How Economic

Democracy Impacts
Workers, Firms, and
Communities
The census of individual workers
in worker cooperatives
Laura Hanson Schlachter
Olga Prushinskaya
Acknowledgments
This project would not have been possible without participating workers. We appreciate
the time and insight of all the individuals and their cooperatives who chose to be part of it.
The Census of Individuals in Worker Cooperatives was sponsored by the Democracy at
Work Institute (DAWI) and conducted in partnership with the U.S. Federation of Worker
Cooperatives (USFWC), Center on Wisconsin Strategy (COWS), and University of Wisconsin
Survey Center (UWSC).
We also extend special thanks to Laura Dresser and Thomas Gunderson of COWS, Kelly Elver
and Griselle Sanchez of UWSC, Joseph Blasi of the Rutgers University Institute for the Study
of Employee Ownership and Profit Sharing, and Ana Martina Rivas of USFWC. This report
is based on research led by Laura Hanson Schlachter in collaboration with Tim Palmer and
Melissa Hoover of DAWI. Olga Prushinskaya, DAWI Metrics and Impact Analyst, synthesized
the key findings contained herein.

2
Contents

Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4

About Worker Cooperatives in the U.S. . . . . . . . . . . . . . . . . . . . . . . . . 7

How Economic Democracy Impacts Individuals . . . . . . . . . . . . . . 14

How Economic Democracy Impacts Firms . . . . . . . . . . . . . . . . . . . . 22

How Economic Democracy Impacts Communities . . . . . . . . . . . . 25

Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

Appendix . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

3
Executive Summary
In this paper, we present findings from a national survey of 1,147 workers in 82
worker cooperatives and follow-up interviews with 15 participants conducted
in 2017. This novel data allows us to explore the impact of workplace democracy
on individuals, firms, and communities across the United States. It also provides
important opportunities for future research about how worker cooperatives can
promote resilience and racial equity in this moment.

As the pandemic drags on, significant challenges persist for small businesses,
workers employed in the service and care sectors, and contract, freelance and
gig workers who cannot access unemployment benefits. The brunt of the
economic losses have been born by Black and brown business owners, low-
wage workers, and entrepreneurs of necessity who have been locked out of
both good jobs and business ownership opportunities. The stresses wrought
by the pandemic, however, simply cast into high relief a much longer arc of
deepening racial and economic inequality, stagnant wages, and concentration
of wealth in the U.S. over the past several decades. The continued division and
dysfunction of American democracy in 2021 is the outcome of concerted efforts
to disenfranchise voters and disempower communities over the past several
decades.

In this moment the proposition of worker cooperatives – that working people

can together own, control, and benefit from their own small businesses – offers
hopeful possibilities. Worker cooperatives can be used to save small businesses
whose owners are retiring or closing, create livelihoods for those with barriers to
work, pool capital among would-be entrepreneurs, and raise standards and firm
competitiveness in industries where a human-centered workplace improves

4
both the job and the service delivery itself. In the process, cooperative business
ownership can build the skills, capacity, and voice of people who otherwise may
not have substantial opportunities to participate in either decisions or profits.

This paper is concerned in part with whether and how worker cooperatives
fulfill this promise for their worker-owners. Our findings suggest that they do.

KEY FINDINGS

1 Individuals benefit materially and psychologically from

workplace democracy.
Across the board, workers in cooperatives make above living wages
and experience high job satisfaction, autonomy, voice, and other non-
financial aspects of job quality.

Overall, workers reported they earn higher wages and receive better
benefits at their cooperative than they did at their previous job.

Worker cooperatives have a very flat wage structure – 77 percent of

firms surveyed had a 1:1 or 2:1 top-to-bottom pay ratio – yet inequalities
persist in terms of ownership status, gender, and race.

2 Workplace democracy benefits firms in terms of

recruitment, retention, and reduced shirking.
Workers in cooperatives value employee ownership and plan to seek it
out in future jobs.

The non-financial benefits of workplace democracy may be more

important than wages to workers in cooperatives relative to those in
traditional firms.

5
 3 Worker cooperatives are woven into the civic fabric of
their communities.
Workplace democracy teaches skills that are transferable to other
workplace contexts and arenas of life.

Rates of political and civic participation are higher among workers in

cooperatives than in the U.S. overall.

