help homework help homework The American Precariat: U.S. Capitalism in Comparative Perspective Kathleen Thelen The address situates the rise of “gig” work

help homework help homework The American
Precariat: U.S.
Capitalism in
Kathleen Thelen

The address situates the rise of “gig” work in the context of
a much longer-term trend toward more precarious forms of
employment. It explores the forces that are driving these
developments and discusses the problems they pose at both
the individual level and the national level. By situating the
United States in a comparative perspective, it identifies the
structural factors that exacerbate the problem of precarity and
intensify its effects in the American political economy.

he theme of this year’s conference is “Democracy
and Its Discontents,” and the way I would like to
contribute to that conversation is by shining a light

on features of my own democracy that I believe do not
receive the attention they deserve from political scientists.
From my perspective, as a student of the comparative
political economy of the rich democracies, one of the
more regrettable casualties of the way we have drawn lines
around the subfields in political science is that this seems
to have sidelined the study of the American political
economy, by which I mean the analysis of American

This is a topic that, at least over the course of my
career, has mostly fallen between the cracks of com-
parative political economy on the one hand and
American politics on the other hand. Scholars like me
who study the political economies of the rich de-
mocracies tend to focus heavily on Europe. For
normative reasons, many of us have been drawn to
the study of capitalism in places like Sweden and
Norway, which have their problems but where we can
still point to the possibility of finding ways to reconcile
successful economic performance with relatively high
levels of economic equality.1

Americanists, for their part, have given us important
insights, especially into the behavioral foundations of
American politics, focusing on public opinion and voting.

UNITE HERE Boston Local 26, June 27, 2018

This photo is reprinted with permission from the UNITE HERE Local 26

President, Brian Lang.

This talk intersects with my work on several collaborative
projects, and I extend special thanks to my co-authors Sabeel
Rahman, Andreas Wiedemann, Bruno Palier, and Pepper
Culpepper. I owe a debt of gratitude as well to a group of
Americanists with whom I have collaborated or taught and
from whom I have learned a great deal about American
politics—Theda Skocpol, Devin Caughey, Jacob Hacker and
especially Alexander Hertel-Fernandez and Paul Pierson.
Many colleagues provided invaluable input into earlier
versions of this paper, including Kate Andrias, Laura Bucci,
Caroline de la Porte, Werner Eichhorst, Dan Galvin, Anton
Hemerijck, Alex Hertel-Fernandez, Martin Höpner, Chris
Howell, Christian Ibsen, Jeff Isaac, Alan Jacobs, Desmond
King, Margaret Levi, Bruno Palier, Paul Pierson, and
Brishen Rogers. I am also grateful to Elizabeth Dekeyser, who
provided expert research assistance, and to Lukas Wolters, who
helped to prepare the manuscript for submission.

Kathleen Thelen was the 114th president of the American
Political Science Association ( She is Ford
Professor of Political Science at MIT and is the author of
numerous books and articles on the political economy of labor
in the rich democracies, including How Institutions Evolve
(Cambridge University Press, 2004, winner of the Woodrow
Wilson Foundation Award) and Varieties of Liberalization
and the New Politics of Social Solidarity (Cambridge
University Press, 2014, winner of the Barrington Moore Book
Award). Her current research focuses on the American
political economy in comparative perspective.


© American Political Science Association 2019 March 2019 | Vol. 17/No. 1 5

Presidential Address
Downloaded from IP address:, on 22 Feb 2019 at 22:43:31, subject to the Cambridge Core terms of use, available at

However, it is rare in the extreme for Americanists to
compare the United States to other countries.2 More
importantly, many of the issues and actors that are central
to the study of the comparative political economy of the
other rich democracies—labor unions, finance, organized
business, wages, working time, skills, education and
training—do not figure at all in mainstream research on
American politics.

In the meantime, we do have some important studies
of high-end inequality in the United States, for example
the influence of the super-rich on public policy, the role
of moneyed interests in American politics, and the
political strategies of the organized business community.3

My emphasis today, however, is different. I am not going
to focus on high-end inequality, but instead on the
“normal,” taken-for-granted operation of markets, in
particular labor markets at the low end of the income
ladder. What I would like to do today is explore what we
can learn about American capitalism—and by extension,
about American democracy—by situating the United
States in a broader comparative context.

