Content uploaded by Noah Zatz
Author content
All content in this area was uploaded by Noah Zatz on May 02, 2016
Content may be subject to copyright.
1
Prison Labor and the Paradox of Paid Nonmarket Work
Noah D. Zatz
UCLA School of Law
zatz@law.ucla.edu
Note: This is a near-final draft of a chapter appearing in Economic Sociology of Work (Nina
Bandelj ed., Emerald Press 2009). The published version is available in print and at
http://www.emeraldinsight.com/Insight/viewContainer.do;jsessionid=672B7F29F5E129
A085C79A8503E2B8A4?containerId=15001469&containerType=BookVolume.
Abstract
Purpose
To use insights from economic sociology to analyze how U.S. employment law
understands and regulates the relationship between prison labor and conventional employment.
Methodology
Legal analysis of all published court opinions deciding whether federal employment laws
such as the minimum wage apply to prison labor.
Findings
Courts decide whether prison labor is an “employment relationship” by deciding whether
it is an “economic” relationship. Most interpret prison labor as noneconomic because they locate
it in a nonmarket sphere of penal relationships. A minority of courts use a different conception
of the economy, one which interprets prison labor as a form of nonmarket work.
Implications
The economic character of prison labor may be articulated using the same theoretical
perspectives and analytical techniques developed to analyze family labor as economically
significant nonmarket work. Doing so, however, too readily accepts the market/nonmarket
distinction. Given the thoroughly social character of market work, prison labor‟s highly
structured, institutionally specific character does not preclude characterizing it as market work,
and some of its features support interpreting it as such.
In this legal context, identifying practices as economic or not, and as market or not, has
concrete consequences for the actors themselves. Rather than using market/nonmarket
distinctions as analytical tools, scholars might treat actors‟ designation of an economic practice
as part of a market or not as a site of conflict, subject to institutionalization, and worthy of
sociological study.
2
Prison labor presents a paradox for scholars studying the relation, and distinction,
between market and nonmarket work. Inmates typically get paid an hourly wage and are tightly
supervised. Their work situation possesses two features often used to define employment
relationships: what in the social science literature Tilly & Tilly label “short-term monetization”
and “time-discipline” (1998: 30-31) and what legal doctrine would call “compensation” and
“control.” Moreover, prison labor lacks features prominent in discussions of nonmarket work,
features often referenced as “intimacy”: work embedded in a highly particularized relationship
among a small number of individuals and often characterized by the worker‟s identification with
or protection of the interests of those who benefit from her labor (Kittay, 1999; Zelizer, 2005c).
Nonetheless, this chapter shows that U.S. courts consistently cast prison labor as a nonmarket
activity and, they reason, therefore not “economic” in nature.
Such classificatory questions have concrete stakes. Because of its nonmarket character,
courts exclude prison labor from the legal category of employment. Only employment
relationships are subject to labor protections like the minimum wage. Therefore, courts‟ account
of economic action becomes linked to whether inmates can challenge working conditions such as
prison wages at or below $1 per hour.
Judicial inquiry into prison labor‟s economic character implicates questions prominent in
economic sociology. Concepts developed by economic sociologists can illuminate legal disputes
over the nature of employment relationships. In turn, these legal boundary disputes suggest
fruitful extensions of sociological research into law‟s role in constructing markets (Dobbin and
Dowd, 2000; Edelman and Stryker, 2005).
Prison labor raises questions about the boundaries of employment that differ from those
usually addressed by legal scholars or sociologists of work. Most research addresses distinctions
3
internal to labor markets. That is, when work crosses the line and exits employment, it
nonetheless remains market work, albeit of a different form. Streeck, for instance, posits a labor
market consisting of “contracts of employment” and “contracts of work” (2005), roughly
tracking the legal distinction between employees and independent contractors. Similarly, casual,
contingent, or informal work may lie outside employment and yet remain within the labor market
(Portes and Haller, 2005). My topic here is a different boundary of employment, across which
lies the domain of nonmarket work.
Conversely, scholarly analysis of nonmarket work rarely leads to interrogating the scope
of “employment” or “labor markets.” Instead, the point, or premise, is that employment is
simply one form of the more general, and basic, category “work.” Nonmarket work, especially
household production, forms another “sector” external to the labor market (Abbott, 2005;
Cameron and Gibson-Graham, 2003). What divides the sectors is the line between paid and
unpaid work. On such an account, employment remains analytically untouched, though it must
share the spotlight with nonmarket work.
My goal here is to analyze a form of subordinated work as economic by questioning how
courts distinguish market from nonmarket work, rather than by simply expanding the economy
“beyond the market” (Abraham and Mackie, 2005). To do so, I look at “nonmarket” work
through the lens of economic sociologists‟ accounts of “markets.” First, I illuminate the
economic character of nonmarket activity and its integration into the workings of market
institutions, drawing on the Polanyian critique of an autonomous market economic sphere
(Krippner and Alvarez, 2007). That boundary crossing is what we see when prison industries
advertise themselves as “the best kept secret in outsourcing” (Unicor Federal Prison Industries,
2008), a way to shift production between market and nonmarket modes. Second, the force of the
4
“non” in nonmarket work comes from its contrast with market activity imagined as asocial,
arms-length bargains. That view of how markets operate runs afoul of the Granovetterian
account of markets‟ embeddedness in social ties (Krippner and Alvarez, 2007) and of social
constructivist accounts of markets more generally (Bandelj, 2008).
If market work is socially embedded work, then what makes it “market” work
nonetheless? And how can this criterion for “market” status distinguish “nonmarket” work
without simply invoking the presence or absence of social embeddedness, which would provide
no distinction at all? My suggestion is that market/nonmarket distinctions should themselves
become an object of study rather than a tool of analysis. Perhaps there is a social process of
differentiating institutions into market and nonmarket spheres that itself requires investigation, a
process that risks being reified when “market” and “nonmarket” are used as simple terms of
description.
To approach these questions, this chapter begins by reviewing basic concepts in
economic sociology and feminist analysis of household labor that give rise to the notion of
nonmarket work and identify activities as such. Next I use these concepts to interpret legal
disputes over whether labor and employment laws apply to prison labor, disputes that turn on
whether prison labor is an “economic” phenomenon. Prison labor may readily be characterized
as economic by conceptualizing it as nonmarket work, contrary to courts‟ reduction of the
economy to markets. Doing so, however, preserves a sharp distinction from employment
understood as market work. I question the stability of that distinction by suggesting how prison
labor also could be interpreted as market work. The chapter concludes with brief reflections on
the practical significance of market/nonmarket distinctions and the prospects for studying them
sociologically.
5
Economic Sociology of Markets and of Nonmarket Work
As I will show, legal disputes over whether to characterize prison labor as an “economic”
phenomenon largely reproduce social scientists‟ notorious formalist/substantivist debate about
the nature and scope of economic action. This section briefly reviews that debate (Polanyi, 2001
[1957]; Cancian, 1966; Granovetter, 2001) and related insights from economic sociology and
allied fields. Of crucial importance are two complementary lines of argument that attack
distinctions between economic and social action. The first emphasizes the social character of
economic action within conventional markets; the second emphasizes the economic character of
social institutions conventionally seen as separate from market relationships.
Formalism identifies the economic with means-ends rationalizing behavior by individuals
seeking to maximize the realization of their discrete interests; these interests are construed
narrowly to prioritize material or financial concerns. In other words, economic action is the
“economizing” behavior associated with market actors in modern Western economies (Cancian,
1966: 466). Insofar as institutions are characterized by other forms of action, they are not
economic in nature. Standing apart from such institutions, the economy is understood as a self-
regulating sphere governed by its laws of supply and demand.
Economic sociology rejects the formalist view, especially insofar as it defines rather than
merely describes economic action. Instead, economic phenomena are conceived broadly as “that
complex of activities which is concerned with the production, distribution, and consumption of
scarce goods and services" (Smelser and Swedberg, 2005: 3; see also Zelizer, 2005c: 3, 13).
The first line of analysis opened up by this substantivist view is a sociology of markets
(Fligstein and Dauter, 2007). This strand of economic sociology takes "bread-and-butter
economic issues" (Swedberg and Granovetter, 2001: 6) as the objects of study and seeks to
6
understand them as fundamentally structured by, or “embedded” in, “ongoing social relations”
irreducible to arms-length bargains between self-interested maximizers (Granovetter, 2001: 51).
The quarrel here is over how to understand market economic action, not over what conduct
constitutes “the economy.”
