kersi d. antia, xu (vivian) zheng, and gary l …moorman/marketing-strategy-seminar... · kersi d....

KERSI D. ANTIA, XU (VIVIAN) ZHENG, and GARY L FRAZIER*

Franchise relationships are prone to conflict. To safeguard the rights ofindividual franchisees, several states have legislated greater franchisordisclosure (registration law) ex ante and/or franchisor "termination forgood cause" (relationship law) ex post. The impact of regulatory oversighton franchisor-franchisee conflict, however, remains unclear. Relying onagency theory arguments, the authors first assess the influence of theregulatory context, both by itself and in combination with the franchiseownership structure, on the incidence of litigated conflict. Conditional onlitigation, they also predict the impact of franchise regulation on both theparties' litigation initiation and resolution choices and the resultingoutcomes. The authors test the hypotheses using a unique multisourcearchival database of 411 instances of litigation across 75 franchisesystems observed over 17 years. The results indicate that the regulatorycontext, by itself as well as in combination with the franchise ownershipstructure, significantly shapes parties' conflict management choices. Theauthors also find evidence of a trade-off between prevailing in theparticular conflict and achieving franchise system growth objectives.

Keywords: franchise registration law, relationship law, ownershipstructure, conflict

Conflict Management and Outcomes inFranchise Relationships: The Role ofRegulation

Franchising is a mainstay of the U.S. economy, with morethan 825,000 franchised outlets accounting for $2.1 trillionof U.S. gross domestic product (GDP) (International Fran-chise Association 2013). The typical "business format"franchise relationship calls for the franchisee to make alump sum and ongoing payments to the franchisor and toabide by the latter's operational stipulations. In return, thefranchisor provides ongoing support to franchisees and

*Kersi D. Antia is Associate Professor of Mariceting, Richard IveySchool of Business, Western University (e-maii: [email protected]). Xu(Vivian) Zheng is Assistant Professor of Marketing, Coliege of Business,City University of Hong Kong (e-mail: [email protected]). Gary L.Frazier is Richard and Jarda Hurd Professor of Distribution Management,Marshall School of Business, University of Southern California (e-mail:[email protected]). The authors are grateful for the feedback fromseminar participants at the 2011 Haring Symposium at Indiana Universityand at the University of Alberta and Oklahoma University. They also thankJan B. Heide and Aric Rindfleisch for their insightful comments on previ-ous versions of this article as well as Tsai-Fang Chen and Neel Khurjekarfor their data collection and coding assistance. Christine Moorman servedas associate editor for this article.

monitors and enforces quality standards across members ofthe franchise system on a continual basis (Shane 2005). Vio-lations or missteps by either party could cause conflict ifone partner perceives the other as indulging in acts thatimpede the achievement of its own goals (Etgar 1979). Ifthe aggrieved partner's claim for redress is denied or con-tested, this constitutes a dispute (Galanter 1983). Left unre-solved, serious disputes may lead to costly litigation, dys-function, and even termination of the relationship.

In an attempt to minimize such costly dysfunctionalconflict and to safeguard the interests of individual fran-chisees, 15 states now require franchisors to register withstate regulators before offering franchise opportunities.Furthermore, 22 states have enacted laws that require fran-chisors to provide "good cause for termination" of theirfranchisees. Although these laws are long-standing (bothwere enacted in 1974), such regulation of franchising hasalways been controversial. Whereas supporters of regula-tion laud the transparency and the "franchisee day in court"elicited by such laws (Benoliel 2009), its detractors cite thehigh cost of compliance and the additional burden it

© 2013, American Marketing AssociationISSN: 0022-2437 (print), 1547-7193 (electronic) 577

Journal of Marketing ResearchVol. L (October 2013), 577-589

578 JOURNAL OF MARKETING RESEARCH, OCTOBER 2013

imposes on the franchisor's task of quality assurance(Brickley, Dark, and Weisbach 1991; Klein 1980).

The marketing discipline has remained, for the most part,on the sidelines of this ongoing debate, despite a rich tradi-tion of scholarship on channel member conflict (see, e.g..Brown and Day 1981; Ganesan 1993; Hibbard, Kumar, andStem 2001). Such reticence may be attributed to two relatedlimitations of prior research. First, although acknowledgingthat conflict is "a dynamic process composed of a series ofconflict episodes" (Brown and Day 1981, p. 263), previousstudies in marketing have relied overwhelmingly on cross-sectional, survey-based assessments that are incapable ofshedding light on unfolding conflict dynamics. A secondsignificant gap in understanding pertains to the "conflictaftermath," or the outcomes associated with both parties'conflict management efforts (Pondy 1967). Prior attemptsto assess conflict management consequences have tended tofocus on immediate and dyad-specific outcomes (Ganesan1993; Hibbard, Kumar, and Stem 2001). Such an emphasisignores the critical notion that "[a]n organization's successhinges to a great extent on its ability to set up and operateappropriate mechanisms for dealing with a variety of con-flict phenomena" (Pondy 1967, pp. 299-300). As a result,the central issues of whether the franchise regulatory con-text serves to reduce or increase the incidence of seriousconflict between franchisors and their franchisee partners aswell as the implications of both parties' conflict manage-ment efforts for dyadic and franchise system-wide out-comes remain unaddressed. Therefore, a rigorous longitudi-nal study of actual conflict episodes based on relativelyobjective accounts of their initiation, resolution, and corre-sponding outcomes is necessary.

The present study addresses this need. We regard litiga-tion between franchise partners as representative of seriousmanifest conflict, defined as open behavioral disagreementsbetween the firms that hamper their respective goal attain-ment (Pondy 1967). We examine both parties' litigation ini-tiation and resolution choices and the dyadic and system-wide consequences attributable to these choices. Relying onagency theory-based arguments, we assess the impact ofregulation on the incidence of litigated conflict betweenfranchisors and their franchisees. We hypothesize the impactof such regulation to vary significantly by (1) the level ofanalysis, either across the franchise system (regulated andnonregulated markets) or within the particular market inwhich the conflict occurs; (2) the specific regulation consid-ered (registration law or relationship law); and (3) the owner-ship structure of the channel system—specifically, the extentto which the franchisor relies on franchisee-owned outlets.We collect multisourced archival data with respect to 411instances of htigation for 75 ñ-anchise systems over a 17-yearobservation window. Conditional on litigation occurrence,we identify the party more likely to initiate the litigation,predict the likelihood of the litigation being settled by alter-native dispute resolution (ADR) rather than adjudication,and assess the dyadic and systemwide outcomes that resultfrom the prior litigation initiation and resolution choices.

In doing so, we make several important contributions to themarketing literature on conflict management and its associ-ated outcomes. First, in assessing how variations in regulatorypractice shape franchisors' and franchisees' litigation-relateddecisions and their attendant outcomes, the present study

speaks to the efficacy of franchise regulation. We are thusable to provide insights into the interplay between regula-tory and distribution structure-related determinants of seri-ous manifest conflict incidence and its management (Dantand Schul 1992). Second, we assess not only the dyadic,more immediate consequences pursuant to channel part-ners' conflict management choices but also the franchisesystem-wide, long-term outcomes attendant to such choices.As we subsequently demonstrate, a single-minded focus onprevailing in the immediate issue at hand could prove counter-productive to long-term aspirations involving the widerchannel system. The present study underscores the lattertrade-off and promotes a nuanced understanding of the intri-cacies of conflict management. Third, our examination ofactual litigation choices and outcomes gleaned from fran-chise disclosure documents (FDDs; formerly known as uni-form franchise offering circulars), public court records, andmultiple franchise industry-specific trade publications mini-mizes the well-known survey data-related concerns andprovides rich insights into the evolving dynamics of seriousconflict and its outcomes.

In the sections that follow, we first provide an overviewof the anticipated role of registration law and relationshiplaw in mitigating conflict between franchisors and theirfranchisees. We then propose hypotheses linking each law,by itself as well as in combination with the franchisor'sownership structure, to franchisor-franchisee conflict man-agement choices and their attendant outcomes. This is fol-lowed by a description of our empirical context and datacollection and analysis methods. We conclude with themanagerial and theoretical implications of our study.

