What's new
		Discussion Paper The Swedish model: Revival after the turbulent 1990s?, by Dominique Anxo, Harald Niklasson Discussion Paper Deepening the Social Dimensions of Regional Integration, by UNU-CRIS Discussion Paper The Influence of the EU on the Evolution of National Employment Models, by Jill Rubery, Gerhard Bosch and Steffen Lehndorff Discussion Paper Harnessing globalization for development: Opportunities and obstacles by Eddy Lee Subscribe to our mailing list and we will keep you informed about our new publications

Business and Society Programme DP/110/2000 ISBN 92-9014-619-2 First published 2000 The New Institutional Economics, Business Associations and Development By Richard F. Doner Emory University Ben Schneider Northwestern University Download PDF Version

Introduction

I. The New Institutional Economics and Development

II. Development Challenges and Institutional Responses: Business Associations

III. Institutional Features of Business Associations

IV. Summary and Conclusions

References

Introduction^{(Endnote 1)}

With the demise of development economics in the 1970s, the academic discipline of economics had little specific theorizing on development to offer practioners and proffered instead universal, liberal nostrums of free trade and free markets (Wing 1990). These universal prescriptions evolved into the first catalogued Washington consensus in the 1980s on the urgency of market-oriented reforms in developing countries (Williamson 1990). In the 1990s, a new connection formed between an emerging institutionalist subfield in economics and the next consensus in Washington after the first generation of market-oriented reforms. The opening of the third annual meetings of the International Society for New Institutional Economics (ISNIE) at World Bank headquarters in Washington, D.C. in September 1999 symbolized this new connection.

The convergence of the New Institutional Economics (NIE) and thinking at multilateral development banks began as disconnected, independent trends. On the side of what is now termed NIE, Ronald Coase, Mancur Olson, Douglass North, and Oliver Williamson, generally viewed as the four original gospels of NIE, were working in the 1960s, 1970s, and 1980s in relative isolation from each other, from mainstream economics, and from the development community. Several things brought them out of isolation and into the mainstream of policy thinking on development. Most visibly, the Nobel committee enshrined the work of Coase and North by Nobel prizes in 1991 and 1993, respectively. In a more institutional sense, the USAID (Agency for International Development) helped Mancur Olson to set up the Center for Institutional Reform and the Informal Sector (IRIS) devoted specifically to studying development problems from an NIE perspective (see Clague 1997a for more background). Presumably too there was some dissemination of NIE through professional channels as students of the founding fathers of NIE found their way into positions in multilateral banks. By the late 1990s, NIE economists were regular invitees at World Bank conferences.

Independent of these theoretical and academic movements, development practioners were grappling with a new generation, sometimes called a second generation, of development challenges created or revealed in large part by the achievements of the first generation of market oriented reforms. These challenges included reinforcing property rights, regulating markets to promote competition, reducing corruption, bolstering political credibility, enhancing the administrative capacity of government agencies to deliver essential social services, and generally creating the institutional infrastructure necessary to make markets work their best. Attention turned from advising states to cease intervening in markets to urging states to provide pro-market institutions. These concerns came across most clearly in flagship publications by the World Bank, from the 1997 World Development Report on The State in a Changing World to the publications of the Latin American group subtitled simply Institutions Matter (Burki and Perry 1998). Publications from the Research group at the Bank also developed NIE themes (see especially Levy and Spiller 1996).

In sum, by the 1990s there was a growing group of theoretically motivated, academic economists interested in comparative, empirical, institutional analysis who had increasing incentives to link up with practically oriented, economists in multilateral development banks who were seeking practical institutional reforms to enhance economic performance. In all, this is a happy, and if anything overdue, convergence. However, the convergence, and potential collaboration, is embryonic and incomplete in many respects, some of which we discuss in this paper.

We might expect the convergence of NIE and practical development thinking to result in attention to a ubiquitous, meso "institution" in developing countries - namely, business associations (also known as employers' associations or trade associations, or business interest associations). From an NIE perspective, business associations are non-state institutions that may or may not reduce transaction costs, promote or restrain competition, extend or restrain rent seeking, and/or lean on the state to protect property rights. In practical terms, then, associations offer at least potentially attractive alternate or supplemental institutions to the weak states so common in developing countries.

Yet neither practitioners nor NIE theorists have paid much attention to business associations. We suspect that biases in the literature on NIE deflect attention away from business associations. First, NIE theorizing about institutions and development is related primarily to overcoming market imperfections and provision of public goods at the macro level, such as regulation, clear property rights, education, judicial services, etc. There is little attention to market failures at the meso or micro level, and the institutional solution is more often the state than collective, non-state institutions like business associations.^{(Endnote 2)}

However, there is nothing in the NIE approach that precludes analysis of business associations, and, as we hope to show, many items in the NIE toolkit are especially apt for understanding these institutions.

The second bias in NIE may be more difficult to overcome, namely, the paradigmatic negative presumption against "special interest" groups. This presumption is inherent in Mancur Olson's work on associations and collective action, and it is reinforced in NIE works, especially North (1990), that identify institutions only as formal and informal rules, such as constitutions, legal systems, and norms, and their enforcement mechanisms.^{(Endnote 3)}

This major if not dominant strain of NIE does not view organizations (such as firms and associations) as institutions, because organizations presumably embody particular interests.^{(Endnote 4)}

The point of departure, then, in much NIE work on business associations is that they represent interests and are thus part of the problem requiring some institutional reform in or by the state. In one of his last publications, specifically on the topic of collective action and development, Olson claimed that the incentives facing special interest groups were the same as those for bandits (1997). In this view, associations are hardly promising candidates for agents of development.

There are serious problems with this perspective. First, it leads to confusion even within the NIE. Organizations are hierarchies, a key part component of Williamson's work, and they are also mechanisms for the enforcement of rules and norms.^{(Endnote 5)}

Second, as mechanisms of enforcement and governance of interdependence, associations affect firm behavior and thus the aggregation of firm interests. None of this to deny that associations may simply reflect their members' (often narrow) interests. But they may not. Indeed, they may pursue a range of interests simultaneously, some narrow, some broad. And even if they are narrowly focused, under certain conditions the pursuit of even narrow interests may yield broader collective benefits.^{(Endnote 6)} In sum, the definitional exclusion of business associations from the realm of institutions leads both to unwarranted assumptions about the nature of their activities and to a lack of attention to their potential benefits.

Our overall goal is thus to fill in some of the missing analysis on business associations within NIE by showing on the one hand, how this literature speaks directly to development challenges and institutional responses, and on the other, how it limits its own explanatory and descriptive reach. We proceed in four steps. Part I is a preliminary effort to define the New Institutional Economics by identifying assumptions and foci common to those working in this field. The goal in section II is to show how recent literature on New Institutional Economics highlights concrete development problems, and then how business associations in some cases constitute or contribute institutional solutions to those problems. In section III we show how NIE can help to identify both problems and potential solutions to the internal capacities of productive associations. Business associations themselves become institutional phenomena to be analyzed.

I. The New Institutional Economics and Development

Any effort to apply the New Institutional Economics to concrete challenges of development must begin by acknowledging that it is, as the editors of a recent volume on The Frontiers of the New Institutional Economics note, "an ambiguous term" (Drobak and Nye 1997: xv). It is not so much a clear school of thought as an "expanded neoclassical economics" (Clague 1997c: 16). It is neoclassical because it focuses on choice, because it begins with an appreciation of the power of neoclassical price theory, and because it has tended to be neo-utilitarian in that it views institutions as largely derivative of interests (Evans 1995: 33). It is "expanded" because its adherents are willing to relax core neoclassical assumptions about perfect information and rationality, and thus perfect markets. It is on this basis that the NIE recognizes the potential benefits of institutions for overcoming the consequences of imperfect information and bounded rationality.^{(Endnote 7)} NIE also elaborates on assumptions of utility maximization by bringing in strategic calculations from game theory; individuals are not just self-regarding, but opportunistic or "self-interest seeking with guile" (Williamson 1996: 327).