A second possibility opens up once markets no longer define the field: economic activity
may exist outside conventional market institutions. This direction has been less prominent
within economic sociology (Abbott, 2005; Zelizer, 2005b). Feminist scholars in sociology and
other disciplines, however, have deployed a substantivist framework to incorporate into accounts
of economic life various activities associated with women and domestic households, ranging
across subsistence agriculture, food preparation, clothing production, housecleaning, and child
care (Benería, 1988; Boydston, 1990; Folbre, 1982). The most adventurous work incorporates
activities for which markets are suppressed or may not exist at all, such as sexuality, friendship,
and maintaining kinship ties (Di Leonardo, 1987; Ferguson, 1989; Zelizer, 2005c).
1
To characterize nonmarket activity as “economic,” the family labor literature identifies
three important phenomena that I will label “circulation,” “substitution,” and “incorporation.”
The first way to articulate nonmarket activity‟s economic character is to trace how work products
circulate across what Arjun Appadurai labels “regimes of value” (1986) or between what
Viviana Zelizer calls “circuits of commerce” (2005a).
2
Consider an industrial homeworker who
1
Volunteering and other forms of (often gendered) civic participation also have been
incorporated into the “care work” framework, which also overlaps with research influenced by
Marxian traditions that take production of use value as basic and that consider work ranging
from domestic production to forced labor (Tilly and Tilly, 1998; Tomlins, 1995).
2
These are not the same concepts, but their considerable overlap suffices for present purposes.
7
works at home in part to coordinate production with child-care and who mobilizes her familial
relationship to child or spouse to enlist them in sorting beads or rolling cigars. The
embeddedness of that work in home and family is compatible with its products‟ subsequent
circulation in wider markets (Boris, 1994). Similarly, when familial housework and caregiving
enable wage-workers to do their jobs (either the next day or when they grow up), intimate
activity helps reproduce the labor traded in markets as a commodity (Fineman, 2004; Folbre,
2001; Rapp, 1982).
Second, even if work products never circulate beyond their original context, they may
substitute for goods and services that otherwise would be obtained from conventional markets.
Thus, subsistence agriculture substitutes for food purchases, home cooking substitutes for
restaurants and take-out, parental child care substitutes for day care centers or nannies, and so on
(Abbott, 2005; Boydston, 1990; Folbre, 2001).
Circulation and substitution undermine the notion, long criticized by feminists, that there
are analytically distinct, albeit complementary, “separate spheres” of market economy and
intimate family (Olsen, 1983). Instead, we see actors participating in both, and in ways that
interact via their economic character.
Thus far, the economically productive character of nonmarket activity remains grounded
in its connection to market transactions. We know that when commercial restaurants hire
workers to prepare meals to sell to customers, this is economic activity; therefore, when family
members do something functionally equivalent, that must be economic, too. Moreover, because
they are anchored in conventional market transactions outside a nonmarket institution,
circulation and substitution tell us nothing about the character of relationships among
8
participants within that nonmarket institution. All we know is that these relationships can be
characterized as economic.
The economic character of nonmarket work also resides within nonmarket relationships.
As Zelizer notes more generally, one consequence of separate spheres thinking is an extensive
"failure to recognize how regularly intimate social transactions coexist with monetary
transactions: parents pay nannies or child-care workers to tend their children, adoptive parents
pay money to obtain babies, divorced spouses pay or receive alimony and child support
payments, and parents give their children allowances, subsidize their college educations, help
them with their first mortgage, and offer them substantial bequests in their wills" (2005c: 27).
Focusing specifically on family labor, legal scholar Joan Williams argues that “unpaid”
housework and caregiving should be understood as one component of a larger complex she
labels “domesticity” (2001). Domesticity institutionalizes an intra-household gendered division
of labor between market and nonmarket work. It also structures (unequally) access both to the
products of nonmarket work (by, paradigmatically, a wageworking husband and minor children)
and wage income from market work (cf. Hasday, 2005).
These domestic relationships illustrate a third way to characterize nonmarket activity as
economic: they incorporate quintessentially economic transactions involving the distribution (to
the worker) of valuable goods, services, money, or other “media” of exchange (Zelizer, 2005c).
Incorporation, in effect, is circulation in reverse: the stuff of conventional market transactions
circulates beyond them and is incorporated into nonmarket relationships.
Examining how incorporation occurs requires us to confront the argument that
understanding intimate relationships as economic reduces them to market transactions and
thereby misrecognizes their nonmarket character. That critique relies on a shift back to
9
formalism. On the one hand, particularistic, affective, noninstrumental relationships among
family members preclude characterizing them as economic. On the other hand, if we do treat
family caregiving as an economic phenomenon, then we must interpret it as (implicitly) a market
bargain between the caregiver and somebody, whether a spouse, the state, or the recipient of
care. The latter, of course, is precisely the intellectual agenda that Gary Becker pioneered within
neoclassical economics (1976). In Zelizer‟s terms, the shift is from “separate spheres” to
“nothing-but”: identifying a relationship as economic fully specifies its essential nature because
being economic entails a specific (market) relational form, and nothing else (2005c).
Those normatively critical of markets worry that highlighting the economic character of
nonmarket relationships will transform their meaning, experience, and social organization,
disrupting or defiling them through “commodification” (Radin, 1996; Walzer, 1983; for a critical
review, see Williams and Zelizer, 2005). This “commodification anxiety” (Williams, 2000) is
triggered by feminist proposals to recognize and restructure nonmarket work‟s economic
character through such institutional changes as linking family caregiving to post-divorce
entitlements, qualification for social security benefits, or contemporaneous state support. Such
reforms seek to reverse the legacy of what historian Jeanne Boydston describes as the
“pastoralization” of women‟s housework as manifesting maternal nurturance (1990); this process
of “turning labor into love” (Silbaugh, 1996) complemented men‟s migration from household
production into industrial wage labor in the antebellum United States. Underpinning the
commodification critique is the notion that intimate relationships‟ pastoral quality will be
undermined by integration with economic transactions, particularly monetary ones; this concern
reflects the “hostile worlds” notion that intimate and economic relationships are intrinsically
incompatible (Zelizer, 2005c).
10
Zelizer‟s work opens up a space in which a relationship‟s incorporation of economic
transactions yields no intrinsic push toward market meanings and forms; to the contrary, such
transactions may be constitutive of particular forms of intimacy. Because it defuses the
commodificatoin critique, Zelizer‟s analysis has been particularly important to feminist legal
scholars seeking to enhance the economic entitlements linked to nonmarket work without
stripping that work of its distinctive nonmarket character (Ertman and Williams, 2005; Hasday,
2005).
Notice, however, that this capacity simultaneously to recognize a practice‟s economic
character and to differentiate it firmly from the market blunts some of the force of calling it
“work.” Both analytically and politically, the point of characterizing family caregiving or
housework as “nonmarket work” is to identify their commonality with employment or “market
work”: both are work, just in different market and nonmarket forms. Identifying this
commonality, however, simply begs the question whether a shared status as “economic” or
“productive” is more important than a differentiated status vis-à-vis the market form. Formalist
distinctions between economic and noneconomic actions or institutions might simply be recast as
distinctions among market and nonmarket forms. These would be two “substantive varieties” of
economic action characterized by different sets of “political factors, cultural understandings,
institutional arrangements, and social ties that shape different systems of economic organization”
(Bandelj, 2008: 11).
I now turn to the concrete case of prison labor to illustrate how concepts of nonmarket
work provide insights beyond the household context and to suggest how some of their limitations
might be overcome.
11
Background: Prison Labor and Employment Law
Prison labor is a ubiquitous feature of incarceration in the contemporary United States.
Of the nearly 1.5 million inmates in state and federal prisons, approximately half participate full-
time in a work program. This work usually involves relatively low profile “state use” activities
on government projects ranging from the iconic task of stamping license plates to baking cookies
for holiday parties. Many inmates perform “prison housework,” doing the laundry, cooking the
meals, and sweeping the floors, thereby maintaining the prison itself. The remaining 10-20% of
inmate labor occurs in “prison industries,” through which either the prison or a contractor sells
inmate-produced goods and services. The ultimate customer sometimes is another arm of
government, which might purchase anything from office furniture to military body armor.
Private firms also use inmate labor for tasks like taking telephone hotel reservations or stitching
graduation gowns. Prison industries generate $2 billion in revenue annually. The value of state
use projects has not been quantified systematically, but hiring equivalent labor probably would
cost billions (Zatz, 2008).
Most prison labor programs pay hourly or daily wages, though some either add or
substitute “good time” credit toward reduced sentence length. For instance, California inmates
staff crews that fight forest fires, for which they receive $1 per hour and one day off their
sentence for each day of work (Arnoldy, 2008). Although rates of pay vary widely, they almost
always fall far below the federal minimum wage that applies to employment covered by the Fair
Labor Standards Act (FLSA), which in 2009 will be $7.25 per hour.
