Experience with Company Unions and their Treatment under the Wagner Act: A Four Frames of Reference Analysis

• DOI: http://doi.org/10.1111/irel.12124
• Date: 01 January 2016
• Published date: 01 January 2016

Experience with Company Unions and their

Treatment under the Wagner Act: A Four Frames

of Reference Analysis

*

BRUCE E. KAUFMAN

This paper reexamines American experience with company unions (also known as

nonunion employee representation plans) before they were banned by the Wagner

Act (1935). For the half-century following the passage of the act, labor historians

and industrial relations scholars painted a bleak portrait of company unions as anti-

union sham organizations. Since the 1980s, additional research has documented a

more positive side; similarly, concern has grown that the Wagner Act’s ban is sti-

fling legitimate employee participation programs. This paper brings new theoretical

and empirical evidence to both historical and legal parts of this debate, including

examination of company unions through individualist, unitarist, pluralist, and radi-

cal frames; demonstration that the pluralists’view of company unions was more

diverse and positive than conventionally portrayed; presentation of new historical

evidence and testimony on the company union experience; and a substantially revi-

sionist assessment of the merits of the Wagner Act’s ban. In particular, the conclu-

sion is that, given any reasonable weighting of the four frames, the company union

ban is overly restrictive and should be modified so companies can implement the

positive side of nonunion employee committees but not the negative. The paper

ends by noting that the unbalanced and narrowly critical treatment of company

unions in the mainline industrial relations tradition is a case study of the field’s per-

haps fatal post–World War II core intellectual-normative contradiction—professed

inclusiveness of all frames of employment relations but, in practice, attention to and

preference for a narrow union-centric version of one frame.

Almost anything that may be said about employee representation will

be true.

—William Leiserson (1928: 119).

*The author’saffiliations are Georgia State University, Atlanta, Georgia and Griffith University, Brisbane,

Queensland, Australia. Email: bkaufman@gsu.edu.

INDUSTRIAL RELATIONS, Vol. 55, No. 1 (January 2016). ©2015 Regents of the University of California

Published by Wiley Periodicals, Inc., 350 Main Street, Malden, MA 02148, USA, and 9600 Garsington

Road, Oxford, OX4 2DQ, UK.

3

Freeman and Medoff (1984) popularized the idea that trade unions have two

faces, which they labeled “monopoly”and “voice.”They only briefly dis-

cussed company unions and assigned them one face, described as “window-

dressing”(p. 108). This dismissive characterization typifies the scholarly

record in the half-century after enactment of the National Labor Relations Act

(NLRA, or Wagner Act). Brody (1994) asserts, for example, the NLRA,

“swept out alternative forms of workplace representation because no com-

pelling case was made for them”(p. 43).

Over the last 25 years the company union experience in the United States

has received renewed scholarly attention, including in some cases a more posi-

tive assessment (Cooper 2011; Fairris 1997; Jacoby 1989; Kaufman 2000;

Nelson 1982; Pencavel 2003; Rees 2007). A major impetus has been concern

that the NLRA’s stringent restrictions on nonunion representation committees

is a contributing cause of a participation–representation gap in the American

workplace (Freeman, Boxall, and Haynes 2007; Freeman and Rogers 1999;

symposium in University of Pennsylvania Journal of Labor and Employment

Law, Spring 2001) and an impediment to employee-involvement programs and

high-performance work systems (e.g., Estreicher 2000; LeRoy 2006). Although

originally a mostly American and Canadian topic (Kaufman and Taras 2000),

the company union experience has become pertinent to broader international

debates in industrial relations (IR) and labor law as researchers examine the

legitimacy and effectiveness of nonunion employee councils in light of the

continuing shrinkage in union density in many countries (Gollan, Kaufman,

Taras, and Wilkinson 2014; Wilkinson, Donaghey, Dundon, and Freeman

2014). Unfortunately, the subject has not entered the management side of

research on human-resource systems and employee involvement/participation

programs even though central to these studies is pinpointing factors which

inhibit or encourage employee voice (e.g., Brinsfield 2014).

The conjunction of North American and international dimensions, and the

reason American labor law contributes to a participation–representation gap,

are well illustrated by the experience of the German auto producer Volkswa-

gen at its Tennessee assembly plant. The company wants to introduce a Euro-

pean-style works council but by the company union ban in the NLRA cannot

do so unless under the aegis of an independent union. The workers in 2014

narrowly rejected the United Automobile Workers in a representation vote

(Greenhouse 2014), leaving the plant with no form of collective voice.

This paper carries the reevaluation project on company unions forward another

step through three innovative features. The first is to examine company unions

through four alternative frames of reference, thus highlighting in a more concep-

tually structured way the diverse purposes, forms, and outcomes of these organi-

zations. The second is to use as input for the frames of reference analysis a

4/ B

RUCE E. KAUFMAN

considerable amount of new material on company unions gleaned from in-depth

mining of historical sources. The third is to use the frames of reference evidence

to evaluate the public interest efficacy of the NLRA’s ban on company unions.

The conclusion is that by any reasonable weighting of the four frames of reference

the NLRA is too restrictive toward nonunion representation plans. Overall, I am

not attempting to establish the truth of any one frame of reference; rather, the les-

son is—per the quotation from Leiserson in the epigraph—the company union

experience was extremely variegated and the verdict depends in part on the cases

examined and the observer’s predisposition toward the subject. Unfortunately for

a balanced treatment of employee representation, however, a vortex of economic

and political events surrounding the Great Depression and New Deal seriously

discredited the entire program, leading Senator Robert Wagner to write the NRLA

with an almost 100 percent ban on the practice. This paper suggests that with the

benefit of hindsight the NLRA’s company union ban is too one-sided and draco-

nian and should be relaxed so companies have wider space to practice collective

but nonunion forms of employee involvement and voice, as has been allowed in

Canada for more than seven decades (McDowell 2000; Taras 2014).

Frames of Reference

A contribution of this paper is to put the company union experience into a

conceptual framework. The framework chosen is from industrial relations and

is popularly known as IR frames of reference. Fox (1966, 1974) originally

proposed three frames: unitarist, pluralist, and radical. Budd and Bhave (2008)

recently expanded this typology with a fourth frame called egoist. Other

authors have constructed different typologies; Harris (1982), for example,

divides employers of the 1930s into three categories: belligerents, sophisti-

cates, and realists. The Budd and Bhave typology, however, is preferable

because it gets to core assumptions about the employment relationship, as

opposed to a descriptive taxonomy of employers’IR strategies. The IR frames

of reference are often cited in the literature but seldom used as an explanatory

device (although see Kaufman 2014a), leading Lyddon (2003) to observe, “If

industrial relations academics cannot use their own discipline’s categories then

it is not surprising that historians take little notice of their work”(p. 108).

Budd and Bhave (2008) define a frame of reference in generic terms as “how

one sees the world”and, more specifically, as a “theory used to guide and eval-

uate behaviors, outcomes and institutions”with respect to the employment rela-

tionship (p. 94). They distinguish frames of reference based on four dimensions

(p. 102): employer interests, employee interests, state interests, and key beliefs.

Company Unions and the Wagner Act /5