DAWI invites the broader research community to: make use of data from this
project to explore important questions ranging from the impact of the form
on individuals, families, firms, industries and societies; to whether cooperatives
may have competitive advantages in certain labor or consumer markets; to
whether shared ownership is better suited to certain industries or localities; to
how cooperatives leverage their aggregation of member labor and capital to
create power for workers; to policy interventions that can support the uptake of
this powerful tool to solve problems of small business retention and resilience;
to the ever-present question of scale, its drivers, barriers, and effects. The
possibilities for future research are vast and rich, and there is now a robust data
set to support future inquiry. The report concludes with instructions for how to
apply for access to the Census of Individuals in Worker Cooperatives.

It is our hope that in addition to spurring additional research, the findings of

this study prove valuable to practitioners developing and supporting worker
cooperatives, advocates making the case to increase supports and resources for
economic democracy, and worker-owners themselves reflecting on their roles in
this dynamic and powerful business form.

6
About Worker Cooperatives in
the U.S.
What is Economic Democracy?
Over the last few decades, people of color, women, and immigrants have
demonstrated that worker cooperatives in the United States can increase access
to wealth and asset building for those traditionally locked out of the economy. As
inequities continue to grow during the ongoing COVID-19 crisis, it is imperative that
we continue to unpack how the model can best support workers, local economies,
and communities who may benefit from it the most.

In our work to build the field of worker cooperative development, Democracy

at Work Institute (DAWI) has seen firsthand the impact of growth in the field
on workers across the U.S. As worker ownership expands across the country
to improve wealth-building opportunities and access to jobs for workers of
color, women, and immigrant workers, it has become increasing important
to understand how workplace democracy impacts individuals, firms, and
communities – a question that was difficult to answer with existing data. DAWI
responded to the sector’s need for high quality data by investing in the most
systematic study of worker cooperatives in the U.S. conducted to date.

The 2017 Census of Individuals in Worker Cooperatives is a national survey of

1,147 people employed in 82 worker cooperatives from across the country.
The dataset also includes in-depth interviews with 15 survey participants in
three cities with vibrant cooperative ecosystems. The survey represents high-
quality baseline data about the experiences, attitudes, and demographics of

7
this unique population of workers. Findings are now informing DAWI and USFWC
efforts to advocate for worker ownership’s inclusion as a tool for rebuilding resilient,
equitable post-COVID economies.

What is a Worker Cooperative?

Worker cooperatives are values-driven businesses that put worker and
community benefit at the core of their purpose. In contrast to traditional
companies, members of worker cooperatives share in the profits, oversight, and
management of the organization using democratic practices.Workers own the majority
of equity in the business and control the voting shares.

Since there is no federal legal category for worker cooperatives in the U.S., which
adopt many different legal structures and operational mechanisms, the Census
of Individuals in Worker Cooperatives drew on DAWI’s knowledge of the field
to develop study eligibility criteria that reflects and celebrates this diversity.
Workplaces were eligible to participate if they employ at least three paid
individuals, are eligible for workplace membership in the USFWC, and are either 1)
majority democratically owned and governed by paid workers, or 2) democratically
governed by paid workers even if they do not necessarily have a separate worker-
owner share class. Enterprises that are wholly owned and democratically governed
by workers according to the principle of one worker, one vote; multistakeholder
cooperatives with a worker-owner share class; consumer and marketing
cooperatives with democratic workplaces; and nonprofits with democratic
workplaces were all eligible to participate in the study.1

1
Employee stock ownership plans (ESOPs) and firms without meaningful democratic worker ownership or governance were not
eligible to participate.

8
Worker Cooperatives in the U.S.
DAWI’s most recent enterprise
survey identified 415 worker
cooperatives and democratic
workplaces that were operational
in 2018.2 Together, these values-
driven businesses employ an
estimated 6,454 workers and
produce about $505 million
dollars in revenue each year.
The typical co-op is small (about
eight workers and $505,000
dollars in gross annual revenue)
and relatively young. There are
also several large and long-
lasting worker cooperatives. For example, Cooperative Home Care Associates
was founded 35 years ago in the Bronx and now employs over 2,000 workers,
most of whom are immigrant women of color.

There has been a steady increase in the identification of existing firms as worker
cooperatives, the launch of new startups, and the conversion of traditional
businesses to democratic worker ownership in the past five years. Between
2013 and 2018, DAWI documented a 36% net increase in the number of worker
cooperatives in the U.S. Expansion of the model can benefit individuals, firms,

2
The total number of worker cooperatives in the U.S. is very likely higher, as some firms may have been excluded in this count
due to survey qualification criteria, including the need to be operating for more than one year at the time of survey completion,
at least three employees, etc. Businesses may also have been missed because they do not have a web presence or do not identify
themselves publicly as a worker cooperative or democratic workplace. DAWI estimates that the number of worker coops in
the U.S. may be closer to 800. For details, see the 2019 Worker Cooperative State of the Sector Report at https://institute.coop/
resources/2019-worker-cooperative-state-sector-report.

