contingent on being students, constitute “education records” subject to

FERPA’s nondisclosure requirements. See Department of Education,

Letter of technical assistance to American Federation of Teachers re:

disclosure of information on teaching assistants, available at:

http://www2.ed.gov/policy/gen/guid/fpco/ferpa/library/aft.html (Aug.

21, 2000) (last viewed Aug. 3, 2016) (hereinafter “Dept. of Education,

AFT letter”). In short, if a student complains about sex harassment by

a student assistant, which may result in academic suspension or expul-

sion, for example, it appears clear that FERPA confidentiality require-

ments would apply to the investigative records, possibly including

witness statements, directly contrary to NLRB law potentially requiring

their disclosure. See fns. 30-31, infra. Because of FERPA’s privacy

requirements, there will undoubtedly be additional conflicts with

NLRB disclosure obligations in other contexts, including union infor-

mation requests to which employers must respond under NLRA Sec.

8(a)(5). See Dept. of Education, AFT letter, supra (information re-

quested by union representing public university student assistants can-

not be disclosed without the student assistants’ consent); see also

http://www2.ed.gov/

policy/gen/guid/fpco/ferpa/library/josephambash.html (Feb. 25, 2002)

(last viewed Aug. 3, 2016) (teaching assistants’ hours of work, stipend,

length of contract, employment category, and selection for layoff are

educational records protected from disclosure by FERPA).

Even before the Board majority decided to apply the NLRA to stu-

dent assistants and thus create inconsistencies with other federal regula-

tions applicable to colleges and universities, the Board’s interpretation

of the NLRA was contrary to other federal agency requirements and

recommendations. For example, the Board’s disclosure requirements

applicable to workplace investigations and witness statements conflict

with guidance from the Equal Employment Opportunity Commission

(EEOC) regarding precisely the same issues. See American Baptist

Homes of the West d/b/a Piedmont Gardens, 362 NLRB No. 139, slip

op. at 12–13 (2015) (Member Johnson, dissenting in part); Report of

the Co-Chairs of the EEOC Select Task Force on the Study of Harass-

ment in the Workplace, available at:

http://www.btlaborrelations.com/wp-content/uploads/2016/06/eeoc-

report-on-sexual-harassment.pdf (June 2016) (last viewed Aug. 4,

2016) (noting “strong support” from stakeholders “for the proposition

that workplace investigations should be kept as confidential as is possi-

ble,” but also observing that “an employer’s ability to maintain confi-

dentiality . . . has been limited in some instances by decisions of the

National Labor Relations Board,” and concluding that the “privacy of

both the accuser and the accused should be protected to the greatest

extent possible, consistent with legal obligations”).

See Piedmont Gardens, supra, 362 NLRB No. 139.

See Banner Estrella Medical Center, 362 NLRB No. 137 (2015).

DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD

My colleagues apparently agree that “promoting equal-

ity of bargaining power is not an aim of higher educa-

tion.”

It is also clear that collective bargaining by stu-

dents is not the focus of the numerous federal laws and

regulations that apply to colleges and universities. These

laws and regulations are designed, directly or indirectly,

to enhance the quality of education, to strengthen equal

access to higher education, and to eliminate potential

obstacles to academic success. There is no reasonable

justification for the Board’s failure to acknowledge the

overriding importance of these non-employment issues

for college and university students.

Nor can the Board freely disregard the fact that the po-

tential resort to economic weapons is part and parcel of

collective bargaining. Therefore, applying our statute to

university student assistants may prevent them from

completing undergraduate and graduate degree require-

ments in the allotted time, which is the primary reason

they attend colleges and universities at such great ex-

pense. It is not an adequate response to summarily dis-

miss this issue, as the majority does, with the common-

place observation that “labor disputes are a fact of eco-

nomic life.” For the students who may find themselves

embroiled in them, labor disputes between universities

and student assistants may have devastating consequenc-

es.

