geographical indications and the competitive provision · pdf filefood standards and grades, a...

GEOGRAPHICAL INDICATIONS AND THE COMPETITIVE

PROVISION OF QUALITY IN AGRICULTURAL MARKETS

GIANCARLO MOSCHINI, LUISA MENAPACE, AND DANIEL PICK

The economics of geographical indications (GIs) is assessed within a vertical product differentiationframework that is consistent with the competitive structure of agriculture. It is assumed that certi-fication costs are needed for GIs to serve as (collective) credible quality certification devices, andproduction of high-quality product is endogenously determined. We find that GIs can support a com-petitive provision of quality and lead to clear welfare gains, although they fall short of delivering the(constrained) first best. The main beneficiaries are consumers. Producers may also accrue some benefitif production of the high-quality products draws on scarce factors that they own.

Key words: competitive industry, free entry/exit, geographical indications, Marshallian stability, qual-ity certification, trademarks, welfare.

The market provision of quality is notoriouslyfraught with difficulties under asymmetric in-formation: when producers cannot crediblysignal the quality of their products, consumers’choices are predicated on the perceived aver-age quality on the market, and this poolingequilibrium has undesirable welfare proper-ties. Following Akerlof’s (1970) seminal con-tribution, such market failures have been theobject of considerable research. One possiblesolution has emphasized the role of firms’ rep-utation as conveyed by their brands (Klein andLeffler 1981; Shapiro 1983). Brand names mustthemselves be informative, of course, and thatin turns requires a credible trademark system.Trademarks thus serve as useful informationtools for consumers by allowing them to morereadily identify the goods of interest, therebyreducing the possibility of consumer confu-sion and economizing on their search costs(Landes and Posner 1987). Given that effect,trademarks also provide an incentive for firmsto produce goods of consistent quality, as ex-pected by consumers, lest they lose consumer

GianCarlo Moschini is professor and Pioneer Chair in Science andTechnology Policy and Luisa Menapace is a Ph.D. student, bothwith the Department of Economics, Iowa State University, Ames,IA. Daniel Pick is Chief of the Specialty Crops and Fiber Branch,Economic Research Service, U.S. Department of Agriculture.

The authors thank two anonymous referees, Giovanni Anania,Harvey Lapan and editor Christopher Barrett for their helpfulcomments. The support of the Center for Agricultural and RuralDevelopment at Iowa State University, and of the U.S. Departmentof Agriculture through a cooperative research project, is gratefullyacknowledged. The views expressed in this article are those ofthe authors and may not be attributed to the Economic ResearchService or the U.S. Department of Agriculture.

loyalty and suffer a loss on their investmentsin trademark development.1

Brands and trademarks are best understoodin an imperfectly competitive setting. Theirrole in agriculture and food production, largelycharacterized by competitive market condi-tions, remains an open question. Individualfirms are typically too small to credibly signalquality to consumers directly, and this is one ofthe justifications for specific types of govern-ment intervention such as the development offood standards and grades, a specific mandateof U.S. federal agencies (Gardner 2003; Lapanand Moschini 2007).2 Alternatively, producerscould bundle together to achieve the criticalmass required for brand name and trademarkdevelopment. A particularly interesting in-stance of such cooperation in the provision ofquality is represented by the use of geograph-ical indications (GIs). This use of geographi-cally based labels to brand products has beenin use for a long time, especially in Europe,but interest in GIs increased considerably af-ter they were recognized as a distinct form ofintellectual property (IP) rights in the TRIPSagreement of the World Trade Organization

1 This standard result of reputation models was anticipated byAkerlof (1970), (p. 499), who noted that “Brand names not onlyindicate quality but also give the consumer a means of retaliationif the quality does not meet expectations.”

2 Two recent instances of government intervention in food andagricultural products labeling are the introduction of new organicfood standards by the U.S. Department of Agriculture (USDA)in October 2002, and the new regulation for labeling geneticallymodified food and feed products in the European Union (EU) inApril 2004.

Amer. J. Agr. Econ. 90(3) (August 2008): 794–812Copyright 2008 American Agricultural Economics Association

DOI: 10.1111/j.1467-8276.2008.01142.x

Moschini, Menapace, and Pick Geographical Indications and the Provision of Quality 795

(WTO) (Josling 2006). In the context of GIs,quality attributes of interest to consumers arepresumed linked to the specific geographic ori-gin of the good and/or particular productionmethods used in that region (the notion of“terroir”), and such attributes cannot be de-termined through inspection by the consumerprior to purchasing the good. The fundamentalrole of GIs in this setting, therefore, is that ofproviding a credible certification mechanismthat solves a real-world information problem.

Some recent contributions have addresseddirectly some of the specific economic issuesrelated to GIs. Zago and Pick (2004) questionthe desirability of GIs by showing that, withan exogenously determined supply of qual-ity, the welfare implications of a fully crediblecertification system based on GIs are ambigu-ous. In Anania and Nistico (2004), low-qualityproducers can choose to sell their producton the high-quality market (i.e., to cheat).Given an imperfect enforcement mechanism,a GI regulation might be desirable for bothlow- and high-quality producers. A few stud-ies have suggested that GIs can be interpretedas “club goods” (nonrival, congestible, and ex-cludable), as discussed in Rangnekar (2004),chapter 4, and this interpretation is adoptedby Langinier and Babcock (2006). The govern-ment provides GI certification rights to high-quality producers, who are free to decide thesize of the club (i.e., who among the high-quality producers has access to it). Lence et al.(2007) focus on the problem of developingnew GIs. The key to developing such prod-ucts is a fixed cost. Certification is implicitlyfree in their setting, and thus costless imitationis possible, so that some degree of supply con-trol may be necessary to encourage geographicproduct differentiation.

In this article, we emphasize that the naturalinstitutional setting for GIs is that of competi-tive markets. Contrary to standard trademarks,which are owned and used by a single firm, GIsare essentially public goods and are used bymany firms simultaneously. Moreover, the useof a GI cannot be denied to any producer in thespecified geographical area, an issue that hasbeen overlooked by previous work. Indeed,in the European Union (EU) where GIs arewidely used, there are typically no limitationson which or how many firms can use a givenGI (provided that all product specifications, in-cluding the geographical origin, are met). Simi-larly, in the United States where GIs are mainlyprotected as certification marks, any firm thatmeets the certifying standards is entitled to use

the corresponding certification mark. Accord-ingly, the purpose of this article is to investi-gate the impacts of a credible GI certificationsystem in a competitive market setting char-acterized by the possibility of free entry, andwe derive and discuss the welfare effects to beexpected in such a context.

Our analysis complements and adds to exist-ing studies in this area in some novel ways. Forinstance, most studies discussed in the forego-ing (Anania and Nistico 2004; Zago and Pick2004; Langinier and Babcock 2006) assumethat producers are ex ante and exogenouslyidentified as either of the low- or high-qualitytype. In particular, high-quality producers sup-ply the high-quality product regardless ofwhether or not they are certified and/or receivea price premium in the market. We relax thisconstraining assumption and allow the (costly)provision of quality to be endogenously deter-mined. Furthermore, in our model the produc-tion of high- and low-quality goods can coex-ist in equilibrium in the same area, which alsocaptures a feature of the real world where notall producers in a given GI region take advan-tage of their right to supply the GI products.Finally, and perhaps most importantly, we an-alyze explicitly the implications of competitiveentry within a coherent model of quality certi-fication through GIs, an issue that, to date, hasnot been addressed.

In what follows we first review the institu-tional setting for GIs, with emphasis on poli-cies implemented in the EU, a leader in thedevelopment and use of GIs. This allows usto substantiate our premise that both the let-ter of existing regulations and the observedpractice in the predominance of cases sug-gest that the relevant market setting is a com-petitive one. In particular, entry of new firmsthat wish to produce GI-certified high-qualitygoods is possible. Based on that, we thenspecify a model to study how the competi-tive structure of agricultural production af-fects the supply of quality in the presence ofa mechanism that mimics the nature of a GI.The model, although by necessity very stylized,captures the essential elements of the prob-lem at hand. In particular, the demand side ofthe model is rooted in the economics of prod-uct differentiation, which provides an attrac-tive formulation on how consumer preferencesvalue quality. On the supply side, our modelallows for different production costs for high-and low-quality goods and permits the supplyof the high-quality (GI-certified) good to beendogenous.

