Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
17 Assessing the Credibility of Voluntary Codes: A Theoretical Framework Franco Furger Over the past 10 years, there has been a steady increase in the number of voluntary environmental initiatives among private sector organizations. ~ In addition to the now well-known case of Responsible Care, several initiatives have been launched at the domestic and international levels. Exam- ples include Coatings Care by the National Paint and Coatings Association, the Coalition for an Environmentally Responsible Economy (CERES) Principles, the development of the ISO 14000 standards of environmental management by the International Organization for Standardization, and the principles and criteria of sustainable forest management developed by the Forest Stewardship Council, a business-nongovernmental organization (NGO) alliance. The proliferation of voluntary initiatives raises several difficult questions: Can these codes be trusted? How can their credibility be assessed? Should regu- lators rely on private initiatives to meet public policy goals, and if so what principles and criteria should inform agreements between the public and the private sector? In this chapter, I focus exclusively on the second question on how the credibility of voluntary codes may be assessed. I explore what criteria may be used by regulators and the public to determine whether a voluntary program should be considered credible. The discussion is conducted in fairly abstract terms and is not centered on any specific voluntary initiative, although I refer occasionally to specific cases to illustrate a point. Discussions of individual cases may be found in Prakash (2000), Howard et al. (2000), Furger (1997), and Rees (1994, 1997~. The goal of such a general discussion is to develop a concept of credibility that is sufficiently robust to be applicable to a wide range of voluntary initiatives. It should enable deci- sion makers in the public sector to assess the credibility of voluntary initiatives 283
284 ASSESSING THE CREDIBILITY OF VOLUNTARY CODES in a cost-effective way, even when environmental performance data are scarce or unavailable. Finally, it should prove helpful to decision makers involved in the design of voluntary agreements. CLASSIC PERSPECTIVES ON "CREDIBILITY" Public administration scholars usually conceive of credibility (or lack there- of), in formal and procedural terms, rather than substantive terms. For example, voluntary programs have been criticized for not including formal mechanisms of monitoring and compliance assurance for rarely incorporating sanctions short of expelling a member. Furthermore, in many cases voluntary codes are perceived as too generic to be enforceable. Finally, critics have faulted these initiatives for not including measures of environmental performance (see Nash, this volume, Chapter 14, for an in-depth discussion of these issues). Not surprisingly, envi- ronmental groups have labeled some of these initiatives as "greenwash" (Greer and Bruno, 1996~. I submit that a legalistic definition of credibility is largely responsible for obfuscating our understanding of voluntary codes and has hindered a fruitful debate over their role in public policy. Consider the claim that the codes are generic, prone to conflicting interpretations, and therefore impervious to vigor- ous enforcement. This characterization indeed does apply to many codes. How- ever, there is no reason to assume that generic codes cannot be trusted. In fact, the generic nature of voluntary codes may increase rather than undermine their credibility. For example, an association representative recently pointed out that the generic nature of the code published by her trade association the American Textile Institute allows her to customize the program's requirements to her members' specific operational and technological circumstances. From the trade association's point of view, customization has the potential to reconcile what public administration scholars have long thought to be the unavoidable tradeoff between effectiveness and accountability (Light, 1995; Os- borne, 1988~. The code provides a general, common framework for all member firms. It may be characterized as an architecture of environmental management. It enables member firms to systematically manage their environmental impacts. On the other end, the generic nature of the code makes it possible for the trade association manager to tailor the program's requirements to her members' spe- cific operational, organizational, and technological conditions. Yet the code man- ager remains accountable to her members: Customization is based on a set of specific guidelines developed by the trade association and its members. Thus, generic codes may reflect the need to carefully balance effectiveness and ac- countability, rather than suggest deceptive intentions by trade associations. (In this paper, the term "effectiveness" indicates the ability of a trade association manager to customize generic code requirements to the specific operational con- ditions of individual member firms.)