The first FLSA challenge to such wages came from an inmate working for a Michigan
military contractor stamping artillery shell casings during World War II (Huntley, 1948).
3
No
3
My analysis here and throughout is based on my review of the 68 reported judicial opinions
12
further litigation appears in published court opinions until the early 1970s, when decisions began
to address inmate work in pharmaceutical companies‟ facilities within prisons (Sims, 1971;
Hudgins, 1971; Lavigne, 1982; Alexander, 1983; Gilbreath, 1991).
Courts rejected these early claims on the ground that the inmates were not “employees”
of the private entities being sued. Judges reasoned that the prison ultimately controlled the flow
of inmate labor, deciding who could participate and reserving the right to terminate participation.
Moreover, prisons negotiated many of the conditions under which inmates worked, including
their rate of pay. These are standard legal considerations when deciding whether a worker is
employed by one firm or another. In essence, the courts cast the defendant firms as customers of
the prison, with the prison acting either as a subcontractor providing services to the firm or as a
temp agency supplying workers (Weiss, 2001). The Huntley opinion explained that “plaintiffs
were employees of the Michigan prison industries and not of the defendant [military contractor]”
(116).
This line of analysis ran into two difficulties. First, it conflicted with legal developments
meant to prevent firms from using subcontracting and temp agencies to avoid employment law
obligations. Drawing on precedents involving building maintenance workers and home health
aides, a federal appeals court in 1984 rejected the argument that a prison‟s “ultimate control”
precluded an employment relationship between an inmate and the organization for which he
worked (Carter). Instead, an employment relationship might exist based on the defendant‟s role
in selecting inmate workers, structuring and supervising their activities, integrating them into its
larger operations, and paying them. The court further held that so long as these conventional
deciding cases brought by inmates and other detainees under federal employment laws or their
close analogues. For a full listing, see Zatz (2008: 882-84).
13
indicators of employment were present, inmate status alone did not preclude FLSA protection
because nothing in the statute specifically excludes prisoners from coverage.
The second vulnerability of the “ultimate control” analysis appears in Huntley‟s
characterization of inmates as employees of the prison. By the mid-1980s, technical changes in
the FLSA statute and in federal constitutional law made it much easier to assert minimum wage
claims against state government agencies of all sorts (Zatz, 2008). FLSA litigation began to
target state-run work programs and name the prison itself as the defendant, unlike earlier suits
against private corporations. The prison‟s high level of control now supported finding an
employment relationship.
Within a decade of Carter, the earlier doctrines had collapsed, and courts routinely
accepted that considerations of organizational control and integration favored the existence of an
employment relationship. Moreover, they adopted Carter‟s holding that nothing categorically
excluded inmates from the class of employees. Instead, FLSA coverage turned simply on
whether an employment relationship existed as a non-technical matter of ordinary social
description. With these barriers cleared away, several federal courts (and one dissenting judge)
ruled in inmates‟ favor at preliminary stages of litigation, both in FLSA minimum wage suits
(Watson, 1990; Gilbreath, 1991; Hale, 1992) and also in an employment discrimination case
presenting analogous issues (Baker, 1988).
These inmate victories were short-lived. The tide again turned against inmate
employment claims even though courts never revived the pre-Carter analysis of control. Instead,
they amplified and elaborated an intuition that previously had been articulated in a secondary,
subdued fashion: the penal context itself renders inmate labor fundamentally different from
ordinary employment. One influential opinion set out to theorize explicitly “our common
14
linguistic intuitions,” which are “at least strained by the classification of prisoners as
„employees‟” (Vanskike, 1992: 807). Previous analyses focusing on control, the court explained,
explored only one boundary of the employment relationship, that between employee and
independent contractor. Prison labor implicates “a different boundary” (810), the line between
economic and noneconomic activities that is my topic here.
Echoing Formalism: Prison Labor as Nonmarket and Therefore Noneconomic
Courts today generally classify prison labor as a noneconomic practice and therefore not
employment. Vanskike reasoned that inmates were “not in a true economic employer-employee
relationship” (812). Instead, in oft-quoted language, incarceration itself means that “[p]risoners
are essentially taken out of the national economy” and placed into “the separate world of the
prison” (810; Hale, 1993; Henthorn, 1994).
This conclusion that prison labor is noneconomic relies upon a formalist equation of the
economic/noneconomic and market/nonmarket distinctions. If an activity falls outside the
market then it falls outside the economy. “A true employer-employee relationship,” courts insist,
involves a “bargained-for exchange of labor for mutual economic gain” (Harker, 1993: 133).
The concept of “bargained-for” is drawn from the classical contract doctrine of “consideration”
and refers to a mutually instrumental interaction, one in which each party acts for the purpose of
eliciting the other‟s act and does so, furthermore, to obtain commercial advantage. Moreover,
because this instrumental exchange exhausts the entire relationship between the parties, it
necessarily occurs “at arms‟ length” (Harker, 1993: 133), outside any ongoing social connection
or shared institutional context. The FLSA only applies where “labor is exchanged for wages in a
free market” (Hale, 1993: 1394).
15
Measured against this conception of “traditional free-market employment” (Villareal,
1997), prison labor easily falls short. Courts demonstrate separation from the market by pointing
to embeddedness in the prison. An inmate‟s work assignment is only one moment in an ongoing
relationship of incarceration. Rather than an isolated arms-length encounter, it “arises out of the
prisoner‟s conviction for a crime” (McCaslin, 1996: 657) and his “status as an inmate”
(Williams, 1991: 997). The relationship is not open to all comers willing to make a deal; instead,
this “opportunity is open only to prisoners” (Danneskjold, 1996: 43).
Not only is the prison a bounded institution rather than an open market, but this
institution is structured by motivations and interests irreducible to the economic maximization
attributed to market transactions. Opinions finding no employment relationship are replete with
references to the prison‟s “rehabilitative or penological interest in inmate labor” (Danneskjold,
1996: 43-44) and denial of “pecuniary” interests (Hale, 1993: 1395): “Inmates perform work
[for the prison industry] not to turn profits for their supposed employer, but rather as a means of
rehabilitation and job training” (Harker, 1993: 133). Outside the litigation context, too, prison
administrators and supporters of inmate labor insist that the labor process provides inmates with
valuable occupational skills, personal development, and dignity (Flanagan, 1993; Wignall,
2005). Even analysts less sympathetic to prison labor see it as shaping inmates into citizen-
workers and not simply as extracting valuable work (Foucault, 1995; Melossi and Pavarini,
1981). Labor programs also provide a short-term disciplinary mechanism by occupying inmates‟
time and by creating a perquisite that may be offered, continued, or withdrawn based on the
prison‟s judgment of the inmate‟s “good behavior” (Carlson, 2004; Sykes, 1971).
Prison labor‟s integration with the distinctive goals and structures of imprisonment also
shapes how courts interpret inmates‟ subjection to coercion, both in their work and in prison life
16
more generally. When prison authorities require inmates to perform particular work
assignments, the involuntary nature of the work provides a relatively straightforward distinction
from the bargains struck by “free labor” in the “free market.” Thus, a prisoner “legally
compelled to part with his labor as part of a penological work assignment” has not “freely
contracted . . . to sell his labor” (Henthorn, 1994: 686).
Courts adhere to this distinction between prison coercion and market freedom even if
inmates possess some formal control over their work. When inmates exercise choice among
possible work assignments, courts reason that inmates still “have not freely contracted to sell
their labor. Choosing where to work is not the same as choosing whether to work” (Nicastro,
1996: 1447). Even in cases where inmates face no obligation to work at all, courts still conclude
that “[v]oluntary work serves all of the penal functions of forced labor” and thus should be
treated the same; after all, “the prisoner is still a prisoner” (Danneskjold, 1996: 43).
Involuntariness rooted in the prison‟s authority over the inmate simply provides a way of
illustrating an underlying point: “his work at the prison was merely an incident of his
incarceration” (Morgan, 1994: 1293).
Placing prison labor in a separate, noneconomic penal sphere also reflects courts‟
understanding of the market economy. Market and penal relationships are perceived to be
mutually exclusive only because courts imagine market encounters as arms-length, instrumental,
financially-motivated bargains,. On this foundation, courts take what I have labeled an
“exclusive market” approach to analyzing employment‟s economic character: the presence of
any penal character marks the absence of a market relationship (Zatz, 2008: 882).
The exclusive market approach is particularly visible when courts confront aspects of
prison labor that might seem characteristic of market economic transactions. Judges relentlessly
17
rebut any suggestion of continuity with market labor by tying inmate work to its penal context.