THE FRANCHISE RELATIONSHIP AND ITSREGULATION

Franchise relationships are subject to agency problems,whereby incentives exist for both the agent (the franchisee)and the principal (the franchisor) to misrepresent them-selves and/or their abilities before the relationship begins(ex ante adverse selection) as well as to renege on their per-formance obligations (moral hazard) during the course ofthe relationship (Bergen, Dutta, and Walker 1992; Rubin1978). Franchisors safeguard their interests by designingand offering relatively complete and one-sided contracts(Kashyap, Antia, and Frazier 2012) specifying their fran-chisees' obligations before, during, and after the relation-ship. Although such contracts safeguard the franchisor'sinterests well, they remain conspicuously vague about thelatter's performance obligations. Moreover, there is littleroom for negotiating the terms of such contracts (Shane2005). Franchisees may "take it or leave it" and must relyonly on the franchisor's concem for its reputation to serveas its bond (Klein 1980).

Franchisors may frequently misrepresent themselves topotential franchisees with regard to (1) their abilities,resources, and prior success and/or (2) their intentions tocontinue the relationship (Rubin 1978). Whereas the formerreflects an ex ante inability of franchisees to obtain criticalinformation with respect to their franchisors' ability andmotivation to perform, the latter represents franchisees' expost vulnerability to unanticipated changes or relationshipdissolution at the franchisor's whim. Both situations canresult in serious conflict and potentially "the failure of par-

Conflict Management and Outcomes in Franchise Relationships 579

ties' ability to reach an acceptable solution by interpersonalmeans" (Dant and Schul 1992, p. 40). We regard the inci-dence of litigation as indicative of serious manifest conflictbetween franchisor and franchisee, undertaken as a conflictmanagement strategy.

In response, several measures designed to safeguard fran-chisees' interests have been legislated. First, the FederalTrade Commission requires all franchisors to provideprospective franchisees with information about themselvesat least ten days before any franchise agreement is signed(Shane 2005). In addition, individual states have adoptedregistration law and relationship law statutes regulating thefranchise relationship to varying degrees. Figure 1 summa-rizes the regulatory emphasis of each statute. Registrationlaw specifies the information franchisors must disclose topotential franchisees before the induction of the latter intothe franchise system. Such information includes, but is notlimited to, prior bankruptcies and litigation (if any); stipula-tions with respect to initiation, maintenance, and termina-tion of the franchise relationship; posttermination noncom-pete clauses; exclusive territories; and the number offranchisees assigned, transferred, and terminated over thepast three years. Although similar to and drawing from theFederal Trade Commission disclosure requirement, regula-tion law states also require franchisors to file FDDs with thestate's regulatory authorities before they may begin to sellfranchises in the state. In imposing full disclosure require-ments on franchisors, registration law is primarily con-cerned with making as much information about franchisors'operations available to prospective franchisees before thelatter's induction into the franchise system (Murphy 2006).

Relationship law restricts the power of franchisors overfranchise terminations, renewals, transfers, and certain otheraspects of the franchise relationship (Grueneberg 2008).Also known as "termination law," relationship law repre-sents a bid to preclude the capricious termination of fran-chisees by their more powerful franchisor partners byrequiring the latter to have "good cause" for termination.

Figure 1AGENCY PROBLEMS IN FRANCHISING AND THEIR

SOLUTIONS

E3

Pro

lnc

yA

gep

eo

f

co

ecti

Sel

<Dtn

Jven

"^

azar

e

X"5

MO

I

RegistrationLaw

RelationshipLaw

FranchiseContract

Franchisee FranchisorParty Facing Agency Problem

typically interpreted as the franchisee's failure to complywith the franchise agreement (Shane 2005). In contrast tothe emphasis registration law places on addressing the exante adverse selection problem, relationship law focuses onminimizing the potential for ex post franchisor moral haz-ard (see Figure 1). Full disclosure (registration law) andgood cause termination (relationship law) are designed tosafeguard the interests of franchisees within the state wherethe particular regulation is in effect.

RESEARCH HYPOTHESES

Impact of Regulation

Both prospective and existing franchisees have moreinformation available to them pertaining to franchisoroperations in registration law states. Moreover, as a fran-chisor expands outlets in more registration law states, it willlikely pay greater attention to the scope and accuracy of dis-closure within and across each of them (Murphy 2006).Together, franchisee access to information and franchisorattention to the details of disclosure should reduce thepotential for miscommunication and misunderstandingsbetween them (Kashyap, Antia, and Frazier 2012). Thus, theadverse selection problem franchisees face should be allevi-ated to some degree in registration law states (Bergen,Dutta, and Walker 1992). The probable end result of suchaccess and disclosure is a reduction in levels of manifestconflict within the franchise system. That is, greater infor-mation sharing should lead to fewer disagreements betweenthe firms that require litigation. As the number of registra-tion law states in which the franchisor operates increases,the greater the likelihood that franchisees will fall within thepurview of the protections afforded by franchise regulation.

H]: The greater the number of registration law states in whichthe franchisor operates, the lower the systemwide litigationincidence.

When serious, unresolvable disputes occur within regis-tration law states, however, franchisors may be emboldenedby the reduction in uncertainty brought about by franchisedisclosure requirements (Shane 2005). That is, the franchisormay view the increased information disclosure required byregistration law as reducing its costs of litigation andincreasing its likelihood of prevailing in the dispute.

Franchisors that aim to comply with registration lawrequirements explicitly specify performance standards, thusproviding a more objective benchmark of performanceagainst which they may evaluate franchisee efforts. Perhapsmost importantly, franchisee noncompliance with explicitlystated obligations is also more easily verifiable by relevantthird parties (i.e., legal courts) (Drahozal and Hylton 2003).Enhanced verifiability and greater ease of demonstratingfranchisee noncompliance in registration law states mayprompt the franchisor to adopt a less conciliatory stance whenserious disputes arise. This is reflected in an increased like-lihood of the franchisor initiating litigation and a decreasedlikelihood of it settling for anything less than its full claim.

H2: In registration law states, litigation is (a) more likely to beinitiated by the franchisor and (b) less likely to be settled byADR procedures.

In relationship law regimes, franchisors can attempt toterminate franchisees only with good cause (Benoliel 2009).


Although relationship law statutes were created to ensurecontinuity of the relationship and to protect franchiseesfrom termination at the whim of the franchisors, they alsohave two potentially unanticipated outcomes. First, empow-ered by the relationship law regime, franchisees in relation-ship law states are likely to be more cognizant of their legalrights. No longer threatened by the impending likelihood oftermination at will, franchisees are also likely to be morewilling to file suit against their franchisor in the event ofperceived transgressions by the latter.

A second, more insidious possible consequence is that thecredible threat of franchisor termination of the relationshipis weakened. Termination for good cause raises the cost ofterminating franchisees and makes the franchisor's task ofquality assurance more difficult (Brickley, Dark, and Weis-bach 1991). Aware of the increased cost of termination,franchisees are more likely to indulge in shirking, in turnforcing the franchisor to undertake corrective action (Antiaand Frazier 2001). Just as with registration law states, thegreater the number of relationship law states in which thefranchisor operates, the greater the likelihood that goodcause termination clauses protect franchisees. Heightenedmanifest conflict and an increase in the incidence of litigationsystemwide are likely to result as the number of relationshiplaw states in which a franchisor has a presence increases.'

H3: The greater the number of relationship law states in whichthe franchisor operates, the higher the systemwide litigationincidence.

The much sought-after outcome of franchisee empower-ment is particularly salient in relationship law states. Fran-chisees aware of the higher cost of termination imposed bygood cause statutes in these states could become moreemboldened (Benoliel 2009). The natural outcome of suchempowerment is an increased willingness to initiate litiga-tion against the franchisor; after initiating lidgation, fran-chisees are also more likely to take a more aggressive stancewith respect to their litigation resolution choices. Antici-pating a more sympathetic judicial forum in relationshiplaw states, franchises are more likely to opt for a court deci-sion (adjudication) rather than rely on ADR methods suchas negotiation, mediation, or arbitration.

H4: In relationship law states, litigation is (a) more likely to beinitiated by the franchisee and (b) less likely to be settledby ADR procedures.