9
and communities in several ways.

First, worker cooperatives can reduce economic inequality by sharing wealth

and ownership more broadly among workers. The annual Worker Cooperative
Enterprise Census has consistently found that most U.S. worker cooperatives
have a flat wage structure. In 2018, the Enterprise Census found that 86 percent
of co-ops had a 2:1 or 1:1 top-to-bottom pay ratio whereas the average large
U.S. corporation had a CEO-to-worker pay ratio of 303:1. This suggests worker
cooperatives prioritize internal equity over other compensation goals.

Second, workplace democracy can increase economic inclusion by creating and

maintaining sustainable, dignified jobs; generating wealth; improving workers’
quality of life; and making business ownership more accessible to people with
limited access to start-up capital. People who identify as women and Hispanic
or Latinx are overrepresented in worker cooperatives compared to the U.S.
workforce overall.3

Third, worker cooperatives can support civic and community development. For
example, scholars have argued that experience with democracy in the workplace
can increase workers’ interest in politics, propensity to vote, and participation in
the larger democratic process. Cooperatives may also nurture prosocial values like
altruism, empathy, and reciprocity. Place-based, values-driven enterprises help
keep capital locally rooted and practice the cooperative principle of “concern for
community” by donating to local causes, participating in social movements, and
investing in community ventures. For example, sustainability initiatives at Union
Cab in Madison, Wisconsin have ranged from purchasing a fleet of hybrid vehicles
to installing a permeable parking lot to supporting the local affiliate of the climate
change organization 350.org.

3
Statistics for employed people in US workforce represent racial groups as percent of total employed derived from the 2018
Current Population Survey. See Table 5 at https://www.bls.gov/cps/cps_aa2018.htm for annual average employment status of the
civilian noninstitutionalized population by sex, age, and race.

10
In this report, we leverage data from
the 2017 Census of Individuals in
Worker Cooperatives to examine
how workplace democracy impacts
individuals and firms. We also
explore the potential for broader
impacts related to economic
inequality, inclusion, and community
development. But first, let us introduce
you to the workers and cooperatives
that participated in this original study.

Worker Cooperatives in the

Study
The Individual Census was a confidential
study. None of the participating workers
industry, geography, and history.
and firms are individually identifiable
in the data. Nevertheless, participants Additionally, of participating worker co-
answered questions about their own ops, 48 percent offered health benefits,
characteristics – and those of workers in 34 percent offered dental benefits,
each participating firm – that allow us to and 19 percent offered retirement
better understand demographic trends benefits. Fifteen percent of participating
in the sector. businesses offered all three types of
benefits. Ten percent of participating co-
Overall, the 82 worker co-ops that
ops were unionized. Unionized worker
participated in the Individual Census are
co-ops were represented by IWW, UFCW,
quite similar to the 306 firms that were
SEIU, AEA, CWA, and Teamsters.
eligible in terms of governance structure,

11
The size of participating co-ops varied: 29 percent of businesses were micro, 34
percent small, 12 percent mid-sized, 17 percent medium, 7 percent large, and 1
percent very large. This is comparable to data from the 2016 State of the Sector,
which found the average number of workers at worker cooperatives was 9.5. As
these graphs indicate, the Census of Individuals in Worker Cooperatives also
reflects the geographic and industrial diversity of the sector.

All Worker Co-ops by Industry Census Participants by Industry

15% 11%
16% 6% 27%
5%
9% 6%
12%
12% 12% 10%
7%
7% 7%
11% 6% 8% 7% 9%

12
 Workers in the Study
The 82 firms that chose to participate
employ 3,544 people overall. Worker-
owners, full-time workers, English
speakers, and workers who identify as
white and male were more likely to return
a completed survey than their employee,
part-time, Spanish-speaking, nonwhite,
female colleagues. Overall nonresponse
bias is largely driven, however, by a
single large and demographically
unique participating firm. If we exclude
respondents from this firm, worker-owners
are the only group overrepresented in the
sample. Tenure of workers at firms ranged
from 1 month to 42 years, with a mean of 6
years and a median of 3.5 years.