As noted above, voluntariness has been dismissed as not detracting from the fundamentally penal
character of inmate labor. One court goes further and argues that voluntariness reinforces that
penal function. Why? Because voluntary work may “better serve[] institutional needs of the
prison” when “the quality of performance is difficult to measure and may depend on motivation”
(Danneskjold, 1996: 43). Here, a classic “market” rationale for incentive wages becomes a
marker of penality.
Similarly, courts deny that prisons‟ financial interests in inmate labor—from selling the
products or from avoiding other expenditures—implies any pecuniary purpose. Sometimes they
simply point to evidence of nonfinancial goals, as if the two could not coexist (Burleson, 1996).
More aggressively still, these financial consequences can be characterized as advancing the
institutional interests in effective punishment: inmate labor “offset[s] some of the cost of
keeping them [imprisoned]” (Bennett, 2005: 410) and forces “prisoners [to] bear a cost of their
incarceration” (Danneskjold, 1996: 43).
The prison labor caselaw shows a disembedded understanding of economic life to be a
ubiquitous feature of judicial reasoning. Nonetheless, in courtrooms, too, there are alternative
analyses of economic action, to which I now turn.
Echoing Substantivism: Prison Labor as Nonmarket but Economic Nonetheless
Although the exclusive market approach dominates the prison labor caselaw, it has been
challenged by what I have termed the “productive work” approach to employment‟s economic
character (Zatz, 2008: 883). In substantivist fashion, it identifies economic activity with the
production and exchange of goods, services, and money rather than with market dynamics. As a
18
result, it resists dividing “the economy” from other institutions and instead shows how prison
labor interacts with conventional labor and product markets.
Exemplifying the productive work approach is the last major opinion broadly endorsing
employee status for inmate workers. In Hale v. Arizona, the U.S. Court of Appeals for the Ninth
Circuit adopted an exclusive market analysis and concluded that prison labor “is penological, not
pecuniary” (1993: 1394-95); Judge Norris dissented vigorously. Joined by one colleague, he
reasoned that “[t]he economic reality is that [the inmates] work. Their labor produces goods and
services that are sold in the channels of commerce. And [the prison industry] pays them for their
efforts. Common sense tells us this relationship is both penological and pecuniary” (1403).
4
The productive work approach characterizes prison labor as an economic exchange (if not
necessarily a bargain) of financially valuable work and monetary payment. The attribution of an
“economic” character rests on linking the prison-prisoner relationship to the prison‟s interactions
with actors operating in conventionally recognizable markets.
What provides these links are processes of circulation and substitution, the same ones so
central to analysis of nonmarket family labor. Prison labor‟s products are fungible with the
products of conventional market labor, and thus inmate workers are fungible with conventional
employees. The prison mediates between two different forms of economic activity.
4
In related contexts, one court has rejected an “artificial dichotomy” in which “one must be
either a welfare recipient or an employee and cannot be both” (United States, 2004: 94); another
tribunal, though subsequently overruled, held that nothing prevented workers from being
employees “simply because they also are students” (New York University, 2000: 1209). On the
relationship between the prison labor cases and employment law analysis of paid nonmarket
work more generally, see Zatz (2008).
19
Judge Norris‟ dissent relentlessly focuses on the wide circulation of prisoner-made goods.
The inmate workers in question produced clothing, raised livestock, manufactured belt buckles,
and answered customer telephone calls. These products and services either were sold in
conventional product markets or were consumed by private firms that otherwise would have
acquired substitutes from such markets. On his analysis, “[t]he fact that prisoners are forced to
work is irrelevant because the unfair competitive effect is the same regardless whether the
worker is forced to work or free to work” (1403).
Although only a dissenting view in Hale, a similar analysis carried the day in the most
prominent employment law victory for inmate workers. In Watson v. Graves (1990), another
federal appeals court applied the FLSA to the relationship between inmate workers and a private
construction contractor. The arrangement was part of an ad hoc work-release program
orchestrated by the local sheriff, who happened to be the contractor‟s father-in-law. The court
emphasized how inmate labor interacted with the local construction market: “Obviously, [other]
construction contractors in the area could not compete with [the defendant‟s] prices because they
had to pay at least minimum wage for even unskilled labor . . . . [J]ob opportunities for non-
inmate workers in the area [were] severely distorted by the availability of twenty dollar per day
workers from the parish jail
”
(1555).
At the heart of Watson is the observation that circulation allows substitution to occur at
steps removed from the initial context of work. If an inmate produces a garment that the prison
sells in the market, then this circulation causes substitution vis-a-vis the purchaser‟s alternative
source of garments. The purchaser substitutes the prison‟s products for the products of another
(market) garment supplier, and thus the prisoners‟ labor substitutes for the alternative supplier‟s
employees.
20
Prison labor‟s integration into wider markets is most obvious when tangible products
circulate outside the prison, but prison labor also affects external markets by substituting for
them and thereby reducing demand. This, of course, is the mechanism behind the cost-savings
often touted as an advantage of prison labor. When its inmates fight forest fires, California saves
$80 million a year, in large part because it avoids having to expand its regular workforce paid
$10-12 an hour or to extend the hours (and potentially incur overtime) of its existing employees
(Arnoldy, 2008).
Even services internal to the prison yield substitution. Consider the analysis offered by
inmates whose work as prison law clerks was integral to the prison‟s constitutionally mandated
provision of legal services to other inmates: “if no qualified inmates had been available in the
inmate population, the prison would have had to turn to the outside market in the general legal
community [and] therefore the State derived a pecuniary benefit from appellants' work”
(Jovanovich, 1995, *2). Similarly, laundering inmate garments and sweeping prison floors are
tasks that private contractors or noninmate workers could be hired to do. Indeed, the Vanskike
opinion recognized this point but rejected the entire line of argument from substitution because
of its sweeping implications: “carried to its logical conclusion, prisoners must be paid minimum
wage for anything they do in prison that can be considered „work‟” (1992: 811).
In contrast to Vanskike, most courts attempt to reconcile their exclusive market approach
with an acknowledgment that prison labor‟s interaction with conventional markets sometimes
gives it an economic character. To do so, they seek to differentiate those forms of inmate labor
that do and do not affect wider markets. Courts routinely assert that services consumed in the
operation of the prison do not raise “the possibility of „unfair competition‟ between prison and
21
private labor” (Danneskjold, 1996: 44).
5
Some even assert that no substitution occurs when
prison industries sell their products to other units of government, rather than to private firms
(Harker, 1993; Miller, 1992).
6
Opinions written in the exclusive market vein also dampen their conflict with productive
work approaches by downplaying prison labor‟s productivity. In terms redolent with the
racialization of work and crime, courts figure prisoners as economically incompetent,
presumptively unemployed prior to entering prison, and fated to idleness if they finish their
5
Similarly, even opinions using a productive work approach to find an employment relationship
may limit their reach to cases involving circulation of prisoner-produced goods. Watson
suggested that any work done “within the confines of the prison” would avoid competition with
“workers in job markets outside the prison” (1990: 1555). Another judge would have found an
economic relationship when inmates produce goods for sale but not when they perform “prison
maintenance or produce goods used solely by the state” (Gilbreath, 1991: 1334). Courts
considering other forms of nonmarket work have been more expansive, finding employment
relationships where institutional maintenance work was “useful” or conferred “economic
benefits” on the employer (Souder, 1973: 813; United States, 2004: 97; Zatz, 2008)).
6
These efforts to separate prison labor from conventional markets are not solely matters of
judicial discourse. They are institutionalized in the rules and practices that structure inmate labor
programs, employment law aside. The Depression-era Ashurst-Sumners Act criminalizes sale of
inmate-produced goods in interstate commerce. By barring circulation, this process of
“enclaving” (Appadurai, 1986) partially separates prison labor from “the economy.” This ban
does not prevent other mechanisms of substitution, however, and there are numerous exceptions
(Zatz, 2008).
22
sentence without newfound skills. Prison labor thus “trains prisoners in the discipline and skills
of work” (Danneskjold, 1996: 43) and thereby “make[s] them less likely to return to crime
outside” (Bennett, 2005: 410). Such assessments of inmates complement the assertion that the
prison only “has a rehabilitative, rather than pecuniary, interest in [the inmate‟s] labors” (Harker,
1993: 133). Additionally, courts elide prisons‟ receipt of valuable work from inmates by treating
their labor as disembodied, unmediated by inmates‟ agency, and already in the prison‟s
possession from the start. Ubiquitous throughout the prison labor caselaw is the notion that “[an
inmate‟s] labor belongs to the prison and is at the disposal of the prison officials” (Watson, 1990:
1553).