Impact of Franchise Ownership Structure

The high-powered incentives for franchisees to providetheir best efforts on behalf of the franchisor also create agreater potential for franchisee free riding (Lafontaine1992; Lai 1990). To minimize shirking, it is common forfranchisors to own and operate some outlets while relyingon independent franchisees to operate others (Bradach1997; Bradach and Eccles 1989). The knowledge the fran-chisor gains from operating its own outlets may enhance itsability to evaluate franchisee efforts (Parmigiani and

'Note that although both franchisee empowerment and potential fran-chisee shirking would result in a greater likelihood of litigated conflict, theformer is more likely to result in franchisees initiating litigation to safe-guard their rights. In the latter case, the initiator of the litigation is morelikely to be the franchisor seeking compliance with operational stipula-tions. Our data and analysis enable us to draw this critical distinction.

Mitchell 2009), including areas in which the franchisee'sperformance might not be up to standard. Moreover, thepresence of company-owned outlets serves as a crediblethreat of franchisee termination (Dutta et al. 1995;Williamson 1996) because the franchisor "puts partners onnotice ... if desired price and quality are not delivered" (Per-ryman and Combs 2012, p. 374).

The greater the reliance on franchisee-owned outlets, theless the presence of company-owned and -operated outlets.With fewer company-owned outlets systemwide, the fran-chisor lacks information about the appropriate behaviorsand outcomes it may require of its franchisees, resulting ina higher probability of conflict between the parties (Michael2000). The credibility of the threat of replacing marginallyperforming fi-anchisees is weakened (Heide 2003); in addition,with increasing franchisor reliance on franchisee-operatedoutlets, the likelihood of a nearby company-owned outletalso decreases. The result is a probable increase in fran-chisees' propensity to safeguard their interests and a weak-ening of the disclosure-induced reduction in systemwide liti-gation incidence brought about by the franchisors' presencein multiple registration law states.

H5a: The greater the franchisors' reliance on franchisee-ownedoutlets, the weaker the negative association between theextent of franchisors' presence in registration law statesand systemwide litigation incidence.

Notably, franchisors' reliance on franchisee-owned out-lets may also significantly weaken the positive associationbetween the extent of their presence in relationship lawstates and systemwide litigation. The key to understandingthis conflict-dampening effect lies in acknowledging theregulatory context and its (mis)fit with the franchise owner-ship structure (Barthélémy 2008). Consider the situation ofa franchisor with a low level of reliance on franchisee-owned outlets. Recall that one of the pertinent effects ofrelationship law is to reduce the threat of franchisors termi-nating the franchise relationship (Brickley, Dark, and Weis-bach 1991). As the number of relationship law states in whichthe franchisor operates increases, so does the number offranchisees covered by relationship law statutes; in tum, morefranchisees gain a sense of increased security and an expec-tation of continuity (Benoliel 2009). This expectation standsin stark contrast to the credible threat of replacement thatfranchisors' reliance on company-owned outlets poses forfranchisees (i.e., due to lower levels of reliance on franchisee-owned outlets). The lack of fit between expectations engen-dered by relationship law regimes and those implied by thefranchise ownership structure creates the spark for a height-ened likelihood of serious conflict.

As franchisors rely increasingly on their franchisees formarket presence, the credible threat of franchisee replacementby company-owned outlets is weakened (Dutta et al. 1995).Moreover, the de-escalation of this ownership structure-elicited threat is in conformance with the relationship law-induced erosion of the credible threat of termination. Thealignment of franchisor-franchisee expectations and per-ceptions is likely to create a less litigious mood on the partof both parties.

H51,: The greater the franchisors' reliance on franchisee-ownedoutlets, the weaker the positive association between theextent of franchisors' presence in relationship law andsystemwide litigation incidence.


Dyadic (Immediate) Versus Systemwide (Longer-Term)Outcomes

Given that both parties' litigation initiation and resolutionchoices are made with the intent of pressing their unre-solved claims, it is worth considering whether the initiatorof the litigation achieves what it set out to in the conflict athand. The probability of prevailing in the focal conflict is adyadic outcome of immediate relevance to both litigants.

When considering the likelihood that the party initiatingthe litigation will prevail in the conflict at hand, note thatfranchisors' relatively greater experience with litigationenables them to view the decisions to initiate and resolvelitigated conflict as carefully considered strategic choicevariables (Priest and Klein 1984). Rational parties are apt tosteer away from efforts that are likely to be unsuccessful(Priest and Klein 1984). Accordingly, franchisors typicallyonly undertake litigation initiation with expectations of pre-vailing in the conflict. In contrast to franchisors' carefulselection of issues worth litigating (Bebchuk 1984), fran-chisees are likely to be less strategic in their conflict manage-ment. Their relative lack of litigation experience manifestsin "picking fights" for the wrong reasons and conducting liti-gation in a less effective manner (Calanter 1983). We there-fore expect that franchisors initiating litigation will achievegreater levels of success in the immediate conflict comparedwith their less experienced franchisee partners.

Regardless of who initiates the litigation, however, reso-lution by ADR significantly reduces the probability of thefranchisor achieving all its claims. By its very nature, ADRinvolves compromise (Goldberg, Sander, and Rogers 1999).To the extent that the litigants attempt to resolve the focalconflict in a conciliatory manner, it signals their resolve towork things out rather than win at all costs (Macaulay1963). The give-and-take that characterizes compromise(Goldberg, Sander, and Rogers 1999) significantly reducesthe probability of the franchisor achieving all its claims.

Hg : Franchisors that initiate litigation are more likely to pre-vail in the focal conflict.

Hgb: Alternative dispute resolution is less likely to result in thefranchisor prevailing in the focal conflict.

Although prevailing in the immediate conflict is impor-tant, it is by no means the only outcome of interest to thefranchisor. Firms are also keenly aware of the need to man-age their reputation as reliable partners (Macaulay 1963),and frequent recourse to litigation will likely lead to nega-tive perceptions among current and potential partners(Macaulay 1963).2 It is very possible for the litigious fran-chisor to "win the battle but lose the war"—that is, prevailin the conflict at hand and yet damage its reputation andconsequent ability to retain existing franchisee partners(Macaulay 1963) or attract potential franchisees as needed(Justis and Judd 2002). It is this trade-off between achiev-ing dyadic, immediate goals and compromising system-wide, long-term outcomes pursuant to conflict managementthat we hypothesize in this article.

As the incidence of franchisor-initiated litigation increases,the relationships between the franchisor and its existingfranchisees become ever more fractious (Galanter 1983),

^Although this need to protect reputation holds true for both franchisorsand franchisees, the onus of franchise system growth rests on the fran-chisor. Therefore, we consider solely the franchisor's perspective.

leading to increased rates of relationship dissolution. More-over, the franchisor gains a reputation for litigiousness(Macaulay 1963), thus driving away potential franchiseesthat might otherwise have joined the system. The doublejeopardy of losing existing franchisees and becoming lessattractive to new ones translates into unmet growth objec-tives. We expect the spillover from the dyadic, immediateconflict to the systemwide, long-term franchisor aspirationsto be significant and negative.

In marked contrast to the ill effects of a reputation for liti-giousness, the adoption of a conciliatory position sends asignal that the franchisor is willing to negotiate. Existingfranchisees become aware of and appreciate the franchisor'swillingness to compromise rather than prevail at theexpense of its franchisee partner. Likewise, potential fran-chisees may take note of the franchisor's propensity forADR and be more willing to become members of the fran-chise system. This should translate into an increased fran-chisor ability to achieve its system growth objectives.

H7a: Franchisors that initiate litigation are less likely to achievesystem growth goals.

H7b: Reliance on ADR results in greater franchisor achieve-ment of system growth goals.

The regulatory context is likely not only to affect the par-ties' conflict management choices but also to influencedyadic and systemwide outcomes directly. Recall that regis-tration law states are likely to have the administrative andjudicial infrastructure that facilitates third-party verificationof adherence (or otherwise) to the franchise's contractualagreement. We would therefore expect franchisors to lever-age the existing judicial infrastructure to their advantage,resulting in a greater likelihood of their achieving favorableoutcomes (Galanter 1983). This may be a Pyrrhic victory,however, owing to the very disclosure such regulation isdesigned to provide. Specifically, information about thefranchisor's disputes with its franchisees is likely not onlyto give pause to those considering entry into the franchisesystem (i.e., potential new franchisees) but also to persuadeexisting franchisees to exit the relationship (Macaulay1963). The end result is an inability on the part of the fran-chisor to achieve its system growth goals.