Participants at a Glance

1% All Worker Co-ops All Worker Co-ops

34%
4% female 69% full-time
79% 61%
male part-time
20% 35%
non-binary Census Participants
Gender Hours full-time
Census Participants Worked part-time
female
65%
male 31%
non-binary

13
How Economic Democracy
Impacts Individuals
Wages and Benefits
Three-quarters of respondents reported they get paid an hourly wage; about
equal shares are salaried and paid in some other way. We focused our analysis
on salaried and hourly workers because these alternative arrangements ranged
widely from “free membership” to “commission.” Although those who reported
they are paid some other way rarely included a numeric value that allowed us to
compute hourly pay, they highlight the variety of ways cooperatives creatively
meet both economic and social needs.

We imputed hourly pay for all salaried workers in order to compare

compensation across respondents overall. Hourly pay (n=836) ranged from
$6.73 to $125 per hour with a mean of $17.74 and a median of $13.76. This
average hourly wage
figure is consistent Mean Hourly Wage by Demographics

with DAWI’s 2018

Overall
economic census,
which estimated Worker-Owners

mean hourly pay at

Men
$19.67.
Women
Consistent with
previous research, we Whites

found egalitarian pay

People of Color
distributions within

14
 Mean Hourly Pay by Industry (n=835)

firms that participated in the Census of Individuals in Worker Cooperatives.

Among those with more than one respondent who reported hourly pay (n=61
firms), the ratio of highest to lowest pay ranged from 1 to 1 to 5.19 to 1. The
mode was 1 to 1 (at eight firms) and the median was 1.50 to 1. Half of all firms
had a pay ratio of less than 1.5 to 1 and 77 percent of firms had a pay ratio of 2
to 1 or lower. A majority – 65 percent – of respondents said that patronage or
profit sharing is distributed more fairly at their current job than previous job.

We did find inequalities, however, between workers of different characteristics.

On average, worker-owners, men, and whites had significantly higher hourly pay
rates than employees, women, and nonwhites overall.

Given that we also see large hourly pay inequalities across industries, pay gaps
by gender and race may be linked to occupational segregation, which occurs
when demographic groups are not equally represented in all occupations.

15
 For example, all of the 23 high earners making above $50 per hour are
employed by six firms. These cooperatives are in lucrative, stereotypically “male”
industries like construction and professional/ scientific/ technical industries and
predominantly employ men (at least 75 percent of the workforce at four of these
six firms identifies as male). In contrast, women tend to dominate the workforce
in low-wage, stereotypically “female” industries like health care and social
assistance where worker cooperatives have experienced significant growth in
recent years.

A forthcoming analysis of wage inequalities in the Census of Individuals

in Worker Cooperatives by race, gender, and immigration status by Sarah
Reibstein and Laura Schlachter explores these patterns in more depth.

Despite large variation in pay between cooperative firms in different industries,

across the board respondents reported earning more at their cooperative
than at their previous job. This “cooperative wage boost” was $3.52 an hour
at the mean and $2 an hour at the median (n=579). We also analyzed this by
ownership, gender, and race and found that dominant groups do better in terms
of the “cooperative wage boost” in all cases – again, likely due to occupational
segregation.4

4
A vast sociological literature documents the scarring effects of unemployment, but a wage boost is often associated with
voluntary job transition. Future research should compare the “wage boost” of transitioning into cooperatives versus traditional
firms.

16
About half of respondents (n=878) described their overall compensation as
better at their cooperative than at their previous job. We found that actual and
perceived compensation differences generally aligned.

We found similar patterns when we asked respondents to compare other

dimensions of their compensation and benefits. For example, about half said
their job at the co-op provides benefits that better meet their family’s needs
(49 percent; n=933) and better allows them to save money while working (47
percent; n=890) than their previous job.

Wealth Accumulation
Worker cooperatives provide unique opportunities to generate broad-based
wealth by allowing worker-owners to purchase an equity stake in their
firm, accumulate retained equity in an internal capital account, and receive
patronage dividends. The Census of Individuals in Worker Cooperatives suggests
that purchasing an equity stake in a worker cooperative is affordable for most
workers. The data also suggests that wealth accumulation is large in a small
portion of firms.

