United StatesDepartment ofAgriculture

Agricultur$~~ocv~~~lve

CooperativeMarketing

1 ACS Agencies-in-ResearchReport 127 Common

Abstract

Cooperative Marketing Agencies-in-Common

Bruce J. Reynolds

Marketing agencies-in-common (MACs) have been usedby farmer cooperatives for many years to accomplish specificmarketing activities. Relatively scant attention or concern hasbeen given to adequately defining MACs in terms of how theydiffer from other forms of organization, particularly from otherfederated cooperatives.

Distinctions can be made in terms of the developmentand ownership of assets. Members of MACs retain individualmember ownership of assets, with their MAC providing vari-ous supplementary functions such as group communicationsand product selling coordination.

Members often have assets that are highly specific totheir own marketing programs and have developed significantexpertise in their respective industries. With a membership ofthis kind, MACs are governed in a multiple principal structure,in contrast to the usual generalized single principal-agent rela-tionship that prevails in most cooperative forms of organiza-tion.

Keywords: Cooperatives, marketing agencies-in-common,principals, agency, and federated.

ACS Research Report 127

January 1994

Price: Domestic $2.00; foreign $2.50

Contents

Highlights ................................................................................ ii

Objectives of Study ................................................................ 1

Distinctive Form of Cooperative ............................................ .3

Multiple Principals and Single Agent ...................................... 7

Principal-Agent Is a Control Function .................................... 9

Examples of Marketing Agencies-in-Common . . . . . . . . . . . . . . . . . . . . 12

Marketing Agencies-in-Common Versus AlternativeOrganizational Forms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20

Coordinating Strategic Plans ............................................... 27

Appendix Tables ................................................................... 31

Highlights

Marketing agencies-in-common (MACs) can be incorpo-rated as cooperatives or non-cooperatives. As cooperatives,they are a federated type, with organizations rather than farm-ers as direct members.

The purposes of MACs are, in most cases, coordinatingsales of member products and extending the marketing pro-grams that member cooperatives have developed. Thisinvolves relatively slight investment and ownership of assets,which provides an empirical means of identifying MACs fromother types of federated cooperatives. An assets/sales ratiobelow 0.1 was effective in separating MACs from a 1991 dataset of federated cooperatives.

Industrial organization economists, particularly in theantitrust field, have developed a theoretical model and defini-tion of common agency. Their work, while not completely orreadily applicable to cooperative organization of MACs, distin-guishes the more general concept of common agency as hav-ing a multiple principal structure, in contrast to the single prin-cipal-agent relationship of most business and cooperativeorganizations.

The principal in a common agency takes a multiple formbecause assets are primarily retained by each member andinclude many which are highly specific to their individual oper-ations. Each member is a principal in its own right because ofits unique products and expertise. While tracking down feder-ated cooperatives and determining their status as a MAC iscomplicated, there is adequate information on about 10MACs, which demonstrate the multiple principal structure.

MACs are a distinct alternative for cooperatives seekingthe benefits of coordination and economies of size. A compar-ison of several objectives and alternative organizational formsreveals more advantages for MACs when members wantmore interorganizational coordination but have a preferencefor maintaining their separate identities. MACs have some dis-

ii

advantages in being the focus for developing distinctive capa-bilities and innovations.

The marketing coordination function of MACs can beextended to interorganizational planning and decision-makingfor certain areas of operation that involve mutual or interactiveimpacts on members. The commitment to develop and main-tain services to members is a distinct and economically diffi-cult challenge for cooperatives in many cases. Often a lack ofcoordination in planning and in carrying out certain operationsrenders less efficient outcomes for cooperatives than if suchprocesses were handled within the framework of a MAC.

Cooperative MarketingAgencies-in-CommonBruce J. Reynolds

OBJECTIVES OF STUDY

The strategy of forming a cooperative marketing agency-in-common (MAC) was in use before the Capper-Volstead Actwas enacted in 1922. In recent years this strategy has receivedincreased interest among dairy and other commodity groups.

The purpose of this report is threefold: (1) to provide adefinition that adequately differentiates MACs from othercooperatives and federated organizations, (2) to examineMACs in practice and in a strategic decision making context ofalternatives, and (3) to extend the business applications ofMACs to a broader strategic objective of cooperatives’ role inthe farm economy that does not apply to firms operating on anon-cooperative basis.

Many of today’s largest farmer cooperatives started outas relatively small organizations with memberships concen-trated in local geographic areas. Over time, regional expan-sion of membership took place. Many of these cooperativesrapidly grew by consolidating or merging with cooperativesin neighboring areas.

Cooperatives have also pursued an alternative to consoli-dation by forming organizations for intercooperative coordi-nation of certain facets of their operations. This alternative isparticularly attractive, as certain marketing or processingfunctions may benefit from economies of size.

Although a merger would accomplish the sameeconomies, intercooperative organization is advantageouswhen some functions are more effectively carried out in adecentralized structure of smaller and more geographicallylocalized cooperatives.

Intercooperative coordination is usually organized as afederated cooperative. A board of directors consists of produc-ers or managers or both from the member cooperatives.Federated cooperatives are organized for various purposes.Some carry out a substantial array of value-added and capital-intensive processing and marketing activities. Others coordi-nate marketing activities that do not involve intensive capitalformation.

The latter type of federated cooperatives are called mar-keting agencies-in-common. The term “marketing agencies-in-common” is explicitly used in the Capper-Volstead Act as away for cooperatives to coordinate their marketing activities.’

The first section of this report defines a MAC and empiri-cally tests the definition as to its adequacy in identifying fed-erated cooperatives that are MACs. The next two sections,“Multiple Principals and Single Agent” and “Principal-AgentIs a Control Function,” examine the principal-agent relation-ship as it applies to cooperatives, and specifically to MACs.

The principals are members of a cooperative whoseresponsibilities are primarily carried out by a board of direc-tors. Agents are employees of cooperatives, primarily man-agement. The principal-agent relationship establishes the con-trol and objectives of a business.