796 August 2008 Amer. J. Agr. Econ.

The characterization of equilibrium centerson the competitive conditions with free en-try/exit. In the benchmark case, in which allinput costs are parametrically given, the needfor costly certification that involves a fixedcost induces increasing returns to scale atthe industry level. Consequently, the competi-tive equilibrium is not Pareto efficient; specif-ically, it underprovides the high-quality good.This equilibrium, however, does entail welfaregains relative to the absence of GI certifica-tion, and thus, it does ameliorate the infor-mation market failure that motivates interestin GIs. In this setting, some simple policiesthat subsidize the GI certification of qualitywould restore Pareto efficiency to the com-petitive equilibrium. Perhaps not surprisingly,given the long-run nature of the competitiveequilibrium that we consider, the welfare gainsdue to GIs mostly take the form of increasedconsumer surplus. The availability of GIs ben-efits producers only when the production ofthe high-quality good draws on scarce factorsowned by producers.

The Institutional Framework

Whereas recent motives of interest in GIs stemfrom their recognition as distinct IP rights inTRIPS and the ongoing efforts to strengthensuch rights, protection of GIs has a long historyin some European countries and elsewhere.GIs are protected under two similar yet dis-tinct legal notions: appellations of origin andmarks. The primary difference is that an ap-pellation of origin requires the existence of aspecial tie between the quality of the productand its geographical origin, whereas in the caseof a mark such a relation is not necessary.3

The EU framework is rooted in its Coun-cil Regulation (EEC) 2081/92, adopted in1992, which established an EU-wide harmo-nized system of protection of GIs for agricul-tural products and foodstuffs (but excludingwines and spirits).4 This regulation definestwo types of GIs, Protected Designations ofOrigin (PDOs) and Protected GeographicalIndications (PGIs), that differ depending onhow closely a product is linked to geogra-

3 More details and discussion of the GI institutional frameworkmay be found in OECD (2000) and Josling (2006).

4 Regulation 2081/92 was recently updated by Council Regula-tion (EC) 510/2006 to comply with the TRIPS agreement. It abro-gates the “reciprocity principle” and it simplifies the bureaucraticprocedure for application. In particular, it simplifies the procedurefor third-country parties to apply for GI registration in the EUand/or to pursue opposition against the EU registration of any GI.

phy (European Commission 2007). Protectionunder a PDO mandates the more stringentconditions, as it requires the quality or char-acteristics of the product must be essentiallyor exclusively due to natural and human fac-tors characterizing the geographic area of ori-gin (e.g., climate, soil quality, local productionknowledge). Also, for a PDO the entire pro-duction process, including the production andprocessing of raw materials, must occur withinthe defined geographic area of origin. In con-trast, the PGI merely requires that a portion ofa designated product’s characteristics and pro-duction occur within the specific geographicalarea.

PDO or PGI protection can be obtained byan association of producers and/or processors.The process requires the definition of so-called “specifications,” which identify the re-quired conditions for the GI label, includingthe characteristics of the product, the produc-tion method, and the geographic area of pro-duction. In addition, the association seekingprotection must designate a third-party inspec-tion body in charge of the certification and in-spection along the entire supply chain. Suchactivities are meant to ensure that productscarrying PDO or PGI labels comply with thespecifications and to ensure that the informa-tion conveyed via labeling is verifiable, thusbolstering the credibility of the GIs system. Itis critical to note that once a product is reg-istered, all producers within the geographicregion who comply with the product specifi-cations, regardless of whether or not they area member of the association that originally ap-plied for the registration, are entitled to use thePDO or PGI label on their product (Article 8of Regulation 510/2006).5

Over 700 PDO and PGI products are cur-rently registered in the EU. Table 1 reports

5 An example to illustrate the foregoing is the Italian cheeseAsiago. The protection of the Asiago denomination under Ital-ian law dates back to 1954, while the PDO status was obtainedin 1996. The Asiago production area comprises a vast region innorth-eastern Italy, encompassing four provinces (Trento, Vicenza,and parts of the lowland provinces of Padua and Treviso). Physi-cal and sensorial characteristics as well as production procedures,from cow-feeding to the cheese ripening process, are outlined indetail in the production specifications. Local know-how and tra-ditions (documented as far back as 1,000 AD) are deemed to bekey element in the production of Asiago cheese. The “ConsorzioTutela Formaggio Asiago” is in charge of supervision, custody, pro-motion, and development of the denomination. Non-members arefree to brand their product as Asiago PDO as long as productionoccurs according to the specifications and the product is certifiedby the appointed third-party inspection body. Control and inspec-tion activities of Asiago producers (both consortium members andnonmembers) are performed by an independent inspection body(the “Certificazione Qualita Agroalimentare s.r.l.”).


Table 1. Number of PDO and PGI Products in the European UnionBreads Fruits Other

Meat- and Other and Fresh Animal TableTotal Cheese Based Bakery Oils Fish Beer Drinks Vegetables Meat Products Olives Other

Belgium 5 1 2 1 1Czech Republic 6 1 1 3 1Denmark 3 2 1Germany 67 4 8 4 1 2 12 31 2 3Greece 84 20 1 25 1 22 1 10 4Spain 105 19 10 7 20 30 13 3 3France 155 45 4 2 9 2 5 26 51 6 3 2Ireland 4 1 1 1 1Italy 159 32 28 3 38 47 2 2 2 5Luxemburg 4 1 1 1 1Netherlands 6 4 2Austria 12 6 2 1 3Poland 1 1Portugal 104 12 28 6 21 26 10 1Slovenia 1 1Finland 1 1Sweden 2 1 1United Kingdom 28 11 3 2 3 1 7 1

Total 747 159 84 20 103 10 17 39 157 104 24 16 14

Source: Compiled by authors from EU data available at http://ec.europa.eu/agriculture/foodqual/quali1 en.htm (accessed on October 2007).

their distribution by country and by productcategory. The majority of these GIs come fromMediterranean countries—more than 75% ofthe products are registered in five southernEU states (France, Italy, Greece, Portugal, andSpain). Nevertheless, the registration of GIsby northern countries has increased over time.Of the 268 applications for new denominationsthat are currently being considered, more thanhalf come from countries other than the afore-mentioned five southern countries (includingeleven from nonmembers countries).

In most other developed countries outsidethe EU, the trademark system provides a le-gal framework for the protection of GIs. Inthe United States, geographical names can beregistered as certification marks. Certificationmarks are characterized by the fact that theuse of the mark is not restricted to any personor entity, as long as the attributes required forcertification are met. U.S. certification marksare typically administered by a governmen-tal body, the presumption being that such anagency is best positioned for “. . . preservingthe freedom of all persons in the region touse the term and, second, preventing abusesor illegal uses of the mark . . .” (USPTO 2007,undated, p. 3).6 Similarly to the appellations

6 A well-known example of a U.S. certification mark is that ofVidalia onions, which is held by the Georgia Department of Agri-culture (Clemens 2002). Producers must apply for an annual licensefrom the Georgia Department of Agriculture to sell Vidalia onions,providing information regarding the type of onions planted, totalnumber of acres and location. Licenses are free. The productionarea covers all or part of the 20 Georgia counties.

system, the product that is labeled with a cer-tification mark is subject to inspection. Inspec-tion activities are in this case the responsibilityof the mark’s owner and not of a third-party in-spection body (but the implications are anal-ogous because the owner of the certificationmark does not conduct production or commer-cial activity; it merely concedes the use of themark to independent producers).

In some instances, GI protection in theUnited State can also be obtained throughindividual trademarks or collective marks.Specifically, that is possible when one can es-tablish that the geographic term in questionhas acquired a “secondary meaning” to con-sumers. Collective marks identify the productsof many firms belonging to a group (e.g., an as-sociation or cooperative). They are meant “foruse only by its members . . .” (USPTO 2007, un-dated, p. 4), and as such they arguably have thenature of club goods.

TRIPS accords stronger GI protections towines and spirits, and even in the EU winesare treated separately. “Quality wines pro-duced in a specified region” and table wineswith a “typical geographic indication,” ex-cluded from Regulation (EEC) 2081/92, areprotected within the framework of the com-mon market organization for wine (EuropeanCommission 2006). This framework limits thegrape-growing potential of the EU with plant-ing rights restrictions, including a ban on newvine plantings. These instruments have onlybeen partially successful in trying to reduce thechronic overproduction in the EU (over the


last two decades the stocks of the aforemen-tioned protected quality wines have actuallybeen growing at a faster rate than consumptionand exports to third countries). In any event,planting rights restrictions apply to total culti-vation of grapes and do allow shifting wine pro-duction into GIs, if desired. Indeed, over time,planting rights have been allocated or reallo-cated to higher-quality productions, increasingthe incidence of GIs on total wine (EuropeanCommission 2002).