To complete its characterization of prison labor as an economic exchange, a productive
work approach highlights the fact that, as Judge Norris observed, the prison “pays [inmates] for
their efforts” (see also Villarreal, 1997). The prison-prisoner relationship thus incorporates the
quintessentially economic medium of money. Moreover, prisons pay in a form denoted a
“wage” and structured as one: a fixed amount per time worked. As Zelizer has shown, the form
of monetary payments often significantly shapes their role as “definers of symbolic meanings
and social relations” (1996: 486). Indeed, in related legal controversies over workfare,
volunteers, and training programs, courts have placed great weight on the existence of payment
in the form of compensation (Zatz, 2008). Sociological research on welfare work programs
likewise emphasizes conflict over whether to characterize payments to participants as wages or
welfare benefits (Goldberg, 2007). Even courts ruling against inmates rarely resist
characterizing their payments as wages, though some have analyzed them as “gratuities”
(McGinnis, 1975: 1224 n.2 & 1238; Harris, 1968: 1016-17; cf. Henthorn, 1994: 686). Indeed,
one recent appellate case held that inmates have “a property interest in payment for their labor,”
23
albeit in a dispute where the existence of an employment relationship was not at issue (Allen,
1996: 261). Furthermore, although inmates‟ wages are integrated into a distinctive system of
prison accounts subject to restricted access and mandatory deductions (Zatz, 2008), courts do not
rely heavily on this form of earmarking to distinguish prison labor from market work.
Instead, courts integrate wage payments into an exclusive market analysis by focusing on
how these monies are incorporated into inmates‟ lives after payment is received. Contemporary
opinions denying employment status consistently emphasize that the prison provides inmates
with basic necessities of food, shelter, and clothing (themselves often incorporated from extra-
prison economies). This, in turn, differentiates inmates from ordinary employees who rely on
wage income to put food on the table through consumer purchasing. At the most technical level,
this point questions whether the goals of a minimum wage apply to inmate labor. More broadly,
it differentiates inmate labor from employment, where employment is understood as an
institution that connects the employer-employee relationship to specific patterns of consumption
and family life. Employment, in other words, is not just about earning compensation, it is about
“earning a living” (Bennett, 2005: 410; cf. Kessler-Harris, 1990).
7
The irony here is that the paradigmatic “free labor situation” cannot be reduced to a
socially isolated instrumental bargain in which “labor is exchanged for wages in a free market”
(Hale, 1993: 1394). Instead, “free labor” links the receipt of wage payments to a broader, and
specific, form of life involving market consumption and maintenance of an independent
household (cf. Stanley, 1998). Thus, what renders paid nonmarket work conceptually possible is
7
Prison labor thus provides a particularly striking illustration of Zelizer‟s thesis that
monetization does not necessarily marketize a practice‟s meaning and displace its contextual
significance (1997).
24
not simply inmate labor‟s embeddedness in the prison, as opposed to a disembedded market, but
also its deviation from the “substantive variety” (Bandelj, 2008) of embeddedness associated
with the labor market and yet not reducible to “market” mechanisms.
Prison Labor as Market Work
The productive work and exclusive market approaches to prison labor have fought to a
conceptual draw. On the one hand, inmates surely participate in economic relationships in which
they perform productive work. On the other, conceptualizing their efforts as nonmarket work
admits a basic difference from conventional market employment, a difference that resides in the
space for a multiplicity of economic forms.
At this juncture the study of nonmarket work would do well to rejoin the economic
sociology of labor markets. After all, the preceding discussion largely has taken for granted the
concept of “market work.” The labor markets described by economic sociologists bear little
resemblance to the caricature held up by courts adopting an exclusive market approach, and
characterizing inmate labor as nonmarket work leaves that caricature unchallenged. The
carework literature does somewhat better on this point. It occasionally notes that some specific
“nonmarket” feature of carework can be found routinely in the labor market (Williams, 2001), or
vice versa (Zatz, 2006), but it rarely casts a critical eye on the market/nonmarket distinction itself
(Cameron and Gibson-Graham, 2003).
8
This section offers as a thought experiment an analysis of prison labor as market work,
something courts and commentators have not considered previously. Such an would make a
8
In some cases, arguments for continuity between nonmarket and market carework rely upon
treating market carework as exceptional in its deviation from arms-length market bargains
(Kittay, 1999: 95).
25
stronger connection between conventional employment and prison labor than simply
characterizing inmates as “workers” engaged in productive activity. My purpose is not to argue
that prison labor really is market work. Instead, I hope to direct critical attention toward
market/nonmarket distinctions by showing their flexibility once informed by economic
sociological accounts of market embeddedness and nonmarket economies.
The case for treating prison labor as market work rests on two complementary arguments.
The first disputes the account of market work to which prison labor is compared. Conventional
employment possesses many of the “nonmarket” features attributed to prison labor, and so these
features do not differentiate the two. The second, affirmative argument is that prison labor
possesses many of the features (beyond productivity alone) used to place conventional
employment in the market domain.
The first point relies on scholarship that Streeck summarizes well: “Generally,
sociological research and theory maintain that the labor market is not really a market, in the
sense of a universalistic, impersonal, color- and gender-blind mechanism matching supply of,
and demand for, labor. . . . More generally still, economic sociologists have argued that not only
are labor markets not the sort of markets that economists believe they are, but they would not
function if they were” (2005; see also Tilly and Tilly, 1998).
Although different in their particulars, prototypical employment relationships possess the
same general characteristics that courts attribute to prison labor in order to contrast its social and
institutional specificity with market work. Ordinary employees work against a backdrop of
compulsion, including legal compulsion, in their immediate relationships to employers, their
general participation in the labor market, and their background access to economic resources
(Atleson, 1983; Tomlins, 1995; Zatz, 2008). The matching of workers and jobs is highly
26
structured by social networks, mediating institutions, and workers‟ status locations (Granovetter,
1995; Tilly and Tilly, 1998). Rather than a discrete market transaction between individual
actors, employment itself is a historically specific, highly institutionalized form the shape of
which has been influenced pervasively by state policy (Jacoby, 1995; Piore and Safford, 2006;
Stone, 2004). And various actors—owners, managers, employees, unions, and the state—
participate in and shape these institutions in ways irreducible to instrumental financial self-
interest. Instead, employment relationships are influenced by the assertion or development of
occupational identities, commitments to an interplay between work and family roles, beliefs in
work‟s redemptive value in daily life, promotion of employment as a basis of democratic
citizenship, and prevention of social disorder (Abbott, 2005; Kessler-Harris, 1990; Schultz,
2000; Streeck, 2005).
Put simply, the thoroughgoing social character of prison labor is consistent not only with
being (substantively) an economic relationship but also with being a market practice. Moreover,
the social specificity of different forms of employment (and employers and employees) defies the
universal abstraction of “market work.”
None of this denies that employment exhibits processes ordinarily identified with
markets, that financial incentives matter, that supply and demand affect prices, and so on. So if
those processes were absent from prison labor, a clear distinction might still exist. But that is not
the case.
“Market” processes are omnipresent in prison labor, though sometimes it takes some
looking to see them and sociological research on contemporary inmate work programs is limited.
Most obviously, when inmates work directly for private firms, there is every reason to think that
those firms approach these workers roughly as they do ordinary employees. Operators explain
27
their participation in terms of access to low wages, low turnover (because workers cannot simply
quit to take another job), and low pressure to provide job security (in part because layoffs do not
generate unemployment insurance claims), in combination with average work quality (Reynolds
and Rostad, 2005; Weiss, 2001). Programs operated directly by the state often seem to have
similar goals; frequently they are mandated to simulate private business organizations and turn a
profit. Similarly, cost-savings from substitution often explicitly motivate prison labor programs,
as with the California firefighters discussed earlier. In another case, those prisoners working in a
county-run labor program were excluded from a group of inmates transferred from county to
state custody because hiring employees to replace their work as janitors and cooks would have
cost the county $1 million per year (Rupp, 2006). These “market” considerations are not
necessarily the exclusive or dominant ones; the point is simply that they are present and
significant.
In terms of how inmates‟ work is organized, the most obvious market feature is the use of
wages as a financial incentive to work effort. As one prison industry supervisor
straightforwardly put it, “It doesn‟t matter how well you treat them. These guys will never work
hard unless they get paid. Would you?” (Gray and Meyer, 1996: 130) Indeed, in settings where
multiple work programs exist, privately-run programs use higher wages, in combination with
their ability to select among applicants, to attract the better-skilled and more productive inmates
(Miller and Crieser, 1986). Implicit in these phenomena, of course, is that inmates approach
work assignments in part as a financial proposition and adjust their behavior to increase earnings.
