With the introduction of the 13th Amendment to the U.S. Constitution, legal
slavery in the U.S. has been prohibited. Only recently a neo-abolition era has
emerged which calls for de facto slavery to be also prohibited (Allain & Bales,
2012). De-Facto slavery describes slavery-like conditions which do not coincide
with the legal definition of slavery, but are still in conflict with many people’s
One example are the labour conditions in Bangladesh’s garment factories (Abrams
& Sattar, 2017). The collapse of the Rana Plaza building in 2013, resulting in
1,100 deaths, sparked a worldwide call for better conditions for the workers.
A second example, which will be the subject of this paper, is prison labour in
the United States. As the Economist reported, prison labour is supposed to teach
prisoners skills they can use to find employment upon release and allow them
to save money, but they frequently work jobs which seem unlikely to increase
their chances at finding employment and for a very low wage (“Prison labour is a
billion-dollar industry, with uncertain returns for inmates”, 2017). Additionally,
the supposedly high profit margins pose a moral hazard for officials, as the state
benefits from incarceration of citizens.
The Economist argues prison labour to be exploitative and with little benefits
to the individual. But according to the framework of moral intuitions introduced
by Haidt and Graham (2007), these arguments only relate to the political liberal
foundations of morals. Political conservatives have, according to Haidt and Gra-
ham, additional intuitions regarding society and it’s structure they base their moral
This literature review presents aspects of prison labour and will discuss to
what extent prison labour can be morally legitimate, according to an extended
set of moral foundations as presented by Haidt and Graham (2007). To do so,
this paper first introduces the five foundations of moral intuition as worked out
by Haidt and Graham, then presents different aspects of prison labour and finally
discusses how the presented aspects relate to the introduced foundations.
2 Moral Codes, Intuitions and Values
Richard A. Shweder (1990) classified the appeals moral arguments are based
on into three codes of moral discourse. The first code concerns the autonomy of
the individual to make free choices. It includes appeals to harm, rights and justice.
Social justice researchers used to restrict their understanding of natural moral law
to this first code, while they conceived arguments based on duty, hierarchy and
interdependency (code 2) and natural order, sacred order, tradition, sin, and per-
sonal sanctity (code 3) as arbitrary social conventions (Gabennesch, 1990). This
restriction can be explained by their focus on moral reasoning which, as Haidt and
Bjorklund (2008) explain, usually occurs only after the decision has already been
made and does not necessarily align with the actual reasons.
Haidt and Joseph (2004; as cited by Haidt & Graham, 2007) confirmed Shweder’s
theory and concluded that cultures apply varying weigths to five intuitions and au-
tomatic emotional reactions common across cultures. Following is a brief overview
of these five psychological systems based on Haidt and Graham (2007). Further
information can be found in Haidt and Joseph (2004); Haidt and Graham (2007);
Haidt and Bjorklund (2008).
1. harm/care: People appear to dislike seeing other people suffer. They are
sensitive to cruelty and harm as well as the relieve of those. It is not speci-
fied wether this is limited to physical harm and thus this paper understands
this to include psychological factors.
2. fairness/reciprocity: Virtues related to fairness and justice appear in all cul-
tures examined. They developed from emotions motivating reciprocal altru-
ism. A few cultures (Not all of the examined) established individual rights
and equality as a consequence.
3. ingroup/loyalty: A tendency to show strong emotions on the distinction
between one’s own group and outsiders/other groups. This is expressed
through virtues such as loyalty, patriotism, and heroism.
4. authority/respect: Many cultures show virtues related to the structure of
society. They include those for distinguishing between good and bad lead-
ership (magnanimity, fatherliness, wisdom versus despotic, exploitative, or
inept) as well as those for subordination (respect, duty, obedience).
5. purity/sanctity: The human species developed disgust as a form of protec-
tion against diseases and parasites. This has been expanded onto social
ideas, especially visible in religion. Thus today’s purity impacts not only
physical contamination but also ideologies.
Haidt and Graham made the proposition that “the morality of political liber-
als is built on the harm and fairness foundations, while the morality of political
conservatives is built upon all five foundations.” (Haidt & Graham, 2007, p. 107).