FRANCO FURGER 285 A second important criticism has focused on inadequate mechanisms of monitoring and compliance assurance. Skeptics often have pointed out that the absence of formal enforcement mechanisms invariably will translate into "free- riding" behavior by member companies. To be sure, free-riding is an all too real possibility. But there are good reasons to believe that free-riding behavior is not nearly as widespread as this legalistic perspective suggests. The U.S. Environ- mental Protection Agency (EPA) reckons that compliance levels in the private sector average approximately 86 percent (Cohen, l999~. The rather limited efforts of this agency to monitor and enforce environmental laws and regulations cannot explain the high level of compliance with environmental regulations displayed by U.S. firms. There are reasons to believe the EPA estimates may be too high (U.S. General Accounting Office, 1990), but these differences don't invalidate the conclusion that these levels of compliance are inexplicably high. If vigorous law enforcement is a poor explanation of law-abiding behavior, what may account for this outcome? As many legal scholars have shown, law- abiding behavior is not simply a matter of assessing the costs and benefits of complying with the law. Often, law-abiding behavior reflects internalized social norms (Etzioni, 2000; Tyler and Darley, 2000; Cooter, 1996; Tyler, 1990~. Sanctions matter of course, but often it is extralegal sanctions rather than state policing that ensure "compliance" (Posner, 1996, 2000; Bernstein, 1992~: Trade association managers, environmental professionals, top executives, community leaders, environmentalists, bankers, insurers, and large customers all may rely on a variety of extralegal sanctioning mechanisms to ensure the proper imple- mentation of a voluntary code. What kind of extralegal sanctions may be avail- able to industry insiders and to the industry "constituencies" are important ques- tions that I discuss later in this chapter. But what evidence is there that voluntary codes may be effective? In-depth evaluations of these initiatives are few and far between and often inconclusive. Indirect support for the effectiveness of voluntary codes is provided by the U.S. experience with voluntary standard-setting organizations.2 There are many simi- larities between standard-developing organizations (SDOs) and trade associa- tions. Both operate as bridging institutions between business interests and the public (Furger, 2001~. Voluntary codes, just like private standards, are devel- oped by private interests. Like voluntary codes, many of these standards are designed to provide a public good. And as for most voluntary codes, no formal enforcing mechanisms exist to ensure compliance with privately developed stan- dards. Finally, most of these standards are not self-enforcing and therefore are exposed to free-riding. Few attempts have been made to evaluate the credibility of standards devel- oped by SDOs. The most thorough analysis of this question has been offered by Krislov (1997~. This author deals extensively with the issues just introduced- the credibility of private standards, SDOs' enforcing efforts, and free-riding be- havior. Based on the (limited) evidence available, he comes to the conclusion
286 ASSESSING THE CREDIBILITY OF VOLUNTARY CODES that "voluntary regulation is less than draconian, but by no means irrelevant or ineffectual" (Krislov, 1997:62~.3 The preceding discussion suggests that it may be misleading to conceive of credibility in formal terms. The obvious alternative, to evaluate the impact of voluntary codes on the environmental performance of participating firms, is rare- ly a practical option. Compiling reliable and consistent data is fraught with diffi- culties. In addition, no consensus has emerged on how to define environmental performance.4 Finally, these efforts rarely reach unambiguous conclusions and thus are of little value to decision makers in the public sector.5 The lack of good environmental performance data is only partially responsi- ble for our rather limited understanding of voluntary codes. Without a robust theoretical framework, it is impossible to explain why some voluntary codes do contribute to improving the environmental record of participating firms while other comparable codes don't. Explaining these differences requires an in-depth understanding of the motives, beliefs, values, and incentives that inform the implementation of a voluntary code. In other words, to be of any practical use the concept of credibility must be defined in substantive rather than formal terms. The chapter is organized as follows: In the next section, I discuss the imple- mentation of voluntary codes as a case of private provision of a public good. The subsequent section forms the core of this chapter. In that section, I show that under certain circumstances, even large and geographically dispersed groups of firms may be able to provide a public good. I then discuss the limits of a purely economic approach to evaluating the credibility of voluntary codes and demon- strate the importance of negotiations and of processes of mutual learning to achieve credibility. I then summarize key findings and explore briefly how the concept of credibility developed in this chapter may inform the design of volun- tary agreements. THE PROVISION OF PUBLIC GOODS IN LARGE GROUPS The adoption of a voluntary code by a trade association may be thought of as a collective action problem: Adopting the code requires all industrial firms to take costly measures, such as investing in pollution prevention technologies and implementing environmental management systems. Although the costs of adopt- ing the code and thus of protecting the environment are borne by all industry members, the benefits of providing this public good are enjoyed by a much larger population.6 Under these circumstances, standard economic theory predicts that this public good is unlikely to be provided. This is a well-known story, told numerous times since Olson's (1965) clas- sic, "The Logic of Collective Action." Applied to the current situation, the theo- ry of collective action predicts that industry members will adopt a voluntary code only if two conditions are met: Every industry member is capable of moni- toring other member firms, and the industry as a whole is able to sanction "free-