Table 1. Core Analytic Components of the New Institutional Economics in Several Recent Overviews
Drobak and Nye (1997)	Clague (1997)	Burki and Perry 1998 (World Bank)
transaction costs	transaction-cost economics	transaction costs
property rights	property rights	property rights, contracts and enforcement
political economy / public choice	collective action	collective action and public goods
quantitative economic history	economics of imperfect information	asymmetric information and principal - agent problems
ideology and path dependence	institutional innovation and efficiency	institutions as sources of predictability and credibility
	cooperation and norms	exit and voice

Even among practioners, there is no consensus on the precise components of the NIE. As seen in Table 1, authors differ on what Clague terms the NIE's constituent "strands of literature" (1997c).^{(Endnote 8)} However, nearly all attempts to characterize the field note three core sets of problems: transaction costs, principal-agent problems, and collective action dilemmas. In the rest of this section we review these three problem areas and the NIE's views on institutional solutions (Table 2). Unlike much of the NIE, we follow Williamson and treat institutions as both as rules and as organizations.^{(Endnote 9)} Williamson's definition keeps associations in the realm of researchable and variable organizations (as opposed to North's approach which relegates firms to the realm of easily predicted interest organizations). Our list of core components attempts to organize the field of NIE by distilling out three fundamentally different relationships that have attracted the most attention: exchange (transaction costs), control and hierarchy (principal-agent), and cooperation (collective action). Of course, many of the same issues such as information costs and opportunism are common concerns in theorizing about these relationships.

Transaction Costs: New institutionalists emphasize the potential costliness of transactions. This line of analysis began with Ronald Coase's 1937 article posing the deceptively simple but powerful question of why firms exist: Why do certain transactions take place not in arms-length markets but within single firms? Subsequently, Oliver Williamson has devoted much of his career to specifying the costs of transactions, the factors that increase these costs, and the institutional responses to such costs (1975, 1985) generating along with other scholars the subfield of transaction cost economics (TCE). Transactions involve costs because they typically require agents to search for partners, to bargain with partners, and to enforce and to monitor agreements among partners. These costs, according to Williamson, increase with the frequency of the transactions, the specificity of the assets involved, and the uncertainty of the context. The potentially steep costs of information are key to issues of transaction costs. It is these transaction costs that generate market imperfections and contribute to market failures (public goods, externalities), and generate demand for institutions to redress them. The most commonly emphasized institutions in the TCE literature are markets and hierarchies. This kind of specification of transactions, their costs, and institutional responses moves us far from the neoclassical ideal of spot-market trading.

Table 2. Types of Analysis in the New Institutional Economics
	Transaction Costs	Principal/Agent	Collective Action
Relationship	Exchange	Hierarchy and control	Cooperation
Components / Analytic Tools	Information costs; costs of searching, bargaining, and enforcement	Asymmetric information and opportunism	Free riding / Selective Incentives
Manifestations in the private sphere	imperfect markets market failures (externalities, public goods)	suboptimal contracts (e.g., sharecropping) distortion of corporate governance adverse selection and moral hazard	market failure under provision of public goods (e.g., training) tragedy of the commons
Manifestations in the Public Sphere	insecure property rights political instability non-credible policies and time inconsistencies political transaction costs, veto points	bureaucratic dysfunction and expansion corruption regulatory capture other elements in public choice	rent seeking undersupply of public goods clientelist bureaucracy

Coase and Williamson analyzed varying conditions in which firms (or hierarchies) in the private sector could reduce the costs of transacting. Douglass North and others have focused special attention on public rules or institutions, at the macro level, that reduce or exacerbate transaction costs, in particular the strength of property rights in law and custom (see Drobak and Nye 1997 for a recent compendium). Other scholars have extended this analysis to examine political stability and policy credibility (Borner, Brunetti, and Weder 1995 and World Bank 1997). When low, both of these factors increase uncertainty and transaction costs, with negative consequences for overall economic activity.^{(Endnote 10)}

Principal-Agent Problems: A second strand of NIE focuses less on information imperfections than on information asymmetries, and especially how opportunistic agents can use them to confound their principals. This focus was initially stimulated by Stiglitz's reflections on developing country problems and expanded into research on optimal contracts, especially in the context of sharecropping.^{(Endnote 11)}

This approach has also shed light on problems of corporate governance, especially transparency. Principal-agent problems plague the modern corporation where ownership and management are separated, and where, as a result, shareholders (the principals) have difficulties controlling managers (their agents), and where managers exploit information asymmetries to further their interests in appropriating the firm's surplus.^{(Endnote 12)}

In response, governance reformers advocate solutions such as cumulative voting rules, insider trading penalties, and shareholders' rights to sue directors for fiduciary negligence.

Theories of principal/agent asymmetries have also been used to analyze problems in the public realm. Principal/agent analysis is thus a building block of public choice analysis, especially in the works of Niskanen (1971) and Tullock (1965). The core contention in this perspective is that bureaucrats (agents) seek to maximize their budgets and prerogatives at the expense of the goals set by legislatures (principals). Problems of principal/agent have special application to regulatory bodies where the objects of regulation attempt to capture the agency, or wrest control from the principal (variously conceived as the public at large or their legislative representatives). PA analysis is relevant to the study of corruption in the bureaucracy, to the extent that principals (again either hierarchical superiors or voters) do not endorse or encourage this behavior. (Of course, corruption is sometimes planned from the center and then agents comply in executing the plan). Solutions include clearer performance incentives and more effective monitoring and feedback mechanisms that make outcomes more measurable and transparent (Klitgaard 1997; World Bank 1998).

Collective Action: Dilemmas of collective action are central to most analyses of NIE. The advancement of this line of reasoning has been most closely associated with Mancur Olson and his two books, The Logic of Collective Action (1965) and The Rise and Decline of Nations (1982). Olson's essential insight was that common interests within a group would not lead to collective action to further those interests if some members had the opportunity to free ride on the efforts of others to provide the collective goods. Individual rationality would lead to collectively suboptimal outcomes. In the private economy the negative consequences would take the form of market failures and the under-provision of public or collective goods that would otherwise make potential members of the collective better off. Training, for example, is a (positive) externality problem in which firms have incentives not to train workers who could then be hired away by free-riding firms that have not invested in training (Booth and Snower 1996). With low investment in training all the firms are worse off.

Elinor Ostrom's work has tackled another specific kind of collective action problem that results in the "tragedy of the commons." Where goods are non-excludable and yet rival, as in common pastures, fishing areas or irrigation schemes, these goods take the form of "common pool resources" and the collective problem is not free riding but rather over exploitation and insufficient investment in maintaining the resource.^{(Endnote 13)}

NIE research on these kinds of issues has explored a range of institutional solutions, including informal institutions such as social capital, collective forms of property rights, and clear performance-linked incentives (Ostrom 1997; 1999).

Lastly, the dilemmas of collective action have come to figure more prominently in the analysis of policy decisions, especially reforms with redistributive consequences. As in the private sector version, there are many outcomes (public goods) which would benefit the entire collectivity, in this case all citizens. But these may not be undertaken by governments because reform benefits are dispersed and those favoring reform cannot overcome free riding to pressure collectively the government, while the status quo provides immediate benefits that facilitates collective action by those opposed to reform. This is the stylized view of pre-reform economies where the beneficiaries of the old order are well organized and the potential beneficiaries of reform are not.^{(Endnote 14)} In other analyses using the paradigm of collective action, the problems reside within the government itself. Barbara Geddes (1994) argues that it may be in the interests of voters and politicians alike to reform clientelist bureaucracies but legislators cannot overcome their individual incentives to exploit their patronage resources, even if it means collective losses for their party as a whole.