Hg: Litigation in registration law states (a) leads to outcomesfavoring the franchisor and (b) reduces the franchisor's like-lihood of achieving system growth goals.

In contrast to registration law, we expect relationship lawstates to provide a supportive forum to franchisees aimingto address perceived franchisor misdeeds (Benoliel 2009).In such judicial regimes, the franchisee is likely to be giventhe benefit of the doubt (at least marginally), resulting in agreater likelihood of a favorable outcome for the franchisee.Similar to registration law states, we expect a reduction inthe extent to which the franchisor is able to achieve its sys-tem growth goals. Such a reduction, however, probablyresults from the heavier burden of proof placed on the fran-chisor when terminating franchisees. The franchisor mustdeploy greater resources to establish good cause for termi-nation (Brickley, Dark, and Weisbach 1991), resources thatcould have been deployed toward growing the franchisesystem and maintaining existing franchise relationships.


Hg: Litigation in relationship law states (a) leads to outcomesfavoring the franchisee and (b) reduces the franchisor'slikelihood of achieving system growth goals.

RESEARCH METHOD

Data Collection and Unit of Analysis

Our longitudinal examination of litigated conflictrequired us to obtain information about each franchisor'slegal suits. Such information is provided in the FDDs filedwith states' regulatory authorities and is publicly availablefor each franchise firm in our sample. Because each FDDprovides relevant information for a three-year window, weobtained FDDs flled in 1997,2000, and 2003 by a randomlyselected sample of 75 franchisors offering business formatfranchise opportunities to afford ourselves an uninterruptedwindow of observation per firm for the nine-year period of1995-2003. The requirement of litigation disclosure overthe previous ten years enabled us to gain information on liti-gated conflicts involving our sample of franchisors for aneven longer period, 1987-2003.

For each instance of reported litigation, the FDD identi-fies whether the franchisor was the plaintiff (the party bring-ing the claim to court) or the defendant (the party beingsued) as well as information on the date of litigation initia-tion, the jurisdiction, the means of resolution (adjudicationvs. ADR), and the terms of the judgment or settlement. Weverified the reported details and filled in missing informa-tion by referring to case docket information available onPublic Access to Court Electronic Records (PACER) .3 Inaddition, we relied on Bond's Franchise Guide and Entre-preneur magazine's Franchise 500 (we used the latter as aconvergent validity check) to acquire information on the

^Public Access to Court Electronic Records is an online public accessservice that enables registered users to obtain case and docket informationfrom federal appellate, district, and bankruptcy courts as well as thePACER Case Locator index. The federal judiciary provides PACER inkeeping with its commitment to provide public access to court informationthrough a centralized service.

headquarters location, franchising history, royalty rates, andso on for each of the franchisors in our sample. Finally, weobtained information on economic and regulatory condi-tions from a web source (http://www.bea.gov). Table 1 pro-vides information on the data source(s) for each variable inour study.

Our unit of analysis is the individual instance of litigatedconflict. The FDDs and PACER yielded information on 526instances of litigation across the 75 franchise firms in our sam-ple. After removing 89 cases that did not involve franchisor-franchisee conflict and 26 cases with excessive missinginformation, we retained 411 litigated instances of conflict.

Measures

Regulatory environment and litigation incidence. Foreach of the 75 franchisors in our sample, we created a meas-ure REGISNUMit (RELNUMit) representing the number ofregistration (relationship) law states in which the franchisori operated during year t. Furthermore, for each litigated con-flict j contested by franchisor i in our sample, we codewhether the particular jurisdiction was a registration law(REGISjj) and/or relationship law (RELjj) state. Weassigned REGISjj a code of 1 if the conflict was litigated ina registration law state and 0 otherwise. We adopted a simi-lar coding scheme for REL¡j. For each of the franchise sys-tems in our sample, the variable INCIDj, took a value of 1 iflitigation was reported for franchise system i in year t; oth-erwise, INCIDj, = 0.

Franchise ownership structure. Consistent with priorresearch (Perryman and Combs 2012; Srinivasan 2006), weoperationalize the franchise ownership structure in terms ofthe franchisor's reliance on franchisee-owned outlets formarket presence. The variable PROPFEjt reflects the num-ber of franchisee-operated units standardized by the totalnumber of operating units for franchisor i in year t.

Litigation initiation and resolution. We focus on bothparties' litigated conflict management efforts, examininginitiation as well as resolution. We operationalize the fran-

Table 1DATA SOURCES AND VARIABLES

Data Sources Theoretical Constructs Variables

FDDs (1997, 2000, and2003)

Bond's Franchise Guide(1991-2004)

Entrepreneur magazine(1987-2003)

Web sources

Litigation incidenceFranchisee shirking

Franchisor shirking

Litigation initiation by franchisorResolution by ADRRelationship dissolutionPrior litigation successOutcome favorability

Number of registration law states in which the franchisor competesNumber of relationship law states in which the franchisor competesNumber of franchised unitsExtent of franchisor reliance on franchisee-owned unitsAchievement of system growth goals

Franchisor incentive to performFranchisee incentive to perform

Economic environmentRegulatory environment: registration law stateRegulatory environment: relationship law state

Occurrence of litigation (INCID)Index of unpaid dues and trademark misappropriation

(FESH)Index of information disclosure and breach of contract

(FRSH)Franchisor initiating (FRINIT)Use of ADR (ADR)Relationship dissolution (RELDISS)Franchisor win ratio (FRWIN)Outcome favorability (OUTCOME)

Number of registration law states (REGISNUM)Number of relationship law states (RELNUM)Number of franchised units (FRUNITS)Ratio of franchised units to total units (PROPFE)Expansion ratio (EXPAN)

Royalty rates (ROYALTY)Total investment (TOTINV)

U.S. GDP (GDP)Registration law (REGIS)Relationship law (REL)


chisor's decision to initiate litigation with the dichotomousvariable FRINIT, which takes the value of 1 if the franchisorinitiates litigation and 0 otherwise. Given the dichotomousoperationalization, FRINIT = 0 reflects initiation by thefranchisee. Although either party may initiate litigation,both litigants must agree to resolve the litigation by ADRprocedures (arbitration, mediation, or negotiation) (Gold-berger, Sander, and Rogers 1999). We therefore operational-ize resolution choice as binary variable ADR, which takes avalue of 1 when the litigated conflict (regardless of whichparty initiated it) is resolved by ADR and 0 otherwise.

Litigation outcomes. We measure outcome favorabilitywith the variable OUTCOMEjj, coded as 1 if the outcomefavored the franchisor and as 0 otherwise. Two independentcoders examined and coded the preceding three measuresand the reasons underlying the conflict for each of the 411reported conflicts, resulting in intercoder agreement of 95%.The few instances of coding disagreement were resolved bydiscussion. We measured the franchisors' achievement of sys-tem expansion goals on an annual basis for each year in oursample. For each year, we first calculated the number of newfranchise outlets, AFRANUNITS,, as the difference betweenthe number of franchise outlets reported in the current yearand the year before (FRANUNITS, - FRANUNITS, _ i).Then we standardized this difference by the number of newoutlets each franchisor planned to open the following year(PROJUNITS(_ i), reported in Bond's Franchise Guide inyear t - 1, so we could assess the extent to which each fran-chisor's expansion goals stated in year t - 1 were met inyear t. To avoid negative values, we then undertook a naturallogarithmic transformation of the preceding variable afteradding a constant c, resulting in the measure EXPAN =ln[(AFRANUNITS,/PROJUNITS,_,) + c].