GI Product Markets and Competition

The analysis of the institutional framework forthe protection of GIs in the preceding sectionsuggests that, typically, all producers locatedin the relevant specified production area havethe option to produce and market the corre-sponding GI product. Thus, it would seem thatcompetitive entry is a feature of the supply con-text of GI products that is fully consistent withmost current regulations governing GIs.

Despite the possibility of competitive en-try/exit, of course, expanding production ofa given GI may be hampered by limitationson the accessibility of relevant inputs. Giventhe great heterogeneity of existing GIs in thisrespect, no simple assessment is possible onhow much such a consideration matters. Forinstance, if the geographic area identified bya given GI is sufficiently small, and/or the GIproduct accounts for much of the local agri-cultural production (e.g., Champagne), landand/or other factors may seriously affect po-tential supply response. In other cases, such asthose of Greek feta cheese and Italian grappa,the appropriate geographic area encompassesvirtually a whole country. The actual level ofutilization of GI labels within a specified areaof production also varies significantly amongdifferent GIs, and often a significant share oftotal production is commercialized without theGI label. For example, olive oil produced in theItalian region of Lazio involves about 130,000producers who grow olive trees on 195,000acres. A GI label is used on less than 10% of theolive oil that could potentially be branded withany one of the three regional GIs (Sabina PDO,Canino PDO, or Tuscia PGI) (Carbone 2003).Similarly, in the case of the Italian wine sec-tor where a high degree of heterogeneity existsamong different wines, the utilization of GIs isonly about 40% (ISMEA 2005). Thus, consid-erable expansion of production of a numberof GI wines would seem possible, even given

the overall constraint posed by EU plantingrights.

If it were possible to manage GIs as pri-vately owned labels with the power to controltotal supply, as in the notion of farmer-owned brands articulated in Hayes, Lence,and Stoppa (2004), that might create the po-tential for attractive noncompetitive returnsfor GI producers. The lure of noncompeti-tive returns in agriculture is, of course, notnew; it has been of interest to farmers fora long time, as evidenced by the historyof the cooperative movement and market-ing orders in the United States (Crespi andSexton 2003).7 Producer associations with di-rect responsibility for managing GIs (called“consortia” in Italy) are perhaps best posi-tioned to pursue noncompetitive goals, espe-cially when they gather most of the producersof the relevant GI product. In fact, antitrustauthorities have intervened with regard to anumber of prominent GI products: the Ital-ian Parma ham and San Daniele ham, theItalian Grana Padano, Parmigiano-Reggianoand Gorgonzola cheeses, and the French Can-tal cheese (OECD 2000). The anticompeti-tive behavior that was investigated concernedattempts by producer associations to controltotal supply through the imposition of individ-ual production quotas to their membership andthrough market share agreements between theconsortia (OECD 2000). In all cases, after theantitrust intervention, production quota andmarket share agreements were abandoned,and competitive conditions were restored.8

A final consideration that will inform ourmodeling choice concerns the production tech-nology of GI products. Whereas it is true thatthe geographic attributes of GIs are often crit-ical to support their perceived higher quality,it should also be clear that there are other ele-ments of the production technology that arepart of a GI’s specifications and that affect

7 For example, Vidalia onions, mentioned earlier, have had afederal marketing order since 1989. The order’s provisions endowgrowers with some supply control. The effects of the marketingorder, of course, are conceptually distinct from those of the certi-fication mark.

8 Consortia used to carry out monitoring activities to ensure thatmembers’ production satisfies the desired specifications. After theintroduction of the 1992 EU regulation on GIs, however, consortialost any authority they might have had over the control of produc-tion, as well as the responsibility for all inspection activities (whichwere assigned to independent bodies). In particular, when awardeda PDO or PGI, consortia had to give up their property right overthe protected name in exchange for the legal protection of the GIprovided by the European regulation (Nomisma 2001). At present,consortia have custody of the collective brand identifying the GIand grant its use to producers who meet the requirements.


not only quality but also the cost of produc-tion. To illustrate, consider the example ofParmigiano-Reggiano cheese. The specifica-tions for this PDO require production to takeplace in a clearly delimited region of north-ern Italy but also mandate a number of otherproduction constraints. These include restric-tive cow-feeding guidelines; notably, it is for-bidden to feed silage to cows that producethe milk used in manufacturing Parmigiano-Reggiano (by contrast, use of silage is al-lowed in the production process of the othercompeting parmesan-type cheese, the GranaPadano). Such restrictions are deemed es-sential to achieve the desired cheese qualitybut are also known to increase considerablythe cost of milk production, by approximately20% by some estimates (de Roest and Menghi2000). Similarly, PDO brie production requiresmanual techniques that may increase produc-tion costs by approximately 25% (Benitez,Bouamra-Mechemache, and Chaaban 2005).

We conclude that, for the case of most GIproducts, the presumption that GI producershave an effective way to control the aggregatequantity supplied of their product is not ten-able. Thus, in the model that follows we willmaintain the possibility of competitive entry ina setting in which producers can elect to sup-ply either the GI product or its generic coun-terpart, and where the production of the GIproduct entails higher production costs thanits generic counterpart. The implications of thefact that some necessary factors in the produc-tion of GIs may be in scarce supply will also beinvestigated.

A Model for the Competitive Provisionof Quality Using GIs

The specification of the model that followsimplements all the main features that appearto be relevant based on the foregoing re-view of the institutional framework and realworld examples. Specifically, in the model: (a)consumers value quality as in the standard ver-tical product differentiation framework; (b)producers can supply quality by undertakingproduction processes that are costlier thanthose required for the alternative, low-qualityproduct; (c) GIs can serve as (collective) qual-ity certification devices, although for theirfunction to be credible additional promotionand certification costs are required; and (d)producers operate in a competitive industry(with free entry and exit).

Demand: Vertical Product Differentiation

As with other studies in this area, we presumethat the quality to be supplied through the useof GIs is valued by consumers within the verti-cal product differentiation structure of Mussaand Rosen (1978). Specifically, we considerthe simple unit-demand version of the verticalproduct differentiation model whereby eachconsumer buys at most one unit of the goodin question and her preferences are describedby the (indirect) utility function

U ={

�q − p if the good is bought

0 otherwise.(1)

where q ∈ R++ indexes the quality of the good,p ∈ R++ is the price of the good, and thepreference parameter � ∈ [�

¯, �] ⊆ R+ indexes

consumer types. The hypothesis here is that ofheterogeneous preferences for quality so thatthe population of consumers can be character-ized by the distribution function G(�) of thepreference parameter.

More specifically, suppose that there areonly two possible qualities in this market, a“low” quality qL and a “high” quality qH > qL.If these two qualities are available at prices pLand pH , respectively, where pH > pL > 0, thenthe consumer decision problem is to select theaction that yields the highest utility among thethree possible options:

U =

�qH − pH if the high-quality

good is bought

�qL − pL if the low-quality

good is bought

0 otherwise

(2)

To simplify the analysis, as in related studiesin this area, we put further restrictions on thedistributions of consumers. That is, we postu-late that the distribution G(�) is uniform andthat � ∈ [0, 1]. The latter condition, in partic-ular, implies that the market will be “uncov-ered” (i.e., as long as prices are strictly positive,some consumers with a low enough � will notbuy anything). More specifically, let

� ≡ pH − pL

qH − qL(3)

� ≡ pL

qL.(4)


Throughout we will consider the typical casewhere 0 < � ≤ � ≤ 1. For that parametric case,consumers with � ∈ [�, 1] will buy the high-quality product, consumers with � ∈ [�, �] willbuy the low-quality product, and consumerswith � ∈ [0, �] will buy nothing. For the pop-ulation of M consumers, market demand isreadily obtained by integrating the unit de-mand of each consumer given the distributionof consumer types. For the uniform distribu-tion assumption invoked earlier, the aggregatemarket demand functions are

X DH = M

(1 − pH − pL

qH − qL

)(5)

X DL = M

(pH − pL

qH − qL− pL

qL

).(6)

Sometimes it is convenient to work with theinverse demand functions. Inverting (5) and(6), for given quantities Xi ∈ [0, M] (i = L,H) satisfying XL + XH ≤ M, yields

pH = qH − (qL X L + qH X H )M

(7)

pL = qL

(1 − X L + X H

M

).(8)

Equations (7) and (8) display the market’swillingness to pay for the two qualities, forgiven supply levels, but also implicitly definethe willingness to pay for the “additional qual-ity” that the high-quality good provides overthe low-quality one. By using (7) and (8), the(inverse) derived demand for the additionalquality (qH − qL) is

pH − pL = (qH − qL)(

1 − X H

M

).(9)

Note that this (market) willingness to payfor the additional quality depends only onthe quantity supplied of the high-quality good(because this quantity implicitly defines themarginal consumer that is indifferent betweenpurchasing the high- or low-quality good).