More general and applied information about MACs isprovided in the fourth and fifth sections of this report,“Examples of Marketing Agencies-in-Common” and“Marketing Agencies-in-Common Versus AlternativeOrganizational Forms.” These brief case studies and compar-isons of MACs demonstrate the concepts and distinctionsdeveloped in the first part of the report.

A framework for developing new MACs and widerapplications is provided in the last section of this report,“Coordinating Strategic Plans.” This section examines applica-tions of MACs that go beyond the usual specific operating

1 Antitrust Laws: Legal Phases of Farmer Cooperatives, Part 3, ACSInformation Number 100,1983, p. 293.

2

objectives to comprise a broader strategic planning responsi-bility.

The complexity of expanding a cooperative’s businessresources and opportunities, as well as of maintaining itsobjectives and governance integrity as a farmer cooperative,requires a certain amount of coordinated action and planning.It has no adequate parallel or analogy among non-cooperativefirms. MACs can provide an organizational mechanism forcoordinated planning of services to farmers and for strength-ening the use of cooperative methods and forms of organiza-tion.

DISTINCTIVE FORM OF COOPERATIVE

Marketing agencies-in-common can be applied to cooper-atives or non-cooperatives. Concern over potential violation ofantitrust laws may have reduced the number of MACs thatnon-cooperatives would have otherwise formed.

The limited exemptions from antitrust provided by theCapper-Volstead Act have enabled farmer cooperatives tohave more experience with MACs than most other segmentsof American industry. Special legislation has provided someantitrust exemptions for both investor-owned firms (IOFs)and cooperatives to organize export agencies-in-common, butthe numbers formed and the extent of their exporting activityhave been relatively small.

Economic theory of common agency offers many criticalinsights, but much of this theory has not been developed forapplication to farmer cooperative MACs. Rather, theoreticalwork on common agency is to a large extent a special topic inthe larger field of antitrust economics. In this context, the pre-sumption is that a MAC formally organized by a group offirms would be used for purposes of collusion. As a result,economists’ understanding of common agency is much differ-ent from a MAC as established by cooperatives.

Economists view common agency as arising sponta-neously when one agent represents several major firms in anindustry.2 No overt planning or actual federated organizationneeds to exist for establishing common agency in this inter-pretation. Consequently, ideas on effective coordination andgovernance of common agencies to increase efficiencies andservices are neglected topics.

The dearth of adequate definition and studies of theapplications of MACs has also occurred in economics litera-ture on farmer cooperatives. Most studies of cooperative orga-nization point out that a MAC is a type of federated coopera-tive, but its distinguishing characteristics have beeninadequately defined. In fact, it is even unclear how a cooper-ative MAC is different from a farmer cooperative, other thanto say that the former has organizations, while the latter hasindividuals as members.

Some writers have defined MACs as a type of federatedcooperative that does not take title in marketing transactions.3However, the distinction of title transfer is generally not appli-cable because there are far too many exceptions. In manycases, the process of title transfer is necessary for coordinatingthe marketing of a group of cooperatives.

While this definition is too restrictive, it makes an accu-rate assumption that information sharing and coordinatedselling, the function of marketing agencies that do not taketitle to goods, is the major purpose of MACs, and not titletransfer. Unlike many other federated cooperatives, MACs donot serve the purpose of sharing in the acquisition and owner-ship of financial and physical assets needed for adding valuefrom processing or packaging. In other words, marketingagencies-in-common are organized by groups of cooperatives

2 Esther Gal-Or, A Common Agency With Incomplete Information,” TheRand Journal of Economics, Summer 1991, Vol. 22 No. 2, p. 274-86.3 Robert Cropp and Gene Ingalsbe, Structure and Scope of AgriculturalCooperatives, Cooperatives in Agriculture, 1989, p. 49.

4

to coordinate marketing, with each member retaining exclu-sive ownership over a unique set of physical and human capi-tal. The marketing agency may coordinate a wide range ofpossible value-added services. But, it does not itself serve asan organizational entity for acquisition or development of theassets required to produce either products or services thatconstitute participation in any given industry.

The above definition can be tested by examining whethera group of MACs can be identified from a data set of federat-ed cooperatives on the basis of having relatively low assets.Table 1 summarizes membership and financial data for 63 fed-erated cooperatives at the end of fiscal 1991, as surveyed byACS’s Statistical & Technical Services Staff. Federated cooper-atives with mixed memberships, i.e., farmers and organiza-tions, are excluded from the data set.

These data are sorted in descending order on the basis ofthe proportion of assets to sales. Assets and sales are oftenexpressed as a ratio of sales over assets to measure turnover orassets utilization. This particular ratio is meant to remove the

Table I- Membership and financial data in FY 1991 for 63 federatedcooperatives, 19 below and 44 above 0.1 assets/sales.

Members Assets/Sales Assets Sales

63 Co-ops

Averages

Median

56 0.43

10 0.28

19 Co-ops, with assets to sales below 0.1Averages 9 0.05

Median 5 0.05

44 Co-ops, with assets to sales above 0.1Averages 76 0.60

Median 12 0.42

$47.9 $203.8

$8.9 $36.6

$6.7 $247.6

$2.4 $97.3

$65.7 $184.9

$13.3 $29.7

5

distortion of significant differences in the size of federatedcooperatives. The assets of most MACs are confined to receiv-ables and inventories, yet some with high sales volume willnormally have current assets that are as high or higher thanthe total assets of relatively small processing federated coop-eratives.

Out of 63 federated cooperatives, 19 have a ratio of assetsto sales below one-tenth of total sales. Most well-knownMACs are in this grouping and these organizations are listedby name in appendix table A-l. The designation of MAC isdebatable for a couple of these federated cooperatives, butgenerally this definition holds. The 44 federated cooperativesabove the 0.1 ratio threshold do not appear to contain anyorganizations that regard themselves as MACs. These federat-ed cooperatives are listed in appendix table A-2.