In sum, NIE helps to identify a series of obstacles, problems, imperfections, and failures, both in states and in markets that can or should be remedied by various institutional means. The institutional solutions for public sector pathologies tend toward small, clean governments, strong property rights, and honest, efficient judicial systems. In the private sector, the NIE emphasis is on efficient, transparent corporate governance. However, the generic problems covered in this section translate into more particular sets of challenges in the context of poor countries and late development. In the next section we consider some of these specific challenges and review the role of business associations in resolving them. We argue that the NIE is helpful in shedding light on the nature of these problems and on potential institutional solutions. But we also suggest that the NIE's emphasis on market imperfections, its antipathy to organized private interests, and its theoretical and invidious distinction between institutions and organizations, lead it to neglect the ways in which some business associations (but certainly not all) help resolve various development challenges.

II. Development Challenges and Institutional Responses: Business Associations

The preceding section has shown how relaxing strict neoclassical assumptions regarding perfect information and rationality sheds light on three sets of problems commonly encountered by both those in both public and private sectors: transaction costs, principal/agent problems, and collective action dilemmas. Our challenge now is to explore whether and how these more generic problems help us to understand the nature and potential institutional solutions of concrete development challenges. The list of such challenges is of course quite long. In this section we focus on a limited number of issues: property rights, administrative reform, macroeconomic stabilization, trade tensions between upstream and downstream sectors, and worker training.^{(Endnote 15)} Although each of these issues involves both public and private sector interests, we have listed them in order of their relevance for change in the public vs. private sector. For each case, we ask two general questions: First, does recent work the NIE framework help to clarify the nature of the development challenge? Second, does NIE theorizing help to account for associational contributions to solving these development problems? We explore this second question by examining empirically how some associations have addressed these problems.

We begin with issues of public sector performance since these have been a central concern of the New Institutional Economics. In Institutions Matter researchers at the World Bank emphasize the importance for economic growth of the rule of law, contract credibility, transparency, predictability of rule-making and enforcement, etc. The institutional mechanisms through which goods and services are provided are assumed to be almost uniquely public or, in the case of common pool resources, collective (Burki and Perry 1998: 17-18; Clague 1997a). The role of the public sector in reforming itself is certainly key, but non-state actors and institutions have had crucial roles in pushing reform, providing complementary services, and/or substituting for government. Below we review two sets of issues seen by the NIE as important- property rights and administrative reform - and the role of business associations in these areas.

Property Rights: Despite other differences all the major NIE theorists agree on the importance of efficient property rights systems, including contract enforcement, for reducing the costs of doing business and thus economic efficiency. Without the ability to hold, benefit from and transfer resources, and without the ability to enforce agreements, producers risk the loss of assets. Such a risk increases with specialization and involvement in complex exchanges. Yet specialization and complex exchanges is the key to technical change and innovation (Clague 1997c: 69).

Insecure property rights are, of course, one of the key characteristics of developing economies. Indeed, precisely because expropriation and other threats are issues that cut across industry and sectoral cleavages, defense of property rights has been a focus of a number of developing country associations. And even the defense of property rights undertaken for narrow sectoral, local or ethnic group interests has benefited other property holders. This is reflected in Nigerian associations efforts to defend members from predatory government measures (Lucas 1993). But the NIE has also acknowledged the potential benefits of a variety of property rights structures. It can thus encompass the successful efforts of ethnic-Chinese dominated business associations to establish joint ventures with foreign investors as de facto protection of their own property rights (Hewison 1989). NIE-inspired research has also begun to explore the role of associations in contract enforcement in developing countries. McMillan and Woodruff have found that in both Vietnam and Eastern Europe, trade associations are not only an important source of information on trading partners but frequently also supplement courts in facilitating dispute resolution and arbitration (1999).^{(Endnote 16)}

Several other issues related to property rights have only recently come under the purview of the NIE.^{(Endnote 17)} The most recent (and important for our purposes) involves corporate governance, an issue that has taken on added significance following the Asian economic crisis. Corporate governance refers to the responsibility of directors and managers to other corporate stakeholders (shareholders, creditors, employees, and consumers). In principle, such responsibility is guaranteed by a series of institutionalized rules, including protection of shareholders rights (especially small shareholders), access to accurate and timely information, and the independence of the corporate board.^{(Endnote 18)} These issues seem eminently susceptible to analysis using elements from the NIE toolkit (principal/agent relations, information asymmetries, and transaction and information costs) yet banking crises in Asia and other developing countries revealed how incipient our understanding of corporate governance in developing countries is.^{(Endnote 19)}

Public Sector Administrative Reform: The need for such reform is straightforward, and associations have often played active roles in efforts to reduce corruption and improve transparency. For example, a primary focus of Nigerian associations has been opposition to "corrupt behavior, inefficiency and the politicization of administration" (Lucas 1993). Yet, associational efforts are typically not broadly focused on overall administrative reform or citizens' rights, probably because the benefits of such efforts are fairly dispersed. Instead, one finds reform efforts focused on more

specific areas. Thus, associations in both Kuwait and Thailand were active in pressing for customs reforms during periods of debt and thus increased export pressures (Moore 1998 and Anek 1992). In Latin America, rather than fighting for reform of judicial systems, some associations have been active in developing private arrangements known as Alternative Dispute Resolutions (ADR) systems.^{(Endnote 20)}

However, despite these cases and the potential for similar associational contributions, the NIE literature largely neglects associations in its discussions of public sector reform. The volume Institutions and Economic Development, for example, has four chapters devoted to improving governmental performance. None mentions private sector institutions.^{(Endnote 21)} This is largely a reflection of the chapters' focus on the supply of institutional innovation through, for example, improved incentive schemes, and an almost total lack of attention to domestic demand for institutional change. That focus may itself reflect the assumption that entrepreneurs in most developing countries are satisfied with weak public sector performance and/or incapable of sufficient organization to press for reform.

Macroeconomic Stabilization: One important component of the economic reform challenge to developing countries involves the reduction of inflation. This typically involves fiscal restraint and price stabilization, often in a context of indexation, as occurred in Brazil. The NIE literature is helpful in portraying this stabilization effort as an implicit contract between government commitment to keep spending down and private sector commitment to moderate price increases. The NIE's emphasis on collective action and credible commitment is helpful in explaining the difficulty of implementing these kinds of reforms. Stabilizing prices, for example, constitutes a collective action problem. The problem involves free riding, as reflected in Brazil's experience: Getting everyone to accept deindexation will be difficult if some believe that others are not making similar sacrifices by keeping prices down (Haggard 1997: 123). Problems of credibility due to time inconsistency may undermine government commitments to fiscal restraint. Governments typically have histories of failure to implement commitments to fiscal restraint which might harm important political constituencies. The resulting credibility problem can be exacerbated by impending elections and/or principal-agent problems involving capture of an important part of the bureaucracy responsible for implementing fiscal policy. Lack of government credibility may also derive from the high costs involved in actually gaining access to information about government spending and revenues. In this context, the private sector is justifiably leery of government commitments to fiscal restraint and hesitant to complete its part of the stabilization contract by keeping prices down (and resisting demands for higher wages).

