ArticlePDF Available

Efficiency in Employee-Owned Enterprises: An Econometric Case Study of Mondragon

Article

Efficiency in Employee-Owned Enterprises: An Econometric Case Study of Mondragon

Abstract

We provide the first econometric study of efficiency for a member of the Mondragon group of worker cooperatives. Eroski is a retail distribution chain and, most unusually, there are two distinct types of hypermarkets: (i) cooperatives with significant employee ownership and voice; and (ii) GESPAs with modest employee ownership and limited voice. For supermarkets the chain includes conventional firms with no employee ownership as well. Our key data are a panel of monthly observations from February 2006 through May 2008, a total of 9,800 observations for supermarkets and 2,150 for hypermarkets. By estimating first difference models we find that hypermarket stores with cooperative ownership grow sales significantly faster than GESPA stores. For supermarkets overall we find no significant difference in performance among the three types of stores. However, for a particular segment of the supermarket called SUPERMARKET CITY (a subgroup of small supermarkets for which having "better customer service" employees is particularly important), cooperatives are found to outperform conventional stores. To investigate mechanisms that help explain why cooperatives are better performers we provide additional evidence that takes account of the role of the more extensive opportunities for employee involvement and training, and stronger economic incentives that exist in cooperatives. Finally, while cooperative members are better paid than their peers in comparable firms, individual-level data also show that job satisfaction is actually lower for workers in cooperatives than for GESPA workers. Though this may be a simple reflection of high worker expectation in cooperatives, cooperatives may well be indeed a "high-stress work system." The overall assessment of cooperatives will need to be nuanced.
DISCUSSION PAPER SERIES
Forschungsinstitut
zur Zukunft der Arbeit
Institute for the Study
of Labor
Effi ciency in Employee-Owned Enterprises:
An Econometric Case Study of Mondragon
IZA DP No. 5711
May 2011
Saioa Arando
Monica Gago
Derek C. Jones
Takao Kato
Efficiency in Employee-Owned Enterprises:
An Econometric Case Study of Mondragon
Saioa Arando
Mondragon University
Monica Gago
Mondragon University
Derek C. Jones
Hamilton College, WDI (Michigan),
MCAC (Mondragon) and SKOPE (Oxford)
Takao Kato
Colgate University, CJEB (Columbia),
TCER (Tokyo), CCP (Aarhus) and IZA
Discussion Paper No. 5711
May 2011
IZA
P.O. Box 7240
53072 Bonn
Germany
Phone: +49-228-3894-0
Fax: +49-228-3894-180
E-mail: iza@iza.org
Any opinions expressed here are those of the author(s) and not those of IZA. Research published in
this series may include views on policy, but the institute itself takes no institutional policy positions.
The Institute for the Study of Labor (IZA) in Bonn is a local and virtual international research center
and a place of communication between science, politics and business. IZA is an independent nonprofit
organization supported by Deutsche Post Foundation. The center is associated with the University of
Bonn and offers a stimulating research environment through its international network, workshops and
conferences, data service, project support, research visits and doctoral program. IZA engages in (i)
original and internationally competitive research in all fields of labor economics, (ii) development of
policy concepts, and (iii) dissemination of research results and concepts to the interested public.
IZA Discussion Papers often represent preliminary work and are circulated to encourage discussion.
Citation of such a paper should account for its provisional character. A revised version may be
available directly from the author.
IZA Discussion Paper No. 5711
May 2011
ABSTRACT
Efficiency in Employee-Owned Enterprises:
An Econometric Case Study of Mondragon*
We provide the first econometric study of efficiency for a member of the Mondragon group of
worker cooperatives. Eroski is a retail distribution chain and, most unusually, there are two
distinct types of hypermarkets: (i) cooperatives with significant employee ownership and
voice; and (ii) GESPAs with modest employee ownership and limited voice. For
supermarkets the chain includes conventional firms with no employee ownership as well. Our
key data are a panel of monthly observations from February 2006 through May 2008, a total
of 9,800 observations for supermarkets and 2,150 for hypermarkets. By estimating first
difference models we find that hypermarket stores with cooperative ownership grow sales
significantly faster than GESPA stores. For supermarkets overall we find no significant
difference in performance among the three types of stores. However, for a particular segment
of the supermarket called SUPERMARKET CITY (a subgroup of small supermarkets for
which having “better customer service” employees is particularly important), cooperatives are
found to outperform conventional stores. To investigate mechanisms that help explain why
cooperatives are better performers we provide additional evidence that takes account of the
role of the more extensive opportunities for employee involvement and training, and stronger
economic incentives that exist in cooperatives. Finally, while cooperative members are better
paid than their peers in comparable firms, individual-level data also show that job satisfaction
is actually lower for workers in cooperatives than for GESPA workers. Though this may be a
simple reflection of high worker expectation in cooperatives, cooperatives may well be indeed
a “high-stress work system.” The overall assessment of cooperatives will need to be
nuanced.
JEL Classification: J54, D21
Keywords: employee ownership, producer cooperatives, labor managed firm,
productive efficiency, Mondragon, shared capitalism
Corresponding author:
Takao Kato
Department of Economics
Colgate University
13 Oak Drive
Hamilton, NY 13346
USA
E-mail: tkato@colgate.edu
* Jones and Kato are grateful to support from NSF SES-052117 and Jones is also appreciative of
support from an Ikerbasque fellowship. Arando and Gago are grateful to support from the Saiotek
programme (of the Basque Government). The authors are extremely grateful to a number of
individuals working for the case firm for interviews, comments and access to the data. The data used
in this study are proprietary. The paper has benefitted from comments from Fred Freundlich and Jeff
Pliskin as well as Dietrich Earnhart (discussant) and other participants at the ASSA meetings in
Atlanta, 2010 and seminar participants at Rutgers and Tokyo.
1
Efficiency in Employee-Owned Enterprises:
an Econometric Case Study of Mondragon
I INTRODUCTION
The recent economic crisis has stimulated much interest amongst researchers and policy
makers concerning the possibilities of alternative ways to structure economic organizations. One
option is shared capitalism, characterized by a variety of financial participation programs (such
as profit sharing, gain sharing, employee ownership, and broad-based stock options), all of which
make workers significant stakeholders of the firm (Kruse, Freeman and Blasi, 2010). Perhaps the
most powerful form of shared capitalism is an enterprise in which workers are majority owners,
with responsibility for key strategic decisions, such as worker cooperatives.
Mondragon is often considered the most successful example of such employee-owned
enterprises in the world. Founded in 1956 with some 25 workers in the Basque Country of Spain,
Mondragon was originally a group of mainly industrial worker-owned enterprises. Subsequently
the group has grown to include firms in other areas, notably retail and finance, and it extends
across Spain, Europe and the globe. By 2008, Mondragon comprised about 250 cooperatives,
subsidiaries and affiliated organizations, including 73 manufacturing plants overseas, altogether
employing almost 100,000 (see http://www.mcc.es/.)
There has been a long-standing interest by diverse scholars on various matters concerning
the Mondragon cooperatives (e.g., Johnson and Whyte, 1977, Bradley and Gelb, 1982). Since
these early studies, Mondragon has continued to grow -- by some estimates Mondragon
represents the seventh largest consortium in Spain (e.g., Arando et al., 2011) Moreover, facts
such as no job losses ever having been sustained by cooperative members, including during the
present economic crisis, help explain why other studies of Mondragon have continued to appear
2
regularly (e.g., Joshi and Smith, 2008). At the same time, a conspicuous characteristic of the
vast bulk of the published literature on Mondragon cooperatives is that there have been few if
any applied studies that use standard hypothesis-testing methods. Thus on one of the key
questions of interest to economists, the efficiency of the Mondragon cooperatives compared with
conventional firms, most published work has been restricted to comparisons using efficiency
indicators that are quite aggregated, for example comparisons at an industrial or overall group
level between cooperatives and conventional firms (e.g., Thomas and Logan, 1982.)
The limited nature of the research on Mondragon cooperatives contrasts with the more
general literature concerning employee owned firms. The first econometric studies of the
performance of firms in which there is substantial/majority employee-ownership firms (EOFs),
often considered as producer co-operatives (PCs) or labor managed firms (LMFs) appeared more
than 30 years ago (e.g., Jones and Backus, 1977.) Work has continued to flow including
influential studies of the U.S. plywood cooperatives (Craig and Pencavel, 1995) and, more
recently, on worker cooperatives in Uruguay (Burdin and Dean, 2009.). There is also a growing
literature that usually focuses on more modest examples of employee ownership, including the
shared capitalism literature for the US (e.g., Kruse, Freeman, and Blasi, 2008), European
countries (e.g., Pendleton and Robinson, 2008 for the U.K), transition economies (e.g., Estrin et
al, 2009) and Japan (e.g., Jones and Kato, 1995.) As well as the issue of comparative
performance, this broader literature, notably the “High-Performance-Workplace-Practices”
literature, considers other matters including the mechanisms that underpin performance
differences (e.g., Ichniowski and Shaw, 2003.)
However, the issues that are examined concerning employee owned firms are still not
definitively settled. Thus while the meta analysis by Doucouliagos (1995) concludes that the
3
balance of evidence demonstrates better performance by PCs compared to participatory capitalist
firms, other assessments, including Bonin et al. (1993:1305) and Dow (2003:184) are not so
sanguine. Concerning the performance of PCs, this ambivalence is particularly apparent when
evaluation is restricted to studies that endeavor to make comparisons between PCs and
conventional firms within the same industry (for a review, see, e.g., Dow, 2003:184.) By
providing new evidence for a Mondragon case on comparative performance and underlying
mechanisms we also contribute to many of these debates.
Our study is facilitated because we are fortunate to have access, for the first time, to two
types of primary data. Most important we have data for the population of stores in the Eroski
retail chain, with Eroski by far the largest employer within the Mondragon group today. By using
these new panel data we provide the first econometric study of efficiency for any Mondragon
cooperative. Moreover, growth in the chain has resulted in there being large numbers of
individual stores that have three distinct categories of employee ownership, ranging from
significant (the cooperatives), through limited (known as GESPA) to zero (conventional stores).
Cooperatives also have large numbers of workers who are not (yet) members and in GESPA
stores many workers choose not to become members. These panel data enable us to contrast the
impact on efficiency of both cooperative and limited employee ownership with conventional
ownership and also to investigate the potential importance of some of the key mechanisms that
might account for differences in performance. In addition we have individual level data for
workers in cooperatives and GESPA stores that allow us to examine issues surrounding job
satisfaction.