MACs generally have a much smaller number of membercooperatives than other federated cooperatives. The variancein the size of memberships is quite high, as indicated by thedifference between the average and median number of mem-bers for all 63 federated cooperatives and the subgroups of 19and 44. Member organizations were significantly more numer-ous for federated cooperatives with assets-to-sales ratiosabove 0.1, an average of 76 compared with 9 members forthose below 0.1.

When memberships are large, it is often a case of relative-ly small cooperatives having organized a federation to acquirecritical assets for some type of value-added service that eachcould not feasibly carry out on its own. In contrast, smallmemberships usually involve relatively large cooperativesthat establish a federation to coordinate marketing.

Results reported in table 1 confirm a characteristic ofMACs -- they generally have limited assets compared withother federated cooperatives. Average and median assets arelower and sales are higher for the 19 MAC-designated federat-ed cooperatives, as compared with the group of 44.

There are other ways to use assets as a measure for iden-tifying MACs than an assets-to-sales ratio. For example, thecondition that average or median assets of members be largerthan total assets of the federated cooperative could also iden-tify MACs. This latter approach would capture potential situa-tions of MACs that might have assets-to-sales ratios greaterthan 0.1. Such cases could arise if federated cooperatives wereto acquire assets for some type of specialized processing orpackaging activity, while its members maintain exclusiveownership of much larger portfolios of critical assets. Futureresearch in the area of using financial structures to identifydifferent organizational structures could develop this or othertechniques for analyzing MACs.

MULTIPLE PRINCIPALS AND SINGLE AGENT

A principal-agent relationship in business involves anowner (principal), who has hired a manager (agent) to operatethe business. The notion of a single principal is applied whenthere are multiple owners. A “multiple principals-singleagent” relationship is a distinguishing feature of MACs, andrelates to the condition that resources and assets be accumu-lated by each member and not by their MAC.

In recent theoretical work, Bernheim and Whinston applythe term “multiple-principal/single-agent” to the concept ofcommon agency.4 While common agency includes all types offormal coordination,.in addition to marketing, theseeconomists use a common marketing agency as an example.Their study is of a wholesale trade, where manufacturersestablish contracts with merchandise agents or brokers to mar-ket their products to retailers. One of Bernheim’s and

4 B.D. Bernheim and M.D. Whinston, Common Marketing Agency as aDevice for Facilitating Collusion, 7% Rand Journal of Economics, V. 16,No. 2, Summer 1985. Also see their article. Common Agency,Econometricia, V. S4, No. 4, July 1986.

7

Whinston’s requirements for common agency is that the prin-cipals do not cooperate in selecting and in establishing a feestructure for their common agent, nor do they communicateabout prices. These economists say that when such coopera-tion exists, the arrangement becomes the standard principal-agent model.

Bernheim and Whinston do not address the opportuni-ties provided by the Capper-Volstead Act for cooperatives toestablish a MAC. Nor do they reference the application ofMACs as authorized by the Export Trading Company Act.They appear to assume that most formal or incorporated com-mon agencies, established under certain enabling legislation,would have a structural relationship of single rather than mul-tiple principal.

One advantage of requiring that each participating firmin a common agency not communicate with one another isthat they establish a definite and clear rationale for multipleprincipals. In their view, common agency is an unplannedoccurrence of an agent seizing the opportunity to coordinatethe marketing of several firms’ competing products, withouttheir explicit direction or instructions. Even though thesefirms do not control the system of coordinated marketing,they do monitor the common agent’s performance with regardto their own product sales.

Bernheim and Whinston are the first agency theorists todevelop the idea and conditions for multiple principals-agentrelationships. The distinction between multiple and singleprincipals provides a way to explain and understand some ofthe differences between MACs and other federated coopera-tives or a cooperative with direct producer membership.

This report follows a slightly different line of reasoningfrom Bernheim and Whinston regarding criteria for multipleprincipals. Some cooperatives have characteristics that distin-guish them as separate principals in a common agency rela-tionship without the strict requirement that they not formallyorganize.

8

A broader definition of multiple principals, where theydirectly communicate and coordinate their control of agentperformance, makes the distinction with a single principalstructure less clear-cut. In some instances, the distinction canbe made only on a case-by-case basis. Furthermore, a broaderdefinition raises a question as to why all cooperatives cannotbe viewed as controlled by a multiple principal structure.

Multiplicity among members or on the board of direc-tors, and subsequent conflict of interests, is not uncommon inmost cooperatives. However, the distinction of separate prin-cipal status for each member is not based on diverse interests.Rather, it hinges on the fact that some cooperatives, unlikeproducers, have developed distinctive expertise, asset speci-ficity, and intellectual property over many years.

Each cooperative in a MAC has a unique set of these fac-tors or resources that it would like to augment and protect butnot transfer out of its direct and immediate control. Theresults of sorting federated cooperatives into a MAC categorywhen assets to sales fall below 0.1 (table 1) reflect this special-ization of assets by the reluctance of MAC members to shareor to transfer such assets outside their organizations.

PRINCIPAL-AGENT IS A CONTROL FUNCTION

The principal-agent relationship has not been widelyused in economic studies of cooperatives. Some agriculturaleconomists have used the theory of agency in the form of hav-ing both parties to a cooperative arrangement (members andmanagement) as agents. Under this approach, the board ofdirectors are referred to as the “representatives of the residualclaimants,” rather than use the term “principal.“5

The reluctance to regard members of a cooperative,specifically the board of directors, as a principal, reflects amistaken perception that control responsibilities are relatively

5 John M. Staatz, Recent Developments in the Theory of AgriculturalCooperative, Journal of Agricultural Cooperation, V. 2,1987, p. 85.

9

weak in most cooperatives. In fact, there is more control inmost cooperatives, although exercised in a different manner,than in most publicly held IOFs. It is as a consequence ofactive control by directors that most farmer cooperatives con-fine their activities to those that either facilitate members’operations or benefit them by improving the market value oftheir farm production and assets, but may not yield the high-est potential earnings.