Under certain conditions, business associations have helped to address these problems by reducing the costs of agreeing to and monitoring macroeconomic bargains or pacts. In Mexico, for example in late 1987, as many macro indicators spun out of control, representatives of business, government, and labor met and signed the first of a series of agreements on wages, prices (public and private), exchange rates, and government spending that brought inflation rapidly down from over 100 per cent to manageable levels within a year. Two specific associational activities were especially important to stabilization: coordination of intra-sectoral or intra-industry price differences, and monitoring and enforcement by the retailers' association of the prices of member firms (Kaufman et al. 1994, Schneider 1997, and Biddle and Milor 1998).

By reducing coordination and monitoring costs, these associational activities are consistent with the NIE concerns and prescriptions. However, we find little if any mention of business associations as potential institutional mechanisms for performing these functions. In fact, a common assumption is that macroeconomic stabilization, unlike sustained economic growth, is better served by a centralized executive authority unconstrained by other institutions and unresponsive to interests organized by associations (World Bank 1998; Clague 1997b: 3). While certainly valid for the initiation of reform, this perspective ignores research demonstrating the important role of private sector interests, including associations, in the consolidation of economic reform (Haggard 1997; Haggard and Webb 1994).

Trade Conflicts between Upstream/Downstream Producers: Developing countries have traditionally attempted to promote linkages and to increase local value added. Many see industrial upgrading as moving beyond enclave and "disarticulated" economies in which firms export goods - whether raw materials or manufactured goods - with little local value added. In the case of raw materials such as agricultural commodities, increasing value added comes from locally based downstream processing. In the case of manufactured goods, increasing value added comes in part from better-trained workers, but also from increased local inputs, especially intermediate goods and capital equipment. There are, however, at least short-term and often serious distributional differences between upstream and downstream firms. In agriculture, upstream firms such as sugar planters want higher prices from downstream sugar millers. These kinds of tensions are common in developing countries such as Thailand, where prawn farmers have fought with cold storage operators and exporters, soybean farmers have fought with soybean refiners, soybean refiners have fought with agribusiness users of feedstuffs, and sugar growers have fought with sugar millers ("Governments to Intervene;" Handley 1993; Ramsay 1987).

These problems are especially intense in the manufacturing sector. The problem, labeled "the industrial policy paradox" by Mark Elder, is the following (1997): The promotion of upstream products typically involves trade protection for and/or encouragement of collusion among the producers of such goods. Because these goods are inputs for downstream users, their higher prices result in higher cost burdens for downstream exporters whose final products must be competitive in global markets. Virtually all developing countries attempting to promote higher value added exports, following at least some period of import substitution, have encountered this "industrial policy paradox."

The Thai case is illustrative: Following steep declines in labor-intensive exports due to rising wage costs relative to neighbors such as China and Indonesia, Thai policy makers and exporters of garments, auto parts and electronics have attempted to reduce the costs of their manufactured exports by pushing hard for tariff reductions on upstream inputs such as petrochemicals and steel (Wichit 1997; Fairclough 1994). The process has been slow largely because of opposition from upstream producers who, following past incentives, invested in capital-intensive facilities and who now stand to reap a much lower return if their products were liberalized.

For mainstream economics liberalization of upstream industries is good. From an NIE perspective the resulting upstream-downstream tension is a distributional conflict which raises collective action and transaction costs challenges.^{(Endnote 22)} The collective action aspect has to do with the fact that the beneficiaries of reform are likely to be more dispersed, their political voice thus weaker, whereas those benefiting from protection are likely to be more concentrated, their political voice stronger.^{(Endnote 23)} The transaction costs challenge involves the costs of mobilizing reform winners and of arranging for compensation of reform losers. But in some cases, where value chains are highly complex, it also involves reducing the costs of simply figuring out how to assess tariffs on products that constitute both raw materials and finished products.

In several cases, associations have helped to resolve precisely these kinds of problems, consistent with the NIE's focus. In Taiwan's successful athletic shoe industry, associations of downstream footwear producers played critical roles in unifying the interests of small producers to negotiate with upstream input suppliers (Cheng 1999). Some associations have also been active in agricultural commodities: In the Thai sugar case, after years of conflict over the price of raw sugar, associations of both millers and growers, prodded by state officials, brokered a price formula acceptable to both parties that avoided disruptions of critical sugar exports (Ramsay 1987). And the Thai National Food Institute has been engaged in helping the government to classify for tariff purposes products such as shrimp. For food processors, shrimp is a raw material and should thus have low tariffs; for shrimp growers, it's a finished product and should have high protection (author interview, Bangkok, July 1999).

Most analyses not only fail to consider such associational contributions, they also ignore the possibility that associations of downstream exporters might in fact favor protection of upstream inputs. Yet this is precisely what occurred for some industries in Japan and the East Asian NICs through a combination of compensation for downstream exporters and measures to ensure that upstream firms used policy benefits to improve their own efficiency. In Taiwan, for example, "the weaving industry supported the liberalization of yarn imports while the spinning industry resisted it." The associations worked out a compromise by which the (upstream) spinners association agreed to lower prices and stabilize supplies of yarn in exchange for the (downstream) weavers abandoning the liberalization proposal (Kuo 1995: Elder 1997).

Employee Training: The quality of human resources is becoming a key influence on national competitive advantage. Efforts to provide the training necessary for such skilling typically encounter problems central to the concerns of NIE. The most critical is of course the collective action dilemma due to free riding. Other things being equal, employers want access to a plentiful supply of workers with strong skills. Individual employers are hesitant to train or contribute to training because of the risk that others will benefit from their investment. This can occur in the context of externalities and/or public goods. In either case, firms, fearing free riding by others, will not contribute to training that would benefit each and all.

Associations have addressed this problem in different countries and sectors. In Brazil's Sinos Valley, local producers and suppliers cooperated in an industry-wide association to set up technical training institutions and a technology center that enhanced the local industry's capacity to respond to export opportunities in the early 1970s (Nadvi and Schmitz 1994: 27). In Singapore and Penang, the International Disk Drive Equipment and Materials Association has initiated and implemented a training program designed to provide a "certificate of competence in storage technology." In Thailand, the Garment Manufacturers' Association examined Hong Kong's training facilities and then established its own program to train sewing machine operators (Author interviews, Thai Garment Manufacturers' Association, July 1999). These associational activities address the collective action and transaction costs problems highlighted by the NIE.

The collective design of training programs may encounter further problems susceptible to analysis with NIE tools. For example, there is the possibility that collective training programs may be out of sync with employers' real and changing needs. There is an additional question of how relevant broad collective efforts are for more specific aspects of training. It is unclear, for example, whether a business association that is successful in promoting extensive quantities of training is also going to be successful at ensuring that such training is transferable, i.e. that can be used by competitors in the industry (Johansen 1999). Addressing both of these questions might require better specification of transaction costs, especially the information and bargaining costs of developing vocational curricula that are in fact consistent with firms' needs.

III. Institutional Features of Business Associations

The preceding section emphasized positive responses by business associations to developmental challenges. Such responses are far from pervasive. Indeed, developing countries are full of associations that consist of little more than a telephone listing, a president, an office, a secretary, and a small annual meeting. These associations are generally of little help in the institution-intensive (or contract intensive) challenges of development.^{(Endnote 24)} Lacking not only staff and funds, but also information about their own members, they are unable to broker the kinds of binding agreements among members Associations necessary to pursue what Schmitter and Streeck have termed the two "logics" of associational objectives. For the "logic of membership," associations must "offer sufficient incentives to their members to extract from them adequate resources to ensure their survival, if not growth." For the "logic of influence," associations "exercise adequate influence over public authorities (or conflicting class associations)..." (Schmitter and Streeck 1999: 19). What kinds of institutional attributes characterize associations capable of mobilizing these kinds of resources? We use an NIE lens to examine four conditions that seem necessary and sufficient conditions for high institutional capacity in business associations: 1) valuable selective benefits; 2) high member density; 3) effective internal intermediation among members; and 4) balanced relations between an association's members and its staff (Doner and Schneider 1999).