Conventional work settings feature many examples of

constructive collective-bargaining relationships. Like-

wise, one cannot assume that all or most negotiations

involving student assistants at universities would result

in strikes, slowdowns, lockouts, and/or litigation. How-

ever, there is no doubt that economic weapons and the

threatened or actual infliction of economic injury are

central elements in collective bargaining to which resort

may be made when parties are unable to reach agree-

ment. As I stated in American Baptist Homes of the West

d/b/a Piedmont Gardens, 364 NLRB No. 13, slip op. at 9

(2016) (Member Miscimarra, dissenting in part), one

must “differentiate between what one would prefer to see

in collective bargaining, and what role Congress contem-

plated for economic weapons as part of the collective-

bargaining process.” I elaborated as follows:

What one hopes to see in any collective-bargaining

dispute is its successful resolution without any party’s

resort to economic weapons. But what Congress in-

tended was for the Board to preserve the balance of

competing interests—including potential resort to eco-

nomic weapons—that Congress devised as the engine

driving parties to resolve their differences and to enter

into successful agreements. As the Supreme Court

Majority opinion, supra, slip op. at 7.

stated in NLRB v. Insurance Agents’ International Un-

ion, 361 U.S. 477, 487–489 (1960), employers and un-

ions in collective bargaining “proceed from contrary

and to an extent antagonistic viewpoints and concepts

of self-interest. . . . The presence of economic weapons

in reserve, and their actual exercise on occasion by the

parties, is part and parcel of the system that the Wag-

ner and Taft-Hartley Acts have recognized.”

When the Board transplants our statute into the univer-

sity setting and places students in a bargaining relation-

ship with the university, experience demonstrates that we

cannot assume bargaining will be uneventful. Collective

bargaining may evoke “extraordinarily strong feelings”

and give rise to a “sharp clash between seemingly irrec-

oncilable positions,” and when parties resort to various

tactics in support of their respective positions, “such tac-

tics are indeed ‘weapons,’” and “[n]obody can be con-

fused about their purpose: they are exercised with the

intention of inflicting severe and potentially irreparable

injury, often causing devastating damage to businesses

and terrible consequences for employees.”

As the court

stated in NLRB v. Wire Products Mfg. Corp., 484 F.2d

760, 765 (7th Cir. 1973), “[t]he strike is a potent eco-

nomic weapon which may, and often is, wielded with

disastrous effect on its employer target.”

Id., slip op. at 9 (Member Miscimarra, dissenting in part) (empha-

sis in original and emphasis added).

Id., slip op. at 10 (Member Miscimarra, dissenting in part).

My colleagues refer to what they characterize as “empirical evi-

dence” that, in their view, suggests collective bargaining involving

student assistants has been undertaken successfully at public universi-

ties (where bargaining is typically governed by state public sector labor

laws, which generally restrict or eliminate the right to strike) and at a

private university on a non-NLRB-supervised basis (New York Univer-

sity). My colleagues state that certain agreements provide for “defined

rights concerning courses, course content, course assignments, exams,

class size, grading policies and methods of instruction, as well as grad-

uate students’ progress on their own degrees,” and according to the

majority, this shows that “parties can and successfully have navigated

delicate topics near the intersection of the university’s dual role as

educator and employer.” Majority opinion, supra, slip op. at 9. Co-

lumbia University and other parties have identified cases where bar-

gaining by student assistants “has proven detrimental to the pursuit of

the school’s educational goals,” with “strikes and grievances over

teaching workload and tuition waivers” and “grievances over classroom

assignments and eligibility criteria for assistantships,” but my col-

leagues dismiss these examples as “labor disputes” that are merely a

“fact of economic life.” Majority opinion, supra, slip op. at 10.

I disagree with my colleagues’ selective attachment of significance

to the examples of peaceful negotiations involving student assistants—

none of which involves economic weapons permitted under the

NLRA—and with their summary discounting of examples that go the

other way. In my view, what should be controlling here are two unas-

sailable propositions: collective bargaining under the NLRA involves

the potential use of leverage through threatened or inflicted economic

injury; and even among parties that negotiate in good faith with the best

intentions, disputes involving resort to protected economic weapons by

COLUMBIA UNIVERSITY

Of course, determining that student assistants are “em-

ployees” and have the right to be represented by a union

under the NLRA does not mean they will choose to be

represented. Likewise, as stated above, I am not predict-

ing that most negotiations involving student assistants

will involve resort to economic weapons. Nonetheless,

in this particular context, I believe collective bargaining

and its attendant risks and uncertainties will tend to de-

tract from the primary reason that students are enrolled at

a university—to satisfy graduation requirements, includ-

ing in many cases the satisfactory completion of service

in a student assistant position. And in some cases in-

volving student assistants, it is predictable that break-

downs in collective bargaining will occur, and the result-

ing resort to economic weapons may have devastating

consequences for the students, including, potentially,

inability to graduate after paying $50,000 to $100,000 or

more for the opportunity to earn a degree.