Finally, recall that U.S. law actually does treat inmate work as market employment in a
narrow, somewhat vaguely defined set of circumstances. Legal authorities classify inmates as
employees when “the prisoner voluntarily works outside the prison for a private company that
28
supervises and directly pays the prisoners” (Barnett, 1999: 2). Inmates in work-release programs
with such characteristics have been treated as employees for purposes of the minimum wage
(Barnett, 1999; Reimonenq, 1996; Watson, 1990), employment discrimination law (Walker,
1994; Baker, 1988), collective bargaining by labor unions (Speedrack, 1997), and credit toward
social security eligibility (U.S. Department of Labor, 1999). Even courts ruling against inmate
employment claims generally accept the results of these work-release cases, treating them as
exceptions that prove the general rule that inmate labor is a noneconomic relationship. An
opinion by prominent judge and law & economics scholar Richard Posner recently explained that
“[t]hose prisoners weren‟t working as prison labor, but as free laborers in transition to their
expected discharge from the prison” (Bennett, 2005). Another leading opinion characterized the
work-release cases as ones in which inmates “freely contracted with a non-prison employer to
sell [their] labor” (Henthorn, 1994).
Courts‟ classification of work-release programs as market employment illustrates the
flexibility of market/nonmarket distinctions. These programs easily could be characterized as
nonmarket work, like other prison labor. Each of the factors typically offered to classify work-
release as market employment—voluntariness, physical location, work for the imprisoning
government versus an independent entity—can be and has been interpreted to reach an opposite
result or rejected as arbitrary (Danneskjold, 1996; Henthorn, 1994). I noted this above in
Danneskjold‟s account of voluntariness. Similarly, although courts tend to see an ordinary
commercial contract in work-release, one readily could reinterpret the program as one piece of
larger penal project. The “simulate[d] private business working conditions” in prison industries
have been characterized as serving the goals of rehabilitation, not establishing similarity to
market labor (George, 1993: 586). Indeed, one court held that although an inmate on work-
29
release had an employment relationship with the private firm where he worked, the prison itself
was not a “joint employer” because “the purpose of the program is to prepare inmates upon
release from prison to function as responsible, self-sufficient members of society” (Reimonenq,
1996: 476). In fact, all the rationales offered for prison industries‟ nonmarket “penological
purpose” are routinely used to justify work-release, too (Flavin, 2005; Turner and Petersilia,
1996).
These penal aims, moreover, play an important role in structuring work-release programs.
As in other inmate work programs, the prison often retains partial or total control over the
inmate‟s wages, deducting sums for victim restitution, prison costs, or child support.
Notwithstanding courts‟ characterization of work-release as “voluntary,” inmates may be
disciplined by the prison for quitting work without good cause or refusing to work overtime.
And the prison exercises control over which inmates get to participate, when they participate,
with what employer, and so on.
All of these points have been raised by employers to argue that work-release participants
are not their employees. Courts have rejected them (Walker, 1994; Speedrack, 1997), even
though analogous arguments are used to place other forms of inmate work outside the market
economy. Moreover, this work-release exception has ill-defined boundaries that have shifted
over time. One decision from the 1970s held that because work release is authorized by statutes
governing inmates, nothing in employment law “convert[s] such status to that of an employee”
(Worsley, 1976). The pivotal Carter decision discussed earlier continues to be cited as an
example of work-release employment, and yet a later case with indistinguishable facts was
decided differently by the same court (Danneskjold, 1996).
30
Nonetheless, courts continue to reaffirm that inmates on work-release are employees in
the labor market, even if they do not agree on precisely when or why. The puzzle is how and
why work-release and other forms of inmate work get assigned to different sides of a
market/nonmarket divide.
Would economic sociologists achieve greater consensus on whether work-release
programs are a “market” phenomenon, and why? Such questions seem to have received little
attention. The term “market” remains in use to characterize objects of study—“sociology of
markets”—even while the field rejects economists‟ descriptions of how these markets operate.
Scholars direct their empirical and theoretical energies toward understanding the phenomena
already taken to be “market” institutions and practices, not toward investigating how that market
designation is produced or justified by the social actors in question or by the researcher. Indeed,
sociologists have noted the absence of a clear definition of “the market,” primarily as a critique
of the conceptual foundations of economics (Lie, 1997). But within economic sociology, too, the
question arises of what, if anything, is the content of this still-ubiquitous category.
In their recent review, Fligstein and Dauter grapple with this problem and offer a
sociological definition of markets as a distinct economic form: “[M]arkets imply social spaces
where repeated exchanges occur between buyers and sellers under a set of formal and informal
rules governing relations between competitors, suppliers, and customers” (2007: 113). This
definition is quite capacious if terms like “buyer,” “seller,” “supplier,” and so on are deployed in
a non-question-begging way. Certainly there are repeated exchanges between inmates and their
putative employers in both work-release and other prison labor programs, and these exchanges
are structured by formal and informal rules. Moreover, those exchanges occur within a broader
social space that links relationships among the prison, inmate, buyers of inmate-produced goods
31
and services, and outside entities that contract with prisons to operate work programs; the state
participates as a regulator, a funder of prisons and work programs, and as the initial supplier of
inmates through the criminal justice system.
Competition might seem to be the crucial characteristic of markets (Fligstein and Dauter,
2007; Swedberg, 2005) arguably lacking from prison labor. But perhaps not. As already noted,
there is some competition among inmates for selection into work programs and some
competition among work programs for selection by inmates. Moreover, the dynamics of
substitution mean that prisons place inmate labor into competition with alternative suppliers of
goods and services to the prison, including conventional firms and conventional employees.
Similarly, the dynamics of circulation mean that private firms place their conventional “market”
suppliers of goods and services into competition with prison industries and prison inmates. In
effect, “market” and “nonmarket” sources of goods, services, money, and labor “compete” with
one another when actors bridge these social fields (Zelizer, 2005a) and when these actors also
control the manner and extent to which they draw from one field or the other.
Of course, the “formal and informal rules” that structure these processes of exchange and
competition differ in important ways from those at play in conventional labor markets. But that
point simply exemplifies the key idea in economic sociology that there is no singular market
logic but instead a plurality of markets (Swedberg, 2005).
Thus, the interactions between prison labor and conventional labor markets easily could
be conceived as a relation between two different markets rather than between market and
nonmarket spheres. Similarly, a building contractor making an election between hiring
unionized tradespeople and streetcorner day laborers bridges two differently structured labor
markets. The day laborer may be excluded from the union and the tradesperson may not be
32
welcome on the streetcorner, yet surely they are in competition in the sense relevant to
encompass them both within some larger market field. The puzzle then is why we would group
together day labor and unionized trades as species of the abstraction “market labor” and yet
exclude prison labor. Instead, we might either dispense with “the labor market” as an
overarching category or include within it all three forms of work organization (and many others).
Conclusion
Whether work-release, prison industries, or prison housework is “market” work might
matter little were no more at stake than bounding an academic subdiscipline like the sociology of
markets. But in the legal disputes I have examined, quite concrete consequences flow from
whether a relationship is labeled market employment or not. To some extent, this reflects courts‟
assimilation of the widespread but misguided view that economic activity occurs exclusively
within markets and that it does so solely on the basis of asocial, profit-motivated, means-ends
rationality. Insights from economic sociology undermine that view and provide some support for
courts‟ construction of the productive work alternative: prison labor involves processes of
circulation, substitution, and incorporation, not radical separation from “the economy.”
Furthermore, the law hardly seems unique in rendering significant a practice‟s
designation as a market phenomenon. The very fact that actors understand their action to occur
in a “labor market” may serve to authorize those goals and forms of action associated with
“markets” and to delegitimize others (Bandelj, 2008). Likewise, this fact may shape both
participants‟ and observers‟ interpretations and experiences of their action (Albiston, 2007).
For these reasons, it should be unsurprising that an institution‟s social designation as part
of “the market” may itself be an object of struggle. That is one important lesson to be taken from
Chad Goldberg‟s fascinating recent study of work-relief and workfare programs, in which quite a
33
lot turned on “classification struggles” over whether participants in these programs were
“workers” in the labor market or “welfare recipients” in the domain of state assistance (2007; see
also Krinsky, 2007b) . Similar issues arise in contention over paid home care (Boris and Klein,
2007) and graduate student labor (Singh et al., 2006). In each case, market/nonmarket
distinctions mediate conflict over who is a “worker” (Krinsky, 2007a).