Selective Incentives. The importance of selective benefits was of course a key insight from Olson's early work. Olson's Logic of Collective Action is most commonly remembered for predictions that small, homogeneous groups would be best able to overcome free riding and engage in collective action (although their very narrowness translated this capacity into more distributive than productive efforts). Conversely, large associations, he argued, could only organize "artificially" by providing selective, excludable benefits that compensate members for their investment in collective action. But this perspective obscures the range, the value, and the origins of potential associational outputs. As such, it unnecessarily limits our assumption as to the potential benefits of associational activities.

Olson argued that once selective benefits had engaged collective action, then activities like lobbying were "byproducts" and not therefore necessarily the collective goods desired most by some majority of all members (since the selective benefits were sufficient to keep members from defecting to free ride). Much of this "play" in the original Olson between member interests and association activities faded in subsequent elaborations. It is important because it raises the possibility, confirmed by subsequent work on collective action (Lichbach and Broz 1999), that associations pursue varying combinations of very different goods: public (non-rival, non-excludable), collective or club (non-rival, excludable), and private goods (rival and excludable, as in selective benefits).^{(Endnote 25)}

Also important in this regard is the relative value of selective benefits. In many "artificial" associations the selective benefits were not very valuable (discounted life insurance, for example), nor was the corresponding investment in collective action. But for some associations, investments of money and time are substantial (especially considering the opportunity costs of time for business executives). For instance, the Colombian government created an export tax and delegated its use for the good of Colombian coffee to an association created to administer these funds and the association, Federacafe, built transportation infrastructure, marketing departments, and invested in nearly every aspect of improving coffee production. Associations in Taiwan and Turkey distributed export quotas (Biddle and Milor 1997).^{(Endnote 26)}

These are examples of substantial selective benefits that elicited correspondingly great member investments in collective action. A key type of "investment" or commitment characteristic of strong associations is the provision by members of sensitive information about members' own activities (e.g., cost structure and production organization). Such information is critical to collective efforts not only in industry rationalization but also in technological improvement through benchmarking. In sum, an important point for NIE is that there exists a much wider range in the value of selective benefits, in types of goods, and consequently in investments in collective action than is contemplated in most theoretical treatments. This potential "basket" of goods is in turn significant for its ability to elicit different kinds and levels of collective action among members.

A final point is that this variation in value may correspond to whether the benefits are internally generated or delegated by the state (Doner and Schneider 1999). There appears to be a fairly low ceiling of benefits that an association can provide without government assistance. These benefits are usually joint marketing (publishing of directories of members, hosting trade fairs, representing the sector at other trade fairs), and the joint purchase of uniform goods and services. The services usually include commodity services like insurance.^{(Endnote 27)}

Density. High member density (in terms of proportion of potential output organized) is necessary to lend authority to the association's decisions and to deter exit.^{(Endnote 28)}

High member density (80 per cent or more of output is not uncommon) confers a monopoly of representation on the association. If two or more associations compete to organize the same members, density will be low, minorities tend to take control, and none will be able to speak for the sector as a whole. Some of the weakest business association in the Americas, in terms of institutional capacity, are found in Argentina and the United States, two countries where business associations have historically been multiple, overlapping, and competitive. Maintaining high density is partly a function of selective benefits (to attract members) and partly a function of effective intermediation of members' interests (to keep members from quitting).

One can also assess density in terms not just of membership but of actual participation by members. Indeed, the decision to join an association is only one, and often a minor, form of collective action. Once individuals or firms join an association, they are continually faced with further options for collective action: whether to go to meetings, run for office, vote in elections, provide full information on the member's firm, voluntarily contribute additional material resources, and subject their firm's behavior to collective decisions. Many of these decisions may follow a simple logic of relative compensation in the form of selective or other goods. However, reaching consensus positions within the associations and making agreements that restrict the behavior of members require much more for success, including many of the items identified by NIE such as enforceable agreements, credible commitments and monitoring. And these in turn would seem to vary with the nature of intermediation within the association.

Intermediation Among Members: Effective intermediation within an association requires: proportional representation, transparency, and frequent interaction among members. These dynamics are themselves influenced by relations among members, elected leaders and permanent staff.

Some degree of proportional voting (by sales, dues paid, number of employees) is important to keep big firms from defecting. If large firms are outvoted by a majority of small firms they are likely to defect, if not formally resigning, then by participating in other associations and by withholding active participation. Conversely, special minority rights for smaller firms, or cross subsidies, may be important to keep some reasonable number of small firms in. Although NIE-inspired work has begun to pay attention to issues of interest intermediation, the focus has been on the broader political system, not business associations (Burki and Perry 1998: 31). Transparency is a related issue that is closer to core concerns of NIE. Relative internal transparency in the distribution of association resources and the decision making process of joint commitments can be restated as an effort by the association to reduce information and especially monitoring costs.^{(Endnote 29)}

In principal, proportional voting, transparency, and thus collective action are easier to achieve under the conditions emphasized by Olson - namely, small numbers of members and homogeneous interests. In addition to these factors, and independent of them, repeated interaction increases the ability of the cooperating partners to commit to new and future agreements. As game theory predicts, reiterated interactions lead participants to discount the future less heavily and to cooperate under more varied circumstances. In terms of bargains, members may be willing to forego something now in the rational expectation that they may be able to recoup the sacrifice in a later round of negotiations. Repeated interaction lowers the semi-contractual transaction costs of reaching an agreement among members by reducing information and monitoring costs.

Staff-Member Relations: An association's capacity for intermediation is not simply a function of horizontal relations among members. It is itself influenced by the relationship between members and the permanent association staff. Principal/agent issues thus emerge, especially in large, well-staffed associations with professional managers. In these larger associations, the professional management has incentives to use association resources for their own purposes. The situation is analogous to shareholders and managers in a corporation. Both the shareholders and the association members have infrequent contact with their respective staff and limited knowledge of its activities. And, as in some corporations, association staffs may hijack the association from their principals (Moore and Hamalai 1993). Information asymmetries can be reduced by repeated interaction (in monthly or weekly meetings at the association, in participation on special task forces within the association, in the exercise of leadership roles, or association sponsored trips, for example). The learning curve is presumably very steep as a member goes from devoting, say, one hour a month to devoting five or ten.

Yet, some autonomy of staff from members (agents from principals) is beneficial to the association in its efforts to respond to the common interests of the whole membership and to think proactively (Schmitter 1994). Honest, incorruptible staff, who can maintain themselves above the conflict among members, enhance the ability of the association to collect and process sensitive information from member firms. If members do not trust that staff to keep information from their competitors, the association will not have access to good information. Staff members may also have greater legitimacy speaking for the membership because they are presumed not to be favoring the interests of a particular firm or segment, as would be the presumption of the positioning by any one member of the association. In this respect the analogy to the relative autonomy of the capitalist state or the insulation of the Weberian bureaucracy is more appropriate (Evans 1995). To the extent that the association has significant benefits, the staff distributing such resources themselves become targets for rent seeking by member firms. Some autonomy from individual firms is useful to staff members who are trying to formulate association positions and activities in ways that promote the collective interests of members.