Supply: Competitive Production of Quality

We presume a standard competitive industrypopulated by numerous (actual or potential)producers who behave as price takers, and eachof whom can produce either the high-quality

good or the low-quality good (or zero quan-tity). Initially, we suppose that these producersare identical and are operating with a pro-duction technology that admits cost functionsCH(xH) and CL(xL) for the high- and low-quality goods, respectively, where xi ≥ 0 (i =H,L) denotes the level of firm’s output for ei-ther the low- or high-quality product. We as-sume that the cost functions Ci(xi) are strictlyincreasing and display standard U-shaped av-erage cost curves. In a long-run equilibriumwith free entry and exit, therefore, firms willbe operating at a strictly positive efficient scale.Furthermore, we assume that CH(x) > CL(x),∀x > 0. The presumption that the high-qualitygood requires a costlier production process israther intuitive, as discussed in the precedingsection (e.g., more labor care, need for higher-quality inputs, need for additional inputs, re-strictions on the use of some inputs, etc.).

In addition to production cost, to market thehigh-quality good, producers need to under-take costly activities that credibly certify, in theeyes of consumers, the claimed higher quality.Such activities may relate to marketing, pro-motion, and/or monitoring of production stan-dards. In principle such activities should beopen to each producer individually, as wouldbe the case for firms marketing with individ-ual trademarks, and we therefore allow forthat possibility. But the case for GIs rests onthe presumption that firms may not be able tomuster the required resources to do that indi-vidually, that is, there is scope for producersto act cooperatively in this regard. Hence, weinterpret GIs as a common brand whereby pro-ducers can bundle together to share the mar-keting, promotion, and certification costs thatare necessary for a credible GI. This assump-tion is quite consistent with the existence ofproducer organizations that take an active partin the marketing of GI products, such as theconsortia discussed in the preceding section.Specifically, we assume that producers sharethe GI promotion and certification costs via acharge per unit of output produced, so that thetotal cost of producing the GI-certified high-quality product is CH(xH) + �xH , where � > 0is the unit certification cost.9

9 An alternative assumption might be a cost-sharing rule thattakes the form of a per-firm charge. As long as firms are identical, aspostulated here, the two assumptions would appear largely equiv-alent. The sharing rules that we follow does, however, simplify thecharacterization of long-run equilibrium (because the minimumefficient scale of the high-quality firms is not affected by the sizeof �). Also, the assumed rule might be more appealing when themodel is generalized to allow for firm heterogeneity.


One of the reasons for the existence of anincentive for firms to share the costs requiredfor a credible certification is that what these ac-tivities produce—consumer goodwill towardthe product with the given GI—has the na-ture of a public good from the producers’perspective. Some of the required costs arelargely independent of the aggregate quan-tity of good that is eventually produced; thiswould be the case, for example, for activi-ties connected to marketing, promotion andadvertising, and overhead costs of the pro-ducer organization in charge of performingsuch functions. We measure the cost of suchactivities by F > 0. Other costs, however, arelikely to depend on the amount produced. Wecontend that this is the case, in particular, forthe portion of certification costs that are meantto monitor production standards and preventcheating and free riding. A credible certifica-tion system, in fact, must recognize (and dealwith) the possibility that producers purport-ing to sell a GI product have an incentive tobehave opportunistically (i.e., they may claimto sell the high-quality good while producingthe low-quality good). Producer organizationshave a variety of mechanisms at their disposalto monitor and limit the opportunistic behav-ior of members. In our context, the challengeis to represent such activities explicitly, so thattheir effects on equilibrium can be assessed,and to do so in a parsimonious way that isconsistent with the rest of the model. To thatend, the enforcement mechanism that we pos-tulate is a sequential “auditing game” (e.g.,Rasmussen 2007, pp. 85–87), as follows.

A producer who wants to supply a quan-tity x of certified, high-quality GI product hastwo strategies: to comply with the relevant GIspecifications or to violate them (by producingthe lower-quality good at cost CL(x) < CH(x)instead). In the enforcement mechanism thatwe envision, the monitoring agency moves firstby announcing an inspections policy {�, T},where � ∈ [0, 1] is the probability of inspectionto verify that the product specifications are met(or, more precisely, � is the fraction of produc-ers that will be subject to inspection), and T > 0is a finite penalty that is paid if a producer failsthe inspection. The individual producer thenchooses whether to comply with or to violatethe production specifications. Given this en-forcement mechanism, the total expected costto the producer associated with the “comply”strategy is CH(x) + �x (and the assumption,of course, of no errors at the compliance-verification stage), whereas the total expected

cost is CL(x) + �x + �T if the “violate”strategy is used.10 Clearly, to induce compli-ance the minimum penalty needs to be atleast as large as the production cost difference[CH(x) − CL(x)]. Specifically, for any givenT > [CH(x) − CL(x)] there exists an inspec-tion probability � ≡ [CH (x) − CL(x)]/T thatmakes “comply” a best response strategy forthe producer. Given that, and if the aggregatereturns to producers from everyone complying(net of the cost of inspections) exceeds thoseof tolerating violation, then it is an equilibriumstrategy for the monitoring body to adopt thepolicy {�, T} at the initial move stage.11

The main point of the foregoing is that com-pliance is obtained with an inspection prob-ability that is high enough, given the penaltylevel. But such a monitoring scheme is costlybecause it requires that firms be inspected withsome probability. Specifically, we assume thatthe cost of each inspection that is carried outis proportional to the level of a firm’s output,that is, �xH , where � > 0. Thus, the expectedmonitoring cost for each producer to be certi-fied is ��xH .12 Note that, in this setup, the totalmonitoring cost is increasing with the numberof producers to be certified, an appealing fea-ture that is lost when total certification cost istreated as a fixed cost only.

The remaining question concerns how manyproducer groups we should expect to see ina GI market. As stated earlier, our workingassumption is that the full certification cost,as given by the fixed cost F and the vari-able cost of monitoring, is shared among themembers of the producer organization on aper-unit-of-output basis, with the portion oftotal cost attributable to the certification ser-vice written as �xH . Thus, under full cost-sharing (we will return to this issue later, inthe context of possible policy implications),if there are n producers sharing such costs, itmust be that �xH = � �xH + F/n. Given these

10 The presumption is that there is no error in the inspec-tion/auditing activities. Anania and Nistico (2004) also rely on asimilar simple and error-free monitoring and enforcing scheme.

11 In this Nash equilibrium, the monitoring agency must carryout the inspections even though, in equilibrium, compliance is ob-tained. Thus, we are assuming that the monitoring authority cancredibly commit to carrying out inspections, consistent with theoverall requirement of a certification system that needs to be cred-ible in the eyes of the consumer.

12 Because only the product �T matters to induce compliance,and because � affects the monitoring cost whereas T does not (inequilibrium everyone complies and no penalty is assessed), ideallyone would want to make T as high as possible and � as small aspossible. The existing legal and institutional framework (as well asfirms’ limited financial assets), however, likely puts bounds on howlarge T can be; given that level of T, the inspection frequency �can in principle be calculated.


structural assumptions, the question of howmany producer groups we should expect re-duces to a simple coalition formation problem.Suppose that, in a competitive equilibrium,there are NH producers engaged in the pro-duction of the high-quality good, each pro-ducing the same quantity xH , and considerthe possibility of there being m groups of sizen < NH (so that n ≡ NH/m), each indepen-dently promoting and certifying their high-quality product. Then this would be a stablecoalition structure if no member can gain byswitching coalitions, that is, by leaving its cur-rent group to join another group (making thelatter of size n + 1). Thus, the hypothesizedcoalition structure would be stable if

F

n+ � �xH ≤ F

(n + 1)+ � �xH .(10)

But clearly this cannot hold. The larger coali-tion attains lower unit promotion and certifica-tion costs and pulls in new members, so that inequilibrium we are left with only one (grand)coalition of size NH .