Such struggles are not open-ended, and so with time the “market” or “nonmarket” status
of a particular practice may come to be stabilized and institutionalized. I have argued elsewhere
that labor and employment law itself participates in that process (Zatz, 2008). Subjection to a
particular form of regulation (or not) differentiates certain practices (like prison labor and
ordinary employment) and groups others together (like various forms of employment) in ways
that further affect the “formal and informal rules” by which they operate. For instance, the
regular payment of cash wages and the separation of employer-employee from supplier-
consumer relationships (versus payment in credit at the company store) is itself partly a
requirement imposed by employment law (Steinfeld, 2001; Zatz, 2008); yet the form and basis of
payment also plays an important role in differentiating employment, which is subject to such
laws, from “nonmarket” work and “independent contracting,” which are not.
Of course, law‟s influence here likely will be mediated and transformed by other intra-
and inter-organizational processes, as the burgeoning literature on corporate antidiscrimination
and antiharassment policies has shown (Dobbin and Kelly, 2007; Edelman et al., 1999). It would
be fascinating to know how organizations that employ multiple forms of labor—conventional
employees, inmates on work-release, inmates within prison facilities, student workers,
volunteers, and so on—go about deciding how to classify them for a variety of different
purposes, how they structure working arrangements that differentiate among workers or create
34
similarity, and so on. Analogous studies have begun with regard to differentiation between
“core” employees and those hired through temp agencies or employed by subcontractors (Smith,
1998; Walsh and Deery, 2006), but these address only distinctions among what all would regard
as varieties of market work.
A mainstay of sociological analysis of markets, including labor markets, is that they are
produced and institutionalized over time as “a social and political project that begins without
stable relationships” (Fligstein and Dauter, 2007; Jacoby, 1995; see also Streeck, 2005).
Employment law disputes over prison labor suggest that there also may be much to learn from
studying the social and political project and process of institutionalizing labor practices as
market or nonmarket work.
References
Abbott, A., (2005), "Sociology of Work and Occupations", in Smelser, N. J. & Swedberg, R.
(Eds.), The Handbook of Economic Sociology, 2nd ed. Princeton, Princeton University
Press, pp. 307-30.
Abraham, K. G. & Mackie, C., eds., (2005), Beyond the Market, National Research Council
Panel to Study the Design of Nonmarket Accounts, Washington D.C., National Academy
of Sciences.
Albiston, C., (2007), "Institutional Perspectives on Law, Work, and Family", Annual Review of
Law and Social Science, Vol. 3, No. 1, pp. 397-426.
Appadurai, A., (1986), "Introduction: Commodities and the Politics of Value", in Appadurai, A.
(Ed.), The Social Life of Things: Commodities in Cultural Perspective, Cambridge,
Cambridge University Press, pp. 3-63.
35
Arnoldy, B., (2008), "Battling California Fires, Inmates Find a Chance at Better Role", Christian
Science Monitor.
Atleson, J. B., (1983), Values and Assumptions in American Labor Law, Boston, University of
Massachusetts Press.
Bandelj, N., (2008), From Communists to Foreign Capitalists, Princeton, Princeton University
Press.
Becker, G., (1976), The Economic Approach to Human Behavior, Chicago, University of
Chicago Press.
Benería, L., (1988), "Conceptualizing the Labour Force", in Pahl, R. E. (Ed.), On Work:
Historical, Comparative, and Theoretical Approaches, New York, Basil Blackwell, pp.
372-91.
Boris, E., (1994), Home to Work, New York, Cambridge University Press.
Boris, E. & Klein, J., (2007), "„We Were the Invisible Workforce‟ ", in Cobble, D. S. (Ed.), The
Sex of Class, Ithaca, Cornell University Press, pp. 177-93.
Boydston, J., (1990), Home and Work, New York, Oxford University Press.
Cameron, J. & Gibson-Graham, J. K., (2003), "Feminising the Economy ", Gender, Place, and
Culture, Vol. 10, No. 2, pp. 145-57.
Cancian, F., (1966), "Maximization as Norm, Strategy, and Theory", American Anthropologist,
Vol. 68, No. 2, pp. 465-70.
Carlson, P. M., (2004), "Something to Lose ", in Krienert, J. L. & Fleisher, M. S. (Eds.), Crime
and Employment, Walnut Creek, CA, AltaMira Press, pp. 61-74.
Di Leonardo, M., (1987), "The Female World of Cards and Holidays ", Signs, Vol. 12 No. 3, pp.
440-53.
36
Dobbin, F. & Dowd, T. J., (2000), "The Market That Antitrust Built", American Sociological
Review, Vol. 65, No. 5, pp. 631-57.
Dobbin, F. & Kelly, E., (2007), "How to Stop Harassment ", American Journal of Sociology,
Vol. 112, No. 4, pp. 1203-43.
Edelman, L. B. & Stryker, R., (2005), "A Sociological Approach to Law and the Economy", in
Smelser, N. J. & Swedberg, R. (Eds.), The Handbook of Economic Sociology, 2nd ed.,
pp. 527-51.
Edelman, L. B., Uggen, C. & Erlanger, H. S., (1999), "The Endogeneity of Legal Regulation ",
American Journal of Sociology, Vol. 105, No. 2, pp. 406-54.
Ertman, M. M. & Williams, J. C., (2005), "Preface: Freedom, Equality, and the Many Futures of
Commodification", in Ertman, M. M. & Williams, J. C. (Eds.), Rethinking
Commodification, New York, NYU Press, pp. 1-7.
Ferguson, A., (1989), Blood at the Root, London, Pandora Press.
Fineman, M. A., (2004), The Autonomy Myth, New York, New Press.
Flanagan, T. J., (1993), "A Full Employment Policy for Prisons in the United States ", Journal of
Criminal Justice, Vol. 21, No. 2, pp. 117-30.
Flavin, J., (2005), "Work-Release Programs", in Bosworth, M. (Ed.), Encyclopedia of Prisons &
Correctional Facilities, Thousand Oaks, Ca, Sage Publications, pp. 1055-57.
Fligstein, N. & Dauter, L., (2007), "The Sociology of Markets", Annual Review of Sociology,
Vol. 33, pp. 105-28.
Folbre, N., (1982), "Exploitation Comes Home", Cambridge Journal of Economics, Vol. 6, No.
4, pp. 317-29.
---, (2001), The Invisible Heart, New York, New Press.
37
Foucault, M., (1995), Discipline and Punish, New York, Vintage.
Goldberg, C. A., (2007), Citizens and Paupers, Chicago, University of Chicago Press.
Granovetter, M., (1995), Getting a Job, Chicago, University of Chicago.
---, (2001 [1985]), "Economic Action and Social Structure", reprinted in Granovetter, M. &
Swedberg, R. (Eds.), The Sociology of Economic Life, 2nd ed. Boulder, Westview Press,
pp. 51-74.
Gray, T. & Meyer, J. A. F., (1996), "Expanding Prison Industries through Privatization", in
Mays, G. L. & Gray, T. (Eds.), Privatization and the Provision of Correctional Services:
Context and Consequences, Greenbelt, MD, Academy of Criminal Justice Sciences, p.
125.
Hasday, J. E., (2005), "Intimacy and Economic Exchange", Harvard Law Review, Vol. 119, No.
2, pp. 491-530.
Jacoby, S. M., (1995), Employing Bureaucracy, Mahwah, NJ, Lawrence Erlbaum.
Kessler-Harris, A., (1990), A Woman's Wage, Lexington, University Press of Kentucky.
Kittay, E. F., (1999), Love's Labor, New York, Routledge.
Krinsky, J., (2007a), "Constructing Workers", Qualitative Sociology, Vol. 30, No. 4, pp. 343-60.
---, (2007b), Free Labor, Chicago, University of Chicago Press.
Krippner, G. R. & Alvarez, A. S., (2007), "Embeddedness and the Intellectual Projects of
Economic Sociology", Annual Review of Sociology, Vol. 33, pp. 219-40.
Lie, J., (1997), "Sociology of Markets", Annual Review of Sociology, Vol. 23, pp. 341-60.
Melossi, D. & Pavarini, M., (1981), The Prison and the Factory, New York, Barnes & Noble
Books.
38
Miller, N. & Crieser, R. C., (1986), "Exemplary Components of Prison Industry Operations", in
American Correctional Association (Ed.), A Study of Prison Industry, College Park, MD,
American Correctional Association, pp. 30-50.
Olsen, F. E., (1983), "The Family and the Market", Harvard Law Review, Vol. 96, No. 7, pp.
1497-578.
Piore, M. J. & Safford, S., (2006), "Changing Regimes of Workplace Governance, Shifting Axes
of Social Mobilization, and the Challenge to Industrial Relations Theory", Industrial
Relations, Vol. 45, No. 3, pp. 299-325.
Polanyi, K., (2001 [1957]), "The Economy as Instituted Process", reprinted in Granovetter, M. &
Swedberg, R. (Eds.), The Sociology of Economic Life, 2nd ed. Boulder, Westview Press,
pp. 31-49.