I begin with a trend that we are hearing a lot about
these days, namely the emergence of the so-called “gig”
economy. This term comes out of the entertainment
industry, where musicians rarely hold steady jobs, and
instead are paid for playing individual gigs on a one-off,
pickup basis. In the meantime, the term gig economy now
refers to a phenomenon where instead of being employed
in the usual way (on some sort of normal employment
contract), workers get one-off payments to perform in-
dividual tasks—like deliver your food, or drive you
around, or walk your dog.

Even if you are not familiar with this particular term,
you probably will have heard of – and most likely used—
some of the services on offer in this new gig economy. No
doubt, many of you have an Uber or Lyft app on your
phone, and I imagine some of you have stayed at an Airbnb
on vacation with your family. I know for sure that many of
my colleagues and graduate students run their surveys
through Amazon’s Mechanical Turk (AMT)—an online
marketplace for digital piecework where “Turkers” can go
online, find an interesting task (like filling out your
survey), and earn money.

If you have used any of these services, you know that
this gig economy has been great for consumers. Users
are getting a service that is tailored to exactly what they
need when they need it—and it is typically very in-
expensive because workers are only being paid for the task
they perform. Gig work can also be a great thing for
workers—or at least for particular kinds of workers. For
example, if you have highly marketable skills (e.g., in
software engineering or web design), this kind of free-
lancing can be very liberating; suddenly you can organize
your time however you like. It can also be great for
parents, especially women, who often bear primary

responsibility for children and for whom it may be
convenient to be able to work flexibly, even from home.4

So there are possibilities here for autonomy and maybe
even personal empowerment.
However, it turns out that for everyone who benefits

from the flexibility that this new type of work offers,
there are also those who are pressed into such employ-
ment for lack of alternatives or by the need to supplement
insufficient income from other jobs. A recent study by
the Pew Research Center revealed that over half (56%) of
Americans who earned money on one of these digital
labor platforms in the previous year indicated that this
income was either “essential” or “important” to making
ends meet.5 It turns out that for people who depend on
these platforms in these ways, it is not so easy to make
money. For example, at $4.65 per hour, the median hourly
earnings of American Turkers are well below the federal
minimum wage.6 The problem is not just low wages; it is
also that many of these platform companies are able to
exploit ambiguities in the law to avoid obligations
associated with other forms of employment. Thus, most
gig workers are considered to be independent contractors,
which means that they do not have any of the rights and
benefits attached to a regular job—no minimum wage
regulations, no overtime rules, no unemployment in-
surance, no workers’ compensation if they get hurt on
the job—on all these fronts, they are on their own.7

In this talk, I want to situate the rise of this type of gig
work in the context of a longer-term and much broader
phenomenon, namely the rise of what the French
sociologist Robert Castel has called the new precariat.8 I
define the precariat, as Castel does, as flowing from
a growing fragmentation of work and the breakdown of
the standard employment relationship, by which I mean
stable, long-term employment contracts that include
benefits and that feature regular, predictable hours. As
Castel notes, the problem of precarity is not just about
a “precarious periphery,” but also about a broader “de-
stabilization of the stable.”9 This is a trend that is common
across all the advanced industrial countries and one that
has resulted in what Jacob Hacker has called the “privat-
ization of risk” as more and more economic risks are
shifted from employers and governments “onto the
shoulders of ordinary Americans” and their families.10

In my remarks, I will focus primarily on the impact of
these trends on low-skill, low-income groups, because
they are the ones who experience precarity most in-
tensely. However, it is important to be clear that precarity
is not the same as poverty, and it does not only hit low-
skill workers. It also affects highly skilled groups, in-
cluding professionals such as airline pilots and lawyers
(especially contract law), who also are employed in-
creasingly on flexible contracts.11 We can underscore this
point by observing trends in our own discipline. Figure 1a
shows that the academic labor force has grown massively

6 Perspectives on Politics

Presidential Address | The American Precariat
Downloaded from IP address:, on 22 Feb 2019 at 22:43:31, subject to the Cambridge Core terms of use, available at

since the early 1970s. However, what we also see is that
non-tenure line positions (e.g., adjuncts or lecturers) have
grown much faster than tenure line positions, so that the
profile of our profession is now predominantly highly
skilled but contingent in the sense of employed on an
“as-needed” basis (see figure 1b). Most of our colleagues on
contingent contracts are ineligible for benefits; they are
hired on a course-by-course basis, and often notified of their
schedules only shortly before the semester begins.12

In short, this practice of substitution—swapping more
flexible, more contingent forms of employment in place of
previously more secure jobs—affects many different kinds
of employees in many different sectors and at different skill
levels. Thus, precarity, as I am using the term today,
extends well beyond the gig workers to whom I referred at
the outset.