FRANCO FURGER 287 riders," that is, member firms that enjoy the reputational and economic benefits of adopting the code, but avoid the implementation costs. These two require- ments often have been taken to suggest that only small groups will succeed in providing a public good. The literature on self-governance supports the view that only small groups succeed in managing problems of collective action. Accordingly, this literature usually focuses on small communities. Ostrom and her students have accumulat- ed a wealth of empirical evidence to support the view that many communities indeed are able to successfully manage common-pool resources (a special case of a public good) without external intervention (Ostrom et al., 2002; Ostrom et al., 1999; Ostrom et al., 1994; Blomquist, 1992; Ostrom, 1990~. Some legal scholars have repeatedly demonstrated that tightly knit communities are quite successful at providing public goods to their members. Ellickson, for example, has studied whaling (Ellickson, 1989) and farming communities (Ellickson, 1991~. The assumption that large groups are unable to provide a public good is rooted in a problematic definition of group size: If a group succeeds in providing a public good, it must be small and spatially concentrated. Conversely, failure to provide a public good indicates a large and dispersed population. There is some wisdom to this definition of group size. But just as Newtonian mechanics pro- vides a satisfactory explanation for many, but not all, natural phenomena, the theory of collective action seems unable to explain why some large groups do manage to provide a public good, while smaller ones don't. Empirical evidence suggests that large groups are not as ill equipped to provide a public good as the economic theory of collective action predicts. Under certain circumstances, in- formation can circulate quite efficiently even within large groups. And as I will show, the effectiveness of informal sanctioning mechanisms is not confined to small groups. Consider the following two examples, the international marine industry and the international diamond trade. The marine industry provides an illustration of the claim that information can circulate quite efficiently even within large groups. The diamond traders for their part demonstrate that informal sanctioning mecha- nisms may be very effective at sustaining common norms, such as honest behav- ior, even in large groups. In an interview with this author, a representative of the Salvage association an organization on which the marine insurance industry depends to conduct accident investigations claimed that an instance of collu- sion between a shipowner and a shipyard in Singapore would be common knowl- edge among marine insurers in London within a week. In the international ma- rine industry, information gathering has become the focus of specialized activities. For example, insurance brokers traditionally have played a very im- portant role in providing information to marine insurers and shipowners. How- ever, insurers don't rely exclusively on brokers. They depend on "agents" in many ports of the world to obtain firsthand information about prospective cli- ents. Thev also leverage shipowners' knowledge about their competitors to over-
288 ASSESSING THE CREDIBILITY OF VOLUNTARY CODES come problems of asymmetrical information.7 In sum, in an industrial sector as large as the international marine industry, one may indeed seem to be operating in a "small world." The diamond traders of New York City provide an illustration of how a sizable group may be able to sustain common rules and norms without resorting to external legal action. The industry is organized into so-called trading clubs, also known as "bourses." The New York Diamond Dealers Club (DDC) is the largest in the United States and has approximately 2,000 members, most of whom are also members of the New York Jewish community. The most striking feature of the DDC is the almost exclusive reliance on extralegal rules and norms to conduct its business. Every aspect of the bourse is governed by the bourse bylaws, and business disputes are resolved through arbitration. The arbitrator's decisions are enforced by the business community, and external enforcement actions usually are not necessary (Bernstein, 1992; Coleman, 1988~. Generally speaking, legal considerations and legal enforcement actions play only a minor role in this trade. Neither the international marine industry nor the international diamond trade may be considered small groups, yet they display a surprising ability to address problems of collective action. They demonstrate that group size is a poor predic- tor of a group' s ability to provide a public good. In the next section, I explore the relationship between group size and collective action in more depth. I show that the ability of a group to overcome problems of collective action can be explained by certain features of the social network in which the group members are embed- ded rather than by sheer group size. ASSESSING THE CREDIBILITY OF VOLUNTARY CODES: A STRUCTURAL APPROACH The Small World Phenomenon The small-world (SW) phenomenon undermines the traditional distinction between small and large