K. Karantininis & J. Nilsson (eds.), Vertical Markets and Cooperative Hierarchies, xx – xx.
© 2006 Springer Academic Publishers. Printed in the Netherlands.
COOPERATIVES: HIERARCHIES OR HYBRIDS?
Centre ATOM, University of Paris (Panthéon-Sorbonne), France
Recent developments in organization theory about arrangements that are neither
markets nor hierarchies provide an opportunity to reconsider the nature of
cooperatives and their fundamental characteristics. The concept of “hybrids”
developed by transaction cost economics to encapsulate the properties of these
arrangements may be particularly relevant in that it provides a theoretical framework
in which to embed cooperatives among other modes of governance. This paper goes
in that direction and proposes a characterisation of different regimes among
cooperatives, establishing a typology grounded in theory. An important result of this
approach is that it challenges standard competition policies towards cooperatives.
The importance of cooperatives as a mode of organization cannot be overestimated.
In the European Union (as it was in 2000) they represented over 130,000 firms, with
more that 2,500,000 employees and 85,000,000 members.
They often have a very
significant market share, particularly in the agrifood sector (from 30% in France to
I would like to thank Michael Cook, George Hendrikse, Kostas Karantininis, Jerker Nilson, and
participants to the Chania Conference on “Vertical Markets and Cooperative Hierarchies: The Role of
Cooperatives in the International Agri-Food Industry” for the incentives they provided, the information
they delivered, and the comments they shared with me. I alone remain responsible for errors and/or
These data are from 1996 and have been published in “Statistics and Information on European
Cooperatives.” International Cooperative Alliance, Geneva, December 1998.
83% in the Netherlands), in banks and credit unions (from 25% in the Netherlands to
35% in Finland), and in retailing activities (over 25 millions members in 1996).
Of course, economists have long been aware of that importance. There is
substantial literature on cooperatives, and significant contributions have been
published recently about changes in their status and the challenges that these
changes represent (Cook, 1995). However, and this is somehow paradoxical, there is
not much about the nature of cooperatives as modes of organization. In the standard
economic literature they tend to be considered as relatively strange animals, in that
they depend on an allocation of property rights that do not fit well within the
traditional dichotomy between markets (with autonomous and distinct property
rights of parties involved in exchange) and firms (with property rights unified within
a legally well defined structure). Clearly, cooperatives do not fit well within this
The emergence in the late 1980s and the 1990s of a substantial body of research
on organizational arrangements that are neither markets nor hierarchies may provide
an opportunity to reconsider the nature of cooperatives and to shed light on some of
their major characteristics. It may also help revisiting public policies, particularly
competition policies, in order to reconsider approaches that do not capture the
essence of cooperatives, establishing policies that either park cooperatives in a
special (favored) status, or want to put them in the same basket as fully integrated
firms. Recent debates about the legal status of cooperatives in the European Union
The concept of “hybrids” has been proposed, particularly by economists
grounding their analyses in transaction costs theory, to encapsulate properties of the
family of arrangements that have characteristics significantly distinct from those
underlying market exchanges while they also differ substantially from those
presiding at the organization of transactions within integrated firms. Therefore, a
question naturally comes to mind: would this concept be appropriate for
In what follows, I explore this question. Section 2 introduces very briefly the
theoretical framework underlying the concept of “hybrid” in a transaction cost
perspective. Section 3 examines what differentiates hybrid arrangements from
integrated firms. Section 4 discusses if these traits suit some fundamental properties
observed in cooperatives. Section 5 develops arguments as to why this
characterization matters and may challenge existing public policies. Section 6
concludes with a call for more research in this direction.
2. ANALYTICAL FRAMEWORK: A SHORT REMINDER
The observation that there exist ways for organizing transactions among economic
units that maintain distinct property rights while they share a significant subset of
Rapport Annuel du Conseil Supérieur de la Coopération, Paris, 2000, pp. 120-121
This discrepancy was already noted by Alchian and Demsetz (1972) and is also discussed by Hansmann
(1988). However, Cook et al. (n.d.) note that there is an increasing interest for organizational issues in the
study of cooperatives in the post-1990 period (see their observation 4, p. 23).
their rights of decision is not new. Without going back to the “industrial district”
identified by Marshall (1920), franchising began to attract some attention in the late
1970s (Rubin, 1978; see also Brickley and Dark, 1987). However, it was in the
second half of the 1980s and the 1990s that a growing literature, initially based in
managerial sciences and sociology, focused on networks and similar modes of
arrangements (Thorelli, 1985; for a pioneering survey, see Grandori and Soda,
1995). In my view, the introduction of the concept of “hybrid” by Williamson in
1991 (1996, Ch. 4)
represents a major step forward in that it embedded the large set
of empirical observations on different arrangements in a theoretical framework that
provided an explanation to their existence and gave coherence to their
The model Williamson proposed and that I summarize here with some minor
changes is based on transaction cost economics, which lies at the core of new
institutional economics. A preliminary question that is often raised with that
approach and which deserves attention is: Why attach so much importance to
transactions? Why use transaction costs as a point of entry for analyzing
organizations? Does it mean neglecting, even abandoning the crucial concept of
costs of production, thus turning away from the structuring role that technology
often plays? Coase (1998) provides an answer, in my view a very convincing one, to
this legitimate question. Transactions matter because their organization under
different types of arrangements and under the umbrella of institutions that make
them more or less easily happen determines the capacity of economic activities to
develop and take advantage of the division of labor and of specialization. In that
sense, the choice of a mode of organization for arranging transactions, which is the
transfer of rights among parties to an activity of production or exchange, is crucial.