Evidence of managerial influence on the formulation ofpolicies and programs of a cooperative, and indications thatthese are consistent with possible private agendas of manage-ment, does not discredit the notion of effective member con-trol. It is impractical to expect that managers would not takecourses of action that are in their self-interest. Effective incen-tives and motivation require that managerial self-interest bealigned with member interests. When interests are notaligned, either performance or governance declines.

The primacy of the principal-agent relationship for coop-eratives highlights several important distinctions with IOFs. Itis often pointed out that IOFs tend to serve investors’ interestsbetter when their control is restricted to general financialstructure issues.6 In other words, an IOF can more effectivelymaximize returns to investors by not being directed into spe-cific lines of business, but instead being able to pursue its owndistinct capabilities in any direction that its managementdetermines.

The meaning of control and the concerns over failure toexercise it in cooperatives and business organizations hasoccupied much attention in academic work and in the mediaregarding executive compensation. Much earlier attention fol-lowed in the wake of Berle and Means work from the 1930s onthe historical transition from owner-manager businesses to thestockholders-manager organized corporation. While the sepa-

6 Robert C. Clark, Agency Costs Versus Fiduciary Duties, see, Principals andAgents: The Structure of Business, ed. J.W. Pratt and R.J. Zeckhauser, 1985,p. 56.

10

ration of ownership from management enabled new sourcesof efficiency and growth -- a critical relationship in coopera-tives -- they identified a concomitant trend of stockholdingowners having lost control.

E3erle and Means’ adopted phrase “separation of owner-ship and control” is unfortunate because it has resulted in thenotion of a principal’s responsibility to control as an activitythat management can appropriate. A more accurate expres-sion is that control is lacking. Harold Demsetz noted this dis-tinction when he used the term “control vacuum” to discussthe separation of ownership and control.7

Although managers have certain fiduciary control tasks,the predominant responsibility for control over the directionof a cooperative lies with the board, not the manager.Control involves monitoring managerial performance anddirecting a firm’s course of action. It is applicable to manage-ment only if one wanted to introduce the idea of managershaving self-control.

The responsibility of cooperative membership for controlcan be divided between governance and performance.Governance is widely recognized as a key element of controland is emphasized by ACS, reflecting concerns about theintegrity of a cooperative when professional management pur-sues business opportunities that provide high returns but donot enhance the value of member products. In such coopera-tives; governance control is lacking.

The notion of “performance control” may seem unrealis-tic because performance is subjected to many uncontrollableenvironmental factors. However, in exercising performancecontrol, directors must evaluate how well management carriesout the objectives of their cooperative. Changes in the generaleconomy that have affected earnings should not influencetheir evaluations.

7 H. Demsetz, The Structure of Ownership and the Theory of the Firm,Journal of Law b Economics, V. 26, June 1983, p. 387.

11

Evidence of effective governance and performance con-trol are long-run market development programs that success-fully enhance the value of member products. Members cap-ture part of this gain in increased value of their productionresources. This is preferable to benefits that arise only fromincreases in a cooperative’s earnings.

The above discussion of control applies to all types ofcooperatives, but the multiple principal structure of a MACinvolves some different points of emphasis. A MAC requireslittle governance due to its low accumulation of assets. In con-trast, performance control of MACs is critical and often diffi-cult. When a MAC’s performance is well below its potential, itis often due to each principal’s preoccupation with its ownorganization’s performance or to a lack of incentive for effec-tive direction. Weak performance of MACs can also arise fromdifficulty in achieving consensus on objectives in a multipleprincipal structure.

A multiple principal structure also affects the role ofmanagement. While a cooperative manager is in the role ofagent for farmers, its managerial function requires much ini-tiative and leadership, perhaps more than what is usuallyassociated with the term “agent.” The notion that what agentsdo is to strictly carry out the directions of their principalsinadequately reflects the true dynamic of most principal-agentrelationships. However, the specialized functions and multipleprincipal structure of MACs place their management in therestrictive type of agent role.

Several MACs have effectively overcome the special chal-lenges that result from a multiple principal structure. Some ofthem are reviewed below.

EXAMPLES OF MARKETING AGENCIES-IN-COMMON

Most MACs discussed here were included in the dataanalysis of table 1. All were below the 0.1 assets/sales ratiothreshold. A couple MACs that went out of business are also

12

discussed in terms of how they demonstrate consequenceseither of too large a membership or of involvement with oper-ations more effectively carried out by consolidated organiza-tional structures. Typically, these examples show that MACsare ineffective for building assets and intellectual propertyresources.

Most relatively large cooperatives that have substantialmarketing expertise, innovative processing techniques, andestablished brand name awareness have slight incentive toconsolidate or even federate with other cooperatives. Whenconstraints or suboptimal economies of size are encounteredin isolated facets of their operations, these cooperatives oftenseek some type of strategic alliance-- a MAC or other arrange-ment.

For example, a group of cooperatives may each beinvolved with marketing byproducts that could be more effi-ciently handled by combining this function into a single orga-nization. In addition, they may not want to focus theirresources on byproducts because management is evaluatedpredominantly on how it enhances prices and develops mar-keting programs for the primary product of its members.

Midwest Agri-Commodities, Inc., is an example of aMAC for marketing intercooperative byproducts. It merchan-dises beet pulp and molasses for three sugar beet processingcooperatives. Assuming the relevant range of per-unit cost formarketing beet sugar byproducts declines with increased vol-ume, a MAC reduces total costs in comparison with the aggre-gate cost of each cooperative for the same volume of byprod-ucts sold.

The marketing of beet sugar may have a similar coststructure, but byproducts are an easier first step for intercoop-erative coordination. A couple members of Midwest Agri-Commodities have established a MAC for their sugar, NorthCentral Sugar Marketing Co-op. Currently, the other memberin the byproducts MAC is considering membership in NorthCentral for coordinating its sugar marketing. Louisiana Sugar

13

Cane Products, Inc., a highly effective MAC for cane sugar,has six members.

The relative ease of gaining acceptance of MACs forbyproducts is also demonstrated by contrasting soybean andcottonseed processing. In 1963, several cooperative soybeanand cottonseed processors established a MAC called Soy-CotSales, Inc.