Now that, with today’s decision, student assistants are

employees under the NLRA, what economic weapons are

available to student assistants and the universities they

attend? They would almost certainly include the follow-

ing:

• Strikes. Student assistants could go on strike,

which would mean that Teaching Assistants,

Teaching Fellows, Preceptors, Course Assistants,

Readers, Graders, Graduate Research Assistants

and Departmental Research Assistants would

cease working, potentially without notice, and the

university could suspend all remuneration.

one or both parties arise with regularity under our statute. I have no

quarrel with the notion that colleges and universities should construc-

tively engage with their students, including student assistants, in a

variety of ways. Yet I believe such engagement need not necessarily

take the form of collective bargaining under the NLRA and instead may

take place (and, I am sure, has taken place and is taking place) “without

the intervention of the Board enforcing a statutory requirement.” First

National Maintenance Corp. v. NLRB, 452 U.S. 666, 681 fn. 19 (1981).

The College Board reports that average annual tuition and fees at a

private four-year college total $32,410, which means that four years’

worth of average tuition and fees total $129,640 at private universities.

See https://bigfuture.collegeboard.org/pay-for-college/college-

costs/college-costs-faqs (last viewed July 29, 2016). If a college career

stretches to six years, which is the most common time period used to

evaluate whether incoming students will graduate from a four-year

college (see fn. 3, supra and accompanying text), average tuition and

fees at a private university would total $194,460. These figures do not

include additional expenditures for room, board, and other living ex-

penses.

In the event of a strike or lockout, an employer under the NLRA

has the right to discontinue all wages and other forms of remuneration,

with the sole exception of those wages or benefits that have already

accrued, the payment of which does not depend on the performance of

work. See Texaco, Inc., 285 NLRB 241, 245 (1987) (“[A]n employer

is not required to finance a strike against itself by paying wages or

similar expenses dependent on the continuing performance of services

• Lockouts. The university could implement a lock-

out, which would require student assistants to

cease working, and all remuneration would be

suspended.

• Loss, Suspension or Delay of Academic Credit. If

a student assistant ceases work based on an eco-

nomic strike or lockout, it appears clear they

would have no entitlement to credit for require-

ments that are not completed, such as satisfactory

work in a student assistant position for a pre-

scribed period of time. For example, if a particu-

lar degree required two semesters of service as a

Teaching Assistant, and a student assistant could

not satisfy that requirement because of a strike or

lockout that persisted for two semesters, it appears

clear the student assistant would not be entitled to

receive his or her degree.

• Suspension of Tuition Waivers. In the event of a

strike or lockout where the university suspended

tuition waivers or other financial assistance that

was conditioned on the student’s work as a student

assistant, students would likely be foreclosed from

attending classes unless they paid the tuition.

Thus, the student assistant’s attendance at universi-

ty could require the immediate payment of tuition,

which averages $32,410 annually at private uni-

versities.

• Potential Replacement. In the event of a strike, the

university would have the right to hire temporary

or permanent replacements. If permanent re-

placements were hired during an economic strike,

this would mean that even if a student uncondi-

tionally offered to resume working at the end of

the strike, the university could retain the replace-

ments, and the student assistant would not be rein-

stated unless and until a vacancy arose through the

departure of a replacement or the creation of a new

position. Here as well, one would expect that the

student would be required to pay full tuition in or-

der to be permitted to attend classes, without re-

gard to whatever tuition waiver or other financial

aid was previously provided in consideration of

the student’s services as a student assistant. Simi-

larly, any failure to satisfy degree requirements as-

sociated with a student assistant’s work as a stu-

dent assistant would preclude attainment of the

degree.

for the employer,” absent proof that the benefits in question were “ac-

crued,” which means “due and payable on the date on which the em-

ployer denied [them].”). See also Ace Tank & Heater Co., 167 NLRB

663, 664 (1967).

See fn. 33, supra.

DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD

• Loss of Tuition Previously Paid. If a student assis-

tant paid his or her own tuition (again, currently

averaging $32,410 per year at a private universi-

ty)

and only received a cash stipend as compen-

sation for work as a student assistant, there appears

to be little question that the student’s tuition could

lawfully be retained by the university even if a

strike by student assistants persisted for an entire

year, during which time the student was unable to

satisfy any requirements for satisfactory work in

his or her student assistant position.