The other condition we need to check is thepossibility that a member has of defecting fromthe coalition with the intention of supplyingthe high-quality product on its own. In such acase, the producer has to undertake the entirefixed cost F individually but saves the need formonitoring costs. This possibility is not prof-itable if

pH x1 − CH (x1) − F

≤ pH xH − CH (xH ) − � �xH − F

NH

(11)

where x1 is the scale of production of the firmthat incurs F individually.13 For approximatelyequal production levels (x1

∼= xH) and a reason-ably large number of producers NH , the con-dition is approximately F ≥ � �xH . Thus, aslong as the fixed cost of certification is largeenough relative to the monitoring cost, defect-ing to market the high-quality product withone’s own trademark is not profitable.

In conclusion, a credible certification systemcan be supported by a GI producer associationthat implements a simple monitoring scheme.Assuming that (11) is satisfied, a coalition mayform to supply the high-quality good, and the

13 The production level of the firm that incurs F individuallywould differ from that of the firm sharing costs because its coststructure is changed (it incurs a fixed certification cost instead of aunit certification cost).

process should lead to just one coalition of sizeNH .14 Whereas in equilibrium the scheme mayensure compliance by producers, it will imposeadditional costs on the producers of the high-quality good. In particular, the total cost func-tion for low-quality producers is simply CL(xL)whereas the high-quality firms have a total costfunction of CH(xH) + �xH , where � is the costof GI certification per unit of output; that is,

� ≡ F

NH xH+ � �.(12)

Equilibrium and Welfare

In this section, we consider the long-run par-tial equilibrium conditions that are relevantwhen it is possible for firms to enter and/orexit the industry of interest (e.g., Mas-Colell,Whinston, and Green 1995, chapter 10). Ini-tially we assume no diseconomies at the indus-try level; that is, the prices of all productioninputs are constant and exogenous to the in-dustry. For a given output price pL of the low-quality good, low-quality producers choosethe production level xL that maximizes profitpLxL − CL(xL). The possibility of entry/exitdrives profit to zero, so that each firm will beproducing at the minimum efficient scale x∗

L,that is, at the point that minimizes average cost

x∗L = arg min

xL

{CL(xL)/xL}.(13)

Let cL ≡ CL(x∗L)/x∗

L denote the unit cost forthe low-quality good at this efficient produc-tion scale. Then, the competitive equilibriumprice for the low-quality good must satisfy15

p∗L = cL .(14)

14 Note that the underlying presumption of a competitive mar-ket is maintained throughout. Taking for given the U-shaped coststructure at the farm level that we have assumed, an alternative hy-pothesis would be to allow the merger of several farms/plants to berun as a single firm, thereby allowing the fixed cost F to be sharedover a larger (private) output that could then be marketed with afirm’s own trademark. Such a hypothesis, of course, would lead toan oligopolistic market structure. We rule that out by assumptionbecause such a strategy would raise difficult agency problems of itsown. Allen and Leuck (1998) provide a convincing account of whyfarming has generally not changed from small family-based firmsto large corporate firms. Indeed, the reasons that slant the trade-offbetween moral hazard and specialization in favor of small farm-ing operation are likely to be even more compelling in the contextof producing the kind of high-quality products identified by thetraditional specifications of GI products.

15 Here and in what follows, we abstract from the possible “in-teger” problem (technically, a nonconvexity) that arises when thefirms’ efficient scale is strictly positive, so that, strictly speaking, thelong-run industry supply correspondence is an integer multiple ofthe efficient scale.


As for the high-quality good, whether inequilibrium it will be supplied at all obviouslydepends on the level of the required certifi-cation cost, vis-a-vis the consumers’ willing-ness to pay for high quality. In an equilibriumin which the high-quality good is also sup-plied, for a given price pH individual producerschoose the production level xH to maximizeprofit pHxH − CH(xH) − �xH . The possibil-ity of entry and exit, however, requires thenumber of producers NH to adjust to ensurethe zero-profit condition (which, in turn, af-fects the per-unit certification cost �). Hence, along-run equilibrium needs to specify the equi-librium price p∗

H , the equilibrium productionlevel x∗

H of each high-quality firm, the equilib-rium number N∗

H of high-quality firms, and theequilibrium per-unit certification cost �∗. Therequired conditions are

p∗H = C ′

H

(x∗

H

) + �∗(15)

p∗H x∗

H = CH(x∗

H

) + �∗x∗H(16)

�∗x∗H ≡ F

N ∗H

+ � �x∗H(17)

N ∗H x∗

H = M

(1 − p∗

H − p∗L

qH − qL

).(18)

Equation (15) is the optimality conditionfor firm-level profit maximization, whereasequation (16) displays the zero-profit condi-tion due to the assumed free entry/exit pos-sibility. For any given per-unit certificationcost, these two equations in conjunction es-tablish that the equilibrium production levelx∗

H must satisfy C ′H (x∗

H ) = CH (x∗H )/x∗

H . Hence,as for the low-quality producers, each firm inequilibrium produces at its minimum efficientscale (the point that minimizes average cost).Let cH ≡ CH (x∗

H )/x∗H denote the unit produc-

tion cost (not including the certification cost)of the high-quality product. Then by usingequations (16) and (18), the equilibrium num-ber of high-quality producers N∗

H must satisfy

(qH − qL)(

1 − x∗H N ∗

H

M

)

= cH − cL + F

x∗H N ∗

H

+ � �.

(19)

Thus, the equilibrium condition in (19)equates consumers’ demand for the additional

quality provided by the high-quality good (rel-ative to the low-quality good), as given byequation (9) derived earlier, with the addi-tional (industry) unit cost of producing this ex-tra quality.16

It is useful to note that, at the industrylevel, the per-unit certification cost is declin-ing in the number of firms that produce theGI product (because of the assumed fixed costof promotion and certification F). The right-hand-side of equation (19) effectively definesthe competitive “industry supply” function forthe high-quality good. Under the usual as-sumption that a firm’s individual production issmall relative to industry output, the individualfirm takes the unit cost as parametrically given.Yet, at the industry level the industry’s unitcost of production is decreasing in the num-ber of high-quality producers (i.e., decreasingin industry output). Any given firm exerts apositive externality on all other firms by shar-ing the fixed certification cost F but does notinternalize this benefit in its decision to en-ter/exit the industry. This positive externalityis a source of increasing returns to scale. Thisfact is bound to have relevant implications foran equilibrium, but it is also the case that suchan instance of parametric external economiesof scale are quite consistent with the exis-tence of competitive equilibrium (Chipman1970), although it does give rise to the pos-sibility of multiple equilibria, as discussednext.

Rather than solving for the equilibriumnumber of firms, one can equivalently solve forthe equilibrium aggregate quantity of the high-quality product. Define X∗

H ≡ x∗HN∗

H . Thenfrom equation (19), X∗

H must be a root of thequadratic equation

�q

M

(X∗

H

)2 − (�q − �c − � �)X∗H + F = 0

(20)

where, for notational simplicity, we define�q ≡ qH − qL and �c ≡ cH − cL. The rootsof this equation are given by the standard for-mula:

(�q − �c − � �) ± √(�q − �c − � �)2 − 4F�q/M

2�q/M.

(21)

16 In an equilibrium in which both the high- and the low-qualityproducts are supplied, the zero-profit condition of course ensuresthat firms are indifferent as to which of the two goods they produce.


Figure 1. Equilibrium with F < F

The sign of the discriminant D ≡ (�q −�c − � �)2 − 4F�q/M determines whetherwe have real roots and, if so, whether we haveone or two roots. Note that dD/dF < 0 sothat, given the other parameters of the model,there exists F ≡ (�q − �c − � �)2(M/4�q)such that D = 0 when F = F . In such a case,there is only one real root to the equilibriumequation. When F > F , there are no real roots,that is, certification is just too costly and thecompetitive equilibrium does not include pro-duction of the high-quality good. When F < F ,there are two distinct roots for the quadraticequation, i.e., we have two candidate equi-librium solutions X H and X∗

H . The case ofF < F is illustrated in figure 1, where the lineardownward-sloping curve represents the con-sumers’ willingness to pay for the “additional”quality, and the nonlinear decreasing curverepresents the additional (industry) unit costof supplying the high-quality good.

