BOP and Market Design Economics

Supervisor

Professor Dr. Peter Knorringa

Pourya Darnihamedani

Davood Midari357269

Erasmus School of Economics

October 25, 2015

Contents

1 Introduction 2

2 Bottom of Pyramid at a Glance 4

2.1 BOP Market . . . . . . . . . . . 6

2.2 Entrepreneurship Barriers in BOP 7

3 Transaction Cost Economics at aGlance 7

3.1 Search Cost . . . . . . . . . . . . 8

3.2 Contract Cost . . . . . . . . . . . 9

3.3 Enforcement Cost . . . . . . . . . 9

3.4 Institutions and InstitutionalChanges . . . . . . . . . . . . . . 10

4 Market Design Economics at aGlance 10

4.1 An Introduction to MatchingMarket: Marriage Market . . . . 12

4.2 Unraveling and Market DesignEconomics: Market for New Doc-tors . . . . . . . . . . . . . . . . 13

4.3 Thinness and Market Design Eco-nomics: Kidney Exchange . . . . 14

4.4 Congestion and Market DesignEconomics: School Choice . . . . 16

4.5 Adverse Selection and Market De-sign Economics; Market for NewEconomists . . . . . . . . . . . . 17

4.6 Force; a Solution for Market Fail-ure: Gastroenterologists Example 18

5 Market Design and BOP: A CaseStudy of Agricultural Sector in In-dia 19

5.1 Indian Agricultural Industry . . . 20

5.1.1 Mandis . . . . . . . . . . 20

5.1.2 Mandi Value Chain . . . . 21

5.2 Transaction Costs and MarketFailures in Indian AgriculturalSector . . . . . . . . . . . . . . . 21

5.3 e-Choupal: Impacts of Semi-Clearinghouse in the AgricultureSector . . . . . . . . . . . . . . . 22

5.3.1 ITC e-Choupal . . . . . . 22

1

5.3.2 e-Choupal Value Chain . 235.3.3 Impact of Semi-

Clearinghouse in MarketFailure . . . . . . . . . . . 23

6 Conclusion 24

References 24

Acknowledgment

Professor Dr. Peter Knorringa raised the ques-tion of how economists can contribute to theemerging field of Bottom of Pyramid. This ques-tion motivates me to write this paper. Specialthanks for his question and other contributionsto this paper.

Abstract

Theory of the bottom of the pyramid with-out doubt is one of the most influential the-ories in the business literatures in the pastdecades. Prahalad suggested a new mecha-nism for poverty reduction; large-scale and wide-spread entrepreneurship is at the heart of thesolution to poverty eradication (Prahalad, 2004,Page 28). This idea attracts attention of manyscholars in di↵erent fields. This paper suggeststhat economic theories can contribute to BOPliterature. This paper utilizes two influentialeconomic theories; transaction costs and marketdesign economics, to further enrich the BOP lit-erature.

1 Introduction

Prahalad introduced a new framework forpoverty reduction. In his book, “Fortune at theBottom of Pyramid”, he notified that the poorcan and must involved in market as a customerand a part of production process (Prahalad,2004, Page 28). This idea attracts attention ofmany scholars in di↵erent fields (Young, 2012;Pervez, Maritz, & De Waal, 2013; Halme, Lin-deman, & Linna, 2012; Hall, Matos, Sheehan, &Silvestre, 2012; George, McGahan, & Prabhu,2012; de Almeida Faria, Hemais, & Guedes,

2

2008). The poor not only should not be excludedfrom the market but they should be included inthe market as a profitable customer class andactive participants of value chain. In his bookhe provided examples of companies that success-fully enter into the BOP market. The commonfeature among all of the examples is that compa-nies approach the poor as consumers or part oftheir value chain. Furthermore, these companiescreate a win-win situation in which both compa-nies and low incomes groups gain from economictransactions. One of the example is Casa Bahiathat is a chain store in Brazil and by creatinga sophisticated credit system makes appliancesmore a↵ordable for the inhabitant of the BOP(Prahalad, 2004, Page 117-137). Casa Bahiagives credit to low and even unpredictable in-come streams. The other example is about thirdlargest cement company in the world; CEMEX.CEMEX targets consumers with income $5-15per day. It has revenue of $6.45 billion. Byteaching the poor to save and invest, CEMEXenables them to purchase house and eases theiraccess to credit (Prahalad, 2004, Page 147-160).The other story is about Hindustan UnileverLimited (HUL). HUL is the largest fast mov-ing consumer goods company in India. In thesame token, it is the largest soap producer inIndia. It has more than 60% of market shareand has significantly contributed to health of thepoor in India. In India more than 19 percent-ages of children su↵er from diarrhea and eachyear 2.2 million people die from this disease.Washing hands with soap prevent this disease,however, it is di�cult to educate the poor toconsume soap more often (Prahalad, 2004, Page207-227). HUL with the help of Indian govern-ment launch a campaign and educate the poorto improve their lifestyle. Therein, they educateand encourage the BOP inhabitant to use soap

more often. Furthermore, HUL creates soap insmall packages that makes soap more a↵ordablefor the poor. Consequently, diarrhea is reducedin India, for illustration; in Thesgora diarrheareduces from 36% to 5% through Lifebuoys Helpa Child Reach 5 campaign 1. Literatures thatarise around Prahalad’s book provide numer-ous examples of companies that make the prod-ucts more a↵ordable for the poor (Pervez et al.,2013; Hall et al., 2012; George et al., 2012; deAlmeida Faria et al., 2008, Gewald et al., 2012;Halme et al.; 2012, Pea 2014)(Ponssard ,2014).There are two ways for making products more af-fordable; technological advancement and trans-action costs reduction. Technological advance-ments decrease production costs and make prod-ucts a↵ordable for middle and low incomeclasses. This has been the dominant mecha-nism of cost reduction since the industrial revolu-tion, such as reduction of cars price, computers’price, real estates’ price and many other prod-ucts price. Recently, economists propose a newway for costs reduction by introducing transac-tion cost economics (Coase, 1937; Williamson,2010)(North1992; Williamson1979). Transac-tion costs consist of three main components;search costs, contract costs, and costs of enforc-ing contracts [Page 6-9](North, 1994)(North &Wallis, 1994). Reducing transaction costs cutsproducts’ price and make them more a↵ordablefor those who are excluded from the market. Be-sides that, transaction costs are higher for thepoor. Hence, reduction of transaction costs hasthe same e↵ect as technological advancement.That is, it makes the products more a↵ordablefor the consumers, especially for the BOP in-

1http://www.hul.co.in/mediacentre/pressreleases/2014/Lifebuoy-Reduces-Diarrhoea-from-36-perc-to-5-perc-in-Thesgora.aspx

3

habitant. Moreover, transaction cost economicsprovides robust theoretical framework that en-ables producers to analyze production costs morethoroughly (Coase, 1937; Mitra, Mookherjee,Torero, & Visaria, 2013; North & through Time,1995; North, 1994; Williamson, 2010, 1999, 1994;Smith, Venkatraman, & Dholakia, 1999).Prahalad implicitly discussed transaction costeconomics under the title of “Reducing Cor-ruption: Transaction Governance Capacity”(Prahalad, 2004, Page 103-123). Economic the-ories discuss this topic extensively. For illustra-tion, three Nobel prize winners in economics aretransaction costs theorists, Ronald Coase, Dou-glas C. North, Oliver Williamson (Nobel,2009;Nobel 1993; Nobel, 1991).Market design economics is an emerging field ineconomics that creates a rigorous mechanism forconfronting with transaction costs. This fieldutilizes game-theory, mathematics, and IT tofind solutions for market failures and transac-tion costs. This specialty has resuscitated sev-eral markets that severely su↵ered from transac-tion costs; kidney exchange market, high schooladmission, working position for new doctors andso on (Roth, 2008).This paper links economics theories; explicitlytransaction costs economics and market designeconomics, with BOP literature. This connec-tion can further enriches both fields and con-tributes to life of the poor.The structure of this paper is as follows; the firstsection of this paper is a brief review on BOP.Herein, barriers for entrepreneurial entrance andgrowth in BOP are discussed. In addition, Pra-halad’s solution to these barriers are elaborated.In the second section, di↵erent types of transac-tion costs are explained. Moreover, the mech-anism that institutions change in favor of thepoor is explained. The third section is introduc-

tion to market design economics. This sectionexplain four di↵erent types of market failuresand their market design economics solutions inform of real life examples. The final section aimsat bridging market design economics and BOP.Therein, we provide an example of a market ina developing country that severely su↵ered fromseveral sources of transaction costs and marketfailures. Furthermore, this example shows thatprivate firms by redesigning the market can over-come market failures and contribute to life of thepoor. At the end, conclusion of this paper isputted into words.