• Misconduct, Potential Discharge, Academic Sus-

pension/Expulsion Disputes. During and after a

strike, employees remain subject to discipline or

discharge for certain types of strike-related mis-

conduct. Correspondingly, there is little question

that a student assistant engaged in a strike would

remain subject to academic discipline, including

possible suspension or expulsion, for a variety of

offenses. In such cases, I anticipate that parties

will initiate Board proceedings alleging that stu-

dents were unlawfully suspended or expelled for

NLRA-protected activity, even though nothing in

the Act permits the Board to devise remedies that

relate to an individual’s academic standing, sepa-

rate and apart from his or her “employment.”

It is also a mistake to assume that today’s decision re-

lates only to the creation of collective-bargaining rights.

Our statute involves wide-ranging requirements and ob-

ligations. For example, existing Board cases require em-

ployers subject to the NLRA to tolerate actions by em-

ployees that most reasonable people would find objec-

tionable, and it is unlawful for employers to adopt overly

broad work rules to promote respect and civility by em-

ployees. Therefore, parents take heed: if you send your

teenage sons or daughters to college, the Board majori-

ty’s decision today will affect their “college experience”

in the following ways:

• Non-Confidential Investigations. If your son or

daughter is sexually harassed by a student assistant

and an investigation by the university ensues, the

university will violate federal law (the NLRA) if it

routinely asks other student-assistant witnesses to

keep confidential what is discussed during the uni-

versity’s investigation.

See fn. 33, supra.

See, e.g., Banner Estrella Medical Center, supra, 362 NLRB No.

137. These disclosures, though required by the NLRB, may directly

conflict with nondisclosure obligations under the Family Educational

Rights and Privacy Act (FERPA). See fn. 29, supra.

• Witness Statement Disclosure. In the above ex-

ample, witness statements submitted by your son

or daughter about sexual harassment by a student

assistant must be disclosed to the union, unless (i)

the university can prove that the statement’s sub-

mission was conditioned on confidentiality, and

(ii) even then, the statement must be disclosed un-

less the university can prove that your son or

daughter needs protection, or other circumstances

outweigh the union’s need for the witness state-

ment.

• Invalidating Rules Promoting Civility. The uni-

versity will be found to have violated the NLRA if

it requires student assistants to maintain “harmoni-

ous interactions and relationships” with other stu-

dents.

• Invalidating Rules Barring Profanity and Abuse.

The university cannot adopt a policy against “loud,

abusive or foul language” or “false, vicious, pro-

fane or malicious statements” by student assis-

tants.

• Outrageous Conduct by Student Assistants. The

university must permit student assistants to have

angry confrontations with university officials in

grievance discussions, and the student assistant

cannot be lawfully disciplined or removed from

his or her position even if he or she repeatedly

screams, “I can say anything I want,” “I can swear

if I want,” and “I can do anything I want, and you

can’t stop me.”

• Outrageous Social Media Postings by Student As-

sistants. If a student assistant objects to actions by

a professor-supervisor named “Bob,” the universi-

ty must permit the student to post a message on

Facebook stating: “Bob is such a nasty mother

fucker, don’t know how to talk to people. Fuck

his mother and his entire fucking family.”

• Disrespect and Profanity Directed to Faculty Su-

pervisors. The university may not take action

against a student assistant who screams at a pro-

fessor-supervisor and calls him a “fucking crook,”

a “fucking mother fucking” and an “asshole” when

See, e.g., Piedmont Gardens, supra, 362 NLRB No. 139. These

disclosures, though required by the NLRB, may also directly conflict

with nondisclosure obligations under FERPA. See fn. 29, supra.

See, e.g., William Beaumont Hospital, 363 NLRB No. 162 (2016);

2 Sisters Food Group, 357 NLRB 1816 (2011).

See Flamingo Hilton-Laughlin, 330 NLRB 287 (1999); Lafayette

Park Hotel, 326 NLRB 825 (1998), enfd. 203 F.3d 52 (D.C. Cir. 1999);

Cincinnati Suburban Press, Inc., 289 NLRB 966, 975 (1988).

See United States Postal Service, 364 NLRB No. 62 (2016).

See Pier Sixty, LLC, 362 NLRB No. 59 (2015).

COLUMBIA UNIVERSITY

the student assistant is complaining about the

treatment of student assistants.

The above examples constitute a small sampling of the

unfortunate consequences that will predictably follow

from the majority’s decision to apply our statute to stu-

dent assistants at colleges and universities. The primary

purpose of a university is to educate students, and the

Board should not disregard that purpose in finding that

student assistants are employees and therefore subject to

all provisions of the NLRA.