Workers become owners in a cooperative when they purchase a membership

share, which guarantees one-person-one vote governance rights within the
firm. Participating firms reported that the cost of purchasing a worker-owner
share in 2016 (n=71 firms) ranged from $0 to $18,000 with a mean of $1,979 and
a median of $500. This implies that a handful of firms have share prices on the
upper end of this range that skew the mean upwards. Three-fourths of worker-
owner respondents (n=555) reported purchasing their share for $1,000 or less
and 93 percent reported a share purchase price of $5,000 or less.

Worker cooperatives often use a system of “internal capital accounts” that

enables them to track the contributions and equity holdings of individual

17
members over time. The majority of participating co-ops (60 percent of the
sample; n=63 firms) reported that they use an internal capital account structure
to track retained equity for each worker-owner. Twenty-five percent use some
other type of structure and fourteen percent use a collective capital account
structure. In total, 44 percent (n=608) of individual worker-owner respondents
reported having their own internal capital account.

Among worker-owners who

reported the total value of their
“My co-op has set me up with both
internal capital account in 2016
a retirement account and internal
(n=209), accumulated equity
ranged from $0 to $150,000
capital account... I feel very safe,
with a mean of $10,091.2 and secure, and comfortable at work,
a median of $2,000. Again, a which means work does not give me
handful of respondents have a lot of long-term or chronic stress.”
retained equity amounts at
the upper end of this range
that skew the mean upwards. Approximately three quarters (78 percent) of
respondents report it is $10,000 or less.

Patronage is a dividend or distribution of net income that a worker cooperative

makes to its members. In total, 64 percent of worker-owners (n=588) reported
they have ever been eligible to receive cash patronage or profit sharing; 75
percent of those who were eligible (n=373) reported they have ever received a
payment.

Of these 280 worker-owners who had ever received a cash patronage or profit
sharing distribution, 240 reported receiving it since 2015. The total value of
these recent distributions (n=214) ranged from $5 to $112,000 with a mean of
$8,679.80 and a median of $2,328. Three-quarters received less than $7,000.

18
The five percent of workers who reported receiving a cash distribution of
more than $30,000 in the last two years were employed by two high-tech
manufacturing and construction firms.

The majority of respondents (65 percent; n=1,007) reported they are at

least slightly confident that they could find $2,000 within the next 30
days in the case of an unexpected expense or emergency and 35 percent
reported they are not at all confident they could.

Other Aspects of Job Quality

Worker cooperatives may provide additional dimensions of job quality
to workers relative to traditional businesses. We asked about job
satisfaction, workplace autonomy, psychological ownership, and other
factors in order to better understand these areas of impact.

Job Satisfaction

We measured job satisfaction with questions about overall working

conditions, relationships at work, and professional development.

In terms of overall working conditions, the majority of respondents

described their job security, confidence in the economic stability of the
company, work effort, and job satisfaction as somewhat or much better
in their current job at a worker cooperative as compared to their last job.

The majority of respondents also described their relationship with their

boss and management, workplace conflict resolution, and sense of
common purpose as somewhat or much better at their current than
previous job.

In the area of professional development, the majority of respondents

described the quality of supervision, feedback, and training they received

19
 Overall Working Conditions

Somewhat or
much better than
previous job

About the same

at current and
previous job

Somewhat
worse or
much worse at
current jobe

and their ability to pursue career goals as somewhat or much better at their
current job at a worker cooperative as compared to their last job.

Autonomy and Voice

We measured autonomy and voice

with questions about absolute “Being able to work from home and
autonomy and influence at worker be more present with my family
cooperatives and questions has been a huge shift... [and] had a
comparing how respondents fundamentally positive impact on my
experienced these things in their life. I am happier, healthier, and more
current and previous jobs.
stable than I have been in years....
As a measure of absolute That solidarity is worth the challenges
autonomy and influence, we of not getting paid quite enough.
used standard questions that But it would sure be nice to get paid
asked workers to rate their enough.”
involvement and direct influence

20
in decisions at various levels. The majority said that they have quite a bit or a
great deal of involvement or direct influence at the individual level (70 percent
of respondents, n=1029), at the team level (57.3 percent of respondents, n=966),
and at the enterprise level (50% of respondents, n=975), indicating a high
level of overall autonomy. We also asked respondents to compare their level of
autonomy and influence at their current job and previous job. A majority said
that they have more control over their own schedule and more influence over
workplace decision-making in their current position at a worker cooperative.

Psychological Ownership

Overall, 43 percent (n=1066) of respondents said they feel quite a bit or a great
deal like an owner at their cooperative. When split by ownership status, we see
that 62 percent of worker-owners feel quite a bit or a great deal like an owner
compared to only 12 percent of employees.