2 Bottom of Pyramid at aGlance

In 2002 Prahalad declared that, altruistic ap-proaches to poverty and philanthropy cannotsignificantly contribute to the life of the poor(Prahalad, 2004, ”Preface”). He proposed a newinstrument for poverty reduction. In his arti-cle “The Fortune at the Bottom of the Pyra-mid”,(January 2002), he noted; large-scale andwide-spread entrepreneurship is at the heart ofthe solution to poverty (Prahalad, 2004, ”Pref-ace”). He divided the world population to 4economic classes with regard to their incomes.The last largest socio-economic class that entails4 billion people, who earn less than $1,500 peryear, is called the Bottom of the Pyramid (BOP)(Prahalad, 2004, Page 27-29). This theory sug-gests that, the BOP can and must be involved inthe market as a customer and as a part of pro-duction process. The poor not only should notbe excluded from the market but they shouldbe included in the market as profitable cus-tomers and active participants of value chains.Moreover, large-scale entrepreneurs with help of

4

NGOs and local governments should approachthe poor as consumers. A win-win relation be-tween the inhabitant of the BOP and producersis needed in order to enrich the poor(Prahalad,2004, Page 5-20).Prahalad in his book, “the Fortune at the Bot-tom of Pyramid”, introduced several cases thatlarge companies target the BOP as their con-sumers, as a result, both companies and thepoor gain significantly from transactions. After-wards, large numbers of literatures arise aroundthe topic of BOP (Pervez et al., 2013; Hall etal., 2012; George et al., 2012)(Gradl et al, 2010;UNDP, 2014). They further explore other ex-amples of successful entrance of large companiesinto the BOP. In the introduction of this paperCEMEX and Casa Bahia cases are introduced.Herein, we discuss Annapurna Salt. Annapurnais a company that helps to eradicate IntellectualDevelopmental Disorders (IDD), by creating anew product, marketing, and distribution chan-nel.Iodine deficiency is the worlds leading cause ofmental disorder (WHO, 2015). In India morethan 200 million people are in the risk of IDDand 70 million are su↵ering from mental disor-der (Prahalad, 2004, Page 260). Iodine preventoccurrence of IDD. Everyone including the pooruse salt, hence, it is easy to prevent IDD by con-sumption of iodized salt. However, search costsare high in rural part of India, that is, there islittle access to Internet, TV, Radio, and othersource of information center. High search costsact as barriers for the poor to search and edu-cate themselves about the impact of iodized salt.Consequently, majority of, the poor are not in-formed about the benefits of iodized salt and arenot willing to pay extra margin for iodized salt,in comparison to usual salt.Hindustan Unilever Limited (HUL) is a sub-

sidiary of Unilever that operates in India. It op-erates since 1930 and now it is the largest branchof Unilever, it has more than 52% of the totalwealth of Unilever. HUL Research Centre in-novates products that match to Indian demand.HUL feeled the gap of iodized salt in the In-dian market. Hence, the research centre aimedat creating an iodized salt that match well tothe kitchen of middle and low Indian economicclasses. They found that in Indian cooking io-dine vanishes. Consequently, the research centreaims at a salt that keep the iodine in the saltfor longer period, be storable, as well as do notvanish in the Indian cooking. The result of theresearch was an iodized salt. Hindustan Unileverbrands the new salt as “Annapurna Salt”.The di�culty of selling Annapurna Salt intomarket is educating the poor and convincingthem to pay more for iodized salt. The In-dian villages have di�culties in accessing toTV, Radio, Internet and other social network.Hence, it is extremely di�cult to reach the poorvia advertisement and social media. To over-come this problem, Hindustan Unilever intro-duced a new selling and advertisement method(Prahalad, 2004, Page 265).In order to educate the poor and improve distri-bution channel in rural parts of India HUL intro-duced Shakti Ammas. Shakti Ammas are Indianladies who are member of self-helping groups inrural area in India. Hindustan Unilever, givesthem a fix salary for educating the poor aboutbenefits of iodine salt as well as introducingand advertising the HUL products. Moreover,Hindustan Unilever, provides them with a com-mission for selling their products. Shaktis re-ceive stocks from a rural distributor. HUL pro-vides initial training to Shakti Ammas that edu-cates them about selling techniques, knowledgeabout advantageous of products, bookkeeping,

5

and some others useful skills. Each Shakti isable to generate between net incomes of Rs.700-1000 (Balakrishna, 2011, Page 140).The result of the new advertisement and distri-bution system is that more than 45,000 Shaktisare now working for HUL. Moreover, Shaktis cre-ate 10 percentage of total HUL’s revenue (Bal-akrishna, 2011, 141). Furthermore, investiga-tions suggest that Annapurna Salt is successfulin reducing IDD in India. Due to the successesof this project in India, United Nations Inter-national Children’s Emergency Fund (UNICEF)ask HUL to expand its business to other BOPcountries such as Ghana and Nigeria (UNICEF,2008, Page 22). “This really has been a win-winsituation” says Raphael da Silva, former head ofUnilevers Popular Foods team in Africa.

Unilever has developed a successfulbusiness, and UNICEF and the GhanaHealth Service have achieved muchgreater take-up of iodized salt. Mr.da Silva also declared that salt pro-duced locally has also benefited localemployment and manufacturing. Fur-thermore, in 2002, Panama was de-clared free of iodine deficiency disorders(UNICEF, 2008)

.

2.1 BOP Market

The term bottom of pyramid or base of pyramidor simply BOP was first introduced by Prahaladin his influential paper; The Fortune at theBottom of Pyramid. He divided the worldpopulation to 4 economic classes with regardsto their incomes. The last largest and poorestsocio-economic class who are 4 billions and earnless than $1,500 per year are called the bottom

of the pyramid (BOP), graph 2 (Prahalad, 2004,Page 28). This definition is the most commondefinition of the bottom of pyramid (Hall et al2012). There are also other definitions of BOP,for instance Shah 2008 claimed that BOP arethe 2.5 billion people who earn less than $2.5per day (Pervez et al., 2013, Page 55). Someother researchers define people who earn lessthan $3,000 per year as BOP (Hammond etal., 2007, Page 147-158). Most of the people atthe bottom of the economic pyramid are basedin South Asia, Africa, Eastern Europe, theCaribbean and Latin America (Subrahmanyanet al, 2008).In spite of the fact that individual purchasingpower is low in the BOP the aggregate demandis extraordinary. Hence, large scale productionis profitable and attractive for all companies.The GDP of nine of developing countries, whichinclude many inhabitant of BOP, is about $12.5trillion. This implies that the BOP is a largemarket for entrepreneurs (Prahalad, 2004, Page34).

Graph2

6

2.2 Entrepreneurship Barriers inBOP

Prahalad believed that transparency reducestransaction costs. In the same token, it is anecessity condition for entrepreneurship growth(Prahalad, 2004, Page 107). Nevertheless,majority of the BOP countries su↵ers froman opaque market with high transaction costs(Prahalad, 2004, page 105-106). For instance,for registering a land an Indian farmer has tofirst approach a broker. The broker lobbies withauthorities and does the long bureaucratic pro-cess and eases the transaction. The cost of regis-tering a land for a farmer is the registration costplus unnecessary transaction costs; broker fees,corruption for bribing o�cers and etc (Prahalad,2004, Page 107-109). As it is explained in de-tail in the next section, high transaction costslead to market failures. In the case of highbroker fee and high bribing cost a poor farmercannot a↵ord registration and this market fails.High transaction costs not only disincentivizethe multinational companies to participate in thepoor countries, but also they hinder local en-trepreneurship growth (Zacharakis et al., 1997).C.K Prahalad declared that there are three re-quirements for creating a transparent marketand reducing transaction costs: Firstly, accessto information and transparency in all of trans-actions. Secondly, fast speed in transactions. Fi-nally, high trust level in the system (Prahalad,2004, Page 109).Prahalad suggested that enforcing an appropri-ate rules is the main way for transparent market(Prahalad, 2004, Page 110). However, experi-ences in developing countries suggest that chang-ing the formal rules and regulation if it is notimpossible in the short time it takes long. Thereason for that is powerful organizations that

benefit from opaque market do not allow radi-cal changes (North, 1992, Page 11).A new field in economics that is called “MarketDesign Economics” or “Economic Engineering”provides other alternative for improving mar-ket’s transparency in short run. In the section of“Market Design Economics” it is explained thathow market design economics’ solutions can ful-fill the transparency conditions, which are pro-posed by Prahalad. Before we explain marketdesign economics, di↵erent types of transactioncosts are explained.

3 Transaction Cost Economicsat a Glance

In 1937 for the first time Ronald Coase notedthat the neoclassical results of e�cient marketsobtain only in the absence of any transactioncosts (Coase, 1937). He showd that, institutionsplay a crucial role when there is a substantialtransaction costs in a market. Moreover, the re-sults of economic activities and entrepreneurialgrowth depend on the transaction costs in themarket (Williamson, 2010, Page 217).For each economic activities an agent hasto bear two source of costs, the first one istransformation (production) costs; that is, allof the costs that incur for transforming rawmaterials into the final products. The othercosts are transaction costs; search cost, contractcost, and cost of enforcing contract. Moreover,transaction costs are determined by qualityof judiciary system, bureaucratic system, andin general term all components of businessenvironment. (North,1994, Page 6-8). Mostof participants in an economy do not produceany final goods. Examples of activities that areinvolved in transacting sectors are politicians,

7

brokers, accountants, judges, bureaucrats,and so on. Transaction costs depends on thecomplexity of an economic system (North, 1989,Page 666). The more complex and opaquean economy the higher transaction costs. InAmerica, which is a developed country and hasless complexity in comparison to developingcountries, 45% of GDP in 1970 was spent onthe transaction sectors. This number is byand large higher in developed countries (North,1992, Page 6).The powerful institutions determine economicstructure of each country, and economicstructure determines wealth distribution(Williamson, 2010). In contrast to what neo-classical economists believe, the competitionamong economic participants does not alwaysleads to an e�cient outcome. Those economicparticipants that have the power and informa-tion design the economic structure in such away that they gain the largest share of wealth,even in the expense of others’ losses. If rulesdo not restrict the powerful transacting sectorsthey strive for a structure in which majorityof the wealth is allocated to transacting sectorand tiny share to production. In other words,transacting sectors can hinder economic growth.That is, an inappropriate economic structurereplaces the productive entrepreneurs to inef-ficient transaction sectors, consequently, mostof a nations wealth spends on transaction costs(North, 1992, Page 6-8).