2. The Board’s Processes and Procedures Are Incom-

patible with Applying the Act to University Student Assis-

tants. Another frailty associated with applying the

NLRA to student assistants at universities relates to the

cumbersome and time-consuming nature of the Board’s

processes and procedures, which makes those processes

and procedures especially ill suited to students in a uni-

versity setting.

The Board has engaged in well-publicized efforts to

expedite the handling of representation cases, and in

2014 the Board issued an election rule that dramatically

revised the Agency’s representation-case procedures.

See 79 Fed. Reg. 74, 308 (2014) (Election Rule). How-

ever, notwithstanding the Board’s commitment to resolve

representation cases as quickly as possible, doing so has

sometimes proven difficult in cases involving colleges

and universities. In part, these difficulties and resulting

delays are owing to the fact that the religious affiliation

of a college or university may entirely preclude the

Board’s exercise of jurisdiction.

However, even when

representation cases involve universities that are not reli-

giously affiliated, Board proceedings may still involve

significant time, and the filing of election-related unfair

labor practice charges may delay scheduled elections for

months or years under the Agency’s “blocking charge”

doctrine.

The Board’s handling of alleged unfair labor practices

(ULPs) takes even more time. Our procedures require

See Plaza Auto Center, Inc., 360 NLRB No. 117 (2014).

See, e.g., Seattle University, 364 NLRB No. 84 (2016) (represen-

tation proceedings involving religiously affiliated university where

representation petition was filed February 20, 2014, and the Board’s

decision issued on August 23, 2016); St. Xavier University, 364 NLRB

No. 85 (2016) (representation proceedings involving religiously affili-

ated university where representation petition was filed April 12, 2011,

and the Board’s decision issued on August 23, 2016); Duquesne Uni-

versity, Case 6-RC-80933 (representation proceedings involving reli-

giously affiliated university where representation petition was filed

May 14, 2012 and remains pending before the Board). See generally

Pacific Lutheran University, 361 NLRB No. 157 (2014).

The Board’s treatment of delays associated with blocking charges

was not materially changed in the Election Rule. See 79 Fed. Reg. at

74,455–74,456 (dissenting views of Members Miscimarra and John-

son).

the filing of a ULP charge, which is investigated by one

of the Board’s regional offices, which decides whether to

issue a complaint, and if complaint issues, this is fol-

lowed by a hearing before an administrative law judge,

with posthearing briefing in most cases. After the judge

issues a decision, parties have the right to file exceptions

to that decision with the Board (in other words, they may

appeal), with further briefing by the parties. Ultimately,

the Board renders a decision, which may be appealed to a

federal court of appeals. In addition, when the Board has

found a violation and has ordered backpay and other re-

medial measures, there are additional compliance pro-

ceedings handled by the Board’s regional offices, which

can result in additional hearings before administrative

law judges, additional posthearing briefs, supplemental

decisions by the judges, and further appeals to the Board

and the courts. In spite of everyone’s best efforts, this

lengthy litigation process consumes substantial time and

too often causes unacceptable delays before any Board-

ordered relief becomes available to the parties. Unfair

labor practice cases may easily be litigated for three to

five years before the Board issues a decision, and some

cases take even longer. See, e.g., CNN America, Inc.,

361 NLRB No. 47 (2014) (alleged ULPs requiring 82

days of trial, more than 1,300 exhibits, more than 16,000

transcript pages, and more than 10 years of Board litiga-

tion, and the case still remains pending on appeal);

Dubuque Packing Co., 287 NLRB 499 (1987), remanded

sub nom. UFCW Local 150-A v. NLRB, 880 F.2d 1422

(D.C. Cir. 1989), on remand 303 NLRB 386 (1991),

enfd. in relevant part sub nom. UFCW Local 150-A v.

NLRB, 1 F.3d 24 (D.C. Cir. 1993), cert. granted 511 U.S.

1016 (1994), cert. dismissed 511 U.S. 1138 (1994) (al-

leged ULPs requiring 13 years of Board and court litiga-

tion).