In exploring these issues, I will be pushing back against
the self-congratulatory tone often struck when our
current administration boasts about record low levels of
unemployment and robust job creation. I also hope to
shed some light on one of the more puzzling features of
the American labor market, namely continued real wage
stagnation despite eight years now of steadily declining
unemployment.13 The question for me is not how many
jobs did the U.S. economy generate last month or last year,
but what kinds of jobs are these and how many of these
jobs do people at the low end have to hold in order to make
ends meet?

The rest of my talk proceeds in five steps. First, I want
to look into the forces that are driving the growth of
various sorts of atypical work arrangements. Second,
I will discuss why these developments pose a problem,

Figure 1
Academic labor force, United States

Source: AAUP and the Integrated Post Secondary Education Data System (excluding graduate students)

March 2019 | Vol. 17/No. 1 7
Downloaded from IP address:, on 22 Feb 2019 at 22:43:31, subject to the Cambridge Core terms of use, available at

both at the individual level and at the national level.
Third, I will situate the United States in a comparative
perspective and show that while these trends are pervasive
across all the rich democracies, the problem of precarity is
especially acute in this country. Fourth, I will zero in on
one of the features of the American political economy
that plays a central role in exacerbating precarity and
intensifying its effects in the United States. Fifth and
finally, I circle back to address the way in which the study
of American capitalism relates to the theme of this
conference on democracy and why we as political scientists
should care about it.

The Growth of Atypical Employment in
the Rich Democracies
Gig work of the sort to which I referred earlier is just one
of many forms of “atypical” employment that have been
on the rise across the rich democracies. Atypical work
travels under different names in different national and
sectoral contexts—“temp” work, fixed-term employment,
mini-jobs, bogus self-employment, on-demand work,
“zero-hours” contracts, and many more. The general
point, however, is that across the rich democracies, the
new reality for a growing number of people is employment
that is less steady, less secure, and less well embedded in
traditional social protections.

“Non-standard” employment is of course nothing
new.14 Historically speaking, the standard employment
relationship is the exception. In the early industrial period,
all work was “casual,” which is to say, insecure and
irregular. And of course, in many parts of the developing
world, this is still the case.15 Early labor market reformers
in Europe, including Fabian socialists such as Sidney and
Beatrice Webb, fought against casual labor, viewing it as
a source of poverty and social dependence.16 However,
some conservatives were also concerned—either because
they saw this underemployed underclass as politically
dangerous or because they considered this a highly in-
efficient way to organize the labor market.17

Thanks not least to the efforts of the emerging
organized labor movement, the trend, at least in the rich
democracies, was toward “decasualization” and later the
emergence of a variety of social protections, so that by the
mid-twentieth century, the norm of stable full-time work
attached to a package of benefits had become entrenched
throughout the advanced capitalist world. The model
midcentury industrial firm embodied what the philosopher
Elizabeth Anderson described as a “nexus of reciprocal
relationships.”18 It was a model that concentrated power in
the hands of managers, but managers whose goal was stable
long-term growth. It was underwritten by “patient” capital
(in Europe often provided by banks, in the United States by
passive and dispersed shareholders) that allowed for the
cultivation of long-term relationships and stable gains for all
of the company’s stakeholders, including labor.

However, as especially the sociologist Gerry Davis has
pointed out, many of today’s largest and most dynamic
companies do not conform to this model, but instead look
more like a nexus of short-term contacts.19 Firms have
become increasingly fragmented (or as David Weil puts it,
“fissured”20), as companies construct extensive networks
of subcontracting and franchising that allow them to
outsource all kinds of operations, using cheaper third
party suppliers of goods and services to cut costs, partic-
ularly labor costs. This means that instead of being directly
employed by the big core company (say, GM), workers are
more likely than before to be working either on a free-
lancing basis (as “independent contractors”) or for some
smaller entity in a network of subcontractors or branch
operations that are formally independent but that are all
part of a hierarchy in the service of the core company.21

The Wall Street Journal recently reported that among the
top 20 employers worldwide in 2017, five are outsourcing
(“workforce solution”) companies that do not produce
anything at all but simply supply workers to other
companies on an on-demand basis.22