Control variables. In addition to the preceding hypothe-sized predictors, we also control for the potential impact ofseveral other factors on both parties' litigation initiation andresolution decisions. Franchisees making significant invest-ments (TOTINV) are more likely to safeguard these invest-ments and may do so by litigation, if necessary. In times ofplenty, the franchise relationship is less likely to experiencestress, and parties are more likely to adopt a cooperativestance. We therefore control for the GDP growth rate (GDP)and its probable negative (positive) impact on litigation ini-tiation (ADR). Prior research suggests the royalty rate(ROYAL) to be an important determinant of franchisorbehavior (Lafontaine 1992); therefore, we control for it aswell. Franchisors that have had more (less) franchisingexperience, as reflected by the years elapsed since theybegan franchising (DUR), are also more (less) likely to ini-tiate litigation (choose ADR).

Furthermore, we hypothesize prior litigation success(FRWIN) to play a critical role in determining a franchisor'sconflict management approach. To the extent that the fran-chisor has been able to prevail in prior conflicts, its expec-tation of obtaining a favorable outcome pursuant to conflictescalation is correspondingly higher. We reflect this expec-tation by computing the cumulative number of conflicts inwhich the fi-anchisor has prevailed, standardized by the totalinstances of litigation involving the franchisor up to, but notincluding, the current case. Finally, the dissolution of a rela-tionship (RELDISS), whether at the behest of the franchisoror the franchisee, likely reduces any tendency for forbear-ance (Heide and Miner 1992). We account for this possibil-

ity with a dichotomous variable that takes the value 1 ifeither party terminates the relationship before htigation.

We also controlled for FRUNITSjt, the natural log-transformed number of franchisee-owned units of franchisei in year t. In addition, we included the incidence of shirk-ing by both parties. For franchisee shirking, we coded thebinary variable UNPAID as 1 if unpaid dues by the fran-chisee were the reported reason for the conflict (0 other-wise); we coded the variable TRADEMK as 1 if the reasonreported for the conflict was trademark misappropriation bythe franchisee (0 otherwise). We then computed franchiseeshirking (FESHjj) as the sum of UNPAID and TRADEMK.Similarly, we coded INFO as 1 if the franchisor's failure toprovide adequate information disclosure or provision ofmisleading information was the cited reason for the litiga-tion and as 0 otherwise. We coded the variable BREACH as1 if breach of contract was claimed by the franchisee and as0 otherwise. We then computed franchisor shirking(FRSHij) as the sum of INFO and BREACH. Table 2 pro-vides the descriptive statistics and correlation matrix for allvariables included in the study.

Model Specification

The incidence of litigation is by no means a random occur-rence. Rather, parties "select" litigation as a means to achievean end. It is therefore necessary to account for the selection oflitigation with a first-stage Heckman (1979) selection model.Conditiotiing on litigation incidence, we must then account forfour additional characteristics of our data. First, we have a mixof binary (FRINIT, ADR, and OUTCOME) and continuous(EXPAN) dependent variables. Second, the litigation initia-tion and resolution decisions and their corresponding dyadicand systemwide outcomes are probably related, requiringthe specification of a correlated error structure among thefour outcomes of interest to us (Greene 2003). Third, toassess the impact of litigation initiation and resolutionchoices on both outcome favorability and the franchisor'sachievement of its expansion goals, recognition of the endo-geneity of the former regressors is required. Fourth, we mustalso account for the clustering of individual observations(litigated confiicts) within franchise systems (Hsiao 2003).

The first-stage sample selection model (see Equation 1)includes all cases of franchisor-franchisee htigation observedfor all 75 firms in our sample, each observed over multipleyears. Although this initial sample comprised 1,133 obser-vations, missing data on franchisors' presence across multi-ple states and ownership structures resulted in 622 completeobservations for the selection regression estimation.^ Con-ditioning on litigation incidence enables us not only toaccount for litigation selection but also to estimate theprobability that either party may initiate litigation; whereasp represents the probability of the franchisor initiating liti-gation against a franchisee, (1 - p) denotes the likelihood ofthe franchisee initiating litigation against the franchisor.^

••Paired t-tests conducted on the incidence of litigation across the obser-vations excluded and included in the first-stage selection model yieldednonsignificant differences (p > .10), suggesting no systematically missingvalues (Little and Rubin 2002).

5Had we not specified the first-stage Heckman selection model in Equa-tion 1, (1 - p) would confound the odds of the franchisee initiating litiga-tion with those of observing litigation for franchise system i in year t. Inour model, the latter is explicitly and separately accounted for in Equation1, which enables us to account for two-sided moral hazard.


Xcr

mertrOO

CM Q

c/3m>

OCOCOLUQ

OOO in -

I I »o (N

(Nmc^oocs— o o — — o

r r \ \ \ 'ooc^ — (NinesO O — o — o

r r ' r ' r

VO — "^m-^vûTtror^ OO"pop — oooop *n*r]

inoo — ooopop' \ [ ' \ ' ' \ \ '

p ppppp

r r r ' r * rp — p p O O O O O C N O > 0 —;CN

I I* ' r ' en iri

O O O O O O — O — O — O O O — r

I p p — p p p p — p — p p p

r r r r r

I I I I I I

2 i ! 8 I | 2 I | 8 5

152

Q

ooONO —

QO

I

9u

'S


We have 411 conflicts reported by 61 of the 75 franchisorsin our sample (see Equations 2-5). We therefore specify thefollowing recursive system of five equations:

(1) INCIDit = ßio + ßiiREGISNUMj, + ßijRELNUMi,

+ ßnPROPFEit + ßi4PR0PFEi, x REGISNUM¡,

-I- ß,5PR0PFE|t xRELNUMit + I^ßControls + con-

Conditional on INCIDu = 1 (n = 411),

(2) FRINITij = ß2o + ß2,REGISij +

(3)

(4)

(5)

where

+ ß32REL|j +

3°ij + S35°ßControls -H (ûy,,

INITij + ß42

ij -I- C04i, a n d

INCIDj, = incidence of litigation for franchisor i in year t;REGISNUMi, = count of the registration law states in which

franchisor i operates during year t;RELNUMjt = count of the relationship law states in which

franchisor i operates during year t;FRINITjj = franchisor i initiating litigation in conflict j ;

jj = conflict j involving franchisor i settled by ADR;

REGISjj = conflict j involving franchisor i in registrationlaw state;

REL|j = conflict j involving franchisor i in relationshiplaw state;

PROPFEj, = franchisor i's reliance on franchisee-owned unitsin year t;

OUTCOMEjj = outcome of conflict j in favor of franchisor i;EXPANjj = extent of desired expansion achieved by

franchisor i in the year following conflict j ;Controls = the vector of coefflcients corresponding to

ROYAL, TOTINV, GDP, DUR, RELDISS,FRWIN, FRUNITSi,, FESHy, and FRSHy,respectively; and

the variance-covariance matrix Cl ~ MVN(H, (û2).We jointly estimate the preceding system of equations

(including the selection model) using Roodman's (2009)conditional mixed-process regression procedure. Thisenables a simultaneous estimation of all five equations anduses a simulated maximum likelihood algorithm (Geweke1989; Hajivassiliou and McFadden 1998; Keane 1994) todirectly estimate the cumulative higher-order likelihoodfunction (for a similar approach and additional details onthe method, see Kashyap, Antia, and Frazier 2012) using aseemingly unrelated regression estimator.

RESULTS

Table 3 displays the conditional mixed-process regressionestimates. The results provide considerable support for ourhypotheses. The significant Wald chi-square statistic of145.2 (j> < .001) demonstrates that across the system ofequations estimated, the predictors have satisfactoryexplanatory power.