To distinguish between the two candidateequilibria when F < F we appeal to stabil-ity conditions, but the choice of the relevantcondition requires some care. Two conceptswith a long history, conventionally labeled asWalrasian stability and Marshallian stability,differ in terms of what variable is viewed aschanging in a situation of disequilibrium.17

17 Walrasian stability posits a price change in response to excessdemand at that price, whereas Marshallian stability supposes thatquantity adjusts when supply and demand prices differ at that quan-tity (e.g., Silberberg 1990, chapter 19).

Whereas the two stability concepts agree whendemand and supply functions have the usualslope, they yield conflicting conclusions whenthe supply curve is sloping downward (inour case the equilibrium associated with X His Walrasian stable, whereas the equilibriumassociated with X∗

H is Marshallian stable). Animportant element, in such a situation, con-cerns why the supply function is downwardsloping. When the negative slope reflects theexistence of industry-wide external economies(the so-called forward-falling supply curve, asopposed to the case of individual backward-bending supply curves), Marshallian stabilityis arguably more appropriate, and indeed sup-ported by strong experimental evidence (Plottand George 1992). Accordingly, in this studywe rely on Marshallian stability and thus iden-tify X∗

H as the stable equilibrium of interest.We should also note that the Marshallianstability concept, with its reliance on out-put adjustment, is appealing in a productioncontext such as ours that allows for firms’ en-try and exit. For example, if the supply ofhigh-quality product were to be to the leftof X∗

H , then high-quality producers would bemaking positive profits, which would stimulateentry and thus expansion of the high-qualitysupply.

The competitive stable equilibrium satis-fies some intuitive comparative static prop-erties. In particular, for the case of F <F , ∂ X∗

i /∂ M > 0, i = H, L (a ceteris paribusincrease in the market size increases the


Figure 2. Consumer welfare in equilibrium

equilibrium quantities of both goods);18

∂ X∗H/∂ F< 0 < ∂ X∗

L/∂ F(e.g., a decrease in thefixed cost of certification F increases theequilibrium level of the high-quality good);∂ X∗

H/∂�q > 0 > ∂ X∗L/∂�q (e.g., a larger qual-

ity markup for the GI product increases theequilibrium level of this good and decreasesthat of the low-quality good); and ∂ X∗

L/∂� >0 > ∂ X∗

H/∂� (e.g., a larger unit monitoring costdecreases the equilibrium quantity of the high-quality product and increases that of the low-quality product). The impact of �c is, of course,qualitatively the same as that of the monitoringcost parameter �.

Welfare

One way to articulate the welfare implica-tion of a GI mechanism is to suppose thatthe high-quality good is technologically feasi-ble but institutional constraints (e.g., lack oflegal protection for the right to use the GI)prevent the establishment of a coalition ofproducers that can credibly deliver the high-

18 The comparative statics properties can be used to further illus-trate the choice of the relevant stability concept by noting that theWalrasian-stable solution X H would produce rather counterintu-itive results. For example, ∂ X H /∂ M < 0 (an increase in the marketsize decreases the equilibrium quantity of the high-quality good).

quality good in a competitive fashion. Relax-ing such constraints would bring about a newequilibrium with both goods being supplied.Before the introduction of a GI, only the low-quality good is supplied, with the competitiveequilibrium condition p∗

L = cL. After the in-troduction of a GI, consumers who do buy thehigh-quality good in equilibrium are better off,whereas consumers who continue to buy thelow-quality good are unaffected. The welfareproperties of the GI equilibrium can be illus-trated as in figure 2, which relates the equilib-rium outcome that we have characterized tothe vertically differentiated demand structureof the model. The downward-sloping lines offigure 2 depict the marginal utility functionsof the population of M consumers, as impliedby the preference structure in (2) (along withthe assumption that the preference parameter� is uniformly distributed on [0, 1]). Total con-sumer surplus is given by the shaded areas.

Of course, to compute the gain in consumersurplus due to the GI mechanism one needsto consider that consumers who buy the high-quality good, in this equilibrium, would stillenjoy some surplus if only the low-qualitygood were supplied. The difference betweenthe two measures is positive whenever the GIequilibrium entails both types of goods being


Figure 3. Gains in consumer surplus

supplied. This could be readily established an-alytically, given the structure of our model, buta graphical illustration can suffice. Specifically,the shaded area in figure 3 illustrates the wel-fare (consumer) gains from the introductionof a GI by using the demand for the addi-tional quality of equation (9) employed earlierto characterize equilibrium.

In conclusion, the foregoing analysis has es-tablished that the following implications arederived from the model. First, there are noprofits to producers in equilibrium (as onewould expect in a long-run competitive modelwith entry). Second, consumer surplus is af-fected by the availability of the high-quality GIproduct. Any institutional change that makesGIs feasible could results in sizeable benefitsto consumers (even without returns to pro-ducers). Finally, only consumers of the high-quality good derive additional welfare fromthe establishment of a GI.

Pareto Efficiency

Not surprisingly, given the existence of (in-dustry) external economies in this setting, thecompetitive equilibrium fails to deliver theconstrained first-best outcome. What we meanby the qualification “constrained” here is thechoice X0

H that a benevolent social plannerwould implement, conditional on having to un-dertake the same certification costs as in com-petitive equilibrium. To derive such a first-bestallocation, denote with �W the gain in wel-fare brought about by production of the

quantity XH of the high-quality good, relativeto zero quantity of this good (the no-credible-certification situation). Given the structure ofthis model,

�W = �q

[2 − X H

M

]X H

2

− (�c + � �)X H − F.

(22)

The optimality condition for a maximum of�W reduces to equating the marginal benefitof the high-quality product to its marginal cost(provided that �W ≥ 0), yielding the first-bestsolution

X0H = (�q − �c − � �)

�qM.(23)

It is readily verified that, at X0H , �W ≥ 0 re-

quires the fixed costs of certification to satisfyF ≤ F 0, where

F0 = M

2�q(�q − �c − � �)2

.(24)

Hence, if the fixed certification costs are toohigh (i.e., F > F 0), provision of the high-qualitygood is not desirable. But for F ≤ F 0 it is so-cially desirable to supply the high-quality goodby the given quality-certification technology,and in that case the optimal provision of thehigh-quality good ought to be at the efficientlevel X0

H given by equation (23).It is now apparent that the competitive equi-

librium falls short of the first-best allocation in


two ways. First, because F0 = 2F > F , then ifthe fixed cost parameter falls in the domainF ∈ (F, F0), the competitive equilibrium en-tails X∗

H = 0 and yet it is strictly socially desir-able to have some high-quality good supplied.Second, even when a competitive equilibriumexists with X∗

H > 0 because F ≤ F , the compet-itive equilibrium delivers a suboptimal level ofoutput, that is, X∗

H < X0H , as can be readily veri-

fied by comparing the solution in equation (23)with the larger of the two roots in equation (21)(see figure 3).

The failure of the competitive equilibrium todeliver the first-best outcome could be reme-died by simple subsidy policies. In the domainF ≤ F , the underprovision of the high-qualitygood is due to the fact that producers who paya share F/X∗

H of the fixed costs of certifica-tion treat that as a marginal cost of production(and specifically do not internalize the contri-bution of their decision to enter the industryon the other firms’ cost of production). Oneway to support the first-best outcome via thecompetitive equilibrium would be to provide alump-sum subsidy to the producer association(e.g., the consortia) equal to the fixed cost Fof quality promotion. Alternatively, the gov-ernment could subsidize production by a unitsubsidy s ≡ F/X0

H , thereby offsetting the por-tion of certification costs due to existence of afixed cost of certification.

The suggestion is sometimes offered that, toprovide incentive for producer organizationsto engage in the type of marketing and pro-motion required for a successful GI, it mightbe desirable to grant market power (i.e., theright to control supply) to producer associa-tions in charge of GIs. In the model of Lenceet al. (2007) this result arises from the assump-tion that a fixed cost is required to developsuch products (very much as in our setting)and that there are no certification costs per se.Because costless imitation is possible in thatcontext, some degree of supply control maybe necessary (depending on the size of the re-quired fixed cost) to encourage producers todevelop a geographically differentiated agri-cultural product. Auriol and Schilizzi (2003),on the other hand, emphasize that certifica-tion costs are critical to achieve credibility. Ourmodel explicitly accounts for the monitoringcosts needed for credible certification, and inthis context we find that market power can-not improve welfare. Specifically, if F ≤ F , acompetitive equilibrium exists, although it un-derproduces relative to the first best; grantingmarket power to a club of high-quality produc-ers would not help, and actually would make

matters worse by further reducing the quantitysupply away from X∗

H (and raise the thornyquestion of who, among the ex ante identi-cal producers, should benefit from the ensuingnoncompetitive profit). Similarly, if F > F , thecompetitive equilibrium entails X∗

H = 0, but inthis parametric case the right to control sup-ply is worthless to a producer association (theindustry average cost is everywhere above therelevant consumer demand).