3.1 Search Cost

One of the transaction costs is search cost andthere are two types of search costs; internal andexternal. Internal search cost is the cognitivee↵ort buyers must engage into direct search in-quiries, sort incoming information and integrate

with stored information to form decision evalu-ations (Smith et al., 1999). The external searchcosts are the costs that are beyond the consumerdirect control and consumers have to decide tobear the cost or not. There are two sorts of ex-ternal costs; primo, cost of acquiring the infor-mation. Secundo, opportunity cost that is spenton searching rather than working. The exam-ple of external search cost is waiting time that isthe time that a customer has to spend in orderto achieve information. In addition, during thepast decades the opportunity cost for the household has been increased, due to the increase inthe women workforce and labour working hours(Jacoby et al, 1976).Consumers make decision on level of their searchbased on the costs and benefits of searches.Moreover, consumers continue the searching pro-cess till the marginal cost of search be equal tothe marginal benefit of search (Smith et al., 1999,Page293). Hauser et al. (1993) claimed that,buyers maximize the value of search subject toa budget constraint. This implies that, a buyermaximize the search with respect to the totaltime that he needs to allocate for a search, whichsources to search, and how much time to spendat each search source.Studies have shown that knowledgeable buyersincur lower search costs and therefore searchmore; less knowledgeable buyers incur highersearch costs and search less (Srinivasan et al.,1991). High knowledge consumers have sophisti-cated decision-making structure. They analyzethe information faster and; as a result, comparethe quality of di↵erent products easier. In sum-mary, high knowledge consumers have two ad-vantages in comparison to low knowledge coun-terparts; first of all, the cost of analyzing infor-mation is lower for them. Moreover, they aremore able in linking internal and external infor-

8

mation. Consequently, they can better managethe search level. If the external search costs,such as waiting time, be susbtantial, high knowl-edge consumers are able to combine the cogni-tion with little external search. Moreover, in thesituation where the external search cost is lowthey use more external searches (Smith et al.,1999, Page 293-296).In comparison, low knowledge consumers su↵erfrom higher search cost. Hence, they approachpurchase decision with help of rules of thumband heuristics (Betmenn et al., 1980).There is a negative relation between search costand search level. The relationship between priorcategory knowledge and search level is positive(Smith et al., 1999).Consumer in BOP, the poor, faced with highersearch cost in comparison to the middle and highclass. The reason for that is, first of all the in-ternal search cost is higher for the poor. That is,middle and high social classes enjoy high qual-ity education system and this helps them to de-velop their cognition and information processingskill. The result is that internal search costs arelower for high-income class. Furthermore, thepoor su↵er from higher external costs. This canbe explained by lower access to the Internet, TV,radio and other social communication among thepoor in contrast to the people in the middle ortop of the pyramid. Besides that, majority ofthe poor live in the rural area that restrict theiraccess to the information and make external in-formation less available.

3.2 Contract Cost

There are personal and impersonal transactions.In personal transactions family ties, friendship,loyalty and certainty of repeating transactionsdecrease opportunistic behaviors. Consequently

the need for detail specification in contract de-creases. Nonetheless, in impersonal transactionsthere is risk of opportunistic behaviors. Theneed for detail specifications in contract increasecontract costs (North 1994; Wiliamson 1979;Williamson, 2010). Hence, there is a need fordetailed and o�cial contracts. This suggests,substantial costs have to be devoted for makingcontracts. Often, there is a need for employing alawyer and registering a contract in an organiza-tion with power of punishment and enforcement(North, 1992, Page 8-10).Costs of contract are higher for the poor in com-parison to the rich counterpart. The reason forthat is, the poor usually cannot a↵ord lawyerand knowledge about the rules and regulations.

3.3 Enforcement Cost

In case of weak judiciary system shirking andcheating pay o↵. As a result, transactionsare too costly and agents are less willing toparticipate in economic transaction. Strongjudiciary system that enforce contract is essen-tial for economic growth. The Western worldscourts, legal system, and strong judicial systemguarantee economic specialization and permitdevelopment of complex systems. E�cientmarkets are the consequence of institutions thatprovide low-cost measurement and enforceablecontracts (North, 1992, Page 8-10).Because of the fact the judiciary systems areweaker in the developing countries enforcementcosts are by and large higher, in comparison todeveloped countries. The low-income class haslower access to the judiciary system and deemedto bear higher enforcement cost.

9

3.4 Institutions and InstitutionalChanges

As it is mentioned before, the poor bear highersearch costs, contract costs, and enforcementcosts, in contrast to high income classes. Thehigh transaction costs lead to market failures.High-income classes push their interests throughchanging the institutions, nonetheless, becauseof the fact that the poor do not have access topower they cannot take the same approach. Thissuggests that, markets are not automaticallypersuade the interests of the poor and marketsfail for the low-income classes. Consequently, themain question is how to reduce transaction costsfor those who are excluded from the market.In the bellow theories of instituitionalists aboutchanging the institutions are explained. Insti-tuitionalists do not explain a feasible methodfor changing the institutions. Nonetheless, mar-ket design economics is an emerging field thatproposes feasible tools for changing the institu-tions and reducing transaction costs. This fieldredesigns several markets and substantially re-duces transaction costs.North defined institutions as the structures thathuman create in dealing with each other (North,1992, 10). Furthermore, he claimed that insti-tutions as well as the technological changes de-termine transactions and production costs. Ide-ally, institutions should provide transparencyand information to market participants. How-ever, powerful institutions can manipulate theinformation that they have and create an opaquemarket. Furthermore, Institutionalists define or-ganizations as players that play in the gamethat its rules are determined by the institutions(Williamson, 1994, 1999).Institutions determine opportunity sets in a so-ciety (North, 1989,665-667). Consequently, eco-

nomic opportunities determine kinds of organi-zations that come into existence. Organizationsaimed at maximizing their survival possibilities.They learned skills and knowledges that pay o↵under the institutional rules. North wrote that:

If the highest rates of return in a so-ciety are from piracy, then organiza-tions will invest in knowledge and skillsthat will make them better pirates; ifthe pay- o↵s are highest from increas-ing productivity, then firms and otherorganizations will invest in skills andknowledge that achieve that objective.(North, 1994, Page 10)

Two agents have power to change the insti-tutions; economic entrepreneurs and politicalagents (North, 1992,11). Motivation of en-trepreneurs for changing the institutions is mar-gins and profits that they earn from changingthe structure of the institutions. For example,entrepreneurs have incentives to change the in-stitutions in order to reallocate the margin frommiddle mans to entrepreneurs.

4 Market Design Economics ata Glance

Market design economics is about understandingof how market works, why they are failed, andhow to fix them (Roth, 2008). Alvin Roth inves-tigates transaction costs and market failures inseveral markets. He proposes solutions for themarket failures(Jackson, 2013, Page 620). Dueto his significant contributions to di↵erent mar-kets, the Sveriges Riksbank Prize in EconomicSciences in Memory of Alfred Nobel in 2012 wasawarded to Alvin Roth, the founders of this field(”Nobel Prize”, 2010).

10

Alvin Roth believes that market failed due tothree main reasons: 1. Thinness: A market thatdoes not attract su�cient proportion of poten-tial market participants will fail. A successfulmarket needs to be thick enough.2. Congestion: In some cases of thick markets,there is a problem of congestion. That is, themarket might not be able to make transactionsfast enough so that market participants can con-sider enough alternative possible transactions toarrive at the satisfactory one.3. Safeness: Make it safe to participate in themarket as simply as possiblea. as opposed to transacting outside of the mar-ketplace orb. as opposed to engaging in strategic behaviorthat reduces overall welfare (Roth, 2008, Page80).A well functioning market fulfills the three trans-parency requirements that Prahalad proposed.Many policies and institutions aimed at solv-ing the aforementioned forms of market failures.One mechanism that fulfills the requirements iscentralized clearinghouse. An example of mar-ket with clearinghouse is exchange markets. Inexchange markets buyers observe all of productsand choose the best possible option.He showed that an appropriate clearinghouseovercomes market failures (Roth, 2002). A clear-inghouse is a centralized matching mechanism.In the same token, he elaborated that the cru-cial features of modern clearinghouse in creatinga transparent market are (Roth, 2008; Niederleet al., 2008; Erev et al. 2010c):1) Overcome Thinness problem: a market hasto create the possibility of participation of manyplayers in a virtual place. Furthermore, it shouldreduces cost of physically participation in themarket. In other words, it aggregates demandand ensure that there are enough participants in

a market (Coles et al. 2010).2) Overcome Congestion in market: the process-ing ability of Internet makes it possible to pro-cess several transaction in a market without hav-ing congestion. Furthermore, it collects informa-tion and has faster decision process (Roth, 2008,Page 5-7).3) Provides trusted intermediaries; it creates asafe process participant to participate and re-veal their information truthfully. A middlemanis trusted from both sides (Roth, 2002, Page 13).4) Computers evolve auction markets intomatching market. This means that the ine�-ciencies in the auction market will be replacedby certainty in the market (Roth, 2002, Page13).The majority of the markets that market design-ers have involved are two sided matching market.That is, an agent not only has to choose but ithas to be chosen. The examples of these sorts ofmarkets are marriage market, hospitals and res-idences that both have preferences about eachother, and etc. A centralized clearinghouse hasbeen used in order to solve market failures andreduce transaction costs.The structure of this part of the paper is as fol-lows, first of all, the introduction to two sidedmatching market is provided. After that, fourdi↵erent forms of market failures and their mar-ket design solutions are explained. These mar-ket failures and their solutions are examples ofreal markets that market design economists haveinvolved in. Finally, it is explained that to over-come market failures it is sometimes necessaryto utilize force.