groups. This term refers to a familiar situation in which two strangers discover to their surprise that they share a common acquaintance. The term was coined in 1967 by Milgram. He showed that within the U.S. population, the average distance between any two individuals, measured by the average number of links separating them, is approximately six hence the now famous "six degrees of separation" hypothesis (Milgram, 1967~. SW networks intuitively can be described as large groups that retain key characteristics of small groups. Members of SW networks experience their social environment as a small group. As for any other small community, the more cohesive this environment is, the more likely it is that an individual's conduct becomes public information. In addition, in cohesive groups, individuals care about their reputation and avoid actions that may damage it. Finally, more cohe-
FRANCO FURGER 289 sive groups are more likely to sanction a violation of a group norm. The degree of cohesion of the social environment of an individual in the network is mea- sured by the so-called clustering coefficient.8 It is not unreasonable to assume that the more cohesive these clusters are, the more likely they will be disconnected. Individuals belonging to different clusters are very unlikely to know each other, that is, to be connected through a link. This means that the average distance between two individuals each belong- ing to a different cluster may be very large.9 Thus, one can expect that large social groups are characterized by a high clustering coefficient and a long char- acteristic path length, or by a small clustering coefficient and a short characteris- tic path length, but not by a high degree of clustering and a short characteristic path length. Watts has shown that this intuitive result is inaccurate. There exists a large class of networks with both high degree of clustering and a short charac- teristic path length. He has labeled these networks small-world networks (Watts, 1998, 1999a, l999b). This counterintuitive result has a fairly straightforward explanation: High clustering and short characteristic path length are properties of highly clustered networks connected with one another through a small number of random ties. It is precisely these ties that account for the small-world phenome- non (Watts, l999b). The SW phenomenon is relevant to the current discussion for one main reason: Trade association members embedded in SW networks are much more likely to adopt and credibly implement a voluntary code. In addition, these trade groups will display a superior ability to solve problems of collective action- independently of group size and geographical dispersion. Let's discuss briefly why this may be the case. Earlier I argued that efficient information exchange among industry mem- bers is a key requirement for a trade group to credibly implement a voluntary code. If the personal and professional ties among the members of a trade associ- ation constitute an SW network, information about individual members is more likely to be common knowledge among all members even if the membership is large and spatially dispersed. This is so because the shorter the characteristic path length, the easier it is for a bit of information to circulate outside the limited realm of a small cluster of member firms. I have also argued that a high degree of clustering increases the likelihood that another firm in the same cluster will apply some form of extralegal sanctions to the "defector." Thus, the higher the degree of clustering, the more likely it is that another industry member will sanction improper implementation practices. Little currently is known about the network structure of specific industry groups.~° Generally speaking, one can expect the structure of these networks to vary considerably from one trade association to the next. Nevertheless, one can predict whether the members of a trade group are embedded in an SW network by examining a few structural variables. For example, mature industries charac- terized by a stable industrial membership are more likely to be embedded in a
290 ASSESSING THE CREDIBILITY OF VOLUNTARY CODES well-developed SW network: Some degree of stability is needed for an SW network to emerge. Conversely, young industries characterized by fast tech- nological change and with significant turnover in their membership may not be conducive to the formation of SW networks. Another important factor affecting the formation of SW networks is the labor market. A closed labor market, that is, a labor market that has significant entry and exit barriers but displays significant internal turnover, greatly facilitates the emergence of an SW network. On the other hand, an open labor market is likely to impede the formation of SW net- works. The reliance on consultants and other professionals by the member firms also may favor the emergence of SW networks. The preceding discussion was based on several simplifying assumptions. I didn't distinguish between firms and individuals operating within these firms, and I haven't examined the role played by the trade association in the implemen- tation process. In addition, I have not examined how information about an indus- try member may circulate, and what kind of extralegal sanctions may be avail- able to various industry constituencies. In the next section, I begin to discuss these issues. Information Gathering in Large Groups The credibility of a voluntary code depends to a significant extent on the ability of a trade association to identify "recalcitrant" members those industrial members with a poor record of code implementation. As noted earlier, identify- ing "defectors" is a difficult task because often codes do not include any moni- toring mechanism. Under these circumstances, a code