And costs that result from this choice largely establish, beside the technological
factors, how these activities will be structured and, therefore, the turf on which
production (and its costs) develops.
As is now well known, Williamson went a step further in the direction opened by
Coase, with a contribution that made the transaction cost approach operational. His
powerful intuition, which was later developed in a heuristic model (Williamson,
1985, ch. 4; Riordan and Williamson, 1985), is that a few characteristics or
”attributes” of transactions, namely their frequency (F), the uncertainty (U)
surrounding their arrangement, and the specific investments (AS) they require,
determine their costs. This relationship between transaction costs and the attributes
of transactions can be expressed functionally as:
TC = f ( F, U, AS)
- + +
with signs indicating the direction in which transaction costs vary when the related
variable increases. The next step in building the model consists of linking the choice
of a mode of governance (GS) to these costs and, therefore, implicitly to the
attributes of the transactions at stake. We can summarize these links in figure 1:
Actually the notion of hybrid was already at work in Williamson (1985) but was considered a transitory
and relatively unstable mode of organization. For an analysis of Williamson’s evolution on this, see
(F, U, AS) TC GS
Figure 1: Relationship between characteristic, costs, and governance of transactions
Under some simplifying assumptions, particularly the idea that in choosing a mode
of governance, agents intend to minimize their costs, Williamson expressed these
relationships in what is often called the heuristic model, explaining the trade-off
between organizing a transaction within the firm (“hierarchy”) and relying on
markets for doing so.
A few years later (Williamson, 1991 [1996, ch. 4]), he extended the model in
order to encapsulate organizational arrangements that were neither hierarchies nor
markets, and labeled them ”hybrids”. Taking the specificity of assets (or investment)
as the key variable that explains the choice among alternative modes of organization
(a proposition already substantiated by several econometric tests: see a review in
Joskow, 1988), he developed an analysis in which increasing costs of governance for
market transactions leaves the way to interfirm agreements before ending up in
vertical integration when mutual dependence becomes so strong that it puts these
agreements at too high a risk.
Since the model is now well known, I do not reiterate
its details here. I stick to its geometric representation, summarized in Figure 2
below, in which the trade-off between the three alternative modes of organization is
indicated in bold lines, with the lower envelope showing the most adapted mode for
the corresponding level of investments specific to the transaction(s) at stake.
Contractual hazards increase when specific investments create mutual dependence because of an
underlying assumption (explicitly made): agents tend to behave opportunistically and to take advantage of
governance Markets Hybrids Hierarchies
Figure 2. Modes of governance. Source: adapted from Williamson, 1996, p. 108
Based on propositions derived from this model, hundreds of tests have been
published, most of them supporting the predictions made by the theory (for surveys
and discussions, see Joskow, 2005, and Klein, 2005). However, in order to go further
and to provide a full explanation of why one mode of organization is preferred over
another for certain transactions,
it is necessary to make one more step and explore
the internal characteristics of these different modes. In other terms, their respective
advantages (and costs) must be assessed. From a technical point of view, this
comparative approach raises important difficulties (Gibbons, 2003; Joskow, 2005).
One condition it must fulfill is the careful examination of the properties of each
mode along lines that allows comparisons. Initial progress in that respect focused on
firms (see Ménard, 2005a). More recent research have contributed to a better
knowledge of some basic properties of hybrid arrangements. In order to discuss
whether cooperatives belong to that mode of organization or not, I now turn to a
review of some of these distinct properties.
3. WHAT DIFFERENTIATES HYBRIDS FROM HIERARCHIES?
At first sight, the arrangements that have been identified as “hybrids” in the
literature form a strange collection. They extend from subcontracting to franchise
That is, how is it that hybrids or integrated firms can monitor contractual hazards better than markets
when there are more specific investments and/or more uncertainty?
This section draws from Ménard (2004).
systems, collective trademarks, partnerships, alliances, and so forth. The vocabulary
itself tends to reflect this uncertain state of affairs: beside hybrids, which is the term
I use in what follows because it refers to a well defined theoretical framework (see
above), we find more descriptive expressions such as “symbiotic forms”, “clusters”,
“supply chain”, “networks”. However, notwithstanding the apparent heterogeneity of
this “bestiary”, the combination of a transaction cost perspective with what we have
learned from the empirical literature delineates some fundamental properties.