Our method is an insider econometrics case study (for a review of this and closely related
methods see Ichniowski and Shaw, 2003 and Jones and Kato, 2011). The overwhelming bulk of
4
the literature on EOFs, especially majority EOFs, has adopted a firm-level approach. While this
is a valuable method, as is widely recognized there are potential problems with this empirical
strategy including issues surrounding measurement error, endogeneity and omitted variables. The
case study approach enables more thorough investigation of the ramifications of important
institutional realities in cooperatives, e.g. the co-existence of member and non-member workers,
issues which typically are not afforded central attention in larger firm-level studies. Thus by
providing what is apparently the first econometric case study of a firm with substantial employee
ownership, potentially we provide a valuable complement to the body of findings derived from
firm level studies, as well as a contribution to a literature that is of growing importance.
The plan of this paper is as follows. The next section highlights key institutional features
of our case and uses new data to provide descriptive statistics. In the third part we review key
theoretical and empirical literature in areas of interest. In the main section of the paper we first
describe the first difference approach that is the basis of our estimating framework as well as our
econometric findings. We follow this by providing additional evidence that relates to some of the
mechanisms that underpin our key empirical findings on comparative performance. We also
analyze new data on job satisfaction for workers in cooperatives and GESPAs. In the final
section we offer concluding remarks and discuss the implications of our study.
II. The Case, Institutional Framework and Descriptive Statistics
Our case is the Eroski chain which is headquartered in Northern Spain. The Eroski group
is a diversified company with different activities including sport and leisure outlets though its
core businesses are supermarkets and hypermarkets, which are the focus of our investigation.
5
Both supermarkets and hypermarkets stores sell similar items, although there is some variation in
the range of items sold, since the outlets are of different size ranging from floor space of 1950
sq.m. to 12, 853 in hypermarkets and 162 to 2500 sq. m. for supermarkets. Smaller supermarkets
carry a product assortment essentially in the food area that is a subset of the product mix offered
by larger stores. Each hypermarket is divided into three basic divisions-- food, clothing and
domestic goods. A key strategy is to sell rather standard products to a wide range of customers
with all items in stock on display, and self-service is the main form of service in most
departments. For the bulk of employees the main tasks are to receive goods, shelve items, and
maintain the appearance of their department. In departments such as specialized foods it appears
that customers are apt to call on the expertise of sales clerks more often than in other
departments such as basic foods.
The overwhelming bulk of these retail outlets are in Spain (2398 of 2441), with the
remainder in France and Andorra, though amongst those 2398 units it is the 109 hypermarket
stores and 705 supermarket stores that are at center stage. Total employment in 2007 was 50, 587
and on the basis of total retail space Eroski was the third largest retail chain in Spain in that
year.1
To enrich our understanding of the institutional realities at the case we read various
materials that the company provided and we also made repeated visits to the firm headquarters
where we have had extensive discussions with senior personnel. Many of these personnel had
made frequent and recent visits to branch stores and thus had intimate knowledge of these
branches. We have also spent considerable time on field trips including multiple visits to outlets
1To allow for adjustment costs, in our study we will focus on hypermarket and supermarket stores
that have been part of the Eroski chain for at least 6 months. There is no newly acquired store in our sample
of stores for which we have reliable data over the period of February 2006-May 2008.
6
of this company in the Basque region as well as different areas of Spain; during each of these
visits we interviewed the store manager and employees.
The retail chain began operations in Northern Spain in 1969. Most of the cooperatively
owned stores are in the Basque region and in these cooperatives employees have substantial
employee ownership. To sustain growth in the chain (and with an eye to becoming a leader in
retailing throughout Spain), in 1997 Eroski began to acquire or open stores in other parts of
Spain. Some of these stores are cooperatives. In others, known as GESPA, employee members
have ownership stakes, though they are more modest than in cooperatives. Other stores remain as
conventional firms. Hence, from the perspective of the extent of employee ownership, there are
three distinct types of store. In turn, these ownership differences result in considerable
differences in the structure and functioning of stores in the three categories.2
These differences are most apparent for worker-members in Eroski cooperatives who,
compared to workers elsewhere in Eroski, have unusual opportunities to participate in both
ownership and decision-making. Equally it is important to realize that, usually, there are non-
member workers in the workforce. While many of these “non-members” are prospective
members on probation, they also include workers on temporary contracts. In the main it appears
that workers on temporary contracts are quite low skilled, in such positions as cashiers. As is
shown in the descriptive statistics reported in Table 1 in hypermarkets this non-member
workforce in Eroski cooperatives, averages about 24% of the total workforce and is somewhat
higher in supermarkets.
Effectively all coop workers who work under permanent contracts are expected to
2 We should also note that Eroski, as are other MCC cooperatives, is supported by a web of
institutions (see, e.g., Joshi and Smith, 2008; Arando et al, 2011).
7
become members and, as such, significant worker owners. While there is no fixed probationary
period, at some point (usually not less than six months) the immediate supervisor of the
candidate for membership, after soliciting opinions of other coop workers, makes a
recommendation with the ultimate decision concerning membership being made by the store’s
manager. In selecting prospective members, a key requirement is the willingness and ability of
the candidate to commit to a substantial capital contribution, a sum that currently is about 6000
Euros which in 2009 amounted to about 30% of the average annual remuneration in an Eroski
store3. While this initial stake remains substantially individually owned, about 20 % of it is
allocated to collectively owned reserves. The member’s stake receives an interest rate that is
related to, though usually set above, the market rate --in 2008 it was 8%. Over time, as Eroski
makes profits, these individual stakes grow as distributions from surpluses are credited to these
individuals’ capital accounts.4
Cooperative members also have opportunities to participate in decision-making in ways
substantially beyond those available to workers in other Eroski stores. Thus worker-members are
able to attend the AGM (Annual General Meeting), though the large number of members
requires that this is done on a representative basis. More importantly perhaps, worker-members
are able to be elected to the Governing Council (the Board of Directors) and the Social Council
(the body responsible for determining many matters of interest to workers, such as working
From Table 1 we see that in hypermarkets the average size of
these stakes was quite considerable having grown to 33,295 Euros and in supermarkets to more
than 26,000 Euros by March of 2008.
3 However the capital contribution requirement is not as onerous as it appears since it can be
spread over 5 years. In addition, possibilities exist for some new members to use previously paid
unemployment premia towards these stakes, since cooperative members are guaranteed job security.
4 For more detail on these features of Mondragon cooperatives, see Thomas and Logan, 1982 and
Bradley and Gelb, 1982 and Arando et al. (2011)
8
conditions.). At the same time, the potential influence of worker members in Eroski cooperatives
is necessarily circumscribed since, unlike in manufacturing cooperatives, there is another large
group of members in Eroski, namely consumer-members. The Governing Council comprises
equal numbers of representatives for worker- and consumer-members. Finally cooperative
members participate in joint labor-management meetings at the store level. From Table 1 we see
that, as a proportion of scheduled working hours spent in such meetings per month (INVOLVE)
averaged 0.24 percent for workers in COOP Hypermarket stores.
The understandings and implicit policies concerning job security and remuneration are
potentially of crucial importance to the functioning of cooperatives. A key incentive for workers
to want to become and to remain as coop members is job security—no coop members have ever
been laid off.5 In addition, cooperatives have wage structures that are much more compressed
and more flexible than in firms outside the group. Thus the norm is for coop members in non-
managerial positions to receive a premium of at least 20% over their outside counterparts. Also
the internal wage differences are compressed, with the usual ratio of top-bottom not exceeding
5:1.6
Turning to the Eroski stores known as GESPA, as in cooperatives individuals can
become members and have individual ownership stakes though membership in GESPA which
requires a capital stake that is about half as large as in a coop, about 3,000 Euros. This represents
about 25% of the average annual earnings for workers in a GESPA store. However, and unlike in
As such top managers tend to receive lower earnings than do managers in conventional
retail stores (by some estimates about 30% below outside rates.)
5 There have been occasional instances of store closures. In such instances members are always
offered comparable employment in nearby Eroski stores.
6 There are some strategic positions for which the gap is higher, sometimes approaching within-
business differences of 8:1. Still these differences are much more compressed than in comparable
capitalist retail chains.
9
cooperatives, not all permanent workers in Eroski stores are required/expected to become
members. In particular, when the stores were acquired, existing workers were not obliged to
become members. However, new workers who are offered permanent contracts are expected to
become members. These individual membership stakes earn an interest rate determined in a
similar way to individual stakes of coop members. Over time, as Eroski makes profits, these
individual stakes grow as distributions from surpluses are credited to individuals’ capital
accounts. However reflecting lower initial levels and shorter average lifethe first GESPA
began in 1997, whereas cooperatives have existed since 1969—as of March 2008 the average
individually owned stake in GESPA was substantially less than in Eroski cooperatives. Thus
from Table 1 we see that in GESPA hypermarkets this STAKE averaged only a little over 2,500
Euros.7
7 The average stake of 2,500 Euros is actually lower than the amount of the initial capital contribution
required for GESPA membership (3,000 Euros). This seemingly anomalous finding is largely due to the fact
that the initial capital contribution required for GESPA membership can be spread over five years, and that as a
result of the relatively young age of GESPA stores many GESPA members have not completed their required
capital contribution.
Membership in GESPA, as with membership in cooperatives, provides what is
effectively 100% job security—no GESPA members have ever been laid off, and in the few
instances of GESPA store closures, members have always been offered alternative employment
nearby. However, while GESPA members are able to be elected to the Social Council, they are
ineligible to attend the AGM or serve on the Board. As such the scope and nature of a member’s
opportunities to participate in control and membership in a GESPA is substantially below that for
members in cooperatives. Indeed in many interviews we heard views expressed that GESPA
membership was widely regarded as a “second class” form of membership. Hence,
unsurprisingly, as we see from Table 1, membership levels in GESPAs were much lower than in
COOPs—averaging almost 61%. Also participation in joint labor management committees was
10
much less developed (from Table 1 we see that average INVOLVE in GESPAs was only about
one tenth as large as in COOPs.)
From the perspective of employee ownership, all stores with conventional ownership in
the Eroski chain, and unlike other Eroski stores, do not provide opportunities for employee
ownership or special structures through which employees can participate in decision making. At
the same time it is important to emphasize that all of these stores represent acquisitions of what
were capitalist firms. As such many workers in these stores had worked for the previous
capitalist owners. Now they experience working as part of a cooperative chain one feature of
which is for all workers in all stores to be subject to key features of the same set of HR policies.
The language contained in various internal company documents and the associated institutional
arrangements strongly suggests that by working in a store within a coop chain, the lot of these
continuing workers might be expected to have improved --arguably they are subject to better
working conditions and better treatment by managers than previously. For example, the workers
in these conventional stores are encouraged to participate in meetings and there are policies
concerning meetings between employees and supervisors and annual development discussions
even though they cannot, as in other stores in the group, become owner-members. Also there is a
raft of policies that encourage training and skill formation. 8
So far as wage setting and employment are concerned in stores with conventional
ownership, all workers (including non-members in cooperatives and GESPA and workers in
8 The need for such policies emerges from our discussions with managers at the chain who
emphasized that the firm’s way of operating, or a key part of its competitive strategy, was that employees’
discretionary effort mattered for company performance. Employees are needed for more than just being
there, and high turnover of employees is not desirable. This is reflected in the company’s written HRM
strategy which emphasizes skill development of employees and the management capabilities of
supervisors, and also career development and job rotation. These issues were also raised in our
discussions with management members, suggesting genuine commitment to the HRM strategy. This
perspective applies to all types of stores.