manager depends on his or her ties to industry insiders to assess the credibility of self-audits, or to obtain information about specific industry members. If a code manager operates in a large trade group, it is unlikely that she will have direct ties to all member firms. However, this is not necessarily a serious impediment to obtaining relevant information about the member firms. The dis- cussion of SW networks suggests that direct ties to all members firms are not necessary. If the code manager is embedded in an SW network, she is likely to be separated from the relevant source of information by a small number of links. In plain English, even though she may not have a colleague working for the member firm she is seeking information about, chances are she will know some- body who does. If the second-order acquaintance is unable to provide the infor- mation needed, he may be able to help by tapping into his own network of colleagues and acquaintances. Interestingly, a direct tie with the firm under scru- tiny may not be the most effective way to obtain sensitive information about that firm. A company official may find it problematic to provide sensitive informa- tion to a trade association representative, especially if that information can em- barrass his company. Code managers don't depend only on their professional networks to obtain
FRANCO FURGER 291 relevant information. Competitive dynamics among member firms may produce unexpected results. Interviews with representatives of the National Paint and Coatings Association (NPCA) have shown that information about free-riding behavior may be reported to the trade association by firms that have been put at a competitive disadvantage by free-riders: Participating firms may have a strong incentive to ensure that their competitors won't gain an unfair competitive ad- vantage by poorly implementing the code. How rival firms learn about their competitors' standards of implementation is not always clear, and the modality varies from industry to industry. For example, it appears that in the pharmaceuti- cal industry, senior vice presidents for environment, health, and safety entertain cordial relations with each other and exchange relevant information. Representa- tives of large corporations have made similar statements on occasion. Skillful code managers may renew and expand their SW networks in several ways: by organizing training events focused on implementation of the code; by convening meetings dedicated to regulatory, technical, and other issues of com- mon interest; and by participating in annual industry meetings, national confer- ences, professional meetings, and social gatherings. Whether in reality code man- agers consciously pursue this strategy is an important question that has not yet received much attention. But who is in a position to provide good, reliable information about a firm's efforts to implement a code? The most likely source of reliable information is neither the top executive nor the line personnel, but the mid-level manager re- sponsible for the program's implementation. Individuals in top management po- sitions are unlikely to share embarrassing or damaging information with trade association representatives. On the other extreme, line personnel are in a much better position to share relevant information with outsiders. However, they are not likely to have strong ties to trade association managers or to other industry practitioners. By contrast, mid-level managers are embedded in networks of pro- fessional ties that may include several other firms outside their local community. Unlike top management, their position may allow them to share information about their firm' s deficiencies with their professional peers and with trade asso- ciation representatives.~3 Efficient information exchange is obviously an indispensable condition for maintaining the credibility of an industry code. But information alone doesn't guarantee that the members of a trade group will properly implement the code. In the absence of some form of sanctioning, the code is unlikely to become an effective tool of environmental management. In the next section, I examine pos- sible sources of extralegal sanctioning and discuss their relevance. Sources of Extralegal Sanctioning The absence of explicit sanctions from most voluntary codes shouldn't be interpreted as a demonstration of untrustworthiness. As this section shows, the
292 ASSESSING THE CREDIBILITY OF VOLUNTARY CODES credibility of voluntary codes is not predicated upon the trade association impos- ing formal, highly visible sanctions. Several other organizations operating in a firm's institutional environment may play an important role in ensuring the code's credibility.~4 Extralegal sanctions may come in at least three forms. The improper imple- mentation of the code' s requirements may trigger guilt or shame, may damage a firm's reputation, and may have negative economic consequences. Guilt and shame speak to the moral dimension of individual choices, to the internalization of moral values and norms. The economic literature tends to dismiss this element of social regulation, but there is little doubt that over the past 20 years, the business community as a whole has dramatically changed its attitudes towards the environment. Of more immediate relevance to this discussion are sanctions that may damage a firm's or an individual's reputation. A large body of literature suggests that reputational considerations are a key aspect of many social inter- actions, including business dealings.