In the time it takes a typical NLRA case to be litigated

and decided by the Board and the courts, the academic

world may experience developments that dramatically

change or even eliminate entire fields of study. Moreo-

ver, not only does a student assistant’s position have a

fixed duration, but the student status of the individual

Sec. 10(j) of the NLRA authorizes the Board’s General Counsel to

initiate proceedings in federal district court seeking interim injunctive

relief in certain cases, but the ultimate resolution of those disputes does

not occur until the Board’s disposition on the merits (subject to further

appellate review). Moreover, the General Counsel is necessarily selec-

tive when evaluating whether particular cases warrant efforts to seek

Sec. 10(j) relief, and there is no certainty that such relief, when sought,

will be granted by the court. See, e.g., Osthus v. Ingredion, Inc., Case

No. 16-CV-38-LRR, 2016 WL 4098541 (N.D. Iowa July 28, 2016)

(denying Sec. 10(j) petition on the basis that the Board “failed to meet

its burden to demonstrate that irreparable injury is likely to occur ab-

sent injunctive relief”).

DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD

occupying that position may itself come to an end long

before a Board case affecting him or her is resolved.

Students generally attend university for the purpose of

doing something else—i.e., to obtain post-graduation

employment, or to go on to post-doctoral or other post-

graduate studies. Moreover, it is not uncommon for stu-

dents to change majors, and faculty members also come

and go. In these respects, treating student assistants as

employees under the NLRA is especially poorly matched

to the Board’s representation and ULP procedures.

3. Other Considerations Undermine the Appropriate-

ness of the Petitioned-For Bargaining Unit. I believe the

Board should find that student assistants are not employ-

ees for purposes of Section 2(3) of the Act. Therefore, I

need not reach whether the bargaining unit sought in the

instant case is an appropriate bargaining unit. Nonethe-

less, I will address two considerations that render the

petitioned-for unit particularly problematic.

Preliminarily, however, I address an issue that is prior

to appropriate-unit considerations: the majority’s deci-

sion to reject not only Brown University but an unbroken,

decades-old line of precedent holding that research assis-

tants are not employees under Section 2(3) of the Act.

Research assistants are graduate students, usually in the

hard sciences, who conduct research projects funded by

private institutions or the government, and Columbia

requires this research to be directly related to the research

officer’s dissertation. The Board has consistently de-

clined to find student research assistants to be employees

under the Act. In Adelphi University, the Board declined

to include graduate student research assistants in a unit of

regular faculty on the basis that the research assistants

were “primarily students,” 195 NLRB at 640, and it dis-

tinguished student research assistants from a research

assistant deemed eligible in another case who “was not

At various points, my colleagues analogize student assistants to in-

termittent workers, “seasonal” workers, and “workforces . . . with sig-

nificant turnover.” These analogies fail to reflect substantial differ-

ences that exist between conventional employees whose work may be

sporadic and student assistants. Even if conventional employees per-

form sporadic work, their employment most often contemplates that

they will remain in the workforce, often in the same line of work and

with the same employer. Student assistants nearly always occupy their

positions on a short-term basis, with plans to permanently abandon

their status as student assistants to complete their education, graduate,

and obtain other positions. (My colleagues admit as much, noting that

“student assistants possess a long-term goal of achieving employment

elsewhere.” Majority opinion, supra, slip op. at 21 fn. 131). This is

merely one additional reason that the Act is such an imperfect fit for

student assistants. See Saga Food Service, 212 NLRB 786, 787 fn. 9

(1974) (finding a unit comprised solely of part-time student cafeteria

workers would not “effectuate the purposes of the Act” “[i]n view of

the nature of their employment tenure and our conclusion that their

primary concern is their studies rather than their part-time employ-

ment”).

simultaneously a student but already had his doctoral

degree,” id. at 640 fn. 8. Similarly, in Leland Stanford

Junior University, 214 NLRB at 621, the Board again

concluded that student research assistants “are primarily

students [and] not employees.” Id. at 623. Even during

the brief period when the Board considered student in-

structors to be employees under the Act, the Board ad-

hered to precedent holding that student research assis-

tants are not Section 2(3) employees. See NYU, 332

NLRB at 1209 fn. 10 (applying Leland Stanford and

finding that student research assistants were not employ-

ees); see also Brown, 342 NLRB at 483 (graduate student

assistants, including research assistants, are not employ-

ees under Section 2(3) of the Act).

The facts regarding the research officers here differ in

no material respect from those of the student research

assistants in Leland Stanford, NYU, and Brown. Here, as

in each of those cases, the students perform research as

part of their progress towards a degree and are primarily

students. Accordingly, based on a line of precedent that

remained unbroken for more than 40 years, I believe the

Board cannot reasonably find that research assistants are

employees for purposes of the Act.