11
these conventional firms,) receive no less than the wage rates that are set out in the collective
agreement that is negotiated by the retail workers trade union and which applies to all retail
workers.
Based on the preceding discussion we believe that our case is a good one in which to test
propositions in three related area. Our key interest is in comparative performance, especially how
cooperatives will perform compared to conventional firms (with no employee ownership). We
are also interested in how well cooperatives perform compared to firms with modest degrees of
employee ownership. The second questions concerns mechanisms--if there are productivity
differences, what accounts for these differences? For example, what will be the impact of
varying degrees of employee ownership and participatory practices (which vary across the three
types of stores) on organizational performance? Third, what are the implications of these
differences in organizational form for worker outcomes? While we know that cooperative
members receive higher earnings than their peers in conventional stores, what is the situation
concerning job satisfaction among workers in different stores? In the next section we review
theoretical and empirical work emerging from different literatures that relates to these questions.
III. Theory and Previous Empirical work
In this section our review of theoretical and empirical evidence in two areas is interwoven
with references to the previous discussion of key institutional features at Eroski so that we end
up with specific predictions for different types of stores
(1) Comparative Performance
For majority EOFs, the economic theory of the LMF yields conflicting predictions
about the productivity effects of worker participation in ownership and is thus inconclusive
12
concerning the expected comparative performance of EOFs and conventional firms.9
Turning to the empirical literature,
Much of
the early influential theoretical work (e.g., Vanek, 1970) argued that co-operative firms would
generate very strong incentives for labor resulting in high technical efficiency of labor. By
contrast other studies were more pessimistic concerning the expected performance of PCs.
Alchian and Demsetz (1972) and Jensen and Meckling (1979) argue that productivity will be
lower in a cooperative because efficient monitoring of workers requires the monitor to be the
claimant on the firm's profits and that the cost of monitoring increases with the number of
monitors. Another influential paper is Holmstrom (1982) who argues that effort level is expected
to be beset with free-rider problems and thus sub-optimal when work takes place in teams (as is
expected to be the case in manufacturing PCs). These pioneering theoretical papers have elicited
a voluminous amount of responses and theoretical objections. Thus Macleod (1988) shows how,
in a repeated game framework, effort supply in LMFs need not be below that in conventional
firm. Others point to different benefits of PCs. Thus cooperatives are expected to be more
productive than conventional firms because incentives (financial participation), peer group
pressure (horizontal monitoring) and the close identification of cooperative members with the
firm will elicit greater effort from workers (Jones and Svejnar, 1985; Fitzroy and Kraft, 1987).
Subsequent work (reviewed in Bonin et al. 1993; Dow, 2003) also recognized that predictions
are more nuanced once it is recognized that “real world” LMFs may not have 100% membership.
10
9 See Bonin, Jones, and Putterman (1993), Jones and Pliskin (1991) and Dow (2003) for surveys.
often the relative performance of conventional firms
and PCs has been estimated by comparing subsample means of measures such as value added per
10 One general point to note is that studies of technical efficiency in retailing, are thin on the
ground. For cooperatives, an early investigation of British retail cooperatives is Jones (1987). So far as
studies of the impact of HR on business performance in retailing is concerned prior work include Ben-Ner
et al. (1999) and Jones, et al.,( 2006, 2009).
13
worker using data on both conventional firms and cooperatives.11 Most econometric evidence
has been obtained from samples exclusively of producer cooperatives.12, 13,14 A few studies have
estimated production functions using data on both conventional firms and cooperatives.15
Amongst these perhaps most noteworthy is Craig and Pencavel (1995) who carefully gathered
data for plywood cooperatives and conventional firms in the Pacific Northwest in that industry.
The authors estimate separate Cobb Douglas production functions for several types of firms
including cooperatives and conventional firms. They find that cooperatives are between 6 and
14% more efficient than the principal conventional firms though there is little difference between
the efficiency of the unionized and classical mills. More recently Jones (2007) also assembles
data for conventional firms and PCs in the same industry, namely the Italian construction sector.
However no evidence is found that cooperatives are more efficient than conventional firms.16
11For example, George (1982).
In
sum, it would seem that a reasonable conclusion based on the research to date for majority EOFs
is that there is no strong evidence that either cooperatives or conventional firms have a sizeable
and persistent significant edge in performance over other organizational forms. Equally it is
12 Some exceptions are Jones (1987), Lee (1988), Berman and Berman (1989), Estrin (1991) and
Craig and Pencavel (1995).
13For example, see Defourney, Estrin, and Jones (1985), Estrin, Jones, and Svejnar (1987), and
Jones and Svejnar (1985).
14 Using estimates of how the productive efficiency of firms varied with respect to measures of
financial and decision making participation, the authors of these studies estimated the efficiency of a
typical cooperative relative to a firm with no worker participation. Since the samples of cooperatives
often exhibited considerable variation over both firms and time in the degree of worker participation, the
estimated productivity effects might be reliable. However, other things remaining the same, one would
prefer a sample of both conventional firms and cooperatives since the variance of the prediction errors is
lower for observations that are similar to those in the sample than for atypical ones.
15 George (1982), Jones (1987, 2007), Conte and Svejnar (1988), Lee (1988), Berman and
Berman (1989), Estrin (1991) and Craig and Pencavel (1995). But only the papers by Jones, Lee, Estrin,
Craig and Pencavel (1995) have focused on the relative technical efficiency of cooperatives.
16 For Mondragon cooperatives, the only econometric study is Martin (2000). This unpublished
study reports findings for a few firms (in machine tools) that do not yield a clear picture.
14
apparent that there is a need for more targeted research. For example, the most frequently cited
comparative study is probably that of Craig and Pencavel (1995). However, while the quality of
the data the authors use is most impressive arguably the robustness of the findings are somewhat
diminished by the relatively small size of the data set (170 observations for 34 mills), and the use
of a problematic measure of capital in the production function estimates. Also not all studies
have taken into account key features of the institutional realities, such as the variation in
membership amongst firms.
While economic theory is ambiguous and empirical findings are unclear, based on that
literature together with our knowledge of the institutions at Eroski (as discussed in section II), we
make the following prediction concerning the performance of Eroski co-ops. Even though the
formal arrangements in Eroski cooperatives are somewhat short of what is envisaged in the pure
theory case of the LMF, Eroski cooperatives do provide high levels of ownership and
participation as well as substantial job security for members. Compared to non-members, front
line workers benefit from receiving wage premia and high levels of training. We expect that
these arrangements will lead to a more committed and motivated workforce who would expend
more discretionary effort and work harder and smarter than the workforces in GESPA or
conventional stores so that cooperatives are expected to display much higher levels of efficiency.
Another source of performance advantage of cooperatives might stem from the possibility that
high levels of peer monitoring in cooperatives would lead one to expect coop labor forces to
have fewer layers of supervisors than in conventionally owned stores.17
For firms with minority EO and/or more limited employee involvement a body of
17 Agirre et. al (2010) argue that organizational commitment in cooperatives facilitates market
orientation which leads to better results.
15
theoretical and empirical literature suggests that an individual change in organizational design,
such as modest employee ownership, is expected to be sufficient to produce sustained benefits to
the firm.18 By contrast other literature argues that, for sustained benefits, complementary
measures are needed. An individual initiative when introduced alone may be insufficient to lead
to persistent gains. For example, employees might need more sharing of enterprise rewards
through financial participation, such as employee stock ownership to accompany teams lest their
commitment to teams becomes undermined (Milgrom and Roberts, 1995, Ben-Ner and Jones,
1995, and Kato and Morishima, 2002) The empirical literature often reports that the
productivity-enhancing effects of individual practices introduced alone may be short-lived since
initiatives lack a complementary mechanism, such as also delegating power to front-line workers
(e.g., for quality circles, Levine, 1995 and Jones and Kato, 2011). By contrast, and broadly
speaking, the empirical literature also finds evidence in support of such complementarities (e.g.,
between employee ownership and employee involvement, such as the shared capitalism studies
of Freeman et. al, 2008, or for Japan Kato and Morishima (2002) or for UK firms, Pendleton and
Robinson (2008).19
While economic theory for firms with minority ownership and control is ambiguous and
empirical findings are unclear, our knowledge of the institutions at Eroski (as discussed earlier in
section II) and that literature leads us to make the following prediction concerning performance
in Eroski GESPAs. While GESPAs do provide a reasonable level of employee ownership and
job security for members, the extent of employee involvement and financial participation is
18 See for example reviews in Blinder (1990) and in Blair and Kochan (2000).
19 However, it is also important to recognize that there do not appear to be any studies that have
investigated these propositions within a business that is a cooperative and part of an elaborate set of
institutions outside the firm.
16
rather modest. In addition, front line workers in GESPA stores do not enjoy the wage premia that
coop members receive. Also GESPA workforces tend to suffer from more divisions than do coop
workforces because of lower membership ratios—not all permanent workers accept the offer of
membership. For reason such as these, the synergies between ownership and participation,
especially compared to coop members, are muted and we would not expect GESPA stores to do
nearly as well as cooperatives. Furthermore, while some of the institutional arrangements might
be expected to result in worker members expending more discretionary effort and working
harder and smarter than workforces in conventional stores, these feelings of being “second class
members” might undermine such forces. Hence we have no strong expectations concerning
GESPA performance compared to conventional stores.
(2) Mechanisms
While as we have indicated above, the literature in the LMF tradition does draw attention
to mechanisms that might account for underlying differences in performance (such as the role of
peer monitoring), since the more recent literature on High Performance Work Systems (HPWS)
addresses this topic more forcefully we will lean on that literature to guide this part of our
empirical work. It points to the real possibility of an establishment boosting its performance by
adopting a variety of complementary new work practices (often called High Performance Work
Practices) and tapping into the ability of frontline workers to produce valuable local knowledge
through their collective efforts; and dealing with local shocks autonomously through
collaboration among themselves. Such diverse HPWPs are expected to be especially potent
within a majority EOF. The following three key elements of the HPWS are often emphasized20
20 See, for instance, Kochan and Osterman, 1994, Appelbaum, et. al. 2000 and Boning,
Ichniowski and Shaw, 2007. In addition, the literature sometimes stresses a synergy between the use of
17
First, in the HPWS, front-line workers will be given opportunities to exert discretionary
effort, acquire useful local knowledge, and share it with their co-workers, and higher-level
managers. The importance of providing such opportunities is self-explanatory. After all, a key
objective of the HPWS is to tap into frontline workers’ discretionary effort and ability to produce
valuable local information and deal with local shocks. Without such opportunities, there will not
be any performance gain.