11

4.1 An Introduction to MatchingMarket: Marriage Market

This section explains two sided matching mar-ket structure. The simplest form of matchingproblem is Two Sided Matching, also knownas marriage market. In the simplest form ofthis market there are two agents; one proposesa match and one either accepts or rejects theproposal. The marriage market refers to setsof men and women that want to marry to apreferred partner.The assumptions of this market are; first, eachwoman can only marry to a man and eachman can only marry to a woman. Each woman(man) has strict preferences over men (women).Each participant prefers to match rather thanbeing unmatched.The main question in this market is how to finda match that is stable. Gale and Shapely definedstable match if no pairs of agents have incentiveto change the current match (Roth, 2008, Page3). The emphasize is on the pair, because asingle participant might want to change thematch, however, there is no pair that are willingto change the match. The other question, forcreating an e�cient matching market, is howto find a stable match that is also a dominantstrategy for all pairs of players. Gale andShapely (1962) proved that di↵ered acceptancealgorithm always leads to stable matches andis dominant strategy for pairs. The marriagemarket example help to better understand thematching market and its solution. Imagine acountry with three men and three women. Table1 explain the preferences of participants:

The two stable matches are (M1,W1), (M2,W3),(M3,W2) and (M1,W2), (M2,W3), (M3,W1).The mechanism that each agent achieves thisresult is that, in the first place a man proposemarriage proposal to woman, or other wayaround. The proposer propose to the mostpreferred choice. Receiver select the mostpreferred proposer. In a case that receiverreceives a proposal from his most preferredproposer, he or she accepts and there will be amatch. Otherwise, there would be no match andreceivers remain single. Each proposer who isnot in a match in the first round proposes to themost preferred choice, which is still available, inthe second round. If there is no choice availablea proposer remain single. This process repeattill the point where proposers either run out ofacceptable receivers or are engaged.To illustrate the e↵ects of algorithm in theaforementioned example, at the first round M1proposes to W2, and M2 and M3 is proposeto W1. The consequence of first round will be(M1,W2) and (M3,W1), and M2 proposal isrejected. Hence, the result will be the stablematching of (M1,W2), (M2,W3), (M3,W1).This algorithm can be extended to a marketwith many participants and still leads to thestable match result. If we were to reverse theroles of women and men in the algorithm, we

12

would actually find the other stable matchingof (M1,W1), (M2,W3), (M3,W2). In case thatwomen be the proposer the stable match will be(M1,W1), (M2,W3), (M3,W2).Suppose rather than di↵ered acceptance algo-rithm, the algorithm of (M1,W1), (M2,W2),(M3,W3) be used. This match will be blockedby the pair (M1, W2), that is M1 and W2 haveincentive to break the current match and matchtogether. Then the match will be changed to(M1,W2), (M2,W1), (M3,W3). The currentmatch block by (M3, W2), by rematching thepartner the match will be (M1,W3), (M2,W1),(M3,W2). The achieved match is again block by(M3,W1), as a result the match will change to(M1,W3), (M2,W2), (M3,W1). As this match isblocked by (M1,W1) the match will be changeto (M1,W1), (M2,W2), (M3,W3).The di↵erence between stable and unstablematch can be found in aforementioned exam-ples. In the first example there is stable match,this implies, there is no pair of match that havean incentive to break the match. Although W3match with her least preferred match, but thereis no man who is willing to break his relationand start a new relation with W3. Nonetheless,as we can see in the second algorithm there isalways possibility of finding a pair that changesthe match.In summary there is always exists a stable matchin matching market and the stable match(es)can only be found by deferred preferencesalgorithm. Moreover, a random algorithmmight leads to endless cycle. In other words,in a decentralized matching market there isno guarantee that the market achieve stablematch. A centralized clearinghouse is necessarycondition for a market to achieve stable match.Furthermore, a decentralized match can increasethe cost of transactions in market. That is,

participants faced with high risk of unsuccessfulmatch and high cost of transactions. Hence,they might show reluctant to involve in amarket with decentralized matching system. Incase of marriage market, a boy face with therisk of not finding his perfect match and hemight not marriage at all. However, if a marketguarantees that the match mechanism leads tothe best possible result boys are more willing toparticipate in this market (Jackson, 2013, Page620-630).

4.2 Unraveling and Market DesignEconomics: Market for New Doc-tors

This example elaborates the problem of unrav-eling in job market for new doctors and accord-ingly the solution for this problem.Alvin Roth is the first one who found applicationof Gale and Shapely in real world market. Thefirst time that he found Gale and Shapely algo-rithms application was in 1984 in the job mar-ket for new doctors. He proved that since 1951deferred acceptance algorithm has been used inmatching new doctors to hospitals.New doctors have to pass two years of intern-ship after their graduation, which is called res-idency. Residency provides a crucial workforcefor hospitals and is considered as an importantperiod that influence medical students’ futurejob career. Hence, there is a fierce competitionamong hospitals to hire medicine students grad-uate. From 1900 to 1945, there had been compe-tition among hospitals for hiring residences. Asa result hospitals o↵ered positions to studentstwo years before students graduation.Today we know that this market su↵ered fromunraveling, which is a common and costly formof market failure. This implies that, students

13

had to make decisions before being well informedabout their preferences and hospitals quality.Hence, students were not able to make the bestpossible decision.Market design economists have discussed prob-lem of unraveling and have proposed several so-lutions for it. For instance, Roth and Xing(1994) noted that as in thirteenth century un-raveling can be eradicated by not allowing trans-actions before goods are o↵ered in a market. Af-ter 1945, in order to avoid unraveling, Ameri-can medical schools released students informa-tion just one year before graduation. Conse-quently, a new problem arised; if most preferredstudents rejected hospitals o↵er, the other can-didates might accepted o↵ers of other hospital.That is, the market did not guarantee that thehospitals and students find their most preferredmatches. Moreover, there was a risk of not find-ing enough residences for hospitals. Hence, hos-pitals proposed to increase numbers of students.The reason for that was hospitals wanted to en-sure that they reach as many number of studentsas possible. This leaded to congestion that is an-other forms of market failure.Markets di�culties and failures brought hospi-tals, students, and schools together to solve thecurrent market failures. Therefore, they de-signed a centralized clearinghouse to improvemarket coordination. The clearinghouse systemworks in the following manners, in the first roundstudents apply and hospitals interview with thestudents. Afterwards, students rank the hospi-tals that they have interviewed and hospitals alsorank the students that they have interviewed.Finally, a centralized clearinghouse, which iscalled today National Resident Matching Pro-gram (NRMP), matches hospitals and students.In 1984 Roth proved that the National ResidentMatching Program (NRMP) employs deferred

preference algorithm, which was created by Galeand Shapely (1962), and it always results in astable match. This suggests, under the currentmatch system there are no student and hospi-tal that prefer mutually to match together butthey do not. In other words, there is no pair ofstudent and hospital that can break the resultedmatch. Roth further showed that 95 percentageof matches in this market were done by clearing-house and particularly by deferred acceptancealgorithm.One might suggests that the successful matchescan be the result of any algorithm and not merelydeferred acceptance algorithm. Roth 1990 and1991 launched a study in England and comparedthe di↵erent region that apply di↵erent algo-rithm. He found that the centers that applydeferred acceptance algorithm produced stablematching and keep the algorithm for a long pe-riod. In contrasts, the other centers that useother kinds of algorithm created instable match.In the same token, Kagel and Roth, in lab experi-ment, proved the hypothesis that deferred accep-tance algorithm eliminate unraveling, however,other algorithm produce unraveling matches.This examples suggests that, centralized clear-inghouse is a solution for unraveling, which isone of the costly form of market failure. Fur-thermore, deferred acceptance algorithm leads tostable match (Roth, 2008, Page4-10).

4.3 Thinness and Market Design Eco-nomics: Kidney Exchange

The kidney exchange market is the example ofmarkets that has been failed for several yearsdue to market thinness. Kidney transplantationis the final remedy of patients with diseases re-lated to kidney. Moreover, the life of many kid-ney patients depend on kidney transplantation.