This kind of fissurization has been facilitated by
technological change. There are classic Coasian efficiency
gains to be made. As technology has reduced the trans-
action costs of monitoring partner firms and employees,
the incentives to subcontract have gone up. However,
while they are enabled by technology, these strategies are
motivated by employer efforts to reduce labor costs by
replacing full time workers with more flexible temporary
or short-term workers, thus externalizing the costs of
adjusting to ups and downs in demand.23

These practices are not unknown in manufacturing
where, for example, automobile companies have long
engaged in subcontracting, and are increasingly using
temps as well. However, these trends have been much
accelerated by the long-term shift in employment out of
manufacturing into services.24 Atypical work has flourished
especially in labor-intensive services such as the retail and
hospitality industries, where both fissurization through
franchising and the use of temporary agency workers are
widespread.25 Moreover, even where core companies em-
ploy workers directly, e.g., major retailers such as Target or
Walmart, new technologies allow employers to adjust their
staffing levels to match consumer demand in real time.
What in management circles is called “lean staffing” often
translates into “just-in-time” scheduling, where workers
find out their schedules hours before they are expected to
show up. Fissurization and lean staffing—singly or, often,
in combination—generate huge labor savings for employ-
ers, and at the same time, huge uncertainty for workers.
In this context, digital platforms such as Uber,

Upwork, and AMT have simply elevated such strategies
to new levels by foregoing employment contracts alto-
gether and using “independent contractors” instead,
where the core firm bears no responsibility for wages or

8 Perspectives on Politics

Presidential Address | The American Precariat
Downloaded from IP address:, on 22 Feb 2019 at 22:43:31, subject to the Cambridge Core terms of use, available at

benefits or hours or anything else.26 This is why platform-
based companies are at pains to call their employees
anything other than workers: they are taskers, Turkers,
providers, “partners,” “heroes,” or “ninjas.”27 And it is
why what these people are doing is referred to as everything
but jobs: gigs, tasks, micro tasks, rides.28

The core logic and the employment implications of
these new enterprises were best summarized by the CEO
of a major crowdsourcing platform in an unscripted
moment of candor: “Before the Internet, it would be
really difficult to find someone, sit them down for ten
minutes and get them to work for you, and then fire them
after those ten minutes. But with technology, you can
actually find them, pay them a tiny amount of money,
and then get rid of them when you don’t need them

However, as we have seen, it is not just platform
businesses that are engaged in such practices. The upshot
of the strategies of all of the firms I have been talking
about is that they are increasingly abdicating the re-
sponsibilities once attached to the standard employment

The Precarity Problem
Why is the decline of the standard employment relation-
ship a problem? Is it even a problem? This is a question
that needs to be answered at two levels—the individual
level and the national level.

At the individual level, for some people, it is not
a problem at all. As mentioned at the outset, it can be
a plus if you are doing these small jobs just for fun or on
the side, and if you have healthcare and social security
through some other avenue. However, for those who
depend on these flexible jobs for their livelihood, it can be
a big problem, because this type of work is highly irregular,
insecure, and typically does not come with the full package
of benefits. For a large number of workers, especially low-
skill workers, the developments I just outlined do not
translate into greater autonomy but instead into an
especially harsh form of sheer precarity as they face
recurring unemployment or chronic underemployment.30

Thus, for example, despite low unemployment, un-
deremployment is actually a significant problem in the
United States. Figure 2 comes out of the dissertation of

Figure 2
Average weekly hours for employed men, United States

Source: Conran 2017, 85

March 2019 | Vol. 17/No. 1 9
Downloaded from IP address:, on 22 Feb 2019 at 22:43:31, subject to the Cambridge Core terms of use, available at

one of my former students, James Conran.31 It tracks the
evolution of average weekly working hours for employed
men in the top and bottom income quartiles in the United
States going back to 1980.32 It shows that the number of
hours worked by men in the top income bracket rose in the
1980s and has stayed steady at a rather high level since
then. What is most striking here, however, is that the
hours of the men in the lowest income group declined
significantly over this period, first in a big drop in the early
1980s and then another large drop after the financial crisis.
This phenomenon is part of what Conran calls the great
reversal, where those at the top of the income ladder—
what we used to think of as the leisure class—are now
working longer and longer hours, while low-wage workers
are often suffering chronic underemployment.