The soybean side never effectively coordinated its mar-keting in this organization, but by the early 198Os, competitivepressures forced participants to either achieve an economiesof size threshold or exit the industry. Merging soybean pro-cessing cooperatives or divisions into a non-MAC federation,Ag Processing, Inc., was the outcome. Soy-Cot continues tooperate as a successful MAC, specializing in marketing veg-etable oil for its cottonseed processing members.

Another segmented market situation that lends itself tothe establishment of MACs by cooperatives is the handling ofnon-member sales as a separate operation. This condition ispresent in the cattle artificial insemination (AI) industry Inaddition to a product, AI cooperatives also offer substantialtechnical service to their members for optimizing its applica-tion. Production above member needs is available for non-member sales without providing the same amount of technicalservices. Non-member sales, including exports, is a separableoperation. In terms of separability, it is akin to the handling ofby-products.

Three different groups of AI cooperatives have estab-lished MACs to handle non-member sales and exports. AIcooperatives also have a marketing agreement with WorldWide Sires to export to Europe, Africa, and Asia. By specializ-ing in non-member sales, most of these MACs are organizedas partnerships, without incorporating as cooperatives.

The key point is that value-added components are con-trolled in a more localized or regional structure by individualcooperatives, while activities where control is less of a concernare the components organized into MACs.

A highly effective MAC, Amcot, Inc., provides coordinat-ed marketing for the primary product handled by four largegrower-member cooperatives. Amcot’s success points out thata potential advantage for organizing MACs for by-products orsegmented marketing is just that, an advantage and not a nec-essary condition for a successful MAC.

Amcot illustrates that MACs having cooperatives with asubstantial amount of non-competing products reduces theextent of members’ concern about sharing a marketing agency.It is misleading to regard Amcot members’ cotton as perfectsubstitutes, such as with grain cooperatives. In fact, one of thereasons Amcot works well is that, although each of its fourmembers handle some similar cotton types and qualities, theyare generally offering significantly different growths. Amcot isto some extent a multicommodity cooperative. An agent needsto represent a wide variety of cotton to provide effective rep-resentation in world cotton markets. This situation is perfectfor cotton cooperatives. Limited to the particular kinds of cot-ton in their member region, they can compete internationallywith a more comprehensive offering to textile mills withAmco t.

There are examples of participation in MACs by coopera-tives with highly developed marketing programs and estab-lished brands. Sun-Diamond is the best known. Much of itscoordination responsibilities involve public policy and rela-tions. Sun-Diamond’s marketing responsibilities focus onactivities and functions that are subject to economies of size. Itprovides order entry, coordinates distribution, handlesaccounts payable and receivable, manages a network of for-eign sales representatives, and handles export documentation.

Member cooperatives provide market development,advertising, product improvements and developments, selec-tion of product lines for export, sales objectives, and pricing.These activities influence the core assets of each member.While such activities may also benefit from economies of size,

15

they are too critical and involve special expertise, so thatshared management and control are unacceptable.

Sun-Diamond’s acquisition of Sunland, a proprietarydried fruit and nut company, has brought to the fore the issueof coordination and compromise in the use of brands.Sunland’s brands were vacated. Its product lines have beenmarketed under various trademarks of the members, primari-ly the Sun-Maid brand. Sun-Diamond manages the Sunlanddivision and selects the member brands to be used. This areaof MAC operations has been one of the most contentious formembers. While different from a situation of a MAC manag-ing a members’ brands and the development of its markets, itinvolves decisions about the use of brands that are not underthe complete control of each member organization.

Cooperating Brands, Inc., is a MAC organized to handlebrands of member cooperatives for specialized packaging. Itwas established by six fruit processing cooperatives to man-age a special institutional juice pack.

A small frozen juice cup is a required type of packagingfor a certain segment of the institutional market. None of thecooperatives wanted to invest in this type of packaging, andpreferred co-packing arrangements. However, there are signif-icant economic advantages to having both large volumecopacking arrangements and a comprehensive line of juicevarieties when selling to institutional buyers. Opportunitiesfor special packs of different kinds of fruit snacks have alsodeveloped.

Cooperating Brands has a trademark licensing agreementwith each member cooperative. In a copacking arrangement,the packing plant does not take title, and has no rights to usethe brands on their own products. A confidentiality agreementrestricts the packer or bottler from making any use or releaseof the proprietary juice formulations. Recently, member coop-eratives have avoided this problem for most of their productsby shipping pre-formulated concentrates to the bottlers.

16

These contractual arrangements indicate the extent ofproprietary information, and hence, specific capabilities thatvalue-added processing cooperatives possess. Juice marketingcooperatives are highly successful and competitive with oneanother.

Cooperating Brands gains access to a specialized andsegmented market, while protecting members’ distinctivecompetence from duplication by competitors. Those coopera-tives that have developed intellectual property and specialtechnical expertise in food processing and marketing will useMACs to achieve economies of size for better access to partic-ular market segments. Such cooperatives will typically estab-lish individually designed contracts with the MAC, makingthe multiple principal structure of the organization veryexplicit.

Although examples of forming a MAC to pool memberresources for market development are few in number, Norbeststands out as a potential model. However, it is important tounderstand that its accomplishments are not a recent develop-ment. Norbest was formed in 1929 to handle turkey sales for20 cooperative packing associations. It was an innovator inusing the branded concept for marketing fresh meat, and untilrecently, was the largest marketer of turkeys in the world.

A similar case, with historical roots, is the early years ofLand O’Lakes, Inc. (LOL). It was originally established as afederated cooperative to coordinate the promotion of a singlebrand of butter among several cooperative creameries. It sub-sequently evolved into a diverse food and farm supply pro-duction and distribution association, with a mixed member-ship structure.

In recent years, dairy cooperatives outside the Midwesthave joined LOL to pack butter using its famous trademarkand coordinate the distribution of those particular productlines. Even though these arrangements parallel both its earlyperiod and the Norbest system, LOL’s evolution as a coopera-

17

tive extends far beyond the range of the coordinating activi-ties of a MAC.