Portes, A. & Haller, W., (2005), "The Informal Economy", in Smelser, N. J. & Swedberg, R.
(Eds.), The Handbook of Economic Sociology, 2nd ed. Princeton, Princeton University
Press, pp. 403-25.
Radin, M. J., (1996), Contested Commodities, Cambridge, MA, Harvard University Press.
Rapp, R., (1982), "Family and Class in Contemporary America", in Thorne, B. & Yalom, M.
(Eds.), Rethinking the Family, New York, Longman, pp. 168-87.
Reynolds, M. & Rostad, K., (2005), "Prison Labor Teaches Valuable Job Skills", in Haley, J.
(Ed.), Prisons, Greenhaven Press, pp. 108.
Rupp, L., (2006), "Skilled Inmates to Stay at Penal Farm", Clarion-Ledger, Jackson, Miss.
Schultz, V., (2000), "Life's Work", Columbia Law Review, Vol. 100, No. 7, pp. 1881-964.
Silbaugh, K., (1996), "Turning Labor into Love", Northwestern University Law Review, Vol. 91,
No. 1, pp. 1-86.
39
Singh, P., Zinni, D. M. & Maclennan, A. F., (2006), "Graduate Student Unions in the United
States", Journal of Labor Research, Vol. 27, No. 1, pp. 55-73.
Smelser, N. J. & Swedberg, R., (2005), "Introducing Economic Sociology", in Smelser, N. J. &
Swedberg, R. (Eds.), The Handbook of Economic Sociology, 2nd ed. Princeton, Princeton
University Press, pp. 3-25.
Smith, V., (1998), "The Fractured World of the Temporary Worker", Social Problems, Vol. 45,
No. 4, pp. 411-30.
Stanley, A. D., (1998), From Bondage to Contract, New York, Cambridge University Press.
Steinfeld, R. J., (2001), Coercion, Contract, and Free Labor in the Nineteenth Century, New
York, Cambridge University Press.
Stone, K. V. W., (2004), From Widgets to Digits, New York, Cambridge University Press.
Streeck, W., (2005), "The Sociology of Labor Markets and Trade Unions", in Smelser, N. J. &
Swedberg, R. (Eds.), The Handbook of Economic Sociology, 2nd ed. Princeton, Princeton
University Press, pp. 254-83.
Swedberg, R., (2005), "Markets in Society", in Smelser, N. J. & Swedberg, R. (Eds.), The
Handbook of Economic Sociology, 2nd ed. Princeton, Princeton University Press, pp.
233-53.
Swedberg, R. & Granovetter, M., (2001), "Introduction to the Second Edition", in Granovetter,
M. & Swedberg, R. (Eds.), The Sociology of Economic Life, 2nd ed. Boulder, Westview
Press, pp. 1-28.
Sykes, G. M., (1971), The Society of Captives, Princeton, Princeton University Press.
Tilly, C. & Tilly, C., (1998), Work under Capitalism, Boulder, CO, Westview Press.
40
Tomlins, C., (1995), "Subordination, Authority, Law", International Labor & Working Class
History, Vol. 47, pp. 56-90.
Turner, S. & Petersilia, J., (1996), "Work Release: Recidivism and Corrections Costs in
Washington State", Research in Brief, Washington, D.C., National Institute of Justice,
U.S. Department of Justice.
Unicor Federal Prison Industries, I., (2008), "Unicor Online: Services",
http://www.unicor.gov/services/, October 17, 2008.
Walsh, J. & Deery, S., (2006), "Refashioning Organizational Boundaries", Journal of
Management Studies, Vol. 43, No. 3, pp. 557-82.
Walzer, M., (1983), Spheres of Justice, Basic Books.
Weiss, R. P., (2001), "„Repatriating‟ Low-Wage Work", Criminology, Vol. 39, No. 2, pp. 253-
92.
Wignall, B., (2005), "Prison Labor Facilitates Inmate Rehabilitation", in Haley, J. (Ed.), Prisons,
Greenhaven Press, pp. 114.
Williams, J., (2000), Unbending Gender, New York, Oxford University Press.
---, (2001), "From Difference to Dominance to Domesticity", Chicago-Kent Law Review, Vol.
76, No. 3, pp. 1441-93.
Williams, J. C. & Zelizer, V. A., (2005), "To Commodify or Not to Commodify: That Is Not the
Question", in Ertman, M. M. & Williams, J. C. (Eds.), Rethinking Commodification, New
York, NYU Press, pp. 362-82.
Zatz, N. D., (2006), "What Welfare Requires from Work", U.C.L.A. Law Review, Vol. 54, No. 2,
pp. 373-464.
41
---, (2008), "Working at the Boundaries of Markets", Vanderbilt Law Review, Vol. 61, No. 3, pp.
857-958.
Zelizer, V. A., (1996), "Payments and Social Ties", Sociological Forum, Vol. 11, No. 3, pp. 481-
95.
---, (2005a), "Circuits within Capitalism", in Nee, V. & Swedberg, R. (Eds.), The Economic
Sociology of Capitalism, Princeton, Princeton University Press, pp. 289-321.
---, (2005b), "Culture and Consumption", in Smelser, N. J. & Swedberg, R. (Eds.), The
Handbook of Economic Sociology, 2nd ed. Princeton, Princeton University Press, pp.
331-54.
---, (2005c), The Purchase of Intimacy, Princeton, Princeton University Press.
Cases and Other Legal Authorities
Alexander v. Sara, Inc., 721 F.2d 149 (5th Cir. 1983)
Allen v. Cuomo, 100 F.3d 253 (2d Cir. 1996)
Baker v. McNeil Island Corrections Ctr., 859 F.2d 124 (9th Cir. 1988)
Barnett v. Young Men’s Christian Ass’n, 1999 WL 110547, 175 F.3d 1023 (8th Cir. 1999)
Bennett v. Frank, 395 F.3d 409 (7th Cir. 2005)
Burleson v. State of California, 83 F.3d 311 (9
th
Cir. 1996)
Carter v. Dutchess Community College, 735 F.2d 8 (2d Cir. 1984)
Danneskjold v. Hausrath, 82 F.3d 37 (2d Cir. 1996)
George v. Badger State Indus., 827 F. Supp. 584 (W.D. Wis. 1993)
Gilbreath v. Cutter Biological, Inc., 931 F.2d 1320 (9th Cir. 1991)
Hale v. Arizona, 967 F.2d 1356 (9th Cir. 1992), vacated en banc, 993 F.2d 1387 (9th Cir. 1993)
42
Harker v. State Use Indus., 990 F.2d 131 (4
th
Cir. 1993)
Harris v. Yeager, 291 F. Supp. 1015 (D. N.J. 1968)
Henthorn v. Dep’t of Navy, 29 F.3d 682 (D.C. Cir. 1994)
Hudgins v. Hart, 323 F. Supp. 898 (E.D. La. 1971)
Huntley v. Gunn Furniture Co., 79 F. Supp. 110 (W.D. Mich. 1948)
Jovanovich v. Angelone, 1995 WL 378678, 59 F.3d 175 (9th Cir. 1995)
Lavigne v. Sara, Inc., 424 So.2d 273 (La. App. 1982)
McCaslin v. Cornhusker Indus., 952 F. Supp. 652 (D. Neb. 1996)
McGinnis v. Stevens, 543 P.2d 1221 (Ala. 1975)
Miller v. Dukakis, 961 F.2d 7 (1
st
Cir. 1992)
Morgan v. MacDonald, 41 F.3d 1291 (9th Cir. 1994)
New York University, 332 N.L.R.B. 1205 (2000)
Nicastro v. Reno, 84 F.3d 1446 (D.C. Cir. 1996)
Reimonenq v. Foti, 72 F.3d 472 (5th Cir. 1996)
Sims v. Parke Davis & Co., 334 F. Supp. 774 (E.D. Mich. 1971)
Souder v. Brennan, 367 F. Supp. 808, 813 (D.D.C. 1973)
Speedrack Products Group v. N.L.R.B., 114 F.3d 1276 (D.C. Cir. 1997)
U.S. Department of Labor, Prison Industry Enhancement Certification Program Guideline, 64
Fed. Reg. 17,000 (Apr. 7, 1999)
United States v. City of New York, 359 F.3d 83 (2d Cir. 2004)
Vanskike v. Peters, 974 F.2d 806 (7th Cir. 1992)
Villarreal v. Woodman, 113 F.3d 202 (11th Cir. 1997)
Walker v. City of Elba, 874 F. Supp. 361 (M.D. Ala. 1994)
43
Watson v. Graves, 909 F.2d 1549 (5th Cir. 1990)
Williams v. Meese, 926 F.2d 994 (10th Cir. 1991)