Table 3CONDITIONAL MIXED-PROCESS REGRESSION ESTIMATES

Independent VariablesInterceptREGISNUMRELNUMPROPFEREGISNUM X PROPFERELNUM X PROPFEREGISRELFRINITADR

Control VariablesLn(TOTINV)FESHFRSHGDPROYALLn(DUR)RELDISSFRWIN(t- 1)Ln(FRANUNITS)

Wald x^ (p-value)

n := 622

INCID

-2.02-.32

.41-.77

.45-.51

.18

-.14

145.20

(-1.23)(-2.49)**

(2.50)**(-.76)(2.90)**

(-2.57)**

(2.35)**

(-2.24)*

(.0000)

Dependent Variables

FRINIT

-2.45

1.51

.20-.58

1.80-1.59

.19-.03

.29

.24

.85-.25

(-1.84)*

(2.36)**

(.93)(-2.79)**

(7.94)***(-4.62)***(2.37)**(-.44)

(.81)(1.24)(2.19)*

(-3.72)***

1.44

-.48

-.53.28.55

.07-.01-.20-.37-.78

.01

n =

ADR

(1.96)*

(-.98)

(-2.74)**(1.33)(2.03)*

(.77)(-.53)(-.94)

(-3.22)***(-3.61)***

(.21)

•411

OUTCOMEFR

.00 (.00)

.22 (.82)-.28 (-1.04)

.99 (4.84)***-.95 (-3.07)***

EXPAN

2.35 (19.30)***

-.16 (-2.17)*-.07 (-1.14)-.29 (-3.91)***

.44 (2.59)***

*p < .05 (one-tailed).**i>< .01 (one-tailed).***p < .001 (one-tailed).Notes: z-statistics are in parentheses.


Impact of Regulation and Franchise Ownership Structure

With respect to litigation incidence, we find, consistentwith Hi and H3, that the greater the number of registration(relationship) law states in which the franchisor competes, thelower (higher) the incidence of litigation (b^ = -.32, bi2 =.41, both ps < .01). These strong main effects of the regula-tory context on the franchise system-wide incidence of liti-gation are, however, significantly weakened by the fran-chisor's ownership structure. Specifically, franchisors'reliance on franchisee-owned outlets is associated with asignificant uptick in litigation incidence, notwithstandingthe number of registration law states in which they compete(bi4 = .45,;? < .01). A similar significant weakening of thepositive main effect of relationship law on litigation inci-dence is also in evidence (bi5 = -.51,;? < .01). Together,these results provide strong support for íl¡^ and R^y.

To gain a better understanding of the moderating impactof franchise ownership structure, we conducted tests of thesimple slopes at low (-1 SD), high (-1-I SD), and mean val-ues of PROPFE (Cohen et al. 2003). Figure 2, Panel A (B),represents the simple slope of the number of registration(relationship) law states on litigation incidence for low andhigh reliance on franchisee-owned outlets. The inverseassociation between REGISNUM and INCID (simple slopeof REGIS = -.04) is reversed for franchisors with high lev-els of PROPFE (simple slope of REGIS = .12; see Figure 2,Panel A). For RELNUM, the effect of PROPFE is just as

Figure 2

SIMPLE SLOPES ANALYSIS OF SIGNIFICANT MODERATORS

A: REGISNUM XPROPFE

auz4)U

c

ide

Inci

1Q.IDQ

ted

(0O)

0-.5

-1.0-1.5

-2.0-2.5-3.0-3.5

Low PROPFEHigh PROPFE

Low REGISNUM High REGISNUMNumber of Registration Law States (REGISNUM)

B: RELNUM x PROPFE

9O

= • 0

8 -.5g -1.01 -1-5£ -2.00 -2.5g. -3.0.2-3 5S -401 -4.5

- - Low PROPFEHigh PROPFE

Low RELNUM High RELNUMNumber of Reiationship Law States (RELNUiUI)

pronounced (see Figure 2, Panel B). For franchisors withlow reliance on fratichisee-owned outlets, the associationbetween RELNUM and INCID is positive (simple slope ofREL = .09); we observe the opposite under conditions ofhigh PROPFE (simple slope of REL = -.09). The post hocprobing of the interactions provides strong support for ourhypotheses and facilitates a clearer understanding of thesubtleties of conflict management.

Franchisor-Franchisee Litigation Choices

Conditional on the observance of litigation, we findmixed evidence for the hypothesized effects of the regula-tory context on franchisors' and franchisees' litigation initi-ation and resolution choices. Although franchisors are notmore likely to initiate litigation in registration law states(b2i = .20, n.s.), they are less likely to rely on ADR for con-flict resolution {h-}i=-.53,p < .01). Thus, H2a is rejected,but H2b is supported. We find partial support for H4 as well,because relationship law states elicit greater odds of fran-chisees initiating litigation (b22 - - .58,/; < .01) but have nosignificant impact on their resolution choices (b32 - .28,n.s.). We thus find support for H^^ but reject H^,.

We find strong support for Hg, relating franchisors' litiga-tion initiation (b4i = .99,p < .001) and reliance on ADR (b42 =-.95, p < .001) to their odds of prevailing in the focal con-flict. Furthermore, franchisor litigiousness is associated withlower levels of expansion goals achieved {h¡i - -.29, p <.001). Reliance on ADR, however, does result in increasedachievement of system expansion goals (b52 = .44, p <.001), in support of H7.

Litigation Outcomes

Turning to the direct effects of the regulatory context, wefind evidence in support of only Hgi, (b53 = -.16,p < .05).None of the other hypothesized direct effects of regulationfind support (b43 = .22, b44 = -.28, b54 = -.07, all n.s.). Withrespect to the control variables, we find higher levels offranchisee investment (GDP growth rates) to be associatedwith greater (lesser) odds of litigation incidence (big = .18,p< .01; bi7 = -.14,p < .05). The GDP growth rates are asso-ciated with a significant increase in franchisor odds of liti-gation initiation (b24 =.19,p<.01), but they do not appearto affect ADR reliance (b34 - .07, n.s.). Royalty rate andfranchising experience have no significant impact on par-ties' litigation initiation or resolution choices (b25 = - .03,b35 = - .01, b26 - .29, b36 - -.20, all n.s.). Whereas relation-ship dissolution has no impact on litigation initiation (b27 =.24, n.s.), the termination of the relationship before litiga-tion ensues significantly reduces the likelihood of resolu-tion by ADR (b37 = -.31,p < .001). We also find that thegreater the franchisor's record of prior litigation success, thegreater (lesser) the likelihood of litigation initiation (ADRresolution) (b28 = .85,p < .05; b38 = -.78,;? < .001). Finally,as we expected, the likelihood of the franchisor (franchisee)initiating litigation increases in response to franchisee (fran-chisor) moral hazard (b29 = 1.80, b2io = -1.59, both ps <.001).

Alternative Model Specifications

Could the impact of each regulation (registration law andrelationship law) be amplified by the presence of the other?To test this possibility, we estimated several alternate model


specifications. First, we attempted to include the multiplica-tive interaction of REGISNUM x RELNUM as an addi-tional predictor in the selection model of litigation inci-dence (Equation 1). We also included the multiplicativeinteraction of REGIS x REL in the subsequent models pre-dicting litigation initiation, resolution, and the dyadic andsystemwide outcomes, both separately for each outcome aswell as together (Equations 2-5). Across all alternativemodel specifications, the interaction term remained non-significant and the fit statistics suffered relative to thehypothesized model. At least within the present context,there seems to be no evidence of such amplification of theimpact of regulation.

DISCUSSION

The objective of this research is to provide a better under-standing of the impact of franchise regulation on the inci-dence, nature, and outcomes of franchisor-franchisee con-flict. Our work acknowledges the significant impact of theinstitutional context in which businesses operate (Dant andSchul 1992; Grewal and Dharwadkar 2002) and extendsprior research on channel conflict by identifying the driversand consequences of serious conflict between channel part-ners. We next discuss how the regulatory context framesfranchisors' and franchisees' litigation choices both by itselfand in combination with fi-anchisors' ownership structure andexplore the theoretical and managerial implications thereof.Our findings inform marketers about an important publicpolicy issue regarding a critical sector of the economy.

Does Franchise Regulation Reduce or Promote Conflict?

As with most phenomena, the answer is not straight-forward. Rather, whether franchise regulations serve toratchet conflict up or down depends on (1) whether weadopt a systemwide perspective or choose instead to focuson particular regulatory contexts (regulated vs. nonregu-lated states), (2) the particular regulation we consider (reg-istration vs. relationship law), and (3) the ownership struc-ture of the channel system—specifically, the extent to whichthe franchisor relies on franchised outlets.