Providing workers with such opportunities to produce useful local knowledge and share it
with management is not sufficient. Obviously if the interest of workers is not aligned with that of
the firm, workers will have little incentive to put forth effort and produce performance-
enhancing local information and share it with management. The interest alignment between
workers and the firm in cooperatives is fostered by two types of human resource management
policies: (i) financial participation schemes (such as employee ownership) by which the financial
wellbeing of workers is more tied to the final success of the firm; and (ii) information sharing
mechanisms through which management shares important information with workers, and fosters
their loyalty and commitment to the firm.21
The third element concerns the matter of ability and skills development. Even if frontline
workers are given an opportunity to produce valuable local knowledge and share it with
management AND have the appropriate incentive to do so, such useful local information may
never be generated or shared widely in the firm in the absence of appropriate ability and skill of
workers. As such, careful screening, recruitment and sustained training are often an integral part
information and communication technologies and the HPWS (e.g., Black and Lynch, 2004).
Unfortunately we do not have data on the use of such technologies.
21 In addition, job security, which is a central feature at Mondragon, can be an important
necessary condition for the High Performance Work System to function optimally. For the importance of
job security in the participatory employment system such as the Japanese system, see for example Levine
(1995) and Carmichael and MacLeod (1993).
18
of the High Performance Work System.
Our expectation is that opportunities for employee involvement and training will be more
extensive and economic incentives more powerful in cooperatives compared to other
organizational forms and that, in turn, these differences in key mechanisms will help to account
for differences in organizational performance.
IV. INSIDER ECONOMETRIC EVIDENCE
(1) Do COOP stores outperform other stores?
To capture the performance effects of differences in ownership structure, we estimate the
following first-difference model:
For hypermarkets,
(1) lnQit = βLlnLit + βcCOOPi + βmMARKETit + βyYEAROPENEDi
+ additional controls + εit
For supermarkets,
(2) lnQit = βLlnLit + βcCOOPi + βgGESPAi + βm MARKETit + βyYEAROPENEDi
+ additional controls + εit
where Δ indicates the first difference between month t and t-1; Qit is output (real sales) in store i
in month t; Lit is employment (measured by the number of full-time equivalent workers) in store
i in month t; COOPi is a dummy variable taking a value of 1 if store i is a coop store, 0
otherwise; and GESPAi is a dummy variable taking a value of 1 if store i is a GESPA store, 0
otherwise. In addition to labor (L), store space is often considered crucial capital input (K) in
retail service production (see, for instance, Jones, Kalmi and Kauhanen, 2006). For all Eroski
stores during the time period under study, however, month to month variations of store space are
19
zero and hence in our first-difference model, lnKit = 0.
Note that during the period of time under study no store changed its ownership type, and
that COOPi and GESPAi are time-invariant. In essence, we are estimating the effect on sales
growth (which is equal to lnQit) as opposed to the level of sales (Qit) of ownership types. The
extensive field research that we conducted at Eroski (in particular, repeated interviews with our
key informant as well as multiple managers) led us to believe that sales growth is indeed a
primary business goal of Eroski. Our focus on sale growth as a key performance measure is
consistent with what Eroski uses to gauge each store’s performance.
To make sure that the estimated coefficients on COOPi and GESPAi are capturing the
pure employee ownership effects, we include a number of controls. First and perhaps most
important, a store located in a rapidly growing market with rising population and average
household income will naturally experience a faster growth of sales. If a disproportionately
higher proportion of COOP stores are located in such rapidly growing markets as compared to
other stores and we fail to control for such a location effect, we will not be able to separate the
performance effect of COOP from the locatio effect. To control for such differences in the store
location’s market condition, we include MARKETit where MARKETit is monthly market index
in month t for the area which store i serves. The monthly market index is provided by the
Spanish National Statistical Institute, and is considered one of the most authoritative market
indicators in Spain.
Second, we consider YEAROPENEDi = the year store i was opened. We have been told
by our informants at Eroski that due to the standard lifecycle model of retail stores, younger
stores tend to grow faster than older stores. YEAROPENEDi will control for such a lifecycle
effect as well as any cohort effects of individual stores.
20
Third, we also include constant (to capture an Eroski-wide time trend which is common
to all Eroski stores regardless of its ownership types), monthly dummy variables (to capture
seasonality of retail sales), and year dummy variables (to control for year time effects) as
additional controls.22
Finally, as in the case of any fixed effect/first-difference model, our first-difference
model controls for unobserved time-invariant heterogeneity of stores that may be correlated with
the level of real sales.
Table 2 reports the OLS estimates of Eq. (1).23
22 We also consider a full set of interaction terms involving monthly dummy variables and year
dummy variables. Our main results change little with the use of such a full set of interaction terms though
there is slight efficiency loss. These as well as all other unreported results are available upon request from
the corresponding author.
For hypermarkets, as shown in the first
column, the estimated coefficient on COOP is positive and significant at the 5 percent level,
confirming that COOP stores grow faster than GESPA stores, ceteris paribus. The size of the
estimated coefficient suggests a plausible growth rate advantage of COOP stores over GESPA
stores, i.e., on average each month COOP stores grow faster than GESPA stores by 0.2
percentage points (or 2.4 percentage points per year). The estimated coefficients on the control
variables have the expected signs and are statistically significant. Specifically, the estimated
coefficient on lnLit is positive and significant at the 1 percent level, and the size of the
coefficient implies that the output elasticity of labor in the underlying Cobb-Douglas production
function is a little over 0.5. We also find the estimated coefficient on MARKETit to be positive
and significant at the 1 percent level, confirming that a store located in a rapidly growing market
with rising population and average household income will experience a significantly faster
growth of sales, and hence that it is important to control for the market conditions of the store
23 All standard errors are clustered at individual store levels.
21
location. Finally, the estimated coefficient on YEAROPENEDi is positive and significant at the 5
percent level, suggesting that young stores indeed grow faster than old stores. This finding is
consistent with the lifecycle model of retail stores as expounded by our informant at Eroski.
As shown in the second column of the table, for supermarkets, we find no evidence for
the growth rate advantage of COOP stores -- the estimated coefficient on COOP is very small
(actually negative) and highly insignificant. Our failure to find any significant difference in sales
growth between COOP and GESPA stores for supermarkets did not surprise our key informant
at Eroski who pointed out the large heterogeneity of stores in supermarkets as a possible reason
for the finding. He then suggested us to focus on a particular subgroup of supermarkets, called
Supermarket City. Supermarket City is essentially a group of supermarket stores that are smaller
than other supermarkets and are still somewhat reminiscent of intimate, small neighborhood
groceries. As such, having “better customer service” employees is particularly important for
Supermarket City. A major strength of COOP stores lies in the fact that COOP helps employees
develop a sense of ownership and hence makes them more committed employees who are willing
and capable of providing closer and helpful attention to their customers. Such a COOP advantage
is probably more relevant to Supermarket City than other supermarkets. In short, we are more
likely to detect a positive performance effect of COOP in the market segment of Supermarket
City.
The last column of Table 2 reports the results for Supermarket City. Reassuringly the
estimated coefficient on COOP is now positive and statistically significant at the 5 percent level.
The size of the estimated coefficient implies a considerable 0.7 percentage-point growth rate
advantage enjoyed by COOP stores in the market segment of Supermarket City over
conventional stores in the same market segment (note that there is no GESPA store in this market
22
segment).
(2) How do COOP stores outperform other stores?
Our earlier discussion of the theoretical literature drew attention to the potential role of
various mechanisms within cooperatives that might help to account for a productivity edge. In
this section we highlight the role of three key elements. First are opportunities for employee
involvement. From the discussion in section II we saw that diverse opportunities existed for
employee involvement in Eroski cooperatives. In measuring the extent of variation in employee
involvement opportunities across different types of stores we use INVOLVEi (the monthly
average for store i during the time period under study of the proportion of scheduled work hours
spent on joint labor-management meetings). We use the monthly average during the entire time
period under study to gauge the extent of employee involvement opportunities at each store, for
it is implausible that the strength of employee involvement at each store changes from month to
month.
As shown in Table 1, for both hypermarkets and supermarkets, COOP stores allow for
much more employee involvement than other stores. For instance, the proportion of scheduled
working hours spent on joint labor-management meetings at the store level per month
(INVOLVE) was on average 0.24 percent for COOP Hypermarket stores as opposed to only 0.02
percent for GESPA Hypermarket stores. Likewise, the average INVOLVE was 0.44 percent for
COOP stores in the market segment of Supermarket City as opposed to negligible 0.002 percent
for conventional stores in the same market segment. As such, it is plausible that COOP stores
outperform other stores in part by providing their employee owners with more opportunities to
produce useful local knowledge and respond effectively to local shocks.
The second mechanism is incentives for workers to take advantage of employee
23
involvement opportunities. From our earlier institutional discussion we believe that STAKEi =
average stake of employee owners (monthly average of store i during the time period under
study) represents the most important part of financial incentives at Eroski, though clearly other
incentives exist, such as annual distribution of surplus.24
In addition to STAKEi, our data provide yet another dimension of the overall strength of
incentives, MEMBERi= proportion of workers who are COOP or GESPA members (monthly
average of store i during the time period under study). While STAKE captures the intensity of
incentives (how big a deal it is for the average employee owner to help her store outcompete its
competitors), MEMBER measures the scope of incentives (what proportion of the total labor
force in the store has some stake in the firm). As shown in Table 1, COOP stores tend to have
considerably higher proportions of employees with some stake in the firm.
Table 1 shows that workers in COOP
hypermarket stores have a stake in their firm that is more than thirteen times as big as that of the
average member-worker in a GESPA hypermarket store. (Unsurprisingly for supermarket stores
in the segment of Supermarket City, employees in conventional stores have no stake in the firm).
Again, the HPWS literature suggests that employees in COOP stores have stronger incentives to
take advantage of employee involvement opportunities; produce valuable local knowledge; and
respond quickly and effectively to local shocks without invoking lengthy formal involvement of
supervisors. It follows that COOP stores outperform other stores.
The third crucial feature is the emphasis on training and skill formation in cooperatives.
24 Individual ownership stakes receive a return that is more or less guaranteed--an interest rate
reflecting market rates. Also, if there is a surplus (profit) part of this is distributed to owners as a bonus which
is also proportional to STAKE. There is, however, another relatively small part of the surplus that goes to
workers-- a kind of profit sharing, which is not proportional to STAKE. In sum, it is safe to assume that most
of the annual distribution of surplus that goes to members is proportional to STAKE (in their role as capital
providers), and that STAKE will measure the strength of financial incentive for employees accurately.