Like most MACs, Midwest Agri-Commodities, AMCOT,and Soy-Cot are examples that do not involve major brand-name products and operate in price-competitive commoditymarkets. The milk marketing MACs, Central Milk ProcessingCooperative and the Regional Cooperative MarketingAssociation, have operated pools for members in their market-ing order areas. These pools have helped stabilize prices andearned over-order premiums.

In apple marketing in the Northwest, federated market-ing for cooperative packing houses dates back to 1922, andpersists today with Wenoka Sales. It sells directly to the gro-cery chains for six members. Wenoka is an alternative to con-tracting with a distributor.

The key advantage of this MAC is that members canimprove their earnings by participating in larger volume salesand by having more control over their negotiating positions,than if each were independently contracting with outside dis-tributors.

The challenge of maintaining product commitment frommembers of a MAC, particularly when operating in activelytraded commodity markets, often raises the issue of marketingagreements. Although marketing agreements are notaddressed in this report, their application by MACs in someform is usually necessary.*

Poorly designed marketing agreements have contributedto the failure of some MACs, by allowing some members togradually reduce their product commitment. This hasoccurred in grains with Producers Export Company andFarmers Export Company, as well as in dried beans, withValley Marketing, Inc.

B As an example of terms in a marketing agreement for a MAC, see,Carolyn Liebrand and Karen Spatz, Exporting: A Marketing Agency-ln-Common for Dairy Cooperatives, ACS Research Report 126.

18

Without exploring all the reasons for the demise of theseMACs, it is worth observing that these examples exhibit thesegmentation pattern discussed earlier. In each case, a groupof cooperatives identify a particular marketing function ormarket segment that they are willing to delegate to a MAC.

Valley Marketing consisted of local grain cooperativesthat handled a relatively small volume of beans from some oftheir members and a couple of predominantly specializeddried bean cooperatives. It was inefficient for the grain coop-eratives to each market a small volume, so a MAC workedwell for them.

However, their interests eventually conflicted when thetwo local dried bean cooperatives wanted to build their ownprocessing facilities for frozen beans. This pressured ValleyMarketing to buy non-member beans to maintain customeraccounts. The two largest members opposed the idea of non-member business and dropped out of the organization, leav-ing it with an unsustainable level of volume.

The case of the grain marketing interregionals reflectedthe segmentation pattern around exporting. Regional graincooperatives reasoned that grain exporting, particularly inoperating port elevators, was subject to more economies ofsize than they could individually attain. Hence, the regionalcooperative members handled domestic marketing, while theinterregional specialized in exporting. But grain markets can-not be segmented in this way for effective trading. A success-ful MAC for marketing large volumes of grain would requireresponsibility for both export and domestic marketing deci-sions.

Delegating special activities to a MAC can work in manycontexts, but this approach is destined for failure when coop-eratives have tried to separate responsibility for grain export-ing from control over domestic marketing. However, the dis-advantages of divided responsibility do not apply to morelocalized grain marketing systems. Given the recent reductionin the number of regional grain cooperatives, many relatively

19

small groupings of local cooperatives have established MACsfor coordinating shipments to export locations or to more dis-tant milling destinations. Members use a part of their grainvolume to supply their feed mills and local buyers, allocatingthe balance for marketing by their MAC. This type of decen-tralized system of decisionmaking and of divided responsibili-ty does not encounter the magnitude of risk and inefficiencythat such a segmented system has in the context of regionaland interregional grain marketing.

MACS VERSUS ALTERNATIVE FORMS

When a group of cooperatives consider forming a MAC,they must compare expected benefits with their current sys-tem and evaluate other alternatives for increasing their scopeand coordination of operations. This section compares MACswith general alternatives.

In some cases, however, cooperatives may regard a MACas a potential transitional structure, rather than a mutuallyexclusive alternative. By establishing a MAC, members buytime to obtain more information for evaluating future stepstoward either complete consolidation or partial combinationswithin a federated cooperative.

There is a potential weakness in trying to use MACs as amiddle ground or transitional step. A decision to form a MACcan often be a delaying tactic in that cooperatives postponehaving to make difficult restructuring decisions. Rather thanlay groundwork for making difficult but economically neces-sary adjustments, MACs can become entrenched. In sum, thecoordination that a MAC provides, without requiring consoli-dation or transferring of assets, can work in either positive ornegative ways. It is preferable to choose an alternative “as if”it were a final and irreversible decision. The evaluation pre-sented below considers MACs strictly as a competing ratherthan a transitional alternative.

20

While there are numerous types of benefits and differentconsiderations in ranking the alternatives, this discussionselects a few key points and some general assumptions inmaking an evaluation. It is a simple example of what, in actualpractice, is a complicated process. The following four organi-zational structure alternatives are contrasted and compared:

1. MAC,2. Non-MAC federated cooperative (NMFC),3. Consolidation or merger of cooperatives (COM),4. No intercooperative coordination (NIC).

MACs are primarily responsible for marketing coordina-tion. If the members of a MAC instead formed a COM struc-ture, it would accomplish marketing coordination and handlethe full range of activities and member services otherwise con-ducted on a smaller scale by several cooperatives in an NICstructure. NMFCs take on more functions than MACs, but arealso an incomplete cooperative structure without the linkageto the services and functions of their member organizations.

To compare these four organizational alternatives, theoperations of member organizations must be evaluated as asingle system with their federations, MAC or NMFC, to forma complete structure.

Six objectives used to evaluate each alternative are:1. governance control,2. performance control,3. economies of size,4. use of capacity and assets,5. development of assets and distinctive capabilities, and6. member involvement and loyalty

Governance Control concerns directors’ responsibility tooversee the activities of a cooperative in compliance with theobjectives and bylaws of the association and civil and com-mercial laws and regulations.

21

Table 2-Ranking of organizational SASrnStiVSS by SffSCtiVSnSSS inaccomplishing objectives.