Our study of 75 franchise systems observed over nearlytwo decades yields evidence that the additional disclosureelicited by registration law serves to reduce the incidence ofserious conflict (i.e., litigation) between franchisors and theirfranchisees systemwide. Across regulated and nonregulatedmarkets alike, registration law-induced transparency offranchisors' operations effectively reduces miscommunica-tion and unmet expectations, thereby promoting more har-monious relations. This negative association between regis-tration law and systemwide litigation incidence is, however,reversed when litigation occurs in states subject to registra-tion law. Within registration law regimes, the well-specifiednature of franchisee obligations enables easier and morecomplete verification of (non)performance. Franchisors aretherefore less reticent to press their claims in court.

Relative to registration law, the impact of relationshiplaw on litigation is more straightforward. The increased dif-ficulty of franchisee termination brought about by relation-ship law results in a corresponding increase in the incidenceof franchisor-franchisee litigation. Such an increase may beanticipated either as a function of the empowerment of fran-chisees by relationship law or as a direct result of increased

franchisee shirking and the enforcement response it elicitsfrom franchisors. The present analysis provides greater sup-port for the former thesis. Note that our findings explicitlysuggest that the franchisee is more likely to initiate litiga-tion in relationship law states, typically in response to per-ceived franchisor moral hazard. Although the end result ofincreased litigation is probably not what regulatorsintended, it is evident that relationship law has prompted agreater appreciation among franchisees of their rights.

The preceding direct effects of regulation are, however,significantly tempered by increases in franchisors' relianceon their franchisees for market presence. We find that ahigher proportion of franchisee-owned outlets translate to agreater likelihood of litigation, probably caused by the lackof fit between the reduced cost of franchisor enforcementefforts and the resultant weakening of the credible threat oftermination. In contrast, franchise systems that have a sig-nificant presence in relationship law states and rely on ahigher proportion of franchisee-owned outlets experiencegreater environment-strategy alignment. Specifically, fran-chisors seem to realize that the franchisee has the upperhand not only because the latter owns and operates a higherproportion of outlets but also because relationship lawimplies a "franchisee day in court." The result is a reducedpropensity for litigation between the parties.

Overall, our own assessment of franchise regulation ispositive. We find little evidence of franchisees shirking inrelationship law states. Rather, the data demonstrate fran-chisees to be significantly empowered and more willing totake on the historically more powerful franchisor in relation-ship law regimes. Although franchise operation in multipleregistration law states is significantly associated with low-ered systemwide incidence of litigation, within individualregistration law regimes, franchisors are less prone to relyon ADR methods. It is worth noting, however, that fran-chisors' increased assertiveness is primarily related to theirrealizing unpaid dues by franchisee partners and correctingtrademark violations, both of which are directly related tothe continued health and viability of all franchisees and thefranchise system as a whole. Although too much of anything(including regulation) can be a bad thing, in our opinion, theadditional transparency elicited by registration law and thecountervailing power brought about by relationship lawbode well for franchisor-franchisee partnerships.

Theoretical Implications

Perhaps the most important theoretical implication toemerge from our study is that the decisions regarding whento initiate litigation and how to resolve it are distinct yethighly interrelated strategic choices. Parties making thesechoices seem to be cognizant of the regulatory context inwhich they operate. Specific dimensions of this contextwork to either encourage or discourage a proclivity for liti-gation and its resolution. To the best of our knowledge, ourstudy is the first rigorous, multiyear, multifranchise systemassessment of both pertinent laws. In emphasizing the roleof the regulatory context, we build on recent theoreticaltreatments of how the imposition of regulation might affect"the internal polity (e.g., by influencing decision making inchannels)" (Grewal and Dharwadkar 2002, pp. 89-90).

The results of our analysis provide strong support for thenotion that franchisors adopt a strategic cost-benefit calcu-


lative approach to litigation with their franchisee partners.The sparing use of litigation is noteworthy, as is the discem-ing recourse to ADR as a means of resolving the issue inquestion. Across the franchise systems we studied, we findfranchisors' litigation choices to be consistent with policingand quality assurance: they ensure the appropriate use ofvaluable trademarks underlying their brand equity (Shane2005) and the payment of royalties and advertising fees thatunderwrite franchisors' monitoring and market developmentefforts, respectively (Blair and Lafontaine 2010).6 Note alsothat, all else remaining constant, franchisors initiating legalclaims against their franchisees are more likely to agree toADR rather than insist on adjudication. The picture thatemerges here is not one of a power-hungry litigant that iseager to "punish" even the slightest transgression but ratherthat of a thoughtful partner that is nevertheless willing to dowhat is necessary to safeguard the franchise brand and itspromise (Antia and Frazier 2001).

Our assessment of both immediate and more long-termoutcomes pursuant to both parties' litigation choices alsoextends what is currently known about conflict manage-ment. Each conflict management choice—the initiation oflitigation as well as its resolution—has a significant impacton the decision maker's achievement of objectives.Although the party initiating the litigation (relying on ADR)is more (less) likely to achieve its immediate goal, the situa-tion is completely reversed for longer-term strategic objec-tives. The present research thus identifies a distinct trade-off that channel partners must weigh as they face a dispute,the consideration of which may provide an effective braketo inexorable confiict. Note that, for the most part, the regu-latory environment seems to affect both dyadic and system-wide outcomes indirectly. Contrary to our expectations,only registration law has a negative effect on franchisors'achievement of systemwide expansion goals. Instead, theregulatory context seems to significantly inform both par-ties' conflict initiation and resolution choices and, in tum,their corresponding outcomes. Our findings are consistentwith the notion that the regulatory context serves as a back-drop against which firms conduct their operations (Grewaland Dharwadkar 2002).

Managerial Implications

Our study provides useful guidance to both franchisorsand their franchisees. To franchisors, perhaps the mostimportant leaming point is the strong evidence we provideregarding the trade-off between the achievement of imme-diate and more long-term objectives. Undoubtedly, byadopting an aggressive conflict management approach, thelikelihood of the franchisor achieving its immediate purposeis increased. The "payback" for such aggression, however,is a demonstrated inability to retain and attract suitable fran-chisee partners. Franchisor managers cognizant of thistrade-off' are likely to make better informed conflict resolu-tion decisions.

By identifying when each party is more apt to litigate, weprovide a useful benchmark to franchisors and franchiseesdealing with unresolved conflict. Rather than risking need-

^As pointed out by an astute reviewer, franchisees look to the franchisorfor policing and enforcement of critical performance obligations and arenot likely to oppose such efforts (Lai 1990).

less provocation or being perceived as being weak, eachparty may calibrate its conflict management response effec-tively. Franchisees would do well not to misuse or appropri-ate franchise trademarks and should make timely royaltypayments to their franchisor. For their part, franchisorsshould pay particular attention to the claims they makewhen attempting to obtain new franchisees, because theseclaims tend to be the most commonly observed grounds forfranchisee-initiated litigation. With this knowledge, bothchannel partners can make better decisions about their ownconflict management approaches.

Limitations and Further Research Directions

Our study is subject to some limitations, which provideworthwhile future research opportunities. First, althoughour use of archival data from multiple sources provides richinformation on multiple franchisors' conflict managementchoices over an extended window of observation, we mustrely on indicators for the theoretically relevant variables inthis study. Future studies of conflict management wouldbenefit from integrating archival and more direct (perhapssurvey-based) operational measures of the theoretical con-structs. Second, despite considering both parties' perspectivesof the conflict, relevant information on individual fran-chisees is limited solely to the contents of the franchisor-reported litigation sutnmaries we examined. Further researchcould make use of the court docket information on record torecreate a richer account of each conflict as it evolved.

A third limitation of the present research lies in our con-ceptualization and operationalization of franchise ownershipstructure at an aggregate, systemwide level. We are there-fore unable to speak to the possibility of state-by-state varia-tions in the proportion of franchisee-owned outlets and theirimplications for franchisors' conflict management choicesacross individual regulatory jurisdictions. A rigorous assess-ment of the drivers and consequences of local variations inownership structure would likely prove promising.

REFERENCES

Antia, Kersi D. and Gary L. Frazier (2001), "The Severity of Con-tract Enforcement in Interfirm Channel Relationships," Journalof Marketing, 65 (October), 67-81.

Barthélémy, Jérôme (2008), "Opportunism, Knowledge, and thePerformance of Franchise Chains," Strategic Management Jour-nal, 29 (13), 1451-63.