Underemployment often takes the form of involuntary
part-time employment—i.e., people who are working
part-time jobs when they actually would prefer to be

working full time.33 Involuntary part-time employment is
highly cyclical for all types of workers, rising in economic
downturns as employers scale back on labor. In the United
States, however, it invariably hits women and people of
color with greater intensity than other groups.34 Un-
deremployment can also take the form of insufficient or
unstable hours even in full-time work. And if you want to
know what this looks like in real life, talk to some of the
people who are taking care of us at this conference—those
who are cleaning your hotel room while you attend your
panels, those who are bussing tables and washing dishes at
the restaurants at which you eat, or those preparing the
food for the reception we are about to enjoy.35

These new forms of insecurity also take the form of
recurring unemployment, as workers move between jobs
that offer low or no employment protection. For a com-
parative picture of this, figure 3 compares the United
States to several other countries based on an OECD

Figure 3
Labor market insecurity by education level

Source: OECD

10 Perspectives on Politics

Presidential Address | The American Precariat
Downloaded from IP address:, on 22 Feb 2019 at 22:43:31, subject to the Cambridge Core terms of use, available at

measure of labor market insecurity, displaying results for
high- and low-skill workers.36 As one would expect, low-
skill workers experience higher labor market insecurity
than high-skill workers. However, what is also striking
here is that in some countries (Belgium and Denmark),
social policies mitigate that gap, reducing the level of labor
market insecurity among low-skill workers to levels that
are not that different from high-skill workers. The United
States stands out both for the comparatively high level of
labor market insecurity experienced by low-skill workers
overall, and for the size of the gap that separates them from
high-skill workers on this scale.
What is also clear is that especially for those workers

who do not freely choose these types of jobs, this
employment flexibility is not really a two-way deal.
Employers in sectors such as retail and food services can

use new technologies to optimize scheduling, in extreme
cases calling workers in (or off) on short notice. What this
often means for workers is that they are being paid to work
part-time but on the condition of full-time availability.37

Figures 4a and 4b are based on a survey of low-skill service
employees in Washington, DC—retail, restaurants and
other service industries.38 The figure on the left shows
how far in advance their work schedules are posted
(darker is less than three days in advance, lighter is less
than a week). As indicated, close to 50% of those
surveyed received a week or less notice for when they
would be expected to show up for work. The figure on the
right shows that even after the initial postings of hours,
employers often make further changes.39 The result is
that a significant share of these workers are getting less
than 24 hours’ notice.40

Figure 4
Results of survey of low-skill service employees in Washington, DC

Source: Schwartz et al. 2015, 7

March 2019 | Vol. 17/No. 1 11
Downloaded from IP address:, on 22 Feb 2019 at 22:43:31, subject to the Cambridge Core terms of use, available at

These arrangements obviously make it extremely diffi-
cult to budget, hard to arrange day care, difficult to pursue
any kind of education, or even work a second job. The net
result of these trends has been to promote a new
polarization in the labor market, as Castel puts it, “between
those who are able to combine individualism with in-
dependence because their social position is secure; and
those who experience their individuality as a cross to be
borne because for them it signifies only a lack of attach-
ments and protections.”41 The demographics of the two
groups are radically different in the United States, as
women and minorities are more heavily concentrated in
those sectors in which the flexibility only works for their

The decline of the standard employment relationship is
not just a problem at the individual level; it is a problem
at the national level as well. Whether people are choosing
these jobs freely or pressed into them by necessity, the
growth of these atypical forms of employment poses long-
term problems. These employment trends are especially
problematic for countries such as the United States that
feature job-based, pay-as-you-go benefit systems in which
the contributions that employers and employees make
today are funding entitlements for the future. As firms
turn increasingly to atypical employment arrangements
that do not require contributions, they relieve themselves
of the additional costs associated with standard employ-
ment contracts, which means that these workers are less
well covered or simply left to provide for themselves. In
this way, the increase in atypical employment simulta-
neously expands the economic risks to which individuals
are exposed while at the same time reducing the social
contributions and thus the resources that would be
necessary to cover them.42 Moreover, even for those
workers who do these jobs as a supplement and receive
benefits through some other job, the companies that hire
on this basis are free riding on those other employers who
do provide benefits.43 Such firms in effect are drawing
from the commons and not contributing to it, and that is
not sustainable over the long run.

The more general negative externalities of these
trends in employment are also significant. Where these
new, more flexible forms of work make it difficult for
people to go to school or even just stay home when they
are sick, they have massive knock-on effects for health,
productivity, and social well-being. The United States is
also particularly prone to “risk contagion,” where mis-
fortune in one arena (e.g., sudden …

Looking for this or a Similar Assignment? Click below to Place your Order