A final observation might be appropriate atthis juncture. We have seen that the failure ofthe competitive equilibrium to deliver the first-best outcome is very much related to the exis-tence of the fixed cost F. Insofar as this type ofcost is interpreted as the cost of marketing andpromotion, to convince consumers that indeedthe GI product in question is a high-qualityproduct, the public authorities’ endorsementof the GI system (as with the PDOs in theEU) might be construed as a policy that at-tempts to lower the firms’ fixed cost of pro-motion (by conveying relevant information toconsumers) and thus can contribute to the ef-ficient competitive provision of quality in agri-cultural markets.19

Upward-Sloping Industry Supply

The fact that there are no returns to produc-ers in the foregoing model is predicated ontwo things: the assumed long-run competitivestructure (i.e., with freedom of entry/exit), andthe constancy of unit costs (no diseconomiesat the industry level). The latter is of coursequestionable, and in fact we typically thinkof competitive aggregate supply functions inagricultural markets as being upward-sloping.Upward-sloping supply functions can arisewhen the inputs used by the industry are inlimited supply (e.g., land) and their price is af-fected by the competitive demand of the indus-try of interest. One way to make this conceptoperational is to endogenize the price of the(otherwise homogeneous) input with upward-sloping supply (e.g., Hughes 1980; Lapan and

19 The EU assists member countries in financing measures to pro-mote agricultural products and food, both in the EU internal mar-ket and in third countries; see Council Regulation (EC) 2826/2000and Council Regulation (EC) 2702/1999. These measures includeinformation campaigns on EU quality and labeling systems, in par-ticular on the EU system of PDOs and PGIs, and the EU system ofquality wines produced in specified regions. The EU finances 50%of the cost of these measures, the remainder being met by producerorganizations and/or member states. The current triennial programtargeting the internal market has a total budget of €50.9 million.A third-country program targets the USA, Canada, India, Japan,and China and covers wine, fruit, meat, dairy products, olive oil,and organic product with a total budget of €18.2 million.


Figure 4. Upward-sloping supply and constant marginal cost of high quality

Moschini 2000). Alternatively, as in Panzar andWillig (1978), one can presume that all firmsdiffer in their endowment of a fixed input (e.g.,location or soil quality) that affects productioncosts and that has no alternative use outside theindustry of interest.

Constant Marginal Cost of High Quality

To simplify the approach of Panzar and Willig(1978), suppose that firms either produce at anoptimal efficient scale x > 0 (for either the low-or high-quality product) or stay out of the mar-ket, and that the firms’ individual efficiency isindexed by a scalar �∈ [0, ∞). Specifically, theunit production cost for the low-quality good iswritten as cL(�), with c′

L(�) ≥ 0. The industrysupply curve of the low-quality good, conse-quently, is upward-sloping, because increasedoutput can only come about by the produc-tion of increasingly less efficient firms. Theproduction of the high-quality good requiresadditional costs, as discussed earlier, and in thiscase the unit cost of the high-quality good iswritten as

cH (�) ≡ cL(�) + �(25)

where � is a constant. Hence, here we assumethat the extra cost required to produce the

high-quality good is independent of the effi-ciency parameter �. With this assumption, theequilibrium condition in equation (20) still ap-plies. Specifically, to illustrate the equilibriumwhen the parameters of the model are suchthat both goods are supplied, figure 2 can beadapted as in figure 4.

Note that in this case producers do enjoya non-zero producer surplus in equilibrium(the shaded area in figure 4). But this returnto producers does not depend on the pro-duction of the high-quality GI product andwould be the same even if only the low-quality good were to be supplied. The area(p∗

H − p∗L)X∗

H in equilibrium simply accountsfor the additional production cost requiredfor the high-quality product, and for the needfor marketing and monitoring to deliver acredible GI certification for the high-qualitygood.

Increasing Marginal Cost of theHigh-Quality Good

More generally, one could postulate that thesupply of additional quality (qH − qL) is alsoupward-sloping. To make the implications ofthat condition more transparent, suppose thatthe unit production cost of the low-qualitygood is the same for all firms and equal to


Figure 5. Upward-sloping supply due to increasing marginal cost ofhigh quality

cL, but the production cost of the high-qualitygood depends on the firm-specific efficiencyparameter �, that is,

cH (�) ≡ cL + �(�)(26)

where cL is a constant and �′(�) ≥ 0. In sucha case, there are clearly no aggregate pro-ducer returns to producing the low-qualitygood, and the possibility of offering the high-quality good can bring about positive re-turns to producers (as well as returns toconsumers).

The equilibrium determination of the high-quality production in our setting is best illus-trated via the demand for quality upgradesused to characterize equilibrium. Thus, figure3 is adapted as in figure 5, where the shadedareas denote the changes in consumer and pro-ducer surpluses brought about by the produc-tion of the high-quality GI product. Thus, it iscertainly possible for the introduction of GIcertification to benefit directly the producersof the high-quality product, consistent withthe view of those advocating the use of GIsas a tool for rural development. Our model,however, makes clear that such an outcome isby no mean guaranteed, and it depends criti-cally on the underlying structure of the agri-cultural production sector. Specifically, what isrequired is that production of the high-quality

products requires specialized inputs in scarcesupply. Exactly how that characterizes real-world GI settings, of course, depends on theparticular case at hand.

Conclusions

In this article, we have developed a model thattreats GIs as an effective certification tool forhigh-quality products that attempts to over-come the very real information problem thatconsumers face when quality cannot be readilyascertained prior to purchasing. This problemis arguably particularly relevant to food prod-ucts that originate from a fragmented produc-tion structure, where individual farmers aretoo small to muster a credible quality-signalingeffort. One of our major points is the competi-tive structure that justifies the need for produc-ers to act collectively, as with GIs, also carriesimplications for the market equilibrium thatarises with a credible GI mechanism. Thus,our analysis has emphasized the implicationsof a competitive equilibrium with the produc-tion of GI products, including the freedomof entry/exit in the production of the high-quality good. Our model has also maintainedan attractive cost structure for the case athand (higher-quality GI products are costlierto produce than their generic counterparts),


and explicitly models the promotion and mon-itoring activities required to make the GI acredible certification system. In addition, thedemand for GI products is modeled in a verti-cal product differentiation context. This cap-tures the likely heterogeneity of consumerpreferences vis-a-vis GI products but also per-mits generic products to interact meaningfullywith GI products both in the demand and inthe supply side of the model.

The main conclusions of our analysis can besummarized as follows. First, it is possible tohave competitive provision of quality in agri-cultural markets, through certification devicessimilar to geographical indications. Second, al-though a competitive equilibrium can exist, be-cause the GI certification entails fixed costsshared by all high-quality producers, thereare external economies of scale at the indus-try level and the competitive equilibrium isnot Pareto efficient. In particular, the com-petitive equilibrium underprovides the high-quality good. The failure of the competitiveequilibrium to achieve a constrained first-bestoutcome can be corrected by policies that sub-sidize the certification of the high-quality good.Also, we find that measures that allows for mar-ket power (i.e., supply control) for GI producerassociations in this setting are not desirable. Fi-nally, the implications of entry in a competitiveframework are critical. The possibility of entryhas been neglected in many previous studies,but, as we have shown, its consideration has im-portant implications for the welfare results thatmay be deduced. In particular, whereas theresolution of the “lemons” problem that thecredible certification through GI makes pos-sible clearly benefits consumers, what it doesfor the welfare of producers in a competitivesetting ultimately depends on the presence ofscarce factors that they own.