14

Healthy people can live with one of their kidneysand donate the other one. However, due to rulesand regulation kidney purchasments and sells areforbidden (Roth, 2006, Page 2-6). Moreover, themain problem for kidney donation is that theblood type and the immune system of patientsmight be incompatible with their donors. AlvinRoth with respects to market limitations de-signed a mechanism that facilitate transactionsin this market.In the Unites States of America, more than70,000 patients are in waiting list for a deceaseddonor. Nevertheless, only 11,000 receive a kid-ney. The result is that above 5000 patients diedwhile they are in the waiting lists for a compat-ible kidney. The same situation holds in otherparts of the worlds. For illustration, in UK 6000renal patients are in waiting list for a trans-plantable kidney, nonetheless, only 1240 have achance to operate a transplantation surgery.Healthy people also can donate one of their twokidneys to renal patients. Moreover, kidneytransplantation from a healthy body is by andlarge more helpful for the patients than cadaverkidney transplantation. However, the donationof kidney requires compatible blood type and im-mune system.There is excess of demand in kidney market,however, due to legal system of all countries, it isillegal to purchase and sell kidney. That is, mostof the societies show repugnant toward transac-tion of some specific goods (Roth, 2008, Page10-12).Alvin Roth indicates that the task of economistsas a market designer is to create a mechanismthat facilitate and increase transactions, givenpresent legal restrictions and social repugnancein the market.There are two possible ways to increase numberof transactions in this market, given the current

legal rules; firstly, the exchange of kidney be-tween donor-patient pairs 2. To understand it,assume two patient-donor pairs, donor A wantto donate his kidney to patient A, and donor Bwants to donate his kidney to patient B. Kidneyof donor A is not compatible with patient A andthe same hold for donor-patient B. Nevertheless,donor A has a compatible kidney with patient Band donor B has a compatible kidney with pa-tient B. It is possible that a donor A donate hiskidney to the patient B in exchange for donor Bdonate his kidney to patient A.Secondly; the other possible exchange is betweena donor-patient pair and a patient who is earlyin the waiting list of a cadaver kidney. Again,assume donor-patient pair of A and patient Bwho is early in the cadaver kidney waiting list.Donor A donates his kidney to patient B, andcadaver kidney is given to the patient A.In spite of the fact that both forms of exchangeare feasible and are in aligned with social normsthese sorts of exchange are rarely happen. Forexample, only 5 exchanges had been occurred in14 transplant centers in New England.There are several factors that explain why ex-change has been so rare, or the kidney markethas failed to exchange kidneys. The most domi-nant cause of market failures is that the market isthin. That is to say, there are not enough incom-patible patient-donor candidates for exchangesin the market. Furthermore, when a compatibil-ity test reflects that donor kidney does not matchto his patient, doctors send a way donor with-out recording the information of donors. Besidesthat, in case that donors information is available,due to medical regulations and privacy rules itis not possible to have an access to the informa-

2patient-donor pair, a donor who wants to give hiskidney to a patient

15

tion.Roth, Sonmez, and Unver (2004a) proved that acentralized clearinghouse that gather databaseof incompatible donor-patient pairs make kid-ney transactions more feasible. This means, themore information about di↵erent donor-patientpairs the easier to find match between di↵erentpatients and donors. Furthermore, they showedthat there is no restriction in sizes of pairs. Thatis, the system has potential to match as manypairs as possible. The reason for that is, thereis a cycle of kidney exchange where a first pairgives to the second pairs and a second pairs giveto the third and third give to the fourth and soon. This cyclel continue till the cycle closed, andlast pair donate to the first.Roth and his colleagues sent this idea to manykidney surgeons. As a result, one of the surgeons(who was the medical director of the New Eng-land Organ Bank) accepted the idea and broughtthe idea to 14 kidney transplant centers in NewEngland. In these centers they started to usethe matching software that is designed by Rothand his colleagues. After years, the matchesextended to several transplantation centers andthey applied the software clearinghouse.In summary, the kidney market was failed due tothe lack of market thickness. The market failedto bring what Jevons called double coincide ofwants (Jevons, 1876). A centralized clearing-house that entail database of donor-patient pairsincreases the numbers of kidney exchange.

4.4 Congestion and Market DesignEconomics: School Choice

This is an example of markets that su↵eredfrom congestion and money does not influencetransactions. The school select annually num-ber of students based on sets of criteria. Stu-

dents have some specific preferences and thereare also admission requirements from schools’side. Market thickness is not a problem here,since government provides supply side in accord-ing to number of students. The performance ofschool choice systems depends on two factors;firstly, quality of dealing with congestion; sec-ondly, makes it safe for participants to revealtrue preferences. New York City (NYC) educa-tion system su↵ered from congestion. Formerly,Boston education choice system did not make itsafe for students to reveal their true preferencesand Roth had successfully redesigned the sys-tem and solved the problem (Chen et al., 2006;Erginet al., 2006; Pathak et al.,2008). Hence,NYC education department asked Roth and hiscolleagues to redesign the education choice sys-tem. A centralized clearinghouse solved theproblem of NYC education system.In the following it is explained that how an inap-propriate structure of a system leads to conges-tion and make it risky for participants to revealtheir preferences. Furthermore, the solution forthese problems are explained.In NYC around 90,000 students have to assignto 500 high school program. The old systemmatched students with schools in the followingmanner; First, students filled out a rank orderlist and only mentioned the top five schools thatthey prefered them the most. Second, the pref-erences list was sent to the five schools. Third,schools made three decisions; accept, reject, andput the students in a waiting list. Fourth, the re-sult was sent to the students. Fifth, students ac-cepted one choice and inserted one of the choicesin waiting list. Sixth, Schools that still possessedempty positions sent new o↵ers. This processrepeated only three times and afterwards thosestudents who did not get admission from anyschools were assigned to zoned schools or as-

16

signed in an administrative process.The market su↵ered from several sever problems.First of all three rounds of process were not suf-ficient to allocate 90,000 students, who made 5choice over 500 potential options. This suggeststhat, market failed due to the congestion (Roth1997). Consequently, only 50,000 received o↵ersand top 17,000 students secured with several of-fers. Moreover, more than 30,000 students as-signed to the school that were not in their pref-erence list.Three issues motivated NYC education depart-ment to call Roth for redesigning the choice sys-tem; primo, 30,000 students who were assignedto the schools that they were not prefer it. Se-cundo, students ranked the schools strategically.That is to say, students filled in preferences listuntruthfully. Students ranked the schools thatwere most likely got admission from and not pre-ferred it the most. The reason for that was, stu-dents thought that schools considered the pref-erences lists for assigning the admissions. Tertio,schools played strategically and did not revealedtheir actual capacity to NYC department of ed-ucation. The reason for that was schools afraidthat students who were rejected from the otherschools filled their positions and studied unwill-ingly.Exercising centralized clearinghouse contributesto the quality of choice system. First of all,it overcomes congestion: specifically, NYC ed-ucation department reduced the number of ad-ministratively assigned students from 30,000 to3,000. Secondly, schools learned that it is notprofitable to hide the capacity. That is to say,the new system guarantee that schools positionsare filled in by students who prefer schools themost. In other words, the new system ensuresthat the match is the best possible match. Inthe same token, Roth and his collogue proved

that deferred acceptance and top trading cyclesalgorithms, which are discovered by Shapely andScarf in 1974, are strategy-proof. In the caseof schooling problem this implies that, for fam-ilies it is dominant strategy to reveal their truepreferences. The last but not the least bene-fits of clearinghouse is that, as the preferenceslists are filled in honestly and truthfully; theNYC schools data are used for several politicalresearches and decision-making process. For in-stance, it has been used to identify the low qual-ity schools. (Abdulkadiroglu et al., 2007).This example can illustrates that computers area trusted intermediaries, consequently, both par-ties have incentive to provide true information.

4.5 Adverse Selection and Market De-sign Economics; Market for NewEconomists

Job market for new PhDs is a decentralized andopen market. This means that, PhD graduatesapply for several institutes separately and in-stitutes consider the applications independently.Moreover, the market for PhD graduate is notgeographically concentrated.In the North America this market su↵ered fromtwo sources of market failures. The main prob-lem was adverse selection, this implies, institu-tions that o↵er jobs were not certain that appli-cants considered the position certainly. Usually,applicants considered some positions as their de-fault options and were not interested on the po-sition. The other problem was congestion; thesystem in North America was designed in a man-ner that it triggered congestion in the market.In the following, first the structure of the jobmarket and its problems are explained, after-wards, it is elaborated that how a central signal-ing system solve the adverse selection problem

17

and a centralized market overcome the conges-tion.The job market for new PhDs is decentralized,however, there are some centralized institutionsthat operate in this market such as AmericanEconomics Association (AEA). The job marketstructure is as following: in the early fall jobseekers can see the advertisement of institutionsin economic associations magazine (for examplesee (http://www.aeaweb.org/joe/). PhDs stu-dents send their application and reference letterto institutions. As a result, department receiveshundreds of application in late fall. They haveto go through all details of the applications andmake a selection for 30 minutes interview thathold in large meeting in early January. Thismeeting consists of 51 associations and takesplace in a hotel and applicants have an interviewwith department in suites. Department have toreserve suites two months in advance. Recruit-ing teams interview candidates one after anotherand candidates up and down the flours and haveseveral meeting in a day.After the meetings in early January some of thecandidates are selected for the fly out. That is,a day where candidate will meet faculty memberand dean.The consequence of this system is that by theend of March market faces thinness problem. Forinstance, departments interview with 20 appli-cants, and invite 6 of them to fly out day ando↵er two of them the position. In case thatthe two applicants reject the o↵ers, departmentshave to look at all of the applicants and start allof the procedure again to find a suitable candi-date. The same problems hold for applicants, ifthey were rejected in the last step they have tosearch for new department. This indicates howcostly is adverse selection for the institutions.AEA by creating a central signaling house over-

come partly the adverse selection problem. Stu-dents have to send a list of their preferred de-partments and their profile to AEA. AEA trans-mits only two recommendation letters to two de-partments. Due to the fact that the signals arelimited they provide valuable information. Fur-thermore, it overcomes the congestion problem,because of the fact that only limited numbers ofapplicants are sent to the interview.