24
Fortunately our data allow us to construct TRAININGi = proportion of scheduled hours spent on
training in general (monthly average of store i during the time period under study). Table 1
shows a somewhat mixed picture. For hypermarkets, COOP stores on average devote less time to
training than GESPA stores whereas for supermarkets in the Supermarket-City segment, COOP
stores on average devote more time to training than conventional stores. However, our data do
not provide information on the amount of HPWS-relevant training (such as problem solving;
team work; customer relations). This measurement issue makes it somewhat difficult to interpret
the results on training.
Specifically, we estimate the following first-difference model:
(3) lnQit = βLlnLit + βhHPWPi + MARKETit + YEAROPENEDi
+ additional controls + εit
For HPWPi, as discussed above, we consider INVOLVEi, STAKEi, MEMBERi, and
TRAININGi. Table 3 shows the OLS estimates of Eq. (3) for the Hypermarket segment. Overall
the results are supportive of predictions drawn from the HPWP PARADIGM. All estimated
coefficients have the expected positive sign and two of them are statistically significant at the 1
percent level (INVOLVEi, and STAKEi).
As shown in Table 4, the estimates of Eq. (3) for a subgroup of City Supermarket stores
are less precise and mostly insignificant although the estimated coefficient on MEMBERi is
close to being statistically significant at the 10 percent level.25
25 We also estimated a fully nested version of Eq. (3) with all four HPWP variables considered
simultaneously. The results, which are available upon request, turned out to be quite robust to the use of
such a fully nested specification although the estimates are slightly less precise due to multicollinearity as
expected.
25
V. Worker Outcomes and the Role of Job Satisfaction
Though the main objective of the paper is to investigate the effect on organizational
performance of cooperatives, we will provide suggestive evidence on the effects on worker
outcomes (especially job satisfaction) of cooperatives. What cooperatives imply for worker
outcomes is an important research question in its own right. However, a study of the implications
of cooperatives for worker outcomes will also provide potentially valuable insight on the nature
and sustainability of cooperatives as a high performance work system.
In the literature on HPWS, there is an ongoing debate over the consequences of HPWS
for worker outcomes. While some argue that HPWS produces gains for both firms and workers
(e.g., Appelbaum, et. al, 2000), Godard, (2001 and 2004) argues that HPWS may make work
more intense and stressful, thus resulting in more worker discontent. There is also a literature
that argues that in organizations with well developed systems providing for employee
involvement, work expectations are apt to be unrealistically high. For example in their cross
national study of the extent of employee involvement in Europe in the 1970s, International
Research Group (1981) found that worker satisfaction with arrangements in countries with well
developed formal arrangements for employee participation (such as in Yugoslavia) was lower
than in many places with more rudimentary schemes.
While the available empirical work is rather limited, there appears to be a similar
division in the literature concerning majority EOFs and cooperatives. Most studies have
investigated the US plywood cooperatives. Thus while Greenberg (1986) finds higher levels of
job satisfaction in the cooperatives, Long (1982) finds no relationship between ownership stake
and job satisfaction. While Rooney (1984) finds evidence of lower injury rates among workers
in majority EOFs, by contrast Rhodes and Steers (1981) find no differences in accident rates and
26
Grunberg and Greenberg (1996) find a poorer safety record in cooperatives. To some degree this
murky picture emerging from previous work on job satisfaction and related outcomes in
cooperatives reflects the use of what are tiny and apparently non-representative samples.
For the case of Eroski, we do know that worker-members in cooperatives receive
substantially higher earnings compared to their peers in other stores within Eroski. At the same
time we have no clear cut predictions as to what to expect to find concerning job satisfaction.
While many of the features of cooperative membership, such as high levels of employee
involvement and high wages can be expected to lead to high levels of job satisfaction, such
owner-workers may also work harder and be subject to higher levels of stress than others. They
are also apt to have much higher levels of expectations from working in cooperatives (than do
workers elsewhere), expectations which, in certain circumstances, may be easily frustrated.
To investigate the implications of differences in organizational structure for job
satisfaction, we are fortunate to have access to micro data for all workers in Eroski hypermarkets
(n=4328). These data were collected from a worker s survey conducted in all Eroski
hypermarkets (including both COOP and GESPA stores) in 2008. On a 5 point scale, workers are
asked to assign a numerical value to their level of satisfaction (with 5 being the most satisfied) in
response to each of 68 questions concerning various aspects of working lives and work
environment.
To capture the level of overall worker satisfaction, following Bartel et al. (2003), we use
employee responses to all 68 questions and produce an Employee Attitude Index (EAI) for each
27
individual worker.26
To provide more rigorous evidence on the satisfaction gap between COOP and GESPA,
we estimate the following equation:
The average EAI was 3.36 and 3.60 for workers in COOP and workers in
GESPA respectively, suggesting that the overall level of job satisfaction is actually lower for
workers in COOP stores than in GESPA stores.
(4) EAIi = α + β1COOPi + β2lnQi + β4lnMARKETi
+ β4YEAROPENEDi + β5MALEi + β6BASQUEi
+ additional controls + εi
where EAIi = EAI of worker i; COOPi = 1 if worker i works for a COOP store, zero otherwise;
lnQi = annual average of monthly sales growth rates during 2007 of the store for which worker
i works; MARKETi = annual average of monthly growth rates of market index during 2007 of
the store for which worker i works; YEAROPENEDi = year opened of the store for which
worker i works; MALEi = 1 if worker i is male, zero otherwise; and BASQUEi = 1 if worker i
filled out the questionnaire in Basque, 0 otherwise (each worker was allowed to fill out the
questionnaire either in Spanish or in Basque). Additional controls include occupational dummy
variables; tenure dummy variables; and worker status (temporary or permanent) dummy
variables.
The OLS estimates of Eq. (4) are presented in Table 4. Column (i) confirms that the level
of overall worker satisfaction is indeed significantly lower in COOP than in GESPA, after
controlling for a variety of individual and store characteristics. Column (ii) shows the OLS
estimates of Eq. (4) augmented by an interaction term involving both COOPi and lnQi. The
26 We experimented with alternative ways of capturing job satisfaction by constructing measures that
use a narrower set of questions as well as a principal component analysis. Findings were largely insensitive to
the use of such alternative measures.
28
estimated coefficient on the interaction term is positive and statistically significant at the 1
percent level, suggesting that the negative satisfaction effect of COOP is mediated when store
performance (sales growth) improves. Or conversely the negative satisfaction effect of COOP
will be amplified when store performance worsens.
One interpretation of these findings is that Eroski’s COOP stores can be viewed as
majority EOFs that employ key mechanisms that constitute a high-powered High-Performance
Work System. These co-op stores are more efficient and COOP workers have bigger financial
stakes and voice, and also receive higher wages. However, employee owners with high stakes in
the firm are expected to go beyond routine work and to engage in a variety of problem solving
activities. Consistent with the arguments of those who do not expect HPWPs to produce
consistent gains for both firms and employees (e.g. Godard, 2001), such workplaces can be quite
demanding and stressful. Interestingly in cooperatives, and in contrast to what has sometimes
been found for conventionally owned firms with HPWPs, stress is highest when sales growth is
relatively weak. An alternative interpretation of the finding is that by being significant
stakeholders, COOP workers at Mondragon probably expect more from their work, resulting in
high expectations and a higher likelihood of disappointment. Such workplace disappointment
may be particularly acute when their hard work does not result in performance improvement.
In sum, our evidence on low job satisfaction in cooperatives points to a need for
somewhat nuanced understanding of cooperatives as a HPWS.
VI. CONCLUSIONS
Recent years have seen a massive growth in employee ownership around the world. A
substantial volume of theoretical and empirical evidence has appeared that investigates the
29
performance of such firms, but it is inconclusive concerning the comparative performance of
EOFs and conventional firms, as well as the comparative performance of firms with majority and
minority EO. The evidence also shows how employee ownership has assumed a wide variety of
forms, including the producer cooperative in which majority control and ownership is vested in
the workforce. Amongst such labor managed firms, one example that has attracted close and
sustained attention by researchers are the Mondragon cooperatives. In this paper, we are
fortunate to use the first micro-econometric evidence for a Mondragon cooperative and, since
stores fall into three distinct ownership categories, we are able to contribute to some of these
debates.
By estimating first difference equations we find: (i) hypermarket stores with cooperative
ownership outperform GESPA stores; (ii) for supermarkets, the picture is more nuanced. Small
“city” supermarkets, those with cooperative ownership, are more productive than Eroski stores
that are conventionally owned. However for larger supermarkets, conventional owned stores
grow faster than both cooperatives and GESPA. These findings are supported by a series of
robustness checks.
We also provide additional evidence that bears on the mechanisms that help to explain
why cooperatives are better-performers. This evidence is consistent with those who argue that
cooperatives are better performers because cooperative members work under institutional
arrangements that differ from those facing workers in other firms. Specifically, compared to
workers in other firms, cooperative members have opportunities for substantial employee
involvement and training and also strong incentives because they have a large financial stake in
the firm. Cooperative members also have unusual job security and they work in firms with
earnings differences that are substantially more compressed than in comparable firms. We
30
attribute the failure to find an effect for more moderate combinations of employee ownership and
employee involvement to the absence in GESPAs of many of the factors that underpin the
cooperative advantage.
Overall our findings tend to lend support to those who cast doubt on the unconditional
supremacy of the Anglo-American shareholder model of corporate governance and advocate
employee ownership and shared capitalism as a viable and possibly even superior alternative to
the Anglo-American shareholder model. While it is unclear whether our findings apply equally
to the retail industry in other countries, we do note that evidence for the potential importance of
innovative work practices does exist for retailing elsewhere (e.g. for the UK, Jones, 1987 and for
Finland, Jones et al. , 2009.) At the same time we do not believe that our findings imply that
employee-owned enterprises are a universal panacea.
First, while cooperative members are substantially better paid than their peers in
comparable firms, when we use individual-level data to investigate job satisfaction among
workers in differing forms of organization we find that job satisfaction is lower for workers in
cooperatives than for GESPA workers. Though this may be a reflection of high worker
expectation in cooperatives, cooperatives may well be indeed a “high-stress work system”. The
overall assessment of cooperatives will need to be nuanced.
Second, clearly there are limitations to our approach--econometric case studies cannot
easily address concerns about selectivity and external validity (Jones et al., 2006). In the context
of our study, compared to firms examined in other studies, many potentially important features
of the cooperative model are apparently especially well developed among Eroski stores, such as
the existence of high membership ratios and average ownership stakes that are far bigger than in
other cooperatives outside Mondragon. The particular configuration of coop features available at
31
Eroski might help to explain our findings--the effects of cooperation that exist here than are more
powerful than have been found in other cases. Likewise, even within Mondragon, we are not
entirely sure that our findings are applicable to the industrial and banking sectors of Mondragon,
cases to which we plan to investigate in future work.