Whereas it is hoped that this article has con-tributed to the clarification of some basic eco-nomic effects associated with the use of GIsas quality certification devices for agriculturalproducts, the analysis that we have profferedhas some limitations. In particular, we haveanalyzed the case of a closed economy andconsidered the role of one GI system in iso-lation. Among the interesting additional ques-tions that arise, one may want to consider theinteraction and competition of several GIs,possibly from different geographic regions inthe same country/jurisdiction, and/or the in-teraction of GIs and other quality labeling(e.g., organic food labels), including the issue

of possible excessive label proliferation. Also,as noted earlier, GIs are of interest in the on-going WTO negotiation and their implemen-tation is a question of intense disagreementamong countries (Fink and Maskus 2006). De-veloped countries are themselves divided onthis topic, with a simmering transatlantic dis-pute rooted in contrasting approaches to trade,IP, and agricultural policy (Josling 2006). A va-riety of perspectives are invoked as germanein this setting, ranging from familiar economicarguments for IP protection (Moschini 2004)to the view of GIs as a tool to safeguard cul-tural heritage, and to foster the preservationof traditional methods of production (Broude2005). Thus, it may be desirable to explorethe international trade implications of the ex-panding reach of GIs, addressing explicitly thecurrent WTO negotiation. Such desirable ex-tensions, which are the object of current re-search, should benefit from the benchmarkanalysis presented in this article.

[Received June 2007;accepted December 2007.]

References

Akerlof, G.A. 1970. “The Market for Lemons: Qual-ity Uncertainty and the Market Mechanism.”Quarterly Journal of Economics 84:488–500.

Allen, D.W., and D. Leuck. 1998. “The Nature of theFarm.” Journal of Law and Economics 41:343–86.

Anania G., and R. Nistico. 2004. “Public Regula-tion as a Substitute for Trust in Quality FoodMarkets: What if the Trust Substitute cannotbe Fully Trusted?” Journal of Institutional andTheoretical Economics 160:681–701.

Auriol, E., and S. Schilizzi. 2003. “Quality Signal-ing through Certification. Theory and an Ap-plication to Agricultural Seed Market.” IDEIWorking Paper No. 165.

Benitez, D., Z. Bouamra-Mechemache, andJ. Chaaban. 2005 “Public Labeling Revisited:The Role of Technological Constraints UnderProtected Designation of Origin.” WorkingPaper, University of Toulouse, Paris.

Broude, T. 2005. “Culture, Trade and Addi-tional Protection for Geographical Indica-tions.” Bridges 9(9):20–22.

Carbone, A. 2003. “The Role of Designation of Ori-gin in the Italian Food System.” In S. Gatti, E.Giraud-Heraud, and S. Mili, eds. Wine in theOld World: New Risks and Opportunities, p.29–39. Franco Angeli: Milan, Italy.


Chipman, J.S. 1970. “External Economies of Scaleand Competitive Equilibrium.” QuarterlyJournal of Economics 84(3):347–85.

Clemens R. 2002. “Why Can’t Vidalia OnionBe Grown in Iowa? Developing a BrandedAgricultural Product.” MATRIC Briefing Pa-per 02-MBF.

Crespi, J.M., and R.J. Sexton. 2003. “Competition,US Farmer Cooperatives and Marketing Or-ders.” English translation of: “Concurrence,cooperatives de producteurs et MarketingOrders aux Etats-Unis.” Economie Rurale(277/278):135–51.

de Roest, K., and A. Menghi. 2000. “Reconsid-ering ‘Traditional’ Food: The Case of Parmi-giano Reggiano Cheese.” Sociologia Ruralis40(4):439–51.

European Commission. 2002. “Ex-post Evalua-tion of the Common Market Organization forWine.” Final Report, European CommissionDG-Agriculture.

——. 2006. “Wine Common Market Organiza-tion.” Working Paper, European Commission,Directorate-General for Agriculture and RuralDevelopment.

——. 2007. “European Policy for Quality Agricul-tural Products.” Fact Sheet of the EuropeanCommission, Directorate-General for Agricul-ture and Rural Development.

Fink, C., and K. Maskus. 2006. “The Debate on Geo-graphical Indications in the WTO.” In R. New-farmer, ed. Trade, Doha, and Development: AWindow into the Issues, pp. 201–211. Washing-ton DC: World Bank.

Gardner, B. 2003. “U.S. Food Quality Standards: Fixfor Market Failure or Costly Anachronism?”American Journal of Agricultural Economics85:725–30.

Hayes, D., S.H. Lence, and A. Stoppa. 2004.“Farmer-Owned Brands?” Agribusiness 20(3):269–85.

Hughes, J.P. 1980. “The Comparative Staticsof the Competitive, Increasing-Cost Indus-try.” American Economic Review 70(3):518–21.

ISMEA. 2005. I vini DOCG, DOC e IGT. Aspettinormativi, economici e di mercato, Istituto diServizi per il Mercato Agricolo Alimentare,Rome, Italy.

Josling, T. 2006. “The War on Terroir: Geographi-cal Indications as a Transatlantic Trade Con-flict.” Journal of Agricultural Economics 57:337–63.

Klein, B., and K.B. Leffler. 1981. “The Roleof Marker Forces in Assuring ContractualPerformance.” Journal of Political Economy89(4):615–41.

Landes, W.M., and R.A. Posner. 1987. “TrademarksLaw: An Economic Perspective.” Journal ofLaw and Economics 30:265–309.

Langinier, C., and B. Babcock. 2006. “AgriculturalProduction Clubs: Viability and Welfare Impli-cations.” CARD Working Paper 06-WP 431.

Lapan, H., and G. Moschini. 2000. “IncompleteAdoption of a Superior Innovation.” Econom-ica 67:525–42.

——. 2007. “Grading, Minimum Quality Standards,and the Labeling of Genetically Modified Prod-ucts.” American Journal of Agricultural Eco-nomics 89:769–83.

Lence, S.H., S. Marette, D. Hayes, and W. Fos-ter. 2007. “Collective Marketing Arrangementsfor Geographically Differentiated AgriculturalProducts: Welfare Impacts and Policy Implica-tions.” American Journal of Agricultural Eco-nomics 89:947–63.

Mas-Colell, A., M.D. Whinston, and J.R. Green.1995. Microeconomic Theory. New York: Ox-ford University Press.

Moschini, G. 2004. “Intellectual Property Rightsand the World Trade Organization: Retrospectand Prospects.” In G. Anania, M. Bohman, C.Carter, and A. McCalla, eds. Agricultural Pol-icy Reform and the WTO: Where Are We Head-ing? pp. 474–511. Cheltenham, UK: EdwardElgar.

Mussa, M., and S. Rosen. 1978. “Monopoly andProduct Quality.” Journal of Economic Theory18(2):301–17.

Nomisma. 2001. Prodotti tipici e sviluppo locale:Il ruolo delle produzioni di qualita nel fu-turo dell’agricoltura italiana, VIII RapportoNomisma sull’agricoltura italiana, Edizioni IlSole 24 ore, Milano, Italy.

OECD. 2000. “Appellations of Origin and Ge-ographical Indications in OECD MemberCountries: Economic and Legal Implica-tions.” Working Paper COM/AGR/APM/TD/WP(2000)15/FINAL, Organization for Eco-nomic Cooperation and Development, Paris.

Panzar, J.C., and R.D. Willig. 1978. “On the Com-parative Statics of a Competitive Industry withInframarginal Firms.” American Economic Re-view 68(3):474–8.

Plott, C.R., and G. George. 1992. “Marshallian vs.Walrasian Stability in an Experimental Mar-ket.” Economic Journal 102:437–60.

Rangnekar, D. 2004. The Socio-Economics ofGeographical Indications, UNCTAD-ICTSDProject on IPRs and Sustainable Development,Issue Paper No. 8.

Rasmussen, E. 2007. Games and Information: AnIntroduction to Game Theory, 4th ed. Malden,MA: Blackwell Publishing.


Shapiro, C. 1983. “Premiums for High Quality Prod-ucts as Returns to Reputation.” Quarterly Jour-nal of Economics 98(4):659–80.

Silberberg, E. 1990. The Structure of Economics, 2nded. New York: McGraw-Hill.

USPTO. 2007. “Geographical Indication Protectionin the United States.” United States Patentand Trademark Office, undated document

accessed on May 2007. Available at http://www.uspto.gov/web/offices/dcom/olia/globalip/pdf/gi system.pdf

Zago, M.A., and D. Pick. 2004. “Labeling Policiesin Food Markets: Private Incentives, Public In-tervention, and Welfare Effects.” Journal ofAgricultural and Resource Economics 29(1):150–65.

geographical indications and the competitive provision · pdf filefood standards and grades, a...

Documents