Governance Control1. NIC No Intercooperative Coordination2. MAC Marketing Agency-in-Common3. NMFC Non-MAC Federated Cooperative4. COM Consolidation Or Merger

Performance Control1. NIC2. COM3. NMFC4. MAC

Economies of Sizel.COM2. NMFC3. MAC4. NIC

Use of Capacity and Assets1. COM2. MAC3. NMFC4. NIC

Development of Assets and Distinctive Capabilities1. COM2. NIC3. NMFC4. MAC

Member Involvement and Loyalty1. NIC2. MAC3. NMFC4. CdM

22

The alternative of no intercooperative coordination (NIC)is the highest rank in table 2 regarding ease or simplicity ofgovernance. The fact that a consolidation or merger (COM) isplaced fourth should not imply that smaller cooperatives havesuperior governance. Rather, governance is more complex inlarge cooperatives and federated systems. Even so, with sys-tems for voting and board representation that reflect member-ship diversity, governance can be effectively carried out inlarge organizations.

Governance responsibilities of most MACs, due to lowassets, are usually less demanding than in other cooperativeorganizational forms. In contrast, governance is the major con-trol challenge when cooperatives confront a consolidation ormerger (COM) decision. In fact, when merger discussionsbreak, it is often due to uncertainties about governance con-trol. Questions about a proposed merger creating poor perfor-mance are rare.

Perfomzance Control -- motivating and monitoring cooper-ative staff to meet a cooperative’s objectives at the highest lev-els of achievement.

MACs are ranked last in performance control because ofthe challenges involved with coordination. Cooperatives witha MAC would be expected to outperform those operating withno intercooperative coordination (NIC), but members’ exerciseof control over performance is often weaker. In fact, when aMAC has improved results over each members’ previous NICexperience, such success may mask the potential for betterperformance, if directors were to exercise more control.

While coordination is a MAC’s key performance advan-tage over a NIC, it is difficult to control in a multiple principalstructure. Each participating cooperative remains in existence,retaining their respective assets, so that their primary perfor-mance concern is with their own cooperative. In fact, the prin-cipals that serve on a MAC’s board are usually managers of

23

the member cooperatives, and their concerns can weaken per-formance control.

For example, some members may become concerned ifothers appear to benefit more. They may become preoccupiedwith establishing rules for how benefits and costs will be equi-tably shared or veto proposed operations which they antici-pate would not benefit them as much as other members. In themultiple principal structure of MACs, these governance con-cerns are apt to move to the forefront, while performance con-trol becomes less important.

Economies of Size -- when larger capacity results in lowerper unit costs.

Economies of size is the classic advantage for many largebusinesses. Its effects are usually significant in agriculturalproduction, processing, and marketing. In regard to physicalhandling and processing of agricultural products, transporta-tion costs often limit the extent of economies of size. By con-trast, those aspects of marketing that involve communicationsand movement of documents are likely to have lower per unitcosts with increased capacity.

For example, transactions costs for a large sale are oftenproportionally smaller than costs on a smaller volume sale. Ifa MAC accomplishes fewer but larger sales transactions thanseparate transactions by several different agents for the samevolume, then it has an economies-of-size advantage over thefragmented structure of NICs. COM and NMFC alternativesaccomplish the same economizing of transactions, while alsoachieving more economies of size in physical facilities forvalue-added services than the members of MACs or coopera-tives in an NIC structure.

Use of Capacity and Assets -- involves the concept ofturnover, with increases in sales volume on a fixed capacity orin relation to a firm’s assets resulting in a lowering of per-unitcosts and a higher return on assets.

24

The case for a COM alternative (table 2) is usually basedon the advantages of higher use of member facilities, as wellas economies of size. When use of capacity or assets is the keyobjective, a single organization structure has advantages inavoiding duplication or excess.

MACs are often more effective than other types of feder-ated cooperatives in improving members’ access to more mar-ket outlets, and in this way, helping them increase their vol-ume of throughput. However, this type of comparison is onlymeaningful in terms of each member’s sales volume. It doesnot consider the value added from the operations of a non-MAC federated cooperative (NMFC), which increases marginsper unit of sales.

NMFCs, with significant assets to use, do not function aswell as MACs in the agent mode of helping members findbuyers. In some cases, MACs operate on a type of brokerage-fee basis. If members independently make sales, their MAC iscredited. This operating policy creates a large network of mar-ket contacts. In contrast, many NMFCs have processing sched-ules that require supply commitments from members.

Development of Assets and Distinctive Capabilities -- overtime, a business will accumulate a wide range of assets thatextend beyond those with reported book values to includeintellectual property, such as trademarks and patents, as wellas human capital development of management and other per-sonnel.

Developing unique assets and human resources are thetools for creating a value-added system for member products.It is often believed that adequate financing is sufficient to pro-vide an effective value-added system, and that common agen-cy can coordinate the needed financing. However, this viewneglects a critical ingredient, an organizational environmentfor innovation. While the potential exists for success in usingMACs for value-added marketing strategies, MACs’ multiple

25

principal structure is a disadvantage, as pointed out by someof the examples.

Value-added strategies involve internal development ofstaff capabilities, as well as substantial accumulation of “bookvalue” types of assets. Historically, most cooperatives havecarried out this type of development within a single organiza-tion structure. If they lack adequate resources, cooperativeshave often joined together in a COM structure. When a groupof cooperatives want to coordinate such long-term programsfor value-added activities, their established organization willusually be a NMFC rather than a MAC.

MACs do not provide the kind of managerial leadershipand initiative needed for conducting most value-added pro-grams. Similar to the problems of performance control, a mul-tiple principal structure often involves a “management bycommittee” approach. While such an approach has advan-tages for certain tasks, examined in the last section of thisreport, it can inhibit the necessary risk-taking and the long-term nature of developing products, techniques, and specificexpertise.

Confusion and conflict over property rights are moreprevalent in the multiple principal structure of common agen-cy, which would tend to undermine the development of avalue-added program.

Member Involvement and Loyalty -- involves active patron-age and keeping informed about a cooperative’s policies andprograms.