A similar claim had been made earlier by Shweder (1990, p. 2063), who said
that only “secularized liberal individualists” reduced moral obligations to code 1
moral discourse. Graham, Haidt, and Nosek (2009) later gave empirical evidence
supporting the proposition made in 2007.
3 Moral Aspects of Prison Labour
3.1 Reduced Recidivism
Proponents of prison labour have long emphasised the reduced idleness and
the skills inmates could learn (West, 1913). This supposedly leads to better be-
haviour inside the prison because it helps prisoners to cope with their situation
and a better chance to find employment upon release, which then leads to lower
recidivism. If these claims are true, it is good for all parties involved. Critics
however raise the concern that existing programs do not teach the prisoners the
skills they would need in order to find employment in the free market (Citation).
Instead they argue that the jobs given to prisoners don’t require useful skills and
have already “left the country”.
Whether this practise reduces recidivism is also questionable. While Saylor
and Gaes (1997) have found a 20% increase in the time ex-convicts spend outside
if they have been in work and job skills programs, Bair (2007) argues that they
had methodological flaws. First, Saylor and Gaes’s were looking at the time ex-
prisoners spend outside before recommitting. This is different from a lower rate
of recidivism other studies in this field look at, but not necessarily an inferior
measure. Second, the test group has not been selected randomly. Only prisoners
who were actually offered a job were in the group, while the control group was
chosen among prisoners who were not offered a job. Thus it is not clear if the
improvements in recidivism came through the job or through selection of a test
group that would have showed lower recidivism even without an inside job. Other
studies cited by Bair have not found a considerable difference in the recidivism
3.2 Reduced Costs & Impact on the Free Market
A second point frequently made is the money prison labour can make for the
prisoner as savings/support for his family and to society to repay at least part of
what the prisoner costs. Reynolds (1996) estimated that employing one in four
prisoners would reduce the costs for U.S. taxpayers by a little less than 10%.
There have been claims that prison labour has a negative impact on the com-
petitive market as prison industries are subsidised (West, 1913). Outside firms find
themselves in an inferior position since they have expenses the state as a produced
has not (Define Prison industries). This is partly due to U.S. prison labour not
being covered by the Fair Labor Standards Act (Derrick, Scott, & Hutson, 2004),
thus prisoners can be paid below minimum wages and are not allowed to organise
in order to bargain. This may drive firms and low-skilled employees out of the
market as they become unprofitable and it is usually their labour that is moved
It is questionable whether prison labour is actually moving jobs inside prisons.
Oregon state official Kevin Mannix said “We propose that (Nike) take a look at
their transportation costs and their labor costs” (Erlich, 1995; as cited by Browne,
2007) when arguing for Nike to bring jobs from Indonesia to U.S. Prisons. This
suggests that prison labour is not taking U.S. jobs away, but at the same time
speaks for bad conditions inside prisons (Those conditions will be discussed in a
Derrick et al. (2004) estimated the impact 25% prisoner employment would
have on wages and unemployment in the U.S. to be negligible. While West men-
tioned that “It is not the volume or the percentage that plays havoc, but the nature
of the competition” (1913, p. 48), he also offers a solution: Prison authorities
should help the prison help the society and thus stay out of markets where com-
petition is tense. They should also have regularly advertised contracts, awarded to
the highest bidder.
Citation (year) suggested that prison labour could even lead to an economic
boost in complementing industries as prison production requires raw materials to
be produced and logistics for transports. Kirchhoff (2010) however pointed to
broader research that suggests this to be untrue. Usually there are fewer benefits
than expected and has in some cases slowed growth down.
Just as prison labour is supposed to have beneficial effects on the costs of
incarceration, inmates benefit of their wage which they can save or use to support
their family outside of the prison (Reynolds, 1996). Critics argue that the wage
prisoners receive is too low. They are not protected by minimum wage laws and
are not allowed to collectively bargain (Browne, 2007). In private U.S. prisons
workers receive between $0.17 and $0.50 per hour, while in federal prisons they
can earn $1.25 (Pelaez, 2014).