The "optimal" degree and nature of staff autonomy is of course difficult to specify. But we can suggest at least one consideration - namely, the specific nature of the collective action problem and their attendant monitoring or enforcement requirements. Collective action problems can be divided into at least two different "games," each with different enforcement costs (Noble 1998). Efforts to reduce capacity, to set wages, to invest in upgrading, or to train workers constitute "prisoners' dilemma" games in which there is constant incentive for members to free ride. These kinds of problems involve heavy enforcement costs and thus more active roles by associational staff. Conversely, once members agree on various types of product standards (e.g. Betamax vs. VHS), they have no possibility to free ride and little incentive to cheat. In such "coordination" games, enforcement costs are low and associational involvement is less important. Even here, however, there is variation: Difficult-to-measure quality standards, as in agricultural exports, are still subject to incentives for members to cheat and claim their products are higher quality goods. And process standards, such as ISO or HACCP (in food production), may be too costly for some members, thus undermining the reputation of the whole industry. Yet these kinds of standards are based on firm-specific information that members will only provide if their involvement is extensive enough to generate trust in the associational staff.^{(Endnote 30)}

IV. Summary and Conclusions

Despite the relative scarcity of NIE-inspired research on business associations, we found that several strands of analysis in the NIE can help both to specify more precisely the potential developmental contributions of business associations can make and to illuminate the internal dynamics of these institutions. In order to assess the potential analytic contributions to the study of business associations, it was first necessary to attempt to categorize the varied kinds of theories that are usually grouped together under the amorphous label of New Institutional Economics. Our approach in Section I was to divide the field according to the three primary kinds of relationships (exchange, hierarchy, and cooperation) and their related sub-fields of NIE: transaction costs, principals and agents, and collective action. We argued that these three components are all useful in identifying difficult development challenges, particularly those challenges whose potential solutions are "institution intensive," as many more are at the end of the 20^th century. In Section II we attempted to show how associations can supply some of the institutional solutions that help to lower the transaction costs, mitigate the principal-agent problems, and reduce the obstacles to collective action that are inherent in many contemporary development challenges. In Section III we turned to the internal characteristics of associations, suggesting that the NIE could provide a useful lens into the features characterizing strong associations: valuable selective incentives, high density, effective interest intermediation, and balanced relations between members and staff.

We also noted that despite its relevance for the study of associations, the NIE has largely neglected the potential contributions of associations. One reason is the failure of most NIE literature to go beyond market imperfections. Redressing this problem will require researchers to explore the variety of market failure problems that plague developing countries attempting not only to diversify their economies but also to raise the value added by their industries. Raising value added is in part a function of exposing local firms to market pressure. But, as research on endogenous growth has shown, it is also a function of absorbing and diffusing technologies that, unlike rules for corporate governance, are tacit and non-codifiable. Such diffusion typically involves more meso-level institutions, potentially including business associations.^{(Endnote 31)}

A second problem is the common definition in Northian NIE of institutions as rules and norms, as opposed to organizations as interests. In our judgment, the view of institutions as both rules and organizations developed in the political science and "governance" literatures is of greater utility (Pempel 1999: 225, fn. 21; Hollingsworth and Boyer 1997). Such a view leaves open the question of whether private interest-driven institutions can yield collective benefits. It also encourages an investigation of the internal features that can help us to begin identifying those associations that are able to play a more productive role. Indeed, scholars who view business associations as a type of governance institution have profitably used the transaction-cost economics strain of NIE to examine the "life history" of associations involved in the control of prices and production.^{(Endnote 32)}

Of course even institutionally strong associations may not act productively, i.e. for the broader public good. This issue is beyond the scope of this essay, but suggests a further benefit of understanding associations as institutions. By allowing us to place the associations in a broader economic and political context, an institutional perspective facilitates an understanding of the ways in which market forces, state interests and coalitional pressures affect the way associations use internal resources (Doner and Schneider 1999).^{(Endnote 33)} Only by considering such contextual factors can we begin to understand whether and how associations are able to pursue the "logics" of membership and influence in the interests of developmental objectives.

REFERENCES

Anek Laothamatas. 1992. Business Associations and the New Political Economy of Thailand. Boulder: Westview.

Bates, Robert. 1997. Open Economy Politics. Princeton: Princeton University Press.

Baumol, William J., and Alan S. Blinder. 1991. Economics: Principles and Policy. New York: Harcourt Brace Jovanovich.

Booth, A.L., and D.J. Snower. 1996. Acquiring Skills: Market Failures, Their Symptoms and Policy Responses. New York: Cambridge University Press.

Borner, Silvio, Aymo Brunetti, and Beatrice Weder. 1995. Political Credibility and Economic Development. New York: St. Martin's Press.

Broz. Lawrence. 1999. "Origins of the Federal Reserve System." International Organization. 53: 1 pp. 39-70.

Burki, Shahid and Guillermo Perry. 1998. Beyond the Washington Consensus: Institutions Matter. Washington, D.C.: World Bank.

Campos, José‚ and Hilton Root. 1996. The Key to the Asian Miracle: Making Shared Growth Credible. Washington, D.C.: Brookings Institution.

Cheng Lu-lin. Forthcoming. Sources of Success in Uncertain Markets: The Taiwanese Footwear Industry. In Deyo, Doner and Hershberg, ed., Flexible Production in East Asia.

Christensen, Scott, and Akin Rabibhadana. 1993. "Exit, Voice, and the Depletion of Open Access Resources: The Political Economy of Property Rights in Thailand." Unpublished ms., Thailand Development Research Institute.

Clague, Christopher. 1997a. Institutions and Economic Development: Growth and Governance in Less-Developed and Post-Socialist Countries. Baltimore: Johns Hopkins University Press.

------. 1997b. "Introduction." In Clague, ed., Institutions and Economic Development.

------. 1997c. "The New Institutional Economics and Economic Development." In Clague ed., Institutions and Economic Development.

Doner, Richard and Ben Ross Schneider. 1999. "Business Associations and Economic Development." Working Paper 99-12 of the Institute for Policy Research, Northwestern University.

Drobak, John and John Nye. 1997. "Introduction." In The Frontiers of the New Institutional Economics. San Diego: Academic Press.

Duenden Nikomborirak and Somkiat Tnagkitvanich. 1999. "The Thai Corporate Governance (sic)" From Crisis to Recovery." Bangkok: Thailand Development Research Institute.

Elder, Mark. 1997. "Introduction: The Industrial Policy Paradox." unpublished ms., Harvard University.

"Emergent Components Industries and the Way Ahead: The Example of India." 1995. Motor Business-Asia-Pacific. 4^th quarter: 122-131.

Evans, Peter. 1995. Embedded Autonomy. Princeton: Princeton University Press.

Fairclough, Gordon. 1994. "On Second Thoughts." Far Eastern Economic Review (August 18): 40-41.

Federation of Thai Industries. 1998. "Code of Ethics of Industrial Operators." Bangkok.

Fletcher, W. Miles II. 1996. "The Japanese Spinners Association: Creating Industrial Policy in Meiji Japan." Journal of Japanese Studies. 22:1: 49-75.

Fogel, Robert. "Douglass C. North and Economic Theory." In Drobak and Nye, the Frontiers of the new Institutional Economics. San Diego: Academic Press, 1997.

Geddes, Barbara. 1994. Politician's Dilemma. Berkeley: University of California Press.

"Government to Intervene if Cold-Storage Firms Deflate Prawn Prices." 1990. Bangkok Post: 17.

Haggard, Stephan. 1997. "Democratic Institutions, Economic Policy, and Development." In Clague, ed., Institutions and Economic Development..., pp. 120-149.

------. and Steven Webb. 1994. Voting for Reform: The Politics of Adjustment in New Democracies. New York: Oxford University Press.

Handley. Paul. 1993. "The Falling Price of Success." Far Eastern Economic Review (April 29): 48-49.

Hollingsworth. J. Rogers, and Robert Boyer, eds. 1997. Contemporary Capitalism: The Embeddedness of Institutions. New York: Cambridge University Press.