Cooperatives of all sizes and structures should neverassume member involvement and loyalty, and must makeexplicit efforts to maintain strong member relationships. Ingeneral, a cooperative with a relatively small number of pro-ducer-members and limited geographic scope has moreopportunities for keeping members involved and loyal. Whena group of cooperatives contemplate forming a federated

26

organization or consolidating, the latter alternative often rais-es concerns about potential erosion of member relations.

More than other types of federated organizations, a MAChas an advantage in requiring less commitment of resourcesand allowing member cooperatives more independence withtheir operations. In addition, MACs do not require complicat-ed systems of voting and layered structures of boards andmanagement, usually concomitant with the formation of largecentralized or federated cooperatives. These systems and lay-ered structures facilitate the process of control and the mainte-nance of member relations. But they also involve additionalcosts that a MAC alternative largely avoids.

Table 2 demonstrates how organizational alternativesmight be ranked by a typical group of cooperatives. In anactual study of alternatives the goals or objectives will usuallybe more specific than the very general ones provided in thisdiscussion. Market access conditions vary by commodity, andMACs can be used to conduct diverse business strategies.Their most common service is to provide coordinating mecha-nisms for cooperatives to have adequate volume to build trad-ing relationships with major buyers.

COORDINATING STRATEGIC PLANS

Cooperatives confront many distinct economic chal-lenges that underline their need for coordinated planning andopportunities to form MACs. A cooperative system is a dualbusiness of cooperative operations and of members’ farmenterprises. The major challenge for cooperatives, that IOFsneed not confront, is to simultaneously facilitate and augmentthe economic performance of their members’ businesses,while maintaining their own financial solvency.

A cooperative provides many services that otherwiseeither would not be supplied or offered with less quality orfewer options. While cooperatives only offer services thatmaintain their earnings, they cannot carry out their agent

27

responsibility by strict reliance on the same kinds of profitmeasures and market signals that IOFs use to guide their deci-sions on supplying services. Cooperatives use both traditionaland special financial measures, but by serving the interests ofother business entities, i.e., farms, the monitoring of their per-formance is more complex than that of a for-profit IOF.

Another distinct economic challenge for cooperatives arebenefits they provide that cannot be exclusively captured bythose who have paid for them through membership. Theseexternal effects occur from marketwide actions of cooperativesin negotiating higher prices with processors, managing aneven flow of product deliveries, or in conducting promotionalactivities that enhance prices.

Intercooperative coordination of planning can help coop-eratives manage their distinct challenges. MACs are one vehi-cle for that purpose. MACs increase cooperatives’ knowledgeof one another so their plans and actions are more likely toreinforce rather than adversely affect each other. The latter sit-uation is more apt to happen from extremely independent andisolated planning.

Members of federated cooperatives customarily shareaspects of their business plans that relate to the effective func-tioning of their federation. With the relatively large member-ships of most non-MAC federations, such sharing of informa-tion is often too limited in scope to be regarded as coordinatedplanning. By contrast, the smaller memberships in mostMACs afford more opportunities for member contacts.Although members of MACs maintain confidentiality of theirrespective organizations’ strategic plans, close working rela-tionships foster more sharing of information and marketingideas than would occur without a MAC.

Coordinated planning is often required for MAC opera-tions, particularly with few but large member cooperatives,where each provides a large share of the collective sales vol-

28

ume. In addition, cooperative MACs often become a mecha-nism for coordinated strategic planning.

The group dimension that a MAC brings to the process ofstrategic planning is one of its key advantages, and one that islacking in the organizational alternatives discussed earlier. Incommenting on this dimension of MACs, a cooperative man-ager once made the point:

In a marketing agency-in-common you have moregroups helping you make a decision. Centralized deci-sion-making is for an opportunist, and cooperativesare not in that business.g

The central point of this observation is emphasized inmuch recent business literature on decisionmaking. Oneexample points out the advantages of collective versus indi-vidual decisions:

Groups are likely to outperform individuals onlyto the extent that productive conflict arises amongtheir members and such conflicts get resolvedthrough balanced debate and careful intelligence-gathering. When that happens, a group is likely tounderstand the issues better than an individual, andis more likely to choose wisely.lO

These attributes are effective for all types of organiza-tions or teamwork situations. The difference for MACs is thatlacking a chain of command and leadership role for manage-

g Edward Breihan, interview on May 14,198O. Former manager of SWIG,Inc., and Plains Cooperative Oil Mill, who was also a director for Amcot,Soy-Cot, and the Farm Credit System.lo J. Edward Russo and Paul J.H. Schoemaker, Decision Traps, 1989,p. 145.

29

ment, they are especially dependent upon effective groupdecisionmaking. MACs provide a system where individualexperience and expertise from several organizations isbrought together to address common problems.

But cooperative MACs do not yield easy advantages.This is evident by the relatively small number that have beenestablished in recent history. The highly decentralized natureof a MAC’s leadership, or multiple principal structure, placesgreater demands on group working than exists with singleprincipal structures. When there is a willingness to meet suchdemands, MACs can provide significant benefits from coordi-nated marketing and strategic planning. Marketing agencies-in-common are also more conducive to a wider application ofcooperative methods than if cooperatives are more organiza-tionally independent or isolated.

30

Table A-I- Nineteen Federated Cooperatives with Assets/Sales Below0.1, FY 1991.

Midwest Agri-Commodities CompanyAllied Federated CooperativeAtlantic Processing, Inc.Waterloo Service CompanyNorbest, Inc.National Woof Marketing CorporationSeald-Sweet Growers, Inc.Equity Co-op Livestock Sales Assn., Inc.Wenoka SalesMarketing Association of America CooperativeHorticultural Producers Federated Assn.Cooperating Brands, Inc.Interstate Producers Livestock Assn.Sun-Diamond Growers of CaliforniaSoy-Cot Sales, Inc.Louisiana Sugar Cane Products, Inc.Texas Cooperative Marketing ExchangeChicagoland Dairy Sales, Inc.Amcot, Inc.

31