Turning to appropriate-unit considerations, I believe

the Board cannot find that the broad array of student as-

sistants here share a sufficient community of interests to

warrant their inclusion in a single bargaining unit. The

Petitioner seeks to represent all “student employees” who

engage in “instructional services,” including “graduate

and undergraduate Teaching Assistants,” “Teaching Fel-

lows,” “Preceptors,” “Course Assistants,” “Readers,”

“Graders,” “Graduate Research Assistants” and “De-

partmental Research Assistants.” The students within the

various classifications in the petitioned-for unit vary con-

siderably in terms of their duties, levels of responsibility,

remuneration, and expected length of service. Although

I would decline jurisdiction over the entire proposed unit

on the basis that student assistants are not “employees,”

and therefore I need not and do not reach or analyze the

various issues relating to whether the proposed unit is

appropriate, the evidence in the record demonstrates that

what various student officers do and how they are remu-

nerated vary enormously.

For example, some student assistants teach, and re-

search assistants perform research. Course assistants do

neither: they perform clerical duties, such as filing and

copying, to help faculty administer courses. Generally,

doctoral students have greater autonomy and responsibil-

ity in performing their instructional duties than do mas-

ter’s degree candidates and undergraduates. Some doc-

toral students serve as preceptors, fully designing and

implementing their own courses. By contrast, non-

COLUMBIA UNIVERSITY

doctoral students predominantly grade papers or provide

tutoring to their fellow students in laboratory or discus-

sion sections.

Course assistants perform work that is intermittent in

nature, and they are paid from Columbia’s casual payroll.

Remuneration for master’s degree students and under-

graduates is awarded only during the semesters that the

students actually perform duties as student assistants. By

contrast, doctoral students receive the same funding dur-

ing the entire time spent pursuing their degree, whether

they are performing duties as a student assistant during a

certain semester or academic year or not. In contrast to

the intermittent tenure of the course assistants, doctoral

students generally must spend at least one year teaching,

and sometimes multiple years, in order to obtain their

degree. Undergraduate and master’s degree students are

not required to serve as student assistants in connection

with their degree requirements.

In view of these and other fundamental dissimilarities,

I believe the petitioned-for unit would likely be inappro-

priate under any community-of-interest test, including

the one stated in Specialty Healthcare.

The second consideration that, in my view, undermines

the appropriateness of the petitioned-for unit relates to

the Board’s treatment of temporary employees, who are

generally excluded from petitioned-for bargaining units.

Here, I disagree with my colleagues’ evaluation of the

student assistants “as a group” and their application of a

special rule to all of them—namely, that their tenure “is

not so ephemeral as to vitiate their interest in bargaining

over terms and conditions of employment.”

This

standard inappropriately deviates from the Board’s exist-

ing principles pertaining to temporary employees by cre-

ating a special rule for them. See, e.g., Fordham Univer-

sity, 214 NLRB 971, 975 (1974) (rejecting creation of a

special rule for temporary employee status governing

See Specialty Healthcare & Rehabilitation Center of Mobile, 357

NLRB 934, 942 (2011) (citing, among the factors the Board must ex-

amine to determine if a unit is appropriate, “whether the employees are

organized into a separate department; have distinct skills and training;

have distinct job functions and perform distinct work, including inquiry

into the amount and type of job overlap between classifications; . . .

have distinct terms and conditions of employment; and are separately

supervised”) (citations omitted); see also NYU, 332 NLRB at 1205 fn.

5, 1209 fn. 10 (excluding research assistants funded by external grants

and students who acted as graders and tutors from a unit of graduate

assistants).

Specialty Healthcare does not govern the appropriateness of the pe-

titioned-for unit in this case because there is no argument that the peti-

tioned-for unit inappropriately excludes other putative employees. In

any event, however, I would not apply Specialty Healthcare for the

reasons stated in my dissenting opinion in Macy’s, Inc., 361 NLRB No.

4, slip op. at 31–33 (2014) (Member Miscimarra, dissenting).

Majority opinion, supra, slip op. at 21.

faculty in academia). The Board has recognized that

short-term or finite employment may not be conclusive

as to temporary status, but it may nonetheless require a

finding that certain members of a petitioned-for unit do

not share a community of interests with the rest of the

unit. Compare Boston Medical Center, 330 NLRB at

166 (finding appropriate a unit of house staff, where all

employees in the proposed unit were generally employed

by the employer for 3 to 7 years) with Goddard College,

216 NLRB 457, 458 (1975) (excluding from a unit of

full-time faculty visiting faculty who generally stay for

only one semester or one year). Regarding the peti-

tioned-for unit in this case, I do not believe the Board can

appropriately recognize a single bargaining unit that

combines course assistants who work a few intermittent

hours with doctoral candidates who may teach for several

years.