The central characteristic of hybrids is that they maintain distinct and
autonomous property rights and their associated decision rights on most assets,
which makes them different from integrated firms; however, they simultaneously
involve sharing some strategic resources, which requires a tight coordination that
goes far beyond what the price system can provide and thus makes them distinct
from pure market arrangements. The former aspect translates into the legal status of
hybrids: parties to these arrangements hold decision rights in last resort. The later
aspect translates into common governance for a more or less significant segment of
activities of the partners involved: hybrids look like a coalition of interests. This mix
of autonomy and interdependence defines the three pillars of hybrids: they pool
resources, they coordinate through contracts that provide a framework, and they
combine competition with cooperation. Let me briefly review these three
Three complementary dimensions
Whatever the form hybrid arrangements take, they implement forms of
interdependence through joint investments. Keep in mind the example of
franchising. Hybrids develop because markets are perceived as unable to adequately
bundle the relevant resources and capabilities while integration would reduce
flexibility and weaken incentives. Looking for rents provides the foundation for
accepting the mutual dependence created through investments specific to the
relationship, whether these specific assets consist of equipment, human capabilities,
or a brand name. However, this pooling of resources is restricted to specific
transactions and concerns only some of the assets owned by the parties. Several
consequences and problems follow. First, choosing partners is a key issue. Hybrids
are selective, not open systems: partners’ identity matters. Second, the complexity of
decomposing tasks among partners and of coordinating across organizational
boundaries requires joint planning and governance for monitoring the agreement.
Third, the existence of an adequate information system among parties accepting to
pool part of their resources is central to the survival of hybrids.
inevitable asymmetries among partners maintaining autonomous rights and the risks
of capture of some strategic information periodically threaten the continuity of the
To summarize, pooling resources in hybrids requires that partners accept losing
part of the autonomy they would have in a market relationship without benefiting
Hybrids have even been qualified as “a cooperative game with partner-specific communication”
(Grandori and Soda  p. 185).
from the capacity to control that a hierarchy could provide. Hence a first problem for
hybrids is: how can they secure the coordination of interdependent investments
without losing the advantages of decentralized decisions?
This problem is partially solved through contracts. Relational contracting
provides a framework for creating “transactional reciprocity”. The resulting
cooperation carries advantages but entails risks. Advantages can be expected from
extended market shares, transfer of competencies, and sharing scarce resources (for
example, financial ones). However, contracts are incomplete and subject to
unforeseeable revisions since they contribute to organize transactions involving
specific investments that are often plagued by uncertainties (for example, joint
investments in R&D projects). We have a typical transaction cost problem here.
Contrary to what agency theory predicts, the features of contracts are not
continuously refined in order to obtain an “optimal contract” that would encapsulate
all required adaptation. As shown by recent studies on franchising (Lafontaine and
Shaw, 1999), contracts are not tailored to suit the exact characteristics of
transactions at stake. Plainly, this would be too costly and the source of too many
rigidities. Rather, contracts provide a relatively simple and uniform framework.
Hence, a second problem that is recurring among hybrids: what governance to adopt
for securing contracts against opportunistic behaviors while minimizing costly or
even impossible renegotiations?
This difficulty is amplified by the importance of competitive pressures, which
comes from two sources. First, partners in hybrid agreements often compete against
each other on segments of their activities. This can take different forms. The
agreement can have provisions that recurrently make partners competing, as in
subcontracting. Notwithstanding restrictions (geographical, etc.), hybrids may have
overlapping strategies, for example, they may target customers from the same
subset. Parties may also cooperate on some activities, such as joint R&D projects,
and compete on others. Second, hybrids usually compete with other modes of
organization, including other hybrids. The standard neoclassical explanation of
hybrids as rent seekers shows its limits here. Hybrids tend to develop in highly
competitive markets in which pooling resources is viewed as a way to deal with
significant uncertainties and survive. However, this competitive environment may
have a highly negative side effect for hybrids: if joint investments required in an
arrangement are moderately specific, partners may be tempted to switch among
arrangements, making them highly unstable. Again, the implementation of an
internal mode of regulation and control is a key issue. Hence a third problem for
hybrids is: what mechanism can be designed for efficiently disciplining partners and
solving conflicts while preventing free-riding?
These three dimensions clearly suggest that there are important regularities
underlying the apparent heterogeneity of hybrids. These regularities are rooted in the
way partners are dealing with the mutual dependence created by the specificity of
some of their investments; by the need to guarantee some continuity in their
relationship and, therefore, the frequency of transactions at stake; and by the
importance of containing contractual hazards and reducing uncertainties. They do so
with the mix of competition and cooperation that characterizes and plagues hybrids.
Because they cannot rely on prices or on hierarchy to discipline themselves, partners
need specific devices for dealing with the problems identified above. What are these
mechanisms and what is the logic behind the choice of specific ones?
Variety in governance
Hybrid arrangements develop when specific investments can be spread over partners
without losing the advantages of autonomous decisions, while uncertainties are
consequential enough to make pooling an advantageous alternative to markets.
However, the combination of specific assets and consequential uncertainties
generates risks of opportunistic behavior and miscoordination. If only one aspect (or
attribute) is present, the governance leans towards contract-based arrangements,
close to a market form. When the two attributes combine, the governance becomes
much more authoritarian. Therefore, I submit that it is the combination of
opportunism, or the risk of opportunism, and of miscoordination, or the risk of
miscoordination, that determines the governance characterizing hybrids. Let me
develop briefly before applying this proposition to the analysis of cooperatives.