32
REFERENCES
Agirre, I; Aizpuru, A. and Lara, P. "La Orientación Al Mercado En La Organización
Cooperativa: Antecedentes Y Resultados En El Grupo Mondragon," Paper presented at
XII Congreso Nacional de Marketing, Oviedo. Spain, 2010.
Alchian, Armen A. and Demsetz, Harold. "Production, Information Costs, and Economic
Organization." American Economic Review, 1972, 62(5), pp. 777-95.
Appelbaum, Eileen; Thomas Bailey; Peter Berg and Kalleberg, Arne L. Manufacturing
Advantage: Why High-Performance Work Systems Pay Off. Ithaca and London: Cornell
University Press, ILR Press, 2000.
Arando, Saioa; Freundlich, Fred; Gago, Monica; Jones, Derek C. and Kato, Takao.
"Assessing Mondragon: Stability and Managed Change in the Face of Globalization," E.
Carberry, Employee Ownership and Shared Capitalism: New Directions and Debates for
the 21st Century. Ithaca, NY: ILR Press, Cornell University, 2011, forthcoming.
Bartel, Ann; Freeman, Richard; Ichniowski, Casey and Kleiner, Morris M. "Can a Work
Organization Have an Attitude Problem? The Impact of Workplaces on Employee
Attitudes and Economic Outcomes," National Bureau of Economic Research, Inc, NBER
Working Papers: 9987, 2003.
Ben-Ner, Avner and Jones, Derek C. "Employee Participation, Ownership, and Productivity:
A Theoretical Framework." Industrial Relations, 1995, 34(4), pp. 532-54.
Ben-Ner, Avner; Kong, Fanmin and Bosley, Stacie. "Workplace Organization and Human
Resource Practices: The Retail Food Industry," University of Minnesota, Retail Food
Industry Center, Working Paper 00-01, 1999.
Berman, Katrina V. and Berman, Matthew D. "An Empirical Test of the Theory of the Labor-
Managed Firm." Journal of Comparative Economics, 1989, 13(2), pp. 281-300.
Black, Sandra E. and Lynch, Lisa M. "What's Driving the New Economy? The Benefits of
Workplace Innovation." Economic Journal, 2004, 114(493), pp. F97-116.
Blair, Margaret M. and Kochan, Thomas A., eds. The New Relationship: Human Capital in
the American Corporation. Washington, D.C.: Brookings Institution Press, 2000.
Blinder, Alan S., ed. Paying for Productivity: A Look at the Evidence. Washington, D.C.:
Brookings Institution, 1990.
Bonin, John P.; Jones, Derek C. and Putterman, Louis. "Theoretical and Empirical Studies of
Producer Cooperatives: Will Ever the Twain Meet?" Journal of Economic Literature,
1993, 31(3), pp. 1290-320.
Boning, Brent; Ichniowski, Casey and Shaw, Kathryn. "Opportunity Counts: Teams and the
Effectiveness of Production Incentives." Journal of Labor Economics, 2007, 25(4), pp.
613-50.
Bradley, Keith and Gelb, Alan. "The Replication and Sustainability of the Mondragon
Experiment." British Journal of Industrial Relations, 1982, 20(1), pp. 20-33.
Burdin, Gabriel and Dean, Andres. "New Evidence on Wages and Employment in Worker
Cooperatives Compared with Capitalist Firms." Journal of Comparative Economics,
2009, 37(4), pp. 517-33.
Carmichael, H. Lorne and MacLeod, W. Bentley. "Multiskilling, Technical Change and the
Japanese Firm." Economic Journal, 1993, 103(416), pp. 142-60.
33
Conte, Michael A. and Svejnar, Jan. "Productivity Effects of Worker Participation in
Management, Profit-Sharing, Worker Ownership of Assets and Unionization in U.S.
Firms." International Journal of Industrial Organization, 1988, 6(1), pp. 139-51.
Craig, Ben and Pencavel, John. "Participation and Productivity: A Comparison of Worker
Cooperatives and Conventional Firms in the Plywood Industry." Brookings Papers on
Economic Activity, 1995, ( ), pp. 121-60.
Defourney, Jacques; Estrin, Saul and Jones, Derek C. "The Effects of Workers' Participation
on Enterprise Performance: Empirical Evidence from French Cooperatives."
International Journal of Industrial Organization, 1985, 3(2), pp. 197-217.
Doucouliagos, Chris. "Worker Participation and Productivity in Labor-Managed and
Participatory Capitalist Firms: A Meta-Analysis." Industrial and Labor Relations Review,
1995, 49(1), pp. 58-77.
Dow, Gregory. Governing the Firm: Workers Control in Theory and Practice. Cambridge, the
U.K.: Cambridge University Press, 2003.
Estrin, Saul. "Some Reflections on Self-Management, Social Choice, and Reform in Eastern
Europe." Journal of Comparative Economics, 1991, 15(2), pp. 349-66.
Estrin, Saul; Hanousek, Jan; Kocenda, Evzen and Svejnar, Jan. "The Effects of Privatization
and Ownership in Transition Economies." Journal of Economic Literature, 2009, 47(3),
pp. 699-728.
FitzRoy, Felix R. and Kraft, Korenelius. "Cooperation, Productivity, and Profit Sharing."
Quarterly Journal of Economics, 1987, 102(1), pp. 23-35.
Freeman, Richard; Kruse, Douglas and Blasi, Joseph. "Worker Responses to Shirking under
Shared Capitalism," National Bureau of Economic Research, Inc, NBER Working
Papers: 14227, 2008.
George, D. . "Workers' Co-Operatives in Denmark." Managerial and Decision Economics,
1982, 3, pp. 205-12.
Godard, John. "A Critical Assessment of the High-Performance Paradigm." British Journal of
Industrial Relations, 2004, 42(2), pp. 349-78.
____. "High Performance and the Transformation of Work? The Implications of Alternative
Work Practices for the Experience and Outcomes of Work." Industrial and Labor
Relations Review, 2001, 54(4), pp. 776-805.
Greenberg, Edward S. Workplace Democracy: The Political Effects of Participation. Ithaca,
N.Y., and London: Cornell University Press, 1986.
Grunberg, Leon; Moore, Sarah and Greenberg, Edward. "The Relationship of Employee
Ownership and Participation to Workplace Safety." Economic and Industrial Democracy,
1996, 17(2), pp. 221-41.
Holmstrom, Bengt. "Moral Hazard in Teams." Bell Journal of Economics, 1982, 13(2), pp. 324-
40.
Ichniowski, Casey and Shaw, Kathryn. "Beyond Incentive Pay: Insiders' Estimates of the
Value of Complementary Human Resource Management Practices." Journal of Economic
Perspectives, 2003, 17(1), pp. 155-80.
International Research Group ed. Industrial Democracy in Europe. New York: Clareendon:
Oxford University Press, 1881.
34
Jensen, Michael C. and Meckling, William H. "Theory of the Firm: Managerial Behavior,
Agency Costs and Ownership Structure." Journal of Financial Economics, 1976, 3(4),
pp. 305-60.
Johnson, Ana Gutierrez and Whyte, William Foote. "The Mondragon System of Worker
Production Cooperatives." Industrial and Labor Relations Review, 1977, 31(1), pp. 18-
30.
Jones, Derek C. "The Productive Efficiency of Italian Producer Cooperatives: Evidence from
Conventional and Cooperative Firms," S. Novkovic and V. Sena, Cooperative Firms in
Global Markets Volume 10 of the Book Series: Advances in the Economic Analysis of
Participatory & Labor-Managed Firms. Amsterdam: Elsevier/JAI, 2007, 3-28.
____. "The Productivity Effects of Worker Directors and Financial Participation by Employees
in the Firm: The Case of British Retail Cooperatives." Industrial and Labor Relations
Review, 1987, 41(1), pp. 79-92.
Jones, Derek C. and Pliskin, Jeffrey. "The Effects of Worker Participation, Employee
Ownership and Profit Sharing on Economic Performance: A Partial Review," R.
Raymond and R. Veljko, Ownership and Participation: International Handbook of
Participation in Organizations Vol.2. Oxford: Oxford University Press, 1991, 43-63.
Jones, Derek C. and Backus, David K. "British Producer Cooperatives in the Footwear
Industry: An Empirical Evaluation of the Theory of Financing." Economic Journal, 1977,
87(347), pp. 488-510.
Jones, Derek C.; Kalmi, Panu and Kauhanen, Antti. "How Does Employee Involvement
Stack Up? The Effects of Human Resource Management Policies on Performance in a
Retail Firm," Industrial Relations (forthcoming), 2009.
____. "Human Resource Management Policies and Productivity: New Evidence from an
Econometric Case Study." Oxford Review of Economic Policy, 2006, 22(4), pp. 526-38.
Jones, Derek C. and Kato, Takao. "The Impact of Teams on Output, Quality and Downtime:
An Empirical Analysis Using Individual Panel Data." Industrial and Labor Relations
Review, 2011, 64(2), pp. 215-40.
____. "The Productivity Effects of Employee Stock-Ownership Plans and Bonuses: Evidence
from Japanese Panel Data." American Economic Review, 1995, 85(3), pp. 391-414.
Jones, Derek C. and Svejnar, Jan. "Advances in the Economic Analysis of Participatory and
Labor-Managed Firms: Introduction," D. C. Jones and J. Svejnar, Advances in the
Economic Analysis of Participatory and Labor-Managed Firms. Greenwich, Conn., and
London: JAI Press, 1985, xi-xiv.
____. "Participation, Profit Sharing, Worker Ownership and Efficiency in Italian Producer
Cooperative." Economica, 1985, 52(28), pp. 449-65.
Joshi, Sumit and Smith, Stephen C. "Endogenous Formation of Coops and Cooperative
Leagues." Journal of Economic Behavior and Organization, 2008, 68(1), pp. 217-33.
Kato, Takao and Morishima, Motohiro. "The Productivity Effects of Participatory
Employment Practices: Evidence from New Japanese Panel Data." Industrial Relations,
2002, 41(4), pp. 487-520.
Kochan, Thomas and Osterman, Paul. Mutual Gains Enterprise: Forging a Winning
Partnership among Labor, Management, and Government (Hardcover). Harvard
Business School Press, 1994.
35
Kruse, Douglas; Freeman, Richard and Blasi, Joseph. "Do Workers Gain by Sharing?
Employee Outcomes under Employee Ownership, Profit Sharing, and Broad-Based Stock
Options," National Bureau of Economic Research, Inc, NBER Working Papers: 14233,
2008.
Kruse, Douglas L.; Freeman, Richard B. and Blasi, Joseph R. eds. Shared Capitalism at
Work: Employee Ownership, Profit and Gain Sharing, and Broad-Based Stock Options.
NBER Conference Report series. Chicago and London: University of Chicago Press,
2010.