4.6 Force; a Solution for Market Fail-ure: Gastroenterologists Example

This example illustrates that in some caseslegislation and enforcement of rules are necessi-ties for operation of a successful clearinghouse.Moreover, the solution for market failures canonly be achieved if several powerful players in amarket cooperate.American medical students who aimed at be-coming a gastroenterologist have to pass threeyears of residency in internal medicine and thenapply for a gastroenterology fellow position.From 1986 to 1990 a stable clearinghouse hadbeen used to assign residents to gastroenterologyfellows. Nevertheless, at the beginning of 1991this centralized system was replaced by a de-centralized matching market. The consequencewas the emergence of unraveling in the market.The gastroenterology institutions o↵ers werescattered in the year and they shorten the pe-riod of response. Furthermore, gastroenterologyinstitutions o↵er to students two years beforestudent enter into the job market. These causedthinness of market and most of the studentscontinue to work as a gastroenterologist in thehospitals where they were residences.These issues created a common consensus,among various organizations that involved ingastroenterology market, that the clearinghouse

18

should exist one year before accomplishment ofresidency period.The problem that raised was that programdirectors who ask for recreation of clearinghouse afraid that the other gastroenterologyorganizations do not use the clearinghouse ando↵er students, again, in two years in advance.As a result, they discussed their concerns withRoth and his colleagues. Alvin Roth proposedthat some regulations have to be enforced inorder to disincentivize applicants and employerto transact outside of the clearinghouse. Rothand Niederle 2008, in lab experiments, testedthe impact of rules on the time of agents o↵ersand applicants response. They propose thatthe matching should be occurred in 15th ofApril. Moreover, they proposed to enforcethe following policy; any working agreementswith residence before the 15th of April canbe terminated. Nevertheless, residences areresponsible for the working agreement that aremade after 15th of April and no party can breakan working contract without having agreementof both sides 3.Four major gastroenterology professional or-ganizations accept to implement the policy;American Gastroenterological Association, theAmerican College of Gastroenterology, theAmerican Society for Gastrointestinal En-doscopy, and the American Association for theStudy of Liver Diseases. The resolution ensuresthat the applicants and position are filled onlyafter the 15th of April and there is no need forearly contract.Resolution aimed at: make it unattractive forprogram director to o↵er in advance. Moreover,nudge applicants to do not accept early o↵ers.

3http://www.cgsnet.org/ por-tals/0/pdf/CGSResolutionJune2008.pdf

As a result, 79% of fellowship programs par-ticipate in the matching clearinghouse. Formthe participants 98% match through the clear-inghouse. This implies, the new rules make itsafe for players in the market to participatein the match. Furthermore, the problem oftiming, dispersed o↵ers through the year, andcongestion have been solved by clearinghouse.This example illustrates that, for clearinghousesto work e↵ectively sometimes there is a needto impose rules and implement force. In manycases, clearinghouse is a solution for marketfailures. However, there are agents who ben-efit from decentralized and opaque market.These agents are not willing to participate inclearinghouse or other mechanisms that trans-parent the market. In this situation, a physicalclearinghouse by itself would not make a greatchange in the market. There is a need for usingforce and regulating the market. In these casesprofessional Gastroenterology organization seattogether and enforce the rules.

5 Market Design and BOP: ACase Study of AgriculturalSector in India

In this section it is explained that high transac-tion costs in Indian agricultural industry lead toseveral source of market failures. In the sametime, an Indian private firm creates a semi-clearinghouse and reduces transaction costs inthat market. Consequently, income of morethan one million Indian farmers substantially in-creased and the firms gain large profit (Kumar,

19

2004,Page 47-60). 4. This section explains in de-tail how this change has happened.In the below the structure of Indian agriculturalsector in post-colonial era is explained. After-wards, the market failures and their causes inthis market are elaborated. Besides that, the im-pact of a semi-clearinghouse on the agriculturemarket is discussed.

4It seems too simplistic that an innovation contributesignificantly to life of more than one million poor farmers.However, this innovation has been so influential that it at-tracts attention of several development awards. Some ofthose awards are; United Nations Industrial DevelopmentOrganisation (UNIDO) Award 2008, Ashoka - Change-makers ’Health For All’ Award 2006, the Stockholm Chal-lenge 2006, Innovation for India Award 2006, Social In-novations category for business organizations, Develop-ment Gateway Award 2005 (previously known as the Pe-tersberg Prize) for the trailblazing e-Choupal initiative.Moreover, ITC is the first Indian company and the secondin the world to win this prestigious award. The ’GoldenPeacock Global Award for Corporate Social Responsibil-ity (CSR) in Emerging Economies for 2005’. The Corpo-rate Social Responsibility Award 2004 from The Energyand Resources Institute(TERI) .The inaugural ’WorldBusiness Award’, instituted jointly by the InternationalChamber of Commerce (ICC), the HRH Prince of WalesInternational Business Leaders Forum (IBLF) and theUnited Nations Development Programme (UNDP). Thisaward recognises companies who have made significante↵orts to create sustainable livelihood opportunities andenduring wealth in developing countries.The ’EnterpriseBusiness Transformation Award’ for Asia Pacific (Apac),instituted by Infosys Technologies and Wharton School ofthe University of Pennsylvania. PC Quest’s IT Implemen-tation Award in the ’Best Project’ category. The GoldenPeacock Innovation Award 2004. The NASSCOM awardfor ’Best IT User in FMCG’ in 2003. The Award is arecognition of ITC’s successful integration of its IT usagewith its business processes. The Seagate Intelligent En-terprise of the Year 2003 Award, for the most innovativeusage of Information Technology

5.1 Indian Agricultural Industry

Backward farming technology and inability inweather prediction leaded to shortage of food inpost independence India. Consequently, politi-cians aimed at implementing policies that solvethe food shortage, insecurity, and in the sametime increase the welfare of the poor farmer.The result was a green revolution in India. Aspart of this plan, the Indian government decidedto distribute farms to the end cultivators. Theideology besides this was to make farms moreproductive and redistribute wealth to the poor.Nonetheless, the farms became too small to beprofitable and farmers became extremely poor.

5.1.1 Mandis

Agricultural Products Marketing Acts created agovernmental agency that is called Mandi withthe aim of more equitable distribution of grainsamong consumers, and traders. Each Mandisapproximately covers 700 square kilometers.Farmers bring their productions to Mandis andtraders purchase products in Mandis place.Hence, they are the key in bridging demand andsupply of oilseed, especially for Soya where themain use is for crushing company.There were three ways for farmers to sell theirproducts: Mandis, traders that distribute theproduct to the crushers, direct sale to crushers.Selling through Mandis and traders accountfor 90% of the transactions in the market(Prahalad, 2004, Page 341).In each Mandis there are two commission agentsthat are responsible for lunching the trades.

20

5.1.2 Mandi Value Chain

Graph 3: Agricultural Value Chain at time ofMandi

1.Inbound Logistic: Farmers choose one of thenearest Mandis and transport their products toMandi’s Place.2. Display and inspection: There is no technicalmethod of estimating quality of the products.At first, a buyer inspects a product with eyesand set a random price. Based on that pricethe auction is started and determined the finalprice.3. Auction: After a potential buyer inspectsa product, a commissioner starts the auction.The auction is incermental biding.4. Bagging and Weighing: After the priceis determined famers have to pay the cost ofweighing and bagging. The cost is pay byproduct. That is laborers will take some of theproducts as the cost of the weighing.5. Payment: A trader gives money to a Mandiand Mandi transfer the money to farmers.Mandi pay the money to farmer with delay andin several payments.6. Outbound Logistic: Bags are loaded to endbuyer after auction takes place

5.2 Transaction Costs and MarketFailures in Indian AgriculturalSector

This market su↵ers from four types of mar-ket failures: a) Congestion and Thinnessb)Asymmetric Information c)Unravelingd)Asymmetric bargaining power.Congestion: Costs of searching price of products

and their trends in rural part of India aresubstantial. There is limited access to Internet,TV, Radio, and other communication tools.This implies, farmers face with di�culty infinding best times and the best places (di↵erentMandis and other places) to sell. Hence, farmerssell the products to the nearest Mandi wheneverfruits ripe. Consequently, the market faces withcongestion. That is to say, because farmersdo not have information about prices in otherMandis and other places, all of farmers in avillage sell their products to the nearest Mandiin the same time. In order to ensure that theysell the products they have to stay in queue anight before.Asymmetric Information: Due to lack of ad-vance technology, cost of searching quality ofproducts is high. That is to say, traders detectquality of products by eye, rather than advancetechnology. This means, traders at Mandiplace are not sure about the quality of farmers’products. This implies, there is asymmetricinformation about products between farmersand traders.Unraveling: Because of current inappropriatestructure of the market farmers sell theirproducts to traders without having informationabout price of products in other places. Todaythis is known as unraveling which is costly formof market failure. This is market failure becauseif farmers do not have information about pricesin other places that they can sell their fruitsfor the optimal price. Moreover, information inother places creates the possibility of arbitrage.Asymmetric Bargaining Power: In the currentsystem Mandis pay farmers’ money with longdelay, after several months. Moreover, Mandisand traders do no pay any interest to farmerfor the delay. This implies, there exists anasymmetric bargaining power between farmers