One way to deal with the three problems identified in the previous subsection is
to rely heavily on contracts. A well known mechanism for disciplining partners
while facilitating coordination is the contractual embedment of restrictive
provisions. Restrictions delineate the domain of action of partners, limiting their
autonomy and identifying areas in which collective decisions must prevail. There is
an abundance of literature on vertical restrictions, much less on horizontal ones. The
emphasis is usually on their consequences on prices and how it can distort
competition. This interpretation misses what is often the main goal of these
provisions − to restrict free-riding while facilitating coordination. This point was
made 20 years ago by Williamson (1985, pp. 183-189) on the Schwinn case. It has
been largely substantiated, for example by numerous studies on supply chain
systems, particularly in the agrifood sector in which traceability and quality control
have became major issues (Ménard and Valceschini, 2005). This role of contractual
restrictions as an efficient tool of governance remains underexplored. However, we
already know enough to be aware of the limits of contracts in that respect. First,
restrictive provisions often produce conflicts among parties, particularly with respect
to their interpretation. Second, they generate suspicion among competition
authorities who see them as sources of collusion. Third, their allocation effects are
difficult to evaluate and monitor, so partners tend to rely on other mechanisms.
The tension between contractual hazards and the expected gains from
investments in interdependent assets provides strong incentives to turn to more
powerful modes of coordination than market-based contracts. This is what our
theoretical framework predicts. However, we have to go a step further and check if
our model can help understanding the specific forms this coordination takes. Using
several empirical studies, including some I have been associated with, I have
submitted in several papers (Ménard, 1996; 1997 ; 2004) that hybrid
organizations tend to produce specific modes of internal governance, which I have
suggested be called “authorities” to emphasize their difference from “hierarchies”.
These devices provide the cornerstone in the architecture of hybrids. Their main
characteristic is the pairing of the autonomy of partners with the transfer of
subclasses of decisions to a distinct entity in charge of coordinating their actions.
The presence of hierarchical elements in contractual agreements has been noted
before (Stinchcombe 1990, chap. 6). However, what I want to emphasize here is the
existence of specific organizational devices intentionally designed by partners for
monitoring their network and for controlling their actions. The authority transferred
to these devices involves intentionality and mutuality, maintaining some symmetry
Empirical studies suggest that the more or less centralized power of these
authorities depends on the degree of mutual dependence among partners and on the
complexity and turbulence of the environment in which a hybrid monitors
transactions. Let me illustrate with two polar cases. Raynaud  studied a group
of millers who created a brand name for high-quality bread in France. Members of
this arrangement use only selected wheat from which they produce first rank flour
that they dispatch to franchised bakers that agree to strict rules. However, there are
risks of opportunistic behavior among partners. First, they may be tempted to free-
ride in delivering lower quality flour. Second, some millers are competing: they
supply the same geographical area and have a strong incentive to attract as their
customers as many bakers as possible. In order to monitor this arrangement,
complex internal governance has been implemented. Requirements regarding the
inputs, quality control, and the monitoring of contracts are delegated to an
autonomous entity, created by the millers and that owns the brand name. The millers
have also created an internal “court”, with delegates operating as private judges for
solving conflicts. In this stylized case, the hybrid arrangement coordinates partners
who are on a par. Sauvée  has exhibited a very different model with a
significant asymmetry among partners. In the case he studied, a private firm has
developed a brand name of canned vegetables of high quality. Inputs are provided
by farmers under contracts that contain detailed requirements and provisions. So far,
this is quite standard. The interesting point is that because of its success the firm was
rapidly confronted to the high transaction costs of monitoring thousands of contracts
and farmers. In order to solve this problem, a complex organization was
implemented, with growers grouped in several distinct arrangements delegating the
negotiation of contracts and the numerous adjustments they require to a joint
committee. Surprisingly, this powerful committee was formerly dominated by the
growers with four delegates, while the firm has two representatives. It plays a key
role, filling the blanks in the contracts, organizing transactions, and negotiating the
distribution of quasi-rents.
Numerous variations of such arrangements could be described. They all
substantiate the idea that hybrid organizations have architecture of their own,
distinct from markets or hierarchies. At one end of the spectrum, close to markets,
hybrids rely on trust. Decisions are decentralized and a loose coordination operates
through mutual “influence” and reciprocity. The resulting relationship is not purely
informal: it tends to be highly codified in order to guarantee continuity in the
transactions and is often in the hands of key players. Palay (1985) has provided a
pioneering study in that respect, showing the role of dedicated managers in charge of
monitoring agreements among partners in the rail freight sector. At the other end of
the spectrum, some hybrids are close to a hierarchy. Parties keep legally distinct
property rights and may even compete on segments of their activities. However, a
significant domain of decisions is coordinated through a quasi autonomous entity,
which operates as a private bureau with attributes of a hierarchy. Joint ventures
provide an illustration. Between these polar cases, other forms of “authority”
develop. “Relational networks” have been extensively analyzed by sociologists and
scholars in organization theory. Because of the significance of contractual hazards
they confront, these arrangements need tighter coordination and control than trust,
with formal rules and conventions framing the relationships among partners.
Examples have been studied by Greif (1993) and Powell (1996), among others.
When uncertainty is even more significant and interdependent assets more
important, more constraining structures of governance develop, often under the
leadership of one party. The pioneering study of Eccles (1981) on the construction
industry provides a good illustration, with one firm establishing its authority either
because it holds specific competences or because it occupies a key position in the
sequence of transactions.