Lawler, Edward E. III. High-Involvement Management: Participative Strategies for Improving
Organizational Performance. San Francisco: Jossey-Bass, 1986.
Lee, B. Productivity and Employee Ownership: The Case of Sweden. Stockholm: Sweden: Trade
Union Institute for Economic Research, 1988.
Levine, David I. Reinventing the Workplace: How Business and Employees Can Both Win.
Washington, D.C.: Brookings Institution, 1995.
Long, Richard J. "Worker Ownership and Job Attitudes: A Field Study." Industrial Relations,
1982, 21(2), pp. 196-215.
MacLeod, W. Bentley. "Equity, Efficiency, and Incentives in Cooperative Teams," D. C. Jones
and J. Svejnar, Advances in the Economic Analysis of Participatory and Labor-Managed
Firms. Volume 3. Greenwich, Conn. and London: JAI Press, 1988, 5-23.
Martin, Terence. "The Impact of Worker Ownership on Firm-Level Performance: A
Comparative Study," Unpublished Ph.D. Thesis, Yale University, 2000.
Milgrom, Paul and Roberts, John. "Complementarities and Fit: Strategy, Structure, and
Organizational Change in Manufacturing." Journal of Accounting and Economics, 1995,
19(2-3), pp. 179-208.
Pendleton, Andrew. "Incentives, Monitoring, and Employee Stock Ownership Plans: New
Evidence and Interpretations." Industrial Relations, 2006, 45(4), pp. 753-77.
Pendleton, Andrew and Robinson, Andrew M. "Employee Stock Ownership, Participation,
and Productivity," Paper presented at the IAFEP meeting. Hamilton College, 2008.
Rhodes, Susan R. and Steers, Richard M. "Conventional Vs. Worker-Owned Organizations."
Human Relations, 1981, 34(12), pp. 1013-40.
Rooney, Patrick M. "Worker Control: Greater Efficiency and Job Satisfaction." Economic
Forum, 1984, 14(2), pp. 97-123.
Thomas, H. and Logan, C. Mondragon: An Economic Analysis. Winchester, MA: Allen &
Unwin, 1982.
Vanek, Jaroslav The General Theory of Labor-Managed Market Economies. Ithaca, NY:
Cornell University Press, 1970.
36
Table 1 Summary Statistics
Sources: : Individual store-level monthly data from February 2006 through May 2008 provided by Eroski.
Note: Standard Deviation in parentheses.
Definitions of the Variable
Qit=real sales of store i in time t;
Lit=number of all full-time equivalent workers working in store i in time t
MARKETit=monthly market index of the area served by store i in time t
YEAROPENEDi = year in which store i was initially opened.
INVOLVEi = proportion of scheduled hours spent on joint labor-management meetings (monthly average of store i during the time period under
study)
STAKEi = average stake of employee owners (average of all employee owners of store i as of March 2008)
MEMBERi = proportion of workers who are COOP or GESPA members (monthly average of store i during the time period under study)
TRAININGi = proportion of scheduled hours spent on training (monthly average of store i during the time period under study)
Hypermarkets
Supermarkets
All
SUPERMARKET CITY
COOP
GESPA
COOP
GESPA
Conventional
COOP
Conventional
lnQit
.0021048
-.0004499
.0042155
.0060926
.0053058
.0104688
.0019786
(.166259)
(.234495)
(.159047)
(.169218)
(.174939)
(.182554)
(.116191)
lnL
it
.0003716
-.0016009
.0023519
.0039068
.0024821
.0068902
.0022351
(.04335)
(.066897)
( .09743)
(.061481)
(.087372)
(.139007)
(.102414)
MARKETit
.0034202
.0035085
.002913
.003626
.003759
.0028605
.0012249
(.115210)
(.107146)
(.112875)
(.096741)
(.104789)
(.115623)
(.099853)
YEAROPENEDi
1995.48
1999.904
1998.405
2000.626
1999.364
2000.18
2002.053
(5.46753)
(4.49017)
(4.74845)
(2.77468)
(4.94237)
(2.52702)
(1.84229)
INVOLVEi
.0024464
.0002408
.0033306
.0011858
2.23e-06
.0044161
.0000192
(.004767)
(.000727)
(.005705)
(.003443)
(.000126)
(.008445)
(.000332)
STAKEi
33295.79
2511.332
26270.68
865.6311
1.398661
23030.07
0
(8847.05)
(1010.40)
(8175.98)
(201.354)
(23.56457)
(10545.04)
(0)
MEMBERi
.7589575
.6075966
.7289384
.5180572
0
.6442763
0
(.073878)
(.135189)
(.118557)
(.153238)
(0)
(.1549173)
(0)
TRAININGi
.0074278
.0080867
.0138549
.0102537
.0062481
.0108354
.0059075
(.01298)
(.015204)
(.038580)
(.021452)
(.053389)
(.041531)
(.012875)
N
675
1420
4747
703
8001
967
321
37
Table 2 Sales Growth and Ownership Types: Insider Econometric Evidence
Dependent variable=lnQit
Hypermarkets
Supermarkets
Supermarket City
lnLit
0.552***
[6.57]
0.265***
[4.96]
0.292**
[2.03]
MARKETit
0.645***
[9.92]
0.815***
[19.39]
1.165***
[5.90]
YEAROPENED
i
0.00016*
[1.87]
0.0004**
[2.36]
0.0002
[0.29]
COOP
i
0.0022**
[2.94]
-0.0003
[-0.32]
0.0074**
[2.63]
GESPA
i
-0.0001
[-0.10]
N
2070
10994
1195
R-squared
0.852
0.404
0.311
Sources: : Individual store-level monthly data from February 2006 through May 2008 provided
by Eroski.
Notes:
1. Absolute values of t statistics are in parentheses (t statistics are based on standard errors that
are robust and clustered at the individual store level).
2. All models include constant and monthly dummy and year dummy variables.
3. For Hypermarket, all stores are either COOP or GESPA and hence the omitted reference
category is GESPA. For Supermarket, there are COOP, GESPA and conventional stores and the
omitted reference category is conventional stores. For Supermarket City only, there are only
COOP and conventional stores and hence the omitted reference category is conventional stores.
* p<0.10, ** p<0.05, and *** p<0.01
38
Table 3 Sales Growth and HRM for Hypermarkets: Additional Evidence
Dependent variable=lnQit
(i)
(ii)
(iii)
(iv)
lnLit
0.552***
[6.57]
0.576***
[6.53]
0.552***
[6.57]
0.552***
[6.57]
MARKETit
0.645***
[9.92]
0.653***
[9.51]
0.645***
[9.92]
0.645***
[9.92]
YEAROPENED
i
0.00014
[1.61]
0.0002***
[2.71]
0.0001
[1.16]
0.00007
[0.87]
INVOLVE
i
0.558***
[2.84]
STAKE
i
6.2x10-8***
[2.74]
MEMBER
i
0.0037
[1.01]
TRAINING
i
0.255
[1.15]
N
2070
1889
2070
2070
R-squared
0.852
0.847
0.852
0.852
Sources: Individual store-level monthly data from February 2006 through May 2008 provided by
Eroski.
Notes:
1. Absolute values of t statistics are in parentheses (t statistics are based on standard errors that
are robust and clustered at the individual store level).
2. All models include constant and monthly dummy and year dummy variables.
* p<0.10, ** p<0.05, and *** p<0.01
39
Table 4 Sales Growth and HRM for Supermarket City: Additional Evidence
Dependent variable=lnQit
(i)
(iii)
(v)
(ii)
lnLit
0.292**
[2.03]
0.292**
[2.03]
0.292**
[2.03]
0.292**
[2.03]
MARKETit
1.165***
[5.90]
1.165***
[5.90]
1.165***
[5.90]
1.165***
[5.90]
YEAROPENED
i
-0.0002
[-0.36]
-0.0002
[-0.27]
0.00003
[0.004]
-0.0002
[-0.29]
INVOLVE
i
0.151
[0.48]
STAKE
i
2.26x10-8
[0.29]
MEMBER
i
0.0069
[1.51]
TRAINING
i
-0.047
[-0.67]
N
1195
1195
1195
1195
R-squared
0.311
0.311
0.311
0.311
Sources: Individual store-level monthly data from February 2006 through May 2008 provided by
Eroski.
Notes:
1. Absolute values of t staistics are in parentheses (t staistics are based on standard errors that are
robust and clustered at the individual store level).
2. All models include constant and monthly dummy and year dummy variables.
* p<0.10, ** p<0.05, and *** p<0.01
40
Table 5 Table 5 Worker Satisfaction and Ownership Types
Dependent variable=EAIi
(i))
((ii)(ii)II)
COOP
i
-.168***
[-6.92]
-.202***
[-7.63]
lnYi
-4.853
[-1.54]
-7.937
[-2.44]
COOPi*(lnYi )
24.206***
[3.46]
lnMARKETi
34.335 ***
[3.84]
35.971***
[4.04]
lnLi
10.453**
[2.55]
10.692***
[2.61]
YEAROPENED
i
.005***
[2.68]
.006***
[3.16]
MALE
i
-.066***
[-2.64]
-.067***
[-2.65]
BASQUE
i
-.109
[-1.45]
-.125*
[-1.67]
OCCUPATION
Controlled
Controlled
TENURE
Controlled
Controlled
STATUS
Controlled
Controlled
N
4328
4328
R-squared
0.124
0.126
Sources: Employee Survey conducted by Eroski in 2008.
Notes:
1. Absolute values of t statistics are in parentheses
2. All models include constant.
3. See text for variable definitions.
* p<0.10, ** p<0.05, and *** p<0.01
... About 74% of the workers in the industrial cooperatives are members. During the severe financial and sovereign debt crisis of 2008-21012, layoffs were very limited, although the bankruptcy or liquidation of some large cooperatives forced the group to make intragroup transfers and early retirements (White and White, 1991;Arando, Gago, Jones, & Kato, 2014). ...
... The benefits to the organization in terms of accumulation of specific human capital and the development of dynamic capabilities was matched by the demand for stable employment by members (Bretos, Errasti, Marcuello, 2019). As in many other similar cases, flexibility in workers' pay and working hours, and both the wage and employment insurance functions of reserves are the main tools used to achieve this result (Arando et al., 2014;Borzaga et al., 2021). ...
Article
In connection with the "Great Reset" literature in management and organizational theory, this article discusses human resource sustainability, employment stability, and layoffs in worker cooperatives, a unique organizational form in which employees are members holding residual control rights. Theoretical and empirical contributions show that worker control stabilizes employment better than investor ownership, opening the door to stronger employment protection. The paper leverages key theoretical insights from evolutionary economics and systems theory to discuss the benefits and critical elements associated with limiting member layoffs in worker cooperatives. While strong employment protection can lead to short-term inefficiencies and underperformance due to labour hoarding, imperfections in governance and control mechanisms, it can foster better accumulation and retention of firm-specific human capital and dynamic capabilities, thereby supporting long-term performance.