Participation in Workplace Governance

The most frequent form of participation in firm governance was attending general
assembly meetings. At the mean and median firm level, respondents attend
general assemblies, committee meetings, and meetings of the Board of Directors
or governance body a few times a year. The least frequent form of participation was
voting for the board of directors or governance body, which occurred once per year
at the mean and median.

Overall, personal interest appears to be the most important factor impacting

respondents’ level of participation in firm governance and decision-making; access
to cooperative-specific training and financial compensation appear to be least
important.

21
 HOW ECONOMIC DEMOCRACY
IMPACTS FIRMS
The Importance of Ownership
Overall, findings suggest that workplace democracy entails several benefits for
firms. The majority of respondents (52 percent, n=1072) said that working for
a company with employee ownership is very or extremely important to them.
Additionally, 32 percent (n=1072) said it is very or extremely important that
their next job be in a company with employee ownership. Ownership thus has
implications for firms related to recruitment, talent retention, and shirking.

Many respondents in the Census of Individuals in Worker Cooperatives appear to

value the non-financial benefits of workplace democracy even more than wages.
Overall, 60 percent (n=1,077) of the sample reported that the opportunity to
work in or start a worker co-op – rather than occupational opportunities per se
– initially attracted them to their job. Almost half (45 percent, n=1063) said their
total wages are lower at their co-op than they would be in a comparable job at
another company in their region – and the proportion of respondents willing to
accept lower wages in order to work at a co-op was actually higher among those
who reported they did not deliberately opt into workplace democracy.5

This may partly explain why turnover rates are relatively lower in co-ops than
conventional firms. Overall, over half of respondents (52 percent; n=1,118) said
they were not at all likely to quit in the next year and only 15 percent said they
were very or extremely likely to leave their firm. As we noted above, tenure

5
Among respondents who were initially attracted to their job because of the opportunity to work in their current occupation, 48
percent said their total wages are somewhat or much lower than in comparable jobs compared to only 42 percent of those who
self-selected into workplace democracy.

22
 ranged widely among respondents from one month to 42 years – but the
average worker had stayed at their co-op for six years.

Another longstanding debate

about cooperatives is related
to what economists call the “I love feeling a sense of ownership in
“free rider problem.” The idea is my current job. I feel like the amount
that individuals are more likely of personal and professional freedom
to slack off if they know their I’m ‘allowed’ is responsible for a higher
rewards are tied to collective
level of job and personal satisfaction.
rather than personal effort –
On the other hand, being so invested
and the cost of calling out a
in this business is often very stressful
colleague who isn’t doing their
fair share is higher than the
and I find myself very critical of co-
benefit, so no one will intervene. workers that don’t take their jobs
Yet studies of other shared seriously or treat the workplace and
capitalism arrangements are other workers with respect.
consistent with our finding that
peer monitoring is common in
worker co-ops and shirking is
rare.6 Forty percent (n=1,051) of respondents say it is easy to identify whether
a coworker is working well or poorly, and co-op workers say they are actually
more likely to intervene if they see someone slacking off than they were in their
previous job.

Many of the firm level findings from this data are comparable to findings from
the enterpise census. The economic census also demonstrated data related to
patronage distributed by firms, although comparable information on individual
worker-level patronage receipt is not available. The Census of Individuals in

6
See, for example, Shared Capitalism at Work (2010) by Douglas Kruse, Richard Freeman, and Joseph Blasi.

23
Worker Cooperatives found that 46 percent of firms did not make patronage
distributions in fiscal year 2016. Of those making payments, distribution ranged
from $1,800 to $1,960,000, with an average distribution of $197,942 and a
median distribution of $18,900.

24
HOW ECONOMIC
DEMOCRACY IMPACTS
COMMUNITIES
Training
One way that worker cooperatives enact the principle of “education,
training, and information” is through structured training programs.
The majority of respondents (79 percent, n=1121) reported that their
cooperative offered some kind of formal training, and 63 percent
(n=863) reported they had participated. Many of these skills are
transferable beyond cooperatives to support workers’ participation in
other workplace contexts and arenas of life.

Over half of respondents (54 percent, n=1116) had received some kind
of cooperative-specific training designed to equip workers to engage
more effectively in workplace
decision-making.