To summarize, hybrid arrangements tend to develop specific modes of
governance with significant variances in the degree of control over partners,
depending on the degree of uncertainty and the nature and degree of specific
investments required by the transactions at stake. If we come back to Figure 1, these
forms correspond to those associated to values between K
, with an
increasing intensity in the centralization of their governance.
4. CAN COOPERATIVES BE UNDERSTOOD AS HYBRID FORMS?
I now turn to a most difficult question: is this analysis relevant to better understand
cooperatives? The question is challenging for at least two reasons. First, there is so
much diversity among cooperatives that finding a unified theoretical framework for
explaining this diversity and encapsulating the various properties of the
arrangements involved is not an easy task. Second, and above all, I am not at all a
specialist on cooperatives. In what follows, I rely heavily on contributions from
colleagues who are much more knowledgeable than I am, particularly Cook (1995),
Cook, Chaddad and Iliopoulos (n.d.), Hendrikse and Veerman (2001), Hendrikse and
Bijman (2002), as well as on discussions with participants at the Chania
Therefore, the exploration proposed in this section is very tentative.
In order to discuss the question of whether or not cooperatives are hybrids, I
refer to the characteristics identified above.
Let us start with the central issue of the
status of property rights and their relationship to decision rights. In that respect,
“Vertical Markets and Cooperative Hierarchies”, Chania (Crete), September 3-7, 2004. Several
contributions to this conference are included in this book. Cook et al. (n.d.) review several papers,
explicitly focusing on three alternative interpretations of cooperatives, as firms, as coalitions, and as a
nexus of contracts. Several aspects of their analysis overlap with mine, although there are also significant
In looking at cooperatives as hybrids, I adopt a distinctly different view from Bonus (1986), who
considered cooperatives as pure business enterprises, as well as from Staatz (1989), who looked at
cooperatives from a pure agency perspective.
there is a wide variety of arrangements among cooperatives. At one end of the
spectrum, close to market relationships, we have cooperatives in which property
rights and decision rights are separated. In this case, cooperators formerly hold
“shares” in a cooperative and receive benefits according to its performance. They
behave very much like small shareholders operating through financial markets, with
very little control over the governance of the cooperative. Retailing and marketing
cooperatives are often of that type. They process and sell products through market-
type relationships; those buyers who are cooperators have very little or no control
over the governance. Hence, decision rights are largely isolated from property
rights: one can consider that cooperators in such cases are related to the cooperative
through quasi-market forms of contracts. At the other end of the spectrum, we have
cooperatives owned and governed by their shareholders, as is often the case with
cooperatives grouping producers (or growers in agriculture). This type of
arrangement tends to coordinate tightly the activities of its members, deciding the
variety of goods or services, fixing quantities to be produced, negotiating with
potential buyers, etc. The example of Savéol, which provides an umbrella to three
cooperatives and dominates the market for fresh tomatoes in France, is a case in
point (Sauvée, 1997; Ménard and Valceschini, 2005). Cooperatives with close
membership or that are quasi-integrated fall into this category. We are almost in the
case of classical hierarchies (Bonus, 1986). Between these polar cases, we find a
large number of cooperatives, particularly the traditional, multipurpose cooperatives
that coordinate a network of partners, most of them being cooperatives themselves
that maintain the autonomy of their property and decision rights. For example,
Cana, a French cooperative that operates in the poultry sector, covers a network of
cooperative-partners from growers to chicks and food suppliers as well as
slaughterhouses. Obviously the internal mode of governance of these widely distinct
arrangements varies significantly, depending on closeness between the allocation of
property rights and the allocation of decision rights. However, almost all
cooperatives share something that makes them different from integrated firms as
well as from pure market relationships: the one-person, one-vote rule, whatever the
size of one’s contribution.
This is a characteristic they share with many hybrid
arrangements, in which decision-making rights are allocated on a par. (See the
example of the millers in Section 3.)
Let us now turn to the three dimensions that I have identified as pillars of hybrid
arrangements, in order to exhibit what properties are shared or not by cooperatives.
(1) Pooling resources. This is surely an aspect which is one of the fundamental
motivations for organizing cooperatives. However, it exists with very variable
intensity, so that mutually dependent investments are more or less consequential.
What theory predicts in these circumstances is that the degree and importance of
specific assets shared by cooperators should determine the intensity in selectivity of
members as well as the intensity in control over their activities. (2) The significance
of contracts among cooperators (to the exclusion of contracts with outside partners).