ONCLUSION

There is a remarkable, life-changing procedure availa-

ble for those fortunate enough to undergo it. During this

procedure, the participants remain awake, they are close-

ly evaluated while answering complex questions for an

extended period of time, and they are monitored while

performing other tasks as directed, which includes inter-

acting with others. This life-changing procedure is

enormously expensive, and many individuals receive

financial assistance while undergoing it. The procedure

is so demanding that many participants never complete

it. Yet, research shows that successful completion of the

procedure improves the rest of the person’s life. It pro-

duces substantially more opportunities, higher compensa-

tion, enhanced satisfaction, and greater upward mobility

both for the participants and for future generations.

This describes the role played by colleges and univer-

sities in the United States. My colleagues apply a dis-

torted and highly selective lens to this life-changing pro-

cedure. Dismissing everything else as “not dispositive,”

they conclude that some participants satisfy the defini-

tion of an “employee” because (i) they perform tasks as

directed, and (ii) they receive financial assistance. Even

more erroneous, in my view, is the notion that public

policy favors taking participants who are trying to com-

plete this life-changing procedure and—while it is being

conducted—having them engage in collective bargain-

ing, which is governed by leverage and the potential

clash of economic interests. It will wreak havoc

to have economic weapons wielded by or against partici-

pants during this expensive procedure, especially since

the weapons include strikes and lockouts—which

can stop the procedure in its tracks—and the permanent

replacement of the participants themselves!

DECISIONS OF THE NATIONAL LABOR RELATIONS BOARD

The Board has a responsibility to acknowledge the

enormous complexity, demands and benefits associated

with every student’s potential graduation from a college

and university. In particular, I believe my colleagues

improperly focus on the NLRA and “wholly ignore other

and equally important Congressional objectives,”

espe-

cially the overriding importance of facilitating each stu-

dent’s satisfaction of degree requirements. Given the

importance of this policy objective—which is reflected

in numerous federal statutes and regulations governing

education, and as to which the Board has no expertise—I

believe the Board cannot reasonably apply our statute to

student assistants at colleges and universities “without a

clear expression of an affirmative intention of Con-

gress.”

No such evidence of Congressional intent ex-

ists.

“The ‘business’ of a university is education,”

and

students are not the means of production—they are the

“product.” Their successful completion of degree re-

quirements results from the combined commitment of

faculty, administrators, and the students’ own academic

efforts. It is true that the Board has asserted jurisdiction

over faculty members in private, non-exempt colleges

and universities, notwithstanding the significant differ-

ences that exist between the academic and industrial

worlds.

In my view, however, obstacles to fitting the

square peg of the NLRA into the round hole of academia

become insuperable when the petitioned-for “employees”

are university student assistants.

Southern Steamship Co. v. NLRB, 316 U.S. at 47.

Catholic Bishop, 440 U.S. at 504.

Yeshiva, 444 U.S. at 686.

C.W. Post Center, 189 NLRB 904 (1971) (faculty members are

professional employees who may bargain collectively).

The question here is not whether colleges and universi-

ties should constructively engage their students, includ-

ing student assistants, in a variety of ways.

The ques-

tion is whether Congress intended—and whether our

statute can be reasonably interpreted—to make the

NLRA govern the relationship between students and

their universities merely because students may occupy a

variety of academic positions in connection with their

education. As noted above, for most students including

student assistants, attending college is the most important

investment they will ever make. I do not believe our

statute contemplates that it should be governed by bar-

gaining leverage, the potential resort to economic weap-

ons, and the threat or infliction of economic injury by or

against students, on the one hand, and colleges and uni-

versities, on the other.

For these reasons, and consistent with the Board’s pri-

or holding in Brown University, I believe the Board

should find that the relationship between Columbia and

the student assistants in the petitioned-for unit in this

matter is primarily educational, and that student assis-

tants are not employees under Section 2(3) of the Act.

Accordingly, I respectfully dissent.

Dated, Washington, D.C. August 23, 2016

______________________________________

Philip A. Miscimarra, Member

ATIONAL LABOR RELATIONS BOARD

See fn. 35, supra.