... If we focus on organizations where the ownership is broadly shared, literature suggests that such organizations tend to generate more positive psychological outcomes than those enterprises where the ownership is not broadly shared (Long, 1978(Long, , 1980Wilson and Peel, 1991;Van Dyne and Pierce, 1993;Kruse, 2002;Höckertin and Härenstam, 2006;Reeves, 2007;McQuaid et al., 2012;Poutsma et al., 2015), although there are studies that reach the opposite conclusion (Arando et al., 2015). ...
Article
Full-text available
This research aims to study possible effects or impacts of COVID-19 in the context of a democratic organizational system analyzing how COVID-19 has influenced employees' perception of their participation in decision-making and its impact on some psychological outcomes and emotions. COVID-19 has accelerated the process of implementation of new frameworks at work (digitalization, teleworking, new skills, and abilities) that have generated the modification of culture and employee management practices. Our hypothesis are, on the one hand, that COVID-19 has generated changes in participation structures and internal communication mechanisms, having to make modifications not to deteriorate the perception of employees about their participation in decision making. On the other hand, COVID-19 has generated changes in the psychological outcomes and emotions of the employees. In the study, we analyze a cooperative belonging to the MONDRAGON cooperative group, where participation in decision-making and ownership is in its DNA. Through qualitative (5 focus groups) and quantitative (short questionnaire) methodologies, involving 42 employees, we investigate firstly, how COVID-19 has affected perceptions about participation in decision-making analyzing what role has played internal communication in these perceptions. Secondly, we investigate how COVID-19 has affected psychological outcomes and emotions. In this case, the perceptions arising from participation in decision-making focus on the assessment that participators make of the governance channels and the day-to-day meetings. Therefore, their appropriateness seems to be a key factor in the perception of participation in the COVID-19 era. Differences have been detected between the perceptions of blue and white collar employees. Such differences have also been founded in the psychological outcomes and emotions. Although this is a single case study, the analysis carried out provides elements of reflection to modify and restructure the decision-making and participation mechanisms, adapting them to the needs of blue and white collar employees in order to "guarantee" the expected outcomes.
... The special case of Mondragon is extensively researched, and has been found to comprise a form of capitalist cooperative where the CEO is hired by the co-ops to manage the strategic and operational issues of the co-ops and enhance efficiency and growth (see e.g. Albin and Stein, 1977;Arando et al., 2015;Bretos, Errasti and Marcuello, 2019b;Cheney, 1999;Whyte and Whyte, 1991). ...
Thesis
Full-text available
When people around the globe are increasingly confronted with the challenges of rising economic inequalities and declining democratization, often associated with the spread of globalized capitalism, it becomes difficult to defend a position of business as usual. Worker co-ops are economic associations equally owned and democratically governed by workers with the potential to contribute to economic democracy and social change. This dissertation explores how worker cooperatives, primarily in Sweden, are constructed and organized by co-operators in ways that can resist capitalism, while at the same time having to relate to capitalism as the context in which they operate. The qualitative study shows that worker co-ops challenge capitalism, associated with economic ideals and hierarchical control, by instead enacting social ideals such as equal work relations through friendships, uncommodified work time and freedom to self-govern. The worker co-ops’ very existence demonstrates that such organizing is viable in the here and now. However, this dissertation also shows how worker co-ops risk the reproduction of power and compromise of ideals in order to survive within capitalist market economic contexts, thus highlighting both the possibilities and pitfalls of organizing for social change.
... A positive association has been observed in the relationship between full participation (decision-making, results and ownership) and organizational performance (e.g. Arando et al. 2015;Blasi et al. 2016;Jones et al. 2019;Mullins et al. 2019). A further positive association between financial participation practices (profit-sharing participation and employee ownership participation) and financial performance has also been revealed (e.g. ...
Article
Full-text available
Employability, talent and/or motivation of people can be a source of sustainable competitive advantage; difficult for competitors to imitate. The involvement of people, and more specifically employee participation, has been identified as a key management tool to develop this advantage. Traditionally however, the industrial relations and personnel management streams have treated employee participation from different perspectives. Economic insights have guided the former, while how employees respond to the decisions in the workplace form the basis of the latter. Accordingly, three main employee participation practices are widely recognized in the field: employee management or decision-making, profit-sharing and employee share ownership. In this research, the relationship between 3 practices of participation, employee well-being and firm performance was explored in 278 Basque companies. Objective data was obtained for organizational performance measurement and 1503 employee responses were gathered about participation practices and well-being. After controlling company size and sector (manufacturing and services were tested in this study) the results showed a significant relationship between any form of participation and employee well-being. Interestingly, a significant and negative relationship was revealed for the relationship between employee decision-making participation and labor productivity. No statistical relationship was found between financial participation practices and organizational performance. This study, therefore, confirms the relationship between one of the pillars of HR practices and employee well-being, but fails to show that participation is positively related to higher firm performance (or vice versa). New research lines are opened for scientific contributors and important insights offered for managers.
Article
Full-text available
In 2019, the Mondragon worker cooperatives, which number around 100, employed over 81 000 workers. Based primarily on information from the Mondragon annual reports, this article traces Mondragon’s employment record from 1983–2019. In this period its Spanish employment growth outran that of Spain by a factor of 3.4, and that of the aggregated oecd countries by a factor of 6.3. On top of the Spanish employment, Mondragon cooperatives’ subsidiaries employed about 4300 workers abroad (7% of the total) in 2001, and about 14 500 (18% of the total) in 2019. The article expands on the reasons for this last type of employment. The article also explains why the proportion of cooperative owner-members in the total employment varies over time. Depending on the sector, in 2019 this proportion is 32–45%, and measured as a proportion of the employment in cooperatives 32–74%—the difference being engendered by non-cooperative subsidiaries. Many cooperatives regard these proportions as second-best practices in the search for a modus between competitive pressures and the maintenance of employment within cooperatives.
Article
Full-text available
The scientific literature has shown Mondragon Corporation (MC), with 65 years of history, as a clear example that cooperativism can be highly competitive in the capitalist market while being highly egalitarian and democratic. This cooperative group has focused on its corporate values of cooperation, participation, social responsibility, and innovation. Previous scientific research reports its enormous transformative and emancipatory potential. However, studies on the effects of various types of worker participation on competitiveness and workers’ psychological wellbeing in this cooperative group exist to a lesser extent. Specifically, one aspect that needs further empirical research and that represents a competitive advantage for Mondragon is the degree of commitment and emotional attachment that can be observed in the people who work there. For this reason, this article aims to identify key elements of the democratic participation of workers in these cooperatives that relate to the development of organizational commitment. Based on a communicative and qualitative approach, data collection included 29 interviews to different profiles of workers (senior and junior workers, members and non-members of the cooperative, and researchers involved in the cooperatives) from eight different cooperatives of the Corporation. Through this research methodology, the participants interpret their reality through egalitarian and intersubjective dialogue because their voices are considered essential to measure the social impact. This study found three different ways in which the democratic participation of worker-members in management and ownership contributes to developing affective organizational commitment among those working in Mondragon cooperatives, generating positive psychological and economic outcomes for both workers and cooperatives.
Article
Purpose The purpose of this paper is to give an updated overview of the research on employee ownership. What does the scientific literature reveal about advantages and disadvantages? What can be learned from different models used in Italy, France, Mondragon (Spain), UK and US with many employee-owned firms in contrast to Denmark. Design/methodology/approach A structured review of the literature on employee. The paper identifies different mechanisms leading to effects on productivity, job stability, distribution, investment etc., and reviews the empirical evidence. The main barriers and drivers are identified and different models for employee ownership in Italy, France, Mondragon (Spain), UK and US are reviewed to identify potential models for a country like Denmark with few employee-owned firms. Findings The article gives an overview over the theoretical predictions and the main empirical evidence of the effects of employee ownership. The pros are greater employee identification with the firm and increased productivity reinforced by increased participation. Employee-owned firms have more equal distribution of wages and more stable employment, and they have greater mutual control between employees and fewer middle managers. The motivation effects may be smaller for large firms and lack of capital may lead to lower levels of investments and capital per employee. Originality/value Comprehensive and updated literature review on the effects and successful formats of employee ownership to identify models for implementation in countries with few employee-owned firms.
Article
Full-text available
The relationship between firms’ owners and managers is a quintessential example of costly principal–agent interaction. Optimal design of monetary incentives and supervision mechanisms are the two traditional ways of reducing agency costs in this relationship. In this paper, we show evidence which is consistent with a third mechanism: firms have managers whose economic preferences are aligned with owners' interests. We uncover differences in economic preferences between managers employed in firms controlled by two distinct classes of ‘patrons’: employee-owned firms (worker cooperatives) and conventional investor-owned firms. In a high-stakes lab-in-the-field experiment, we find that co-op managers are less risk-loving and more altruistic than their conventional counterparts. We do not observe differences between the two groups in terms of time preferences, reciprocity, and trust. Our findings are consistent with existing evidence on worker cooperatives, such as their tendency to self-select into less risky industries and their compressed compensation structures.
Article
Purpose The purpose is to review the effects of employee participation (EP) in decision-making, ownership and profit on job quality, worker well-being and productivity, and derive policy recommendations from the findings. Design/methodology/approach The authors summarise results of “declining labour power”, plus theoretical arguments and empirical evidence for the benefits of EP for job quality, satisfaction and productivity. Findings Worker well-being and job satisfaction are ignored unless they contribute directly to profitability. EP is needed to remedy this situation when employers have market power and unions are weak. The result can be a rise in both productivity and well-being. Research limitations/implications The chief issue here is that there are data limitations, particularly on the well-being effects of participation. Practical implications Lots of encouraging examples in many countries need legislative help to multiply. Social implications It is quite possible that there could be major implications for welfare and employment. Originality/value The authors make the case for public sector subsidies for employee buyouts and new cooperative start-ups, as well as legislation for works councils and profit sharing.
Article
In this paper we draw on recent progress in the theory of (1) property rights, (2) agency, and (3) finance to develop a theory of ownership structure for the firm.1 In addition to tying together elements of the theory of each of these three areas, our analysis casts new light on and has implications for a variety of issues in the professional and popular literature, such as the definition of the firm, the “separation of ownership and control,” the “social responsibility” of business, the definition of a “corporate objective function,” the determination of an optimal capital structure, the specification of the content of credit agreements, the theory of organizations, and the supply side of the completeness-of-markets problem.
Article
Human resources, national competitiveness and employee welfare the mutual gains enterprise choice and diffusion of mutual gains strategies human resources and organizational governance the role of labour and worker representation institutions and policies for human resource development toward a mutual gains labour-management policy building a mutual gains coalition.