There are exceptions to this general rule, which is one of the reasons why the status of cooperatives as
distinct from firms is challenged. For an analysis of these changes in ownership status, see Hendrikse and
Again, the intensity of contracting varies widely according to the type of
cooperatives. Contracts tend to be particularly detailed, with important provisions
and sanction clauses in cooperatives that need to tightly coordinate the actions of
their members and/or that must strictly control quality, as with growers or dairy milk
cooperatives. They are much less specific and can even be almost pure formalities
when it comes to agreements among members with no idiosyncratic investments in
the cooperative, as with retailing and marketing cooperatives. Again, our theoretical
framework allows making predictions about the characteristics of contracts
depending on the specificity of assets that cooperators are pooling; for example,
duration of contracts should be much shorter in the later case while in the former
case they are either long term, or short term and automatically renewable. (3)
Competition conditions. They also change significantly according to the type of
cooperatives. The need to tightly control free-riding, when specific assets are at
stake, and to restrain the autonomy of decisions of partners when the reputation of
the whole depends on respect for requirements by each party to the agreement,
seriously reduces competition among members. In these situations, tight
coordination through formal governance prevails, while competition among
members is much more frequent in market-oriented cooperatives that monitor
weakly specific assets, that is, assets easily redeployable from one type of activity to
Based on these casual evidences that need to be substantiated and tested by the
specialists in the field, our model suggests the following application of the
arrangements identified in Figure 2 to various types of cooperatives. (See Figure 3)
If we take the degree of specificity of investments made by cooperators in their
coop as a key variable (uncertainty should be added in a more developed model),
transaction cost theory predicts that costs of governance tend to increase with the
increase in asset specificity, but at a different rate according to the organizational
arrangement, with the costs of using markets increasing more rapidly than hybrids
which also increases more rapidly than hierarchies when investments connected
directly to the relationship become significantly more idiosyncratic. When it comes
to cooperatives what this means is that the more easily redeployable assets are held
by cooperators in their coop, and the closer we are to market arrangements, as with
retailing or marketing cooperatives. Symmetrically, the more specific to the
transactions organized by a cooperative are the assets detained by cooperators, the
tighter the coordination should be, bringing into the arrangement a form of
governance that is very close to full integration. Different modes of organizing
cooperatives fall in between, as suggested by our figure 3. And there are cases when
investments are so specific to the transactions monitored by the cooperative that it is
structured and governed very much like a classic integrated firm.
costs Markets Hybrids Hierarchies
Figure 3. Modes of governance among cooperatives.
This suggested typology obviously needs to be discussed and tested. The emphasis
on the degree in the specificity of investments for determining the mode of hybrid
governance must be substantiated by theoretical arguments and must be assessed
through empirical studies. Moreover, uncertainty is certainly another key variable in
organizing transactions that should be introduced in the model. The advantage of
focusing on the variable “specific investments” is that it puts at the forefront of the
analysis of cooperatives the interdependence between the degree of selectivity in
membership and the intensity required in the control of decision rights on one hand,
and the importance of the degree of coordination needed on the other hand, in order
to determine the mode of governance that can efficiently monitor the type of
transactions at stake. More importantly, it provides a theoretical framework for
examining and classifying cooperatives, which allows predictions that can be tested
5. WHY DOES IT MATTER?
It is legitimate to question why the development of this approach to cooperatives
should matter. My answer is twofold: it might be highly relevant, for positive as well
as for normative reasons. I am mostly emphasizing the second aspect here.
On the positive side, the examination within a well defined theoretical
framework of factors that determine the mode of governance of cooperatives should
help understanding better their differentiated characteristics and properties. More
precisely, finding a model that allows characterizing the nature and variety of the
different modes of organization that exist among cooperatives should provide
important insights for understanding why one form emerges and predominates for
certain types of activities. The transaction cost approach developed in the previous
sections might shed light on two important issues: What are the attributes of
transactions a cooperative wants to organize that can explain why a specific
arrangement fits these attributes better than another one? And what makes
organizing transactions among cooperators more adequate, and therefore more
successful, than using market relationships or integrating within a unified firm?
Referring to adequate concepts for answering these questions may also have
important consequences in a normative perspective. If there is economic
explanation, grounded in solid theory, why do so many cooperatives have the
characteristics of hybrids, and why among the variety of hybrid arrangements do
cooperatives adopt specific forms and choose different modes of governance? The
answers may provide indications about what type of cooperative should be chosen
for organizing specific types of transactions.
Adequate answers to the questions raised above also involve policy issues. If a
substantial subset of cooperatives are hybrid arrangements that exist because they
provide the most relevant arrangement for the type of transactions they are
organizing, and if this capacity to arrange these transactions efficiently depends,
under identifiable conditions, on the implementation of mechanisms of coordination,
control, and discipline over the members involved, the resulting governance may
challenge standard competition policies. This brings into the picture the ongoing
debate about the status of cooperatives that should prevail in the European Union.
Standard competition policies are based on a theory of competition grounded in
the dualism between markets and firms (“hierarchies”). In principle, firms are
allowed to freely develop their activities so long as they conform to some “rules of
the game”, mainly: (i) they respect certain principles in their interactions, a major
principle being that they do not build coalitions (rule #1); and (ii) their activities do
not threaten “normal” market structures, that is, structures that guarantee the
continuity of competition. Therefore, developing strategies that generate market
power over a certain threshold is prohibited (rule #2). Confronted with these
benchmarks, most cooperatives (with the possible exception of retailing
cooperatives) represent a challenge to the two basic rules, particularly rule #1.
The issue is also debated in the U.S., although to my knowledge the Capper-Volstead Act has not really
been challenged so far.
Indeed, they clearly form a coalition of legally autonomous actors. And they often
do so in order to capture part of the market. An important consequence, now argued
in many instances of the European Union, is that with respect to the theory in which
competition policies are grounded, cooperatives are anomalies tolerated for political
reasons, but that sound economic policies should prohibit.