21

and traders. That is, traders and Mandisenjoy higher bargaining power, in comparisonto farmer. As a result traders pay farmers’payment with a long delay. Substantial enforce-ment costs in this market causes asymmetricbargaining power. The Indian judiciary systemis weak and traders are not punished for delayin the payment. That is, it is di�cult for thefarmers to enforce the contract cost.The implication of the above market failuresis that, beside production cost there are othercosts that farmers have to take into account.First of all, they have to bear the congestioncost, that is, staying in a queue a night beforeselling the products. Moreover, there is riskof being late in the selling’s queue and donot selling the products. Secondly, farmershave to take into account cost of asymmetricinformation, that is, the risk of facing withbuyers who estimate products’ quality lowerthan their real quality. Thirdly, farmers haveto consider the opportunity cost of selling tothe Mandis. That is, selling in Mandis place forlower price in comparison to selling in the otherplaces. Finally, there is a substantial delay costin the payment of the products into account.After considering transaction costs into account,a simple-cost benefit analysis suggests that thecost of operating in agricultural sector is by farhigher than what is expected. Consequently,the agriculture sector in India declined. Forinstance, in Khasrod, from 100% of farmersplanting soya only 50% continue farming(Prahalad, 2004, Page 342).

5.3 e-Choupal: Impacts of Semi-Clearinghouse in the AgricultureSector

In the market design economics section it is ex-plained that how clearinghouses overcome sev-eral market failures in di↵erent sectors. Herein,we explain how a semi-clearing house overcomesthe market failures in the agriculture industry inIndia, which is a market of the poor farmer 5.

5.3.1 ITC e-Choupal

ITC is an Indian conglomerate that recentlydiversified into agribusiness and informationtechnology. ITC called its mangers to utilize thenew technology in a way that they overcome tothe problems in the current agriculture market.Consequently, mangers and IT men came upwith e-Choupal.e-Choupal is the a semi-clearingouse andprovides a centralized information centre forfarmers about price of products, its trend, andprice of agricultural inputs. Farmers can decidewhere to sell their products in according tothe information that they have. e-Chouple arein distance of max 5 kilometers to the farmer.It covers more than a million farmers and in11,000 villages and 2000 kiosk across 4 states(Prahalad, 2004, 343). In the following, thevalue chain of ITC is explained. Afterwards,how the clearinghouse solve the market failureis discussed.

5It is called semi-clearinghouse because it containsmany features of clearinghouse but not all of them. It iscentralized information centre, however, it does not applyS&G algorithm to match supply of the products with thedemand.

22

5.3.2 e-Choupal Value Chain

Graph 4: e-Choupal Value Chain

1.Price Setting and Dis-semination: The processtakes place in 4 steps;A- Bench mark Price: Mandi closing price ofprevious day is used for the fair average qualityprice at e-ChoupalB- A commissioner in Mandi send the price listto e-choupleC- Farmers send their products to e-Choupaland Snachalak inspect products and decreasesthe price according to the bench markD- If farmer chooses to sell his produces to ITC,Sanchalak gives him a document that declaresthe quality, village, approximate quantity andconditional price.2.Inbound Logistics:Farmers bring the notes from Sanchalak tonearest ITC hub.3. Inspection and Grading: The inspection thatdetermines a price is inspection that farmerscan evaluate it and understand it. There aretwo inspections: 1) Sanchalak estimations thatdetermines prices 2) Sample of the productssend to a lab. Therein, quality of productsare checked, this test will not a↵ect the priceand it is done to double check the inspectionof Sanchalaks’ estimation. In order to motivatefarmers to produce higher quality ITC givebonus to the farmers that produce higherquality in comparison to the previous products’quality. The bonus give farmers free inputssources.4.Weighing and Payment:Weighing method: In ITC first an entire trolleyis weighted then trolley without the productsis weighted. The di↵erence is the weigh of the

products.Payment: Farmers give the note and get themoney back.A copy of the lab report, agreed rate, andreceipt for farmers’ records will send to thefarmers.5.Logestics and Storage: Farmers transport theproducts to ITC.

5.3.3 Impact of Semi-Clearinghouse inMarket Failure

Introducing semi-clearinghouse in this marketpartly solve 1. Congestion 2. Asymmetric Infor-mation 3. Unraveling 4. Asymmetric BargainingPower.1. Congestion: In the new system farmers com-municate with ITC and appoint an exact time fortransporting the product. Consequently, therewas no problem of congestion and farmers donot need to come a night before and sleep in aqueue.2. Asymmetric information: In this system ITCutilizes new technology to determine the qual-ity of the products. As a result the informationabout the quality of the products reveals for bothsides.3. Unraveling: e-Choupal reduces search costsfor the farmers and they can easily investigatesthe prices and their trends. As a result they selltheir products while they have better informa-tion about the prices in di↵erent places.4. Asymmetric Bargaining Power: ITC is a fa-mous company, hence, it has incentive to keep itsreputation and it avoids rent seeking behaviours(Mueller, 2003, Page 333-350). Consequently,farmers trust ITC easier than other purchaser.Consequently, there is not so much needs ofspecifying details in contract and enforcing con-tracts. The result is contract costs and enforce-

23

ment costs decreases substantially. Moreover, e-Choupal partially solves asymmetric bargainingpower, which is a discriminatory and commonform of market failure.

6 Conclusion

Prahalad notified that extension of market intothe BOP is crucial for poverty eradication. Man-agement science has been dominant in BOP lit-erature. This paper try to utilize economic the-ories to further enrich BOP literature.In this paper it is discussed that, active involve-ment of the poor in the production value chainand approach the poor as consumer is a pow-erful tool for poverty eradication. Institutional-ists rise the point that in the case of high trans-action costs entrepreneurial growth is unlikely.Moreover, companies show reluctant to enter ina market with high transaction costs. There isa common consensus that transaction costs aresubstantial in BOP markets. This implies, it ismore costly for companies to enter into the BOPmarkets in comparison to their counterparts indeveloped countries. Market design economicsis an emerging field in economics that proposessolutions for transaction costs and market fail-ures. This paper provides cases that show clear-inghouses successfully overcome market failures.This article ends by a section that tries to linkmarket design economics and transaction costseconomics with BOP literature. Therein, by anexample, it is suggested that using market de-sign economics’ solutions trigger entrepreneurialengagement in BOP market and contributes tolife of the poor.This paper suggests that market design andtransaction costs can provide robust theoreticalframework to further enrich BOP studies. Con-

necting these two fields can contribute to bothspecialties. In the same token, Roth believesthat the maturity of market design economicsrequires accumulation of experience in severalmarkets. Market design economists have to fur-ther investigate other markets and explain in de-tail complexities, results, and failures of di↵erentmarkets. He suggests that, the next generationof authors should not focus on game theory andabsolute mathematical models, the developmentof this field requires transferring audiences fromscientific journals to societies and participantsof several markets. Furthermore, in spite of thefact that poverty eradication requires multidisci-plinary approach, management view is dominantin BOP literature. Hence, there is a need for fur-ther engagement of the other sciences into thisfield.Long steps have to be taken in order tolabel; an-alyze the implication of market design economicsinto the BOP field. This paper is a small brickfor the bridge that link market design economicsand BOP literature. As Roth said the evolutionof markets are from auction to matching mar-ket. Development of algorithm that can transferBOP market from auction to matching marketmight significantly improves structure of BOPmarkets. Furthermore, market design economicsmaybe able to faster the pace of the poor in-volvement into the production value chain andcustomer.

References

Coase, R. H. (1937). The nature of the firm.economica, 4 (16), 386–405.

de Almeida Faria, A., Hemais, M. W., & Guedes,A. L. M. (2008). Responsabilidade so-cial corporativa: Construindo uma per-

24

Bereby-Meyer, Y., & Roth, A. E. (2006). Thespeed of learning in noisy games: Partialreinforcement and the sustainability of co-operation. The American Economic Re-view , 96 (4), 1029–1042.

Besanko, D., Dranove, D., Shanley, M., & Schae-fer, S. (2000). Economics of strategy. johnwiley&sons. New York .

Bradley, S. W., McMullen, J. S., Artz, K.,& Simiyu, E. M. (2012). Capital isnot enough: Innovation in developingeconomies. Journal of Management Stud-ies, 49 (4), 684–717.

Busch, L.-A., & Horstmann, I. J. (1997). Bar-gaining frictions, bargaining proceduresand implied costs in multiple-issue bar-gaining. Economica, 64 (256), 669–680.

CANON, C., FRIEBEL, G., & SEABRIGHT,P. (2013). Market design and market fail-ure. The Handbook of Rational Choice So-cial Research, 473.

Cees van Beers, P. K. . A. L. (March2013). Understanding frugal innovationin africa: Schumpeter revisited discussionpaper for the international workshop ‘the(mis)fortune of frugal innovation. Greno-ble.

Chataway, J., Hanlin, R., & Kaplinsky, R.(2014). Inclusive innovation: an architec-ture for policy development. Innovationand Development , 4 (1), 33–54.

Christina Gradl, C. K. (2010). endeva (Tech.Rep.). German Federal Ministry for Eco-nomic Cooperation and Development.

Coase, R. H. (1937). The nature of the firm.economica, 4 (16), 386–405.