3.3 Weak Agency, Vulnerability and Extreme Harms
In 2010 Debra Satz laid out a framework for assessing whether markets are
noxious. It includes four dimensions in which a market can be morally unde-
sirable: Weak Agency, Vulnerability, Extreme Harms for Individual and Extreme
Harms for Society. According to Satz, high scores in one of them, or several to-
gether can make a market noxious. Prison labour may score along three of the
It may be a case of Weak Agency as prisoners have little decision making
power themselves. They only choose whether to apply for a working program or
not, which programs are available, their wage etc. are someone else’s decision.
This is paired with the Vulnerability dimension as working in prison is their only
option to support their family outside or save for their release. Thus their deci-
sion to work may be out of desperation. According to Satz an exchange made in
desperation is not one between equals. A third point may be physical and mental
harm, as U.S. prisoners are not covered by the Fair Labor Standards Act (Derrick
et al., 2004). The eighth amendment to the U.S. constitution prohibits “cruel and
unusual punishment” but prisoners have struggled to get their cases approved as a
violations of the constitution (Doughterty, 2007).
When assessing prison labour in the United States with the restricted, liberal,
set of moral foundations of harm/care and fairness/reciprocity the arguments fo-
cus on the exploitation of prisoners. They lack choice, are in a vulnerable, des-
perate position and encounter harm. A wage and improved recidivism rates could
pose a counter argument, but the wages are not covered by minimum wage laws,
prisoners are not allowed to collectively bargain and the effect on recidivism rates
When including the traditional, more conservative moral foundations, the weight
of reduced recidivism rates increases and decreased costs for society become im-
portant. The benefit to the outside population may outweigh the problem of ex-
ploitation in the ingroup/loyalty domain. Additionally the fact that prisoners have
violated the law may further legitimise their exploitation (authority/respect). But
this requires an advantage to the society. Conservative moral foundations include
the liberal ones, thus recidivism rates and financial burden on society have to be
significantly reduced in order to legitimise an individual in a vulnerable, harmful
position. The reviewed literature is currently not allowing a definitive view on
whether this is the case.
It is thus argued that the current state of prison labour in the U.S. can not be
legitimised by conservative moral foundations, because of known issues regarding
exploitation as well as little benefits to inmates and uncertain returns for society.
But this is not an issue with prison labour itself, it is an issue with it’s current
situation in the United States.
To evaluate the extent to which prison labour itself is legitimate according to
Haidt and Graham’s (2007) five moral foundations there have to be more decisive
studies into whether prison labour reduces recidivism rates and financial costs
to society. Additionally, the literature into prison labour is concerning almost
only the United States. Thus, circumstances in other countries have not been
assessed in this paper. Not discussed in this paper are also externalities such as
racist sentencing. According to some critics, the 13th amendment prohibiting
slavery in the United States has a loophole because it states that “neither slavery
nor involuntary servitude is allowed, except as a punishment for a crime.” This
has supposedly been exploited by laws being especially harsh on offences mainly
associated with African Americans (See Jim Crow laws). This paper assumes fair
sentencing. A discussion about how generally allowing prison labor could amplify
racist jurisdiction up to further research.
What can still be said is that the legitimacy of the current state of prison labour
in the United States is controversial and dependent on currently unsubstantiated
belief. In conjunction with the proposition that political conservatives value all
of the five foundations (Haidt & Graham, 2007; Haidt & Bjorklund, 2008) it is
hypothesised that not only liberals but also conservatives think that the current
state of prison labour in the U.S. is morally wrong.
This paper set out to understand to what extent prison labour can be legit-
imised by conservative moral foundations. While it’s current situation in the
United States is an issue for political liberals, who tend to focus on harm and
fairness in their moral ideas, political conservatives may have a more differenti-
ated viewpoint, as they include social benefits in their moral argumentation. But
the literature about prison labour in the United States does not provide conclusive
evidence that social benefits exist. Currently, the situation is not acceptable to
neither liberals nor conservatives.
The case of prison labour in the United States is special. Critics argue it is
immoral because of bad working conditions and low wages, but these two are not
inherent to prison labour itself. As it has been shown, there can be benefits from
prison labour and downsides are not inherent, thus the issue lies within the United
States and not prison labour.