This way of positing the problem tends to ignore the very reason why there exist
non-standard arrangements like cooperatives and, more generally, hybrid forms. In
Sections 2 and 3, I have explained why, in a transaction costs perspective, modes of
organizing transactions legitimately develop that are based on neither market
relationships nor hierarchy, and why these modes tend to adopt inter-firms or inter-
units coordination that impose dome discipline and constraints on parties to the
agreement. What happens when competition policies are implemented that ignore
the ‘raison d’être’ of these non-standard arrangements? What are the consequences
of ignoring the logic that explains hybrids in terms of minimization of transaction
costs? Let me briefly discuss the issue through two stylized examples.
First, what happens if arrangements of the hybrid types are prohibited, for
example, to the motive that they represent a coalition of independent actors? If we
refer to Figure 1, this means suppressing hybrids, so that the lower envelope of the
curve corresponding to the degree of specific investments in the domain [K
eliminated. The result is that transactions are organized either under market
arrangements (when assets have a specificity lower than K
) or within integrated
firms (when specificity of assets involved is higher than K
). This means that costs
of governance for the entire domain defined by [K
] are higher than they would
have been if hybrids would have been allowed. Higher social costs result.
A second stylized example corresponds to a situation in which competition
authorities (or other public entities) who do not properly understand the role of
hybrids in a competitive environment would impose specific restrictions on their
activities (that is, restrictions that are not imposed on market transactions nor on
transactions that are organized by a firm, for example, regarding advertising). Such
constraints translate into higher costs of governance for hybrids, which shifts their
representative curve upwards. As a result, there may be more room for market
transactions on the left side (K
is moved slightly to the right), and there is much
more room for transactions organized under the umbrella of integrated firms (K
moved to the left). The consequence is that an entire area in which transactions
could have been advantageously arranged by hybrids are now transferred to less
efficient modes of organizations. Again, social costs result.
To summarize, the ignorance of the specific nature of cooperatives or, more
generally, of hybrid arrangements have important consequences that are
misunderstood and need further exploration. This is a typical example of how the
institutional environment may have a substantial impact on what modes of
organization are chosen and on the consequences of these choices on economic
efficiency and social welfare.
The following analysis is developed extensively in Ménard (2005b), with specific examples provided
by recent decisions of competition authorities.
This paper has explored some properties of cooperatives in the light shed by a new
institutional approach, with the conceptual apparatus of transaction costs at the core.
There are three main messages from this very preliminary examination.
First, we need a theoretical framework for understanding much better the nature
of cooperatives. The standard neo-classical approach that captures the essence of
organizations through a production function performs very poorly in that respect.
Similarly, the principal-agent approach does not explain why cooperatives exist and
the specific forms they adopt. On the other hand, recent developments in transaction
costs economics suggest very fruitful perspectives and provide powerful tools for
going further in that direction.
Second, there are strong incentives for studying more carefully the observable
characteristics of cooperatives in terms of modes of governance. On the positive
side, it may help us understand why and when certain modes are preferred to others.
On the normative side, it may suggest ways of determining which forms should be
chosen, or should be modified in what direction, in order to fit with the properties of
transactions that need to be organized.
Third, the analysis developed above suggests that there is an urgent need for
policy makers and for competition authorities to introduce transaction costs issues in
their reasoning. It is no more possible to build policies and regulation based on the
simplistic trade-off between markets and integrated firms. And using in a rather
scholastic way provisions of political arrangements, like article 81 of the Rome
Treaty, to justify derogations in favor of hybrids and related arrangements cannot be
considered satisfying anymore. Indeed, recent theoretical developments suggest that
hybrids and similar arrangements are not “derogatory” − they are at the very heart of
a dynamic market economy. In that respect, the theoretical and political status of
cooperatives should be reexamined in a much more positive perspective.
Alchian, A.A. and H. Demsetz. 1972. "Production, Information Costs and Economic Organization."
American Economic Review 62(5):777-795.
Bonus, H. 1986. “The Cooperative Association as a Business Enterprise: a Study in the Economics of
Transactions.” Journal of Institutional and Theoretical Economics 142(3):310-399.
Brickley, J. A. and F. H. Dark. 1987. "The Choice of Organizational Form: the Case of Franchising."
Journal of Financial Economics 18(2):401-420.
Coase, R. H. 1998. “New Institutional Economics.” American Economic Review 88(2):72-74.
Cook, M.L. 1995. “The Future of US Agricultural Cooperatives: A Neo-institutional Approach.”
American Journal of Agricultural Economics 77:1153-1159.
Cook, M.L. and C. Iliopoulos. 2000. “Ill-defined Property Rights in Collective Action: The Case of US
Agricultural Cooperatives.” In C. Menard ed. Institutions, Contracts and Organizations. Perspectives
from New Institutional Economics. Cheltenham: E. Elgar, pp. 335-348.
Cook, M.L., F. R. Chaddad, and C. Iliopoulos. n.d. “Advances in Cooperative Theory since 1990: A
Review of Agricultural Economics Literature”. Working Paper, University of Missouri-Columbia.