Overall, about half (54 percent, “[Working in a co-op] has given me

n=979) also said that access to experience/insight into the work that
training impacted their level
goes on behind the scenes in running
of participation in cooperative
a business. It has given me the
governance. Workers who received
confidence that if I chose to, I could
cooperative-specific training
participated significantly more in
start another business and feel well-
workplace decision-making than prepared for success.”
those who did not, and were also

25
significantly more likely to report their level of participation had increased
over time.

About a third of respondents (34 percent, n=863) had also received some
kind of life skills training through their cooperative.

These types of investments in workers’ collective decision-making,

communication, and other skills not only increase the level of human
capital in communities, but also have the potential to empower workers
to participate more actively off the clock.

Civic and Political Participation

Since the early 20th century,
scholars and observers have
argued that democracy in “What I like most about worker
the workplace “spills over” cooperatives is that they allow you
into democracy on a larger
to live your values in the workplace
scale.7 Data from the Census
instead of working against your
of Individuals in Worker
values on the job and then try to
Cooperatives suggests that
workers in cooperatives are
undo the damage on your free time.”
indeed quite active in civic life.

Overall, 39 percent of respondents (n=1021) reported they had

volunteered through or for an organization in the past year compared to
only 31 percent of employed respondents in the general population in
the 2017 Current Population Survey.8

7
See, for example, Gabriel Almond and Sidney Verba’s The Civic Culture: Political Attitudes and Democracy in Five Nations (1963)
and Carole Pateman’s Participation and Democratic Theory (1970).
8
For details on the 2017 Current Population Survey Volunteering and Civic Life Supplement, see https://www.census.gov/data/
datasets/2017/demo/cps/cps-civic.html.

26
Cooperative workers volunteered with Excluding respondents from
many different types of organizations. one large and demographically
Social or community service groups
unique participating co-op, we
were the most popular followed by
find that …
political parties or advocacy groups.

More than half of volunteers (54

percent) had given their time to
two or more groups and 36 percent
had volunteered with three or more
groups. In comparison, only 18
percent of respondents in the 2015
Current Population Survey Volunteer 60% of co-op workers voted in the
Supplement volunteered for two or 2014 midterms compared to 49% of
more organizations and only six percent workers in the general US population
for three or more organizations.9 and...
Respondents were also active in other
types of civic activities including
serving as a group officer or committee
member outside of work, attending
public meetings, and joining in a
protest. Seven in ten respondents
(n=998) had participated in at least
one of these activities in the past 78% of co-op workers voted in the
year. Almost half of respondents had 2016 presidential election compared
participated in all three. to 77% of workers in the general U.S.
Among those who reported they had population.
participated in a meeting as an officer

9
For details on the 2015 Current Population Volunteer Supplement, see https://catalog.data.gov/dataset/current-population-
survey-volunteers-supplement

27
or on a committee of a group or organization outside of work, half (n=994) had
participated in one to six meetings in the past year and 22 percent had gone to six
meetings or more.

Among those who attended a public meeting in which there was discussion of
community affairs in the past year, about half (n=969) attended one to two public
meetings and about half attended three or more.

Among those who joined in a protest march, rally, or demonstration in the past year,
a slightly larger share participated in one to two protests than participated in three
protests or more.

Rates of voting among respondents in the Census of Individuals in Worker

Cooperatives were comparable to those of other U.S. workers.10

10
For details on the 2016 American National Election Studies Post-Election Questionnaire, see https://electionstudies.org/data-
center/2016-time-series-study/. For details on the 2014 Current Population Survey Voting and Registration Supplement, see
https://catalog.data.gov/dataset/current-population-survey-voting-and-registration-supplement.

28
Conclusion
The Census of Individuals in Worker Cooperatives offers a first-of-
its-kind view into the impacts of economic democracy on workers,
firms, and communities. Further research is needed to draw further
conclusions from this data, and we invite the research community to
engage in this work.

The Democracy at Work Institute is now accepting applications

from researchers interested in using the de-identified data collected
through the Census of Individuals in Worker Cooperatives in their own
work.

The application to access this data is available on our website at www.

institute.coop/censusdataaccess along with instruments and the
interview protocol.

Please send questions about accessing the data to olga@institute.

coop.

29
APPENDIX:
ABOUT THIS PROJECT
Research Design
The Census of Individuals in Worker Cooperatives is a mixed-methods
study. We modeled our research design after the 2010 Shared
Capitalism at Work study by Douglas Kruse, Richard Freeman and
Joseph Blasi. This involved decentralized administration of a paper-
and-pencil survey to a nonrandom sample of workers clustered in
firms and semi-structured follow-up interviews with a purposive
sample of survey respondents.