Bebchuk, Lucian A. (1984), "Litigation and Settlement UnderImperfect Information," RAND Journal of Economics, 15 (3),404-^15.

Benoliel, Uri (2009), "The Expectation of Continuity Effect andFranchise Termination Laws: A Behavioral Perspective," Ameri-can Business Law Journal, 46 (1), 139-77.

Bergen, Mark, Shantanu Dutta, and Orville C. Walker Jr. (1992),"Agency Relationships in Marketing: A Review of the Implica-tions and Applications of Agency and Related Theories," Jour-nal of Marketing, 56 (July), 1-24.

Blair, Roger D. and Francine Lafontaine (2010), The Economics ofFranchising. New York: Cambridge University Press.

Bradach, Jeffrey L. (1997), "Using the Plural Form in the Manage-ment of Restaurant Chains," Administrative Science Quarterly,42 (2), 276-303.


and Robert G. Eccles (1989), "Price, Authority, and Trust:From Ideal Types to Plural Forms," Annual Review of Sociol-ogy, 15,91-11^.

Brickley, James A., Frederick H. Dark, and Michael S. Weisbach(1991), "The Economic Effects of Franchise TerminationLaws," Journal of Law and Economics, 34 (1), 101-132.

Brown, James R. and Ralph L. Day (1981), "Measures of ManifestConflict in Distribution Channels," Journal of MarketingResearch, 18 (August), 263-74.

Cohen, Jacob, Patricia Cohen, Stephen G. West, and Leona S.Aiken (2003), Applied Multiple RegressionlCorrelation Analy-sis for Behavioral Sciences, 3d ed. Hillsdale, NJ: Lawrence Erl-baum Associates.

Dant, Rajiv P. and Patrick L. Schul (1992), "Conflict ResolutionProcesses in Contractual Channels of Distribution," Journal ofMarketing, 56 (January), 38-54.

Drahozal, Christopher R. and Keith N. Hylton (2003), "The Eco-nomics of Litigation and Arbitration: An Application to Fran-chise Contracts," The Journal of Legal Studies, 32 (2), 549-84.

Dutta, Shantanu, Mark Bergen, Jan B. Heide, and George John(1995), "Understanding Dual Distribution: The Case of Repsand House Accounts," Journal of Law, Economics, and Organi-zation, \ I (1), 189-204.

Etgar, Michael (1979), "Sources and Types of Intrachannel Con-flict," Journal of Retailing, 55 (1), 76-78.

Galanter, Marc (1983), "Reading the Landscape of Conflicts:What We Know and Don't Know (and Think We Know) AboutOur Allegedly Contentious and Litigious Society," UCLA LawReview,3H4),4-71.

Ganesan, Shankar (1993), "Negotiation Strategies and the Natureof Channel Relationships," Journal of Marketing Research, 30(May), 183-202.

Geweke, John (1989), "Bayesian Inference in Econometric Mod-els Using Monte Carlo Integration," Econometrica, 57 (6),1314-39.

Goldberg, Stephen B., Frank E.A. Sander, and Nancy H. Rogers(1999), Dispute Resolution: Negotiation, Mediation, and OtherProcesses. New York: Aspen Publishers.

Greene, William H. (2003), Econometric Analysis, 5th ed. UpperSaddle River, NJ: Prentice Hall.

Grewal, Rajdeep and Ravi Dharwadkar (2002), "The Role of Insti-tutional Environment in Marketing Channels," Journal of Mar-keting, 66 {My), &2-91.

Grueneberg, Susan (2008), The FTC Franchise Rule: Analysis andCommentary. Chicago: American Bar Association.

Hajivassiliou, Vassilis and Daniel McFadden (1998), "The Methodof Simulated Scores for the Estimation of LDV Models,"Econometrica, 66 (4), 863-96.

Heckman, James (1979), "Sample Selection Bias as a Specifica-tion Error," Econometrica, 47 (1), 153-61.

Heide, Jan B. (2003), "Plural Governance in Industrial Purchas-ing," Journal of Marketing, 67 (October), 18-29.

and Anne S. Miner (1992), "The Shadow of the Future:Effects of Anticipated Interaction and Frequency of Contact onBuyer-Seller Cooperation," Academy of Management Journal,35 (2), 265-91.

Hibbard, Jonathan D., Nirmalya Kumar, and Louis W. Stern(2001), "Examining the Impact of Destructive Acts in Market-ing Channel Relationships," Journal of Marketing Research, 38(February), 45-61.

Hsiao, Cheng (2003), Analysis of Panel Data. Cambridge, UK:Cambridge University Press.

International Franchise Association (2013), "Franchising,"(accessed June 25,2013), [available at http://buildingopportunity.franchise.org/].

Justis, Robert T. and Richard J. Judd (2002), Franchising. Cincin-nati: South-Westem Publishing.

Kashyap, Vishal, Kersi D. Antia, and Gary L. Frazier (2012),"Contracts, Extracontractual Incentives, and Ex Post Behaviorin Franchise Channel Relationships," Journal of MarketingResearch, 49 (April), 260-76.

Keane, Michael P. (1994), "A Computationally Practical Simula-tion Estimator for Panel Data," Econometrica, 62 (1), 95-116.

Klein, Benjamin (1980), "Transaction Cost Determinants of'Unfair' Contractual Arrangements," American EconomicReview, 70 (2), 356-62.

Lafontaine, Francine (1992), "Agency Theory and Franchising:Some Empirical Results," RAND Journal of Economics, 23 (2),263-83.

Lai, Rajiv (1990), "Manufacturer Trade Deals and Retail PricePromotions," Journal of Marketing Research, 27 (November),

Little, Roderick J.A. and Donald B. Rubin (2002), StatisticalAnalysis with Missing Data. Hoboken, NJ: John Wiley & Sons.

Macaulay, Stewart (1963), "Non-Contractual Relations in Busi-ness: A Preliminary Study," American Sociological Review, 28(1), 55-67.

Michael, Steven C. (2000), "Investments to Create BargainingPower: The Case of Franchising," Strategic Management Jour-nal,21 (4), 497-514.

Murphy, Kevin (2006), Franchise Handbook: A Complete Guideto All Aspects of Buying, Selling, or Investing in a Franchise.Ocala, FL: Atlantic Publishing Company.

Parmigiani, Anne and Will Mitchell (2009), "Complementarity,Capabilities, and the Boundaries of the Firm: The Impact ofWithin-Firm and Interfirm Expertise on Concurrent Sourcing ofComplementary Components," Strategic Management Journal,30 (10), 1065-91.

Perryman, Alexa A. and James G. Combs (2012), "Who ShouldOwn It? An Agency-Based Explanation for Multi-Outlet Own-ership and Co-Location in Plural Form Franchising," StrategicManagement Journal, 33 (4), 368-86.

Pondy, Louis R. (1967), "Organizational Conflict: Concepts andModels," Administrative Science Quarterly, 12 (2), 296-320.

Priest, George L. and Benjamin Klein (1984), "The Selection ofConflicts for Litigation," Journal of Legal Studies, 13(1), 1-55.

Roodman, David (2009), "Estimating Fully Observed RecursiveMixed-Process Models with CMP," Working Paper No. 168,Center for Global Development.

Rubin, Paul H. (1978), "The Theory of the Firm and the Structureof the Franchise Contract," Journal of Law and Economics, 21(1), 223-34.

Shane, Scott (2005), From Ice Cream to the Internet: Using Fran-chising to Drive the Growth and Profits of Your Company.Upper Saddle River, NJ: Prentice Hall.

Srinivasan, Raji (2006), "Dual Distribution and Intangible FirmValue: Franchising in Restaurant Chains," Journal of Market-ing, 10 Quly), 120-35.

Williamson, Oliver E. (1996), The Mechanisms of Governance.New York: Oxford University Press.

Copyright of Journal of Marketing Research (JMR) is the property of American MarketingAssociation and its content may not be copied or emailed to multiple sites or posted to alistserv without the copyright holder's express written permission. However, users may print,download, or email articles for individual use.

kersi d. antia, xu (vivian) zheng, and gary l …moorman/marketing-strategy-seminar... · kersi d....

Documents