Johansen, Lars H. 1999. "Employer Collective Organization and Training: A Theoretical Analysis of Implications and Conditions for Employers' Collective Action." Paper presented at the 11^th International Conference of the Society for the Advancement of Socio-Economics, Madison, Wisconsin. July 8-11.

Klitgaard, Robert. 1997. "Information and Incentives in Institutional Reform." In Clague, ed., Institutions and Economic Development.

Knight, Jack. 1992. Institutions and Social Conflict. New York: Cambridge University Press.

Kuo Cheng-Tien.1995. Global Competitiveness and Industrial Growth in Taiwan and the Philippines. Pittsburgh: University of Pittsburgh Press.

Levy, Brian and Pablo Spiller. 1996. Regulations, Institutions, and Commitment: Comparative Studies of Telecommunications. New York: Cambridge.

Lucas, John. 1993. "State and Society in Nigeria: A Study of Business Associations in Kano." Ph.D. diss., Indiana University.

McMillan, John and Christopher Woodruff. 1999. "Private Order under Dysfunctional Public Order." paper presented for conference on "Private and Public Ordering of Commercial Transactions, University of Michigan Law School, Feb. 11-12.

Moore, Mick and Ladi Hamalai. 1993. "Economic Liberalization, Political Pluralism and Business Association in Developing Countries." World Development 21 (12): 1895-1912.

Moore, Pete."Doing Business with the State: Explaining Business Lobbying in the Arab World." Ph.D. diss., Dept. of Political Science, McGill University.

Niskanen, William. 1971. Bureaucracy and Representative Government. Chicago: Adine

North, Douglass. 1984. "Three Approaches to the Study of Institutions." In David Colander, ed., Neoclassical Political Economy. Cambridge: Ballinger.

------. 1986. "The New Institutional Economics." Journal of Institutional and Theoretical Economics. 142:2, pp. 30-37.

------. 1990. Institutions, Institutional Change, and Economic Performance. Cambridge: Cambridge University Press.

------. 1991. "A Transaction Cost Theory of Politics." Journal of Theoretical Politics 1.

Olson, Mancur. 1965. The Logic of Collective A+ction. Cambridge: Harvard University Press.

------. 1982. The Rise and Decline of Nations. New Haven: Yale University Press.

------. 1997. "The New Institutional Economics: The Collective Choice Approach to Economic Development." In Christopher Clague, ed., Institutions and Economic Development Baltimore: Johns Hopkins University Press.

North, Klaus. 1997. Localizing Global production: Know-how transfer in international Manufacturing. (Geneva: ILO).

Ostrom, Eleanor. 1999. "Reforms, Property-Rights Systems and Development." APSA-CP Newsletter of the Organized Section in Comparative Politics of the APSA. 10:1 (Winter): 20-22.

Pempel, T.J. 1999. Regime Shift: Comparative Dynamics of the Japanese Political Economy. Ithaca: Cornell University Press.

Ramsay, Ansil. 1987. "The Political Economy of Sugar in Thailand." Pacific Affairs. 60:2 (Summer): 248-70.

Roemer, Paul M. 1994. AThe Origins of Endogenous Growth.@ Journal of Economic Perspectives 8:1 (Winter): 3-22.

Saxonhouse, Gary. 1974. "A Tale of Japanese Technological Diffusion in the Meiji Period." Journal of Economic History. 1: 149-165.

Schmitter, Philippe. 1994. "Organized Interests and Democratic Consolidation in Southern Europe." In Richard Gunther, Nikiforos Diamandouros, and Hans-Jurgen Puhle, eds., The Politics of Democratic Consolidation. Baltimore: Johns Hopkins University Press.

Schmitter, Philippe C., and Wolfgang Streeck. 1999. "The Organization of Business Interests: Studying the Associative Action of Business in Advanced Industrial Societies." Koln: Max-Planck-Institut.

Schneider, Ben Ross. 1999. "States and Collective Action: The Politics of Organizing Business in Latin America." Paper presented meetings of the American Political Science Association.

Schneider, Ben Ross and Sylvia Maxfield. 1997. "Business, the State, and Economic Performance in Developing Countries." In Maxfield and Schneider, eds., Business and the State in Developing Countries. Ithaca: Cornell University Press.

Schneiberg, Marc and J. Rogers Hollingsworth. 1991. "Can Transaction Cost Economics Explain Trade Associations?" In Roland M. Czada and Adrienne Windhoff-Heritier, eds., Political Choice: Institutions, Rules, and the Limits of Rationality. Boulder, Col.: Westview Press.

Stiglitz, Joseph. 1974. "Incentives and Risk-Sharing in Sharecropping." Review of Economic Studies. 41: pp. 219-55.

Streeck, Wolfgang and Philippe C. Schmitter. 1985. "Community, Market, State -- And Associations? The Prospective Contribution of Interest Governance to Social Order." In Streeck and Schmitter, eds., Private Interest Governance. London:

Tullock, Gordon. 1965. The Politics of Bureaucracy. Washington: Public Affairs Press.

Wichit Sirithaveeporn. 1997. "Petrochemicals/Import Tariff Reductions: Ministries Split Over Proposal." Bangkok Post November 24): 01.

Williamson, John. 1990. The Progress of Policy Reform in Latin America. Washington D. C: Institute for International Economics.

Williamson, Oliver. 1996. "The Institutions and Governance of Economic Development and Reform." In The Mechanisms of Governance. New York: Oxford University Press.

Wing, Woo. 1990. "The Art of Economic Development: Markets, Politics and Externalities. International Organization 44:3 (Summer): 403-429.

World Bank. 1997. World Development Report: The State in a Changing World. New York: Oxford University Press.

Endnote 1:
We are grateful to Aurelio Parisotto for comments on a previous version.

Endnote 2:
Although often conflated as "market shortcomings," (e.g. Baumol and Blinder 1991: Ch. 29), market imperfections and failures merit distinction. Strictly speaking, market imperfections are departures from the purest form of competition. They involve situations in which, due to particular circumstances such as monopoly or information asymetries, markets fail to sustain "desirable" activities or to stop "undesirable" activities (Bator 1992: 35). Market failures, on the other hand, involve under (over) supply of goods (bads) because the very nature of these goods does not figure in private market calculations, even under conditions of perfect competition. Such goods have spillover effects which result in a divergence between social and private costs.

Endnote 3:
In the view of the World Bank's Institutions Matter, formal institutions are various types of rules and regulations, and their enforcement mechanisms; informal institutions are political norms, values, and levels of trust. Organizations, on the other hand, include legislative chambers, parties, agencies, the judiciary, NGOs, parent-teacher associations (PTA), private firms, trade unions, and business organizations. (Burki and Perry 1998: 11-12). One problem with this distinction is its failure to reconcile its belief in institutions as enforcement mechanisms with the fact that organizations are enforcement mechanisms.

Endnote 4:
David and North (1971) (as cited in Williamson 1996: 325) make a similar distinction between institutional rules and institutional arrangements. Coase and Williamson of course are more interested in NIE as it applies to firms or what Williamson calls the institutions of governance, though their focus almost never includes associations. We return to Williamson's conceptual definitions below.

Endnote 5:
Illustrating this confusion, the World Bank publication Institutions Matter lists "courts" as institutions and "the judiciary" as an example of an organization (Burki and Perry 1998: 11-12). As Kiren Chaudhry notes, while it is possible conceptually to delink a set of rules, say about taxation, from "the coercive and information gathering arms of the tax bureaucracy, in practical terms the two are too closely related to separate, for the 'humanly devised constraints' of taxation have no meaning apart from the actual functioning of the organization." (1997: 10, fn.9).