We invited all 306 U.S. worker cooperatives that met our eligibility
criteria (out of the 326 in DAWI’s database in November 2016) to
participate in the study; 82 ultimately chose to do so. All survey
materials were available in English & Spanish.

Each participating firm identified a “point person” who received a

packet of survey materials, filled out a 13-item questionnaire about
their cooperative, distributed the 58-question worker questionnaire
to their colleagues (typically during an all-staff meeting), and returned
completed surveys to the study team.

Survey Quality
We calculated response rates using the 2016 AAPOR Standard
Definitions and Outcome Rate Calculator: https://aapor.org/Education-

30
Resources/For-Researchers/Poll-Survey-FAQ/Response-Rates-An-Overview.aspx. In total, we
distributed 2,313 worker surveys to 88 co-ops that expressed interest in participating
and received 1,147 completed surveys from individuals in 82 firms. DAWI estimates the
sector employed about 6,000 people in 2017. These 82 firms employ an estimated 59
percent of the target population, a large proportion of individuals in the study universe,
and we achieved an overall worker response rate of 19 percent. This was possible in part
thanks to participation from one large worker co-op that employs one-third of workers
in the target population and contributed one-third of the sample. Response rates within
firms ranged from 9 to 100 percent with a mean of 72 percent and a median of 80
percent.

We used data from DAWI’s annual enterprise census to assess nonresponse bias at the
firm level and concluded it was relatively minimal. Based on firm demographic data
reported by the point person, we determined that worker-owners, full-time workers,
English speakers, and workers who identify as white and male are overrepresented
in our sample. Excluding respondents from the large firm mentioned above, full-time
workers are the only overrepresented group.

Key Survey Measures

We appended firm industry using DAWI’s classification of each participating co-op’s
North American Industry Classification System (NAICS) code: https://www.census.gov/
eos/www/naics/

We classified firm size based on the total number of workers in each co-op, as reported
by the point person. We developed sector-specific categories based on DAWI’s
knowledge of size thresholds that are meaningful for worker co-ops. Using firm-size
threadholds developed for this study, 58 percent of participating firms are “micro” (nine
workers or fewer), 36% are “small” (ten to 49 workers), 5 percent are “medium” (50 to 249
workers), and one is “large” (employs 250 workers or more).

For respondents who reported getting paid by the hour, hourly pay represents workers’

31
hourly wage. We imputed the hourly pay of salaried respondents by dividing their
annual salary by the hours they reported working in a typical week by 52 weeks.

Rather than asking respondents to report absolute values of job quality and
satisfaction, the Census of Individuals in Worker Cooperatives asked respondents
to assess their current job in light of a concrete reference point: their previous job.
Whenever possible, we triangulated these responses with other questions that allowed
us to compare current and previous job characteristics. For example, one question
asked respondents to rate how their overall compensation and benefits compare,
from “much worse” to “much better” at their current (i.e. cooperative) relative to their
previous job. We also mathematically compared reported hourly pay at the current
and previous job of the 579 respondents who answered both questions, then analyzed
how well perceived and actual compensation changes lined up.

The Census of Indivuals in Worker Cooperatives survey included a number of questions

about workers’ frequency of participation in workplace governance activities on a
scale from one (no participation in this activity) to five (high frequency of participation
in this activity) where one indicates “never participating,” two indicates “participating
every two or three years,” three indicates “participating every year,” four indicates
“participating a few times a year,” and five indicates “participating every month or
more.”

A critical mass of questions (27 in total) in the Census of Individuals in Worker

Cooperatives were inspired by the Shared Capitalism at Work instrument, including
questions on psychological ownership and shirking, which opens up opportunities for
comparative research on worker co-ops and ESOPs.

Interviews
The study director conducted 15 semi-structured follow-up interviews with a
purposive sample of survey respondents from 11 co-ops in three of the most vibrant
cooperative ecosystems in the U.S. We recruited interviewees via an email invitation to

32
all 33 co-ops that participated in the survey from the Bay Area, Madison, and New York City.

Those who were interested filled out an online screening questionnaire. In-person
interviews took place in June and July 2017. They focused on key themes that emerged
from our preliminary survey data analysis including self-selection, job quality and
satisfaction, participation in workplace governance, and civic participation. Each
interviewee chose their own pseudonym and we have de-identified the transcripts to
preserve confidentiality.

33