Coase, R. H. (1960). Problem of social cost, the.JL & econ., 3 , 1.

Dahlman, C. J. (1979). The problem of exter-nality. Journal of law and economics, 141–

162.de Almeida Faria, A., Hemais, M. W., & Guedes,

A. L. M. (2008). Responsabilidade so-cial corporativa: Construindo uma per-spectiva pro-governo. Revista de Admin-istracao FACES Journal , 7 (3).

Ensminger, J. (1994). Transaction costs throughtime: The case of the orma pastoralists ineast africa. Anthropology and InstitutionalEconomics, 69–85.

Erev, I., Ert, E., & Roth, A. E. (2010). Achoice prediction competition for marketentry games: An introduction. Games,1 (2), 117–136.

Ezequiel Reficco, P. M. (2009). Inclusive net-works for building bop markets. SAGEPublication.

George, G., McGahan, A. M., & Prabhu, J.(2012). Innovation for inclusive growth:Towards a theoretical framework and a re-search agenda. Journal of managementstudies, 49 (4), 661–683.

Gewald, L. I. P., Jan-Bart. (2012). Transform-ing innovations in africa: explorative stud-ies on appropriation in african societies(Vol. 11). Brill.

Gordon, R. J. (2012). Is us economic growthover? faltering innovation confronts thesix headwinds (Tech. Rep.). National Bu-reau of Economic Research.

Gradl, C., & Jenkins, B. (2011). Tackling barri-ers to scale: From inclusive business mod-els to inclusive business ecosystems. CSRInitiative, Harvard Kennedy School, Cam-bridge MA.

Hall, J., Matos, S., Sheehan, L., & Silvestre, B.(2012). Entrepreneurship and innovationat the base of the pyramid: a recipe for in-clusive growth or social exclusion? Journalof Management Studies , 49 (4), 785–812.

25

Halme, M., Lindeman, S., & Linna, P. (2012).Innovation for inclusive business: In-trapreneurial bricolage in multinationalcorporations. Journal of ManagementStudies, 49 (4), 743–784.

(2014). Retrieved fromhttp://www.hul.co.in/mediacentre/pressreleases/2014/Lifebuoy-Reduces-Diarrhoea-from-36-perc-to-5-perc-in-Thesgora.aspx

(n.d.). Retrieved fromhttp://www.worldbank.org/en/topic/poverty/overview

Ian Ayres, J. B. (1991). Trpartism: Regulatorycapture and empowerment.

Jackson, M. O. (2013). Economic engineeringand the design of matching markets: Thecontributions of alvin e. roth. The Scan-dinavian Journal of Economics, 115 (3),619–639.

Jacoby, J., Chestnut, R. W., Weigl, K. C., &Fisher, W. (1976). Pre-purchase informa-tion acquisition: Description of a processmethodology, research paradigm, and pi-lot investigation. Advances in consumerresearch, 3 (1), 306–314.

Jenkins, B., Ishikawa, E., Barthes, E., & Gi-acomelli, M. (2008). Supporting en-trepreneurship at the base of the pyramidthrough business linkages. In Internationalfinance corporation, international businessleaders forum, and the csr initiative at theharvard kennedy school.

Kharas, H. (2007). Trend and issues in develop-ment aid. Wolfensohn Center for Develop-ment .

Luo, X., & Bhattacharya, C. B. (2006). Cor-porate social responsibility, customer satis-faction, and market value. Journal of mar-keting , 70 (4), 1–18.

Malla, S., & Timilsina, G. R. (2014). House-hold cooking fuel choice and adoption of

improved cookstoves in developing coun-tries: a review. World Bank Policy Re-search Working Paper(6903).

Milgrom, P., & Roberts, J. (1990). Bargainingcosts, influence costs, and the organizationof economic activity. Perspectives on posi-tive political economy , 57 , 60.

Mitra, S., Mookherjee, D., Torero, M., & Visaria,S. (2013). Asymmetric information andmiddleman margins: An experiment withwest bengal potato farmers (Tech. Rep.).

Mosley, P., et al. (1987). Overseas aid: its de-fence and reform. Wheatsheaf Books.

nations development Programme, U. (2014).Barriers and opportunities at the base ofthe pyramid. United nations developmentProgramme.

Niederle, M., Proctor, D. D., & Roth, A. E.(2008). The gastroenterology fellowshipmatch—the first two years. Gastroenterol-ogy , 135 (2), 344–346.

Niederle, M., Roth, A. E., & Unver, M. U.(2013). Unraveling results from compara-ble demand and supply: An experimentalinvestigation. Games, 4 (2), 243–282.

North, D. C. (1992). Transaction costs, insti-tutions, and economic performance. ICSPress San Francisco, CA.

North, D. C. (1994). Economic performancethrough time. The American economic re-view , 359–368.

North, D. C., & through Time, E. P. (1995).Douglass c. north. Lives of the Laureates,3rd edn. Cambridge, MA: MIT Press.

North, D. C., & Wallis, J. J. (1994). Integratinginstitutional change and technical changein economic history a transaction cost ap-proach. Journal of Institutional and The-oretical Economics (JITE)/Zeitschrift furdie gesamte Staatswissenschaft , 609–624.

26

OECD. (2012). Innovation for development adiscussion of the issues and an overviewof work of the oecd directorate for science,technology and industry (Tech. Rep.). Au-thor.

OECD. (2014). Reviews of innovation policycolombia 2014. OECD .

Pereira, P. (2005). Do lower search costs reduceprices and price dispersion? InformationEconomics and Policy , 17 (1), 61–72.

Pervez, T., Maritz, A., & De Waal, A. (2013).Innovation and social entrepreneurship atthe bottom of the pyramid-a conceptualframework. South African Journal of Eco-nomic and Management Sciences , 16 (5),54–66.

Pesa, I. (2014). Fru-gal’innovation’in’africa’ technol-ogy,’entrepreneurship’and’development’in’innovation’processes’.Centre of Frugal Innovation.

Peter Coles, P. B. L. M. N. A. E. R., John Caw-ley, & Siegfried, J. J. (2010). The job mar-ket for new economists: A market designperspective. Journal of Economic Perspec-tives.

Prahalad, C. K. (2004). The fortune at the bot-tom of the pyramid. Wharton School Pub-lishing.

Programme, U. N. D. (2008). Creating value forall: Strategies for doing business with thepoor. Author.

Rao, B. C. (2013). How disruptive is frugal?Technology in Society , 35 (1), 65–73.

Roth, A. E. (2002). The economist as engi-neer: Game theory, experimentation, andcomputation as tools for design economics.Econometrica, 1341–1378.

Roth, A. E. (2006). Repugnance as a constrainton markets (Tech. Rep.). National Bureauof Economic Research.

Roth, A. E. (2008). What have we learned frommarket design?*. The Economic Journal ,118 (527), 285–310.

Roth, A. E., Prasnikar, V., Okuno-Fujiwara, M.,& Zamir, S. (1991). Bargaining and mar-ket behavior in jerusalem, ljubljana, pitts-burgh, and tokyo: An experimental study.The American Economic Review , 1068–1095.

Simester, D., & Knez, M. (2002). Direct andindirect bargaining costs and the scope ofthe firm*. The Journal of Business, 75 (2),283–304.

Smith, A., Fressoli, M., & Thomas, H. (2014).Grassroots innovation movements: chal-lenges and contributions. Journal ofCleaner Production, 63 , 114–124.

Smith, G. E., Venkatraman, M. P., & Dholakia,R. R. (1999). Diagnosing the search coste↵ect: Waiting time and the moderatingimpact of prior category knowledge. Jour-nal of Economic Psychology , 20 (3), 285–314.

Solaiman, M. (n.d.). Branding in bottom ofpyramid (bop): Case of grameen bank(gb).

Soto, D. (2000). The mystery of capital: Whycapitalism triumphs in the west and failseverywhere else. NewYork:BasicBooks.

Spulber, D. F. (2009). The theory of thefirm: Microeconomics with endogenous en-trepreneurs, firms, markets, and organiza-tions. Cambridge University Press.

Stiglitz, J. E. (2002). Information and thechange in the paradigm in economics.American Economic Review , 460–501.

Teece, D. J. (2010). Forward integration andinnovation: transaction costs and beyond.Journal of Retailing , 86 (3), 277–283.

van Tubergen, K. (2013). Partnering up in

27

base of the pyramid (bop) projects — anoverview of the current academic stancesand research gaps from literature comple-mented with insights from bop practition-ers.

Williamson, O. E. (1979). Transaction-cost eco-nomics: the governance of contractual rela-tions. Journal of law and economics , 233–261.

Williamson, O. E. (1994). Transaction cost eco-nomics and organization theory. Organiza-tion Theory: from Chester Barnard to thepresent and beyong , 207–256.

Williamson, O. E. (1999). Public and private bu-reaucracies: a transaction cost economicsperspective. Journal of Law, Economics,& Organization, 306–342.

Williamson, O. E. (2010). Transaction cost eco-nomics: The natural progression. Journalof Retailing , 86 (3), 215–226.

Wooder, S., & Baker, S. (2012). Extracting keylessons in service innovation. Journal ofProduct Innovation Management , 29 (1),13–20.

Young, A. (2012). The african growth mira-cle (Tech. Rep.). National Bureau of Eco-nomic Research.

Philips. (2013). Unlocking new markets.

28