Endnote 6:
For empirical cases, see Broz's account of the broader consequences of actions by private financial interests for central bank development (1999), and Fletcher's (1996) account of the impact of lobbying of Japanese spinners' association. Even Jack Knight's strategic account of institutional origins does not preclude collective benefits as unintended consequences of distributional conflict (1992: 20).

Endnote 7:
See especially North (1986). For a discussion of the evolution of North's views, see Evans (1995: 33). Williamson provides a fuller catalogue of differences between transaction costs economics (TCE) and economic "orthodoxy" (1996: 6-7). There Williamson subscribes to Simon's 1961 definition of bounded rationality as behavior that is "intendedly rational, but only limitedly so."

Endnote 8:
In one of the earliest attempts to synthesize institutional approaches in economics, North (1984) distinguishes three approaches: 1) transaction cost (a label North applies to his own work), 2) public choice and rent seeking (Tullock, Buchanan, and Tollison), and 3) collective action (Olson).

Endnote 9:
Williamson argued that "NIE actually took shape in two complementary parts" (1996: 325). The first addresses "mechanisms of governance" (and more micro analysis primarily addressing transaction costs and organizations or hierarchies like firms). The second concerns itself with the "institutional environment" (also referred to as the constitutional order, background conditions, or rules of the game).

Endnote 10:
North (1991) and others have also extended the use of TCE to an analysis of politics and political exchanges, where transaction costs are especially heavy and always heavier than in economic life.

Endnote 11:
The institutional solution in this case involves contractual formulas, such as proportional shares rather than wages or fixed rents, and long term commitments, that maximize the interests of both landlord and tenant (Stiglitz 1974; Clague 1997c: 19). That optimal contracting is grouped together with principal/agent analysis (Clague 1997c: 19) rather than TCE which also focuses heavily on contracts demonstrates the fluidity of conceptual boundaries within NIE and the risks of drawing hard and fast categories.

Endnote 12:
In addition to principal-agent problems, research on asymmetric information has also shed light more directly on imperfect information regarding the quality of goods and services. The focus of this work has been on problems of adverse selection and moral hazard (Clague 1997c: 19).

Endnote 13:
Olson analyzes pure public or collective goods which are non-excludable and non-rival, like a tariff, where no producer of a protected product can be excluded from the benefit and where one producer's "consumption" of the benefit does not reduce the benefit (of protection) for other producers. Common pool resources are also non-excludable (as in fishing areas) but rival in the sense that the consumption by one person reduces the potential consumption of everyone else.

Endnote 14:
As discussed below, this view has come under a criticism that opens the way for greater participation by private sector interests in the reform process.

Endnote 15:
Other challenges would include market deregulation, reduction of overcapacity, quota allocation, market information provision, diffusion of technology, conflict adjudication, and standards. See Doner and Schneider (1999).

Endnote 16:
McMillan and Woodruff distinguish between "social networks" and "business networks." While the former "have a dampening effect on market forces," the latter have more favorable effects on exclusion (1999: 39).

Endnote 17:
One is the importance of alternative forms of property rights for either "common pool resources" (Ostrom 1999), or where formal property rights law does not exist (Christensen and Akin 1993). Another has to do with the political or coalitional bases of property rights.

Endnote 18:
For a review of these issues, see Duenden and Somkiat (1999).

Endnote 19:
Major issues on this research agenda include: 1) identifying and balancing the interests of diverse groups of stakeholders; 2) sorting out the second-order institutional factors that impede governance reform (e.g. weakness in government tax collection, weaknesses in financial institutions which might otherwise help to monitor corporate governance); 3) distinguishing benefical network linkages from those consisting of nothing more than insider trading, and 4) assessing the potential of business associations to address problems of corporate governance, such as auditing standards (see for example, Federation of Thai Industries 1998). However, there is a danger that concerns with corporate governance "crowd out" attention to inter-firm linkages required for upgrading in the "real" sector. A recent study of the "Governance Aspects of the East Asian Financial Crisis" (Lanyi and Lee 1999), inspired by the work of Mancur Olson, makes no mention of the challenges of upgrading in manufacturing and logistics, despite significant weaknesses in these areas in countries such as Thailand. On technological weaknesses, see the papers presented to the World Bank "Conference on Thailand's Dynamic Economic Recovery and Competitiveness," May 20-21, 1998, Bangkok.

Endnote 20:
In a rare reference to business associations, Coase noted in an interview in the Spring 1999 newsletter of ISNIE, that trade associations might be capable of creating private judicial systems.

Endnote 21:
See Clague (1997a), the four chapters deal with legal reform in transition economies, reform of tax admijnistration, information and incentives in institutional reform, and rational compliance with bureaucracies.

Endnote 22:
See for example, Burki and Perry's (1998: 28) application of the NIE to economic reform.

Endnote 23:
It is also typically assumed that the benefits from reform are less immediate and more dispersed than those of preserving the status quo. In this case, however, downstream exporters are likely to obtain immediate and concentrated benefits from reduced input costs.

Endnote 24:
Where, in contrast, markets and property rights are precarious and transaction costs exorbitant, as for example in parts of Russia and war-lord states of Africa, then even associations with minimal institutional capacity may play a role in pushing for measures to correct state failures and enhance market functions (and, as was seen in the Peruvian case, the associations may develop greater institutional capacity in the course of pressuring government).

Endnote 25:
This also suggests possible divergence between member interests and association behavior or the relative autonomy of association staff (agents) from members (principals), an issue to be addressed below.

Endnote 26:
Another major benefit that states grant associations is institutionalized access to government policy makers and policy forums. The access sometimes comes through arrangements like consultative councils in Asia, or inflation pacts, or seats on the boards of public agencies or state enterprises (historically). This access is a valuable conduit for information and a conduit that can be construed as a selective benefit for the most active members of an association. The information gained through government granted access can have a high value for large firms in the association that therefore have strong incentives to invest in collective action, especially in the form of attending meetings, participating in commissions and working groups, and politicking for leaders and the representatives they appoint to attend meetings with the government. Moreover, firms are more likely to invest more material resources in developing in house information and technical expertise among staff members that can then be called into service to support the association's positions in meetings with government (see Schneider 1999).

Endnote 27:
Customized information and services, if there is a market, are more likely to be provided by private consultants. There may be some lag in the early development of consultant services where the association can detect a demand for a service that no private firm provides. But once the service and demand are developed, it can be privatized and competitors to the association are more likely to emerge.

Endnote 28:
Hirschman's analysis of exit and voice is not usually counted among the core works of NIE, but Burki and Perry (1998) include exit and voice as key concepts in institutional analysis. Hirschman was of course declaiming the importance of institutions long before many of the founders of NIE, but with a more electic set of analytic tools and conclusions.

Endnote 29:
Campos and Root (1996: 101-3) argue that transparency in government allocation decisions (promoted by business/government deliberation councils) reduces opportunities and temptations for rent seeking. The same argument can be applied to associations that distribute resources (usually delegated by governments) where internal transparency would reduce incentives for members to seek rents from their associations.

Endnote 30:
For example, the staff of the Thai Productivity Institute have had to be very cautious in developing sufficient member trust to generate the kinds of production information necessary for benchmarking. Interview, Bangkok, July 1999.

Endnote 31:
On endogenous growth and the need for institutions, see Roemer (1994). On the role of associations in technology diffusion, see North (1997). For an interesting illustrative case, see Saxonhouse (1974).

Endnote 32:
Schneiberg and Hollingsworth (1991) concluded that TCE does not account for the initial emergence or ongoing reproduction of such groups, but that it does explain their organizational development.

Endnote 33:
For a useful study of associational success based on an explicitly political strategy, see the case of the Japanese spinners' efforts to reduce import tariffs on yarn (Fletcher 1996).