Eccles, R. 1981. "The Quasifirm in the Construction Industry." Journal of Economic Behavior and
Gibbons, R. 2003. “Team Theory, Garbage Cans and Real Organizations: Some History and Prospects of
Economic Research on Decision-making in Organizations.” Industrial and Corporate Change
Grandori, A. and G. Soda. 1995. “Inter-firm Networks : Antecedents, Mechanisms and Forms.”
Organization Studies 16(2):183-214.
Greif, A. 1993.“Contract Enforceability and Economic Institutions In Early Trade: The Maghribi
Traders.” American Economic Review 83(3):525-547.
Hansmann, H. 1988. “The Ownership of the Firm.” Journal of Law, Economics and Organization
Hendrikse, G.W.J. and C.P. Veerman. 2001. “Marketing Cooperatives: An Incomplete Contract
Perspective.” Journal of Agricultural Economics 52(1):53-64.
Hendrikse, G.W.J. and W.J.J. Bijman. 2002. “Ownership Structure in Agrifood Chains: The Marketing
Cooperative.” American Journal of Agricultural Economics 84(1):104-119.
Joskow, P. 1988. “Asset Specificity and the Structure of Vertical Relationships.” Journal of Law,
Economics and Organization 4(1):95-117.
___.2005. “Vertical Integration.” In C. Ménard and M. Shirley, eds. Handbook of New Institutional
Analysis. Springer, pp. 319-348.
Klein, P. 2005. “The Make-or-Buy Decisions. Lessons from Empirical Studies.” In C. Ménard and M.
Shirley, eds. Handbook of New Institutional Analysis. Springer, pp. 435-464.
Lafontaine, Francine and Kathrin Shaw (1999) “The Dynamics of Franchise Contracting: Evidence from
Panel Data.” Journal of Political Economy. 107: 1041-1080.
Marshall, A. 1920. Principles of Economics. 8th ed. London: MacMillan. Reprint 1969
Ménard, C. 1996. “On Clusters, Hybrids and other Strange Forms. The Case of the French Poultry
Industry.” Journal of Institutional and Theoretical Economics 152(1):154-183.
___. 2004. “The Economics of Hybrid Organizations.” Journal of Institutional and Theoretical
___. 2005b. “The Inadequacy of Competition Policies: A New Institutional Approach.” In M.
Oppenheimer and N. Mercuro, eds. Law and Economics: Alternative Economic Approaches to Legal
and Regulatory Issues. New York: M.E. Sharpe, pp. 27-54.
___. 2005a “A New institutional Approach to Organization.” In C. Ménard and M. Shirley, eds.
Handbook of New Institutional Analysis. Springer, pp. 281-318.
___. 2005c. “Oliver Williamson and the Economics of Hybrid Organizations.” in M. Augier, J. March
and D. Teece (eds.), Title not yet determined, Oxford: Oxford University Press, in press.
___.1997/2005. “Le Pilotage des Formes Organisationnelles Hybrides”, Revue Economique 48(2): 741-
751. English translation: “The Governance of Hybrid Organizational Forms.” In C. Menard, ed. The
International Library of New Institutional Economics. Cheltenham: Edward Elgar, vol. IV, pp. 105-
Ménard, C., and M.M. Shirley. 2005. Handbook of New Institutional Economics, Berlin-Boston-
Dordrecht-New York: Springer.
Ménard, C. and E. Valceschini. 2005. “Institutions for Governing Agriculture and Rural Areas.”
European Review of Agricultural Economics, in press.
Palay, T.M. 1985. "Avoiding Regulatory Constraints: Contracting Safeguards and the Role of Informal
Agreements." Journal of Law, Economics, and Organization 1(1):155-175.
Powell, W. 1996. “Inter-organizational Collaboration in the Biotechnology Industry.” Journal of
Institutional and Theoretical Economics 152(1) :197-215.
E. 1997. Propriété et exploitation partagée d’une marque commerciale : aléas contractuels et
ordre privé. PhD dissertation, Université de Paris (Panthéon-Sorbonne).
Riordan, M. and O. Williamson. 1985. “Asset Specificity and Economic Organization.”International
Journal of Industrial Organization 3(4):365-378.
Rubin, P. H. 1978. "The Theory of the Firm and the Structure of the Franchise Contract." Journal of Law
and Economics 21(1):223-234.
L. 1997. “Managing a Brand in the Tomato Sector: Authority and Enforcement Mechanisms in a
Collective Organization.” Acta Horticulturae 536: 537-554
___. 2002. “Governance in Strategic Networks.” Working Paper, ISAB: Beauvais.
Staatz, J. M. 1989. Farmer Cooperative Theory: Recent Developments. Washington DC: U.S.
Department of Agriculture, ACS Agr. Coop. Service Research Rep. 84.
Stinchcombe, A. 1990. Information and Organizations. Berkeley: University of California Press.
Thorelli, H.B. 1986. "Networks: Between Markets and Hierarchies." Strategic Management Journal
Williamson, O.E. 1991.  "Comparative Economic Organization: The Analysis of Discrete
Structural Alternatives." Administrative Science Quarterly 36(2):269-296. Reproduced in The
Mechanisms of Governance. Oxford: Oxford University Press.
___.1985. The Economic Institutions of Capitalism. New York: The Free Press-Macmillan.