~5 A damaged reputation can affect an indi- vidual or an organization in a variety of ways. It can translate into a loss of sales and revenues, a lower stock valuation, higher insurance premiums, the loss of friends and colleagues, and in extreme cases, a ruined professional career. Finally, contractual and business relations also may become tools of extralegal sanctions, such as when consumer boycotts occur or when the inadequate implementation of the code translates into higher insurance premiums and more costly access to ^- . financing. Several organizational actors in the institutional environment of a firm may play an active role in ensuring the proper implementation of a code. In addition to the trade association, the parent company, other industry members, large cus- tomers (Walton et al., 1998),16 financial institutions (insurance companies and banks),~7 local communities, accounting firms and professional organizations, national and international NGOs (Bendell, 2000), and international organiza- tions all may play a significant role in promoting a credible implementation of the code (Miles and Covin, 2000~. Empirical studies that assess the role of these constituencies in ensuring a credible code implementation are still rare, but tend to support the view that their aggregate impact can be significant. For example, in developing countries, local communities may play an important role in im- proving environmental protection in the absence of an effective regulatory sys- tem. Empirical studies have shown that industrial plants in communities with active local politics are associated with lower environmental impacts (Arora and Cason, 1999; Pargal and Wheeler, 1995~. Finally, how does a trade association deal with noncompliance? It is well known that trade associations usually have a limited sanctioning capacity. A trade association's ability to sanction noncompliance essentially reflects its bar- gaining power, that is, the costs and benefits associated with staying versus leaving the association. A trade association's sanctioning capacity is strength- ened if the membership provides highly valuable and exclusive services to the
FRANCO FURGER 293 member firms. On the other end, a firm's decision to leave a trade association may be associated with significant costs. Leaving the trade group may result in being perceived as a liability by potential business partners; may be interpreted by the public, environmental groups, and regulators as an indication of a poor environmental record; and may increase the costs of insurance and financing. National regulators are in an excellent position to improve the attractiveness of membership. For example, the recently launched EPA Performance Track pro- gram provides recognition and some degree of regulatory flexibility to a selected number of trade associations (Steve Sides, National Paint and Coating Associa- tion, personal communication, June, 2001~. Absent such enticements, trade association officials must resort to different strategies. For example, these officials may make a strategic use of their position as "information brokers" in the industry network. An information broker can wield considerable indirect sanctioning power by making sensitive and poten- tially damaging information about a recalcitrant member available to its direct competitors and/or to other firms. In essence, their position as information bro- kers confers on them the power to trigger what may be called an extralegal "sanctioning cascade." A sanctioning cascade is not without problems. Depend- ing on the structure of the industry network, it can generate a considerable amount of gossip and outrage. Intense gossip can lead to information distortion and unfairly damage a firm's reputation. From Strategic Behavior to Learning Processes The credibility of voluntary codes cannot be discussed exclusively in terms of information diffusion and extralegal sanctioning mechanisms. The incomplete nature of many voluntary codes reduces considerably the relevance of informal sanctions: For sanctions to be fairly imposed, there must be a consensus on what constitutes a "violation" or an "infraction." Only after the industry members have agreed on what constitutes a proper implementation of the code can the role of informal sanctions be explored. In this section, I focus on the role played by environmental managers in shaping the expectations related to implementation of the code. "Implementation" in this context takes on a specific meaning: It identifies the process by which environmental managers convert incomplete and open-ended code requirements into actual implementation practices. It consists of extensive and repeated discussions among environmental professionals, com- bined with regular information exchange. It is, in essence, a process of mutual learning that leads to a common understanding of what constitutes a proper implementation of the code. It is also a process that facilitates the establishment of new ties among environmental professionals ("horizontal" ties) and reinforces the relationship between these professionals and the trade association ("vertical" ties), thus contributing to the formation of what some scholars have called a "community of practice" (Gherardi et al., 1998; Brown and Duguid, 1991~.
294 ASSESSING THE CREDIBILITY OF VOLUNTARY CODES The term "common understanding" doesn't imply that the code require- ments will ultimately translate into uniform implementation practices. "Common understanding" in this context refers to acceptable ways of interpreting generic code requirements in different technical, regulatory and business contexts. Developing a common understanding of what constitutes acceptable interpreta- tions of generic code requirements is key to reconciling conflicting demands for accountability and flexibility, it enables customization of a consensus about dif- ferences rather than uniformity a demonstration of the code' s ability to recon- cile effectiveness and accountability.~9 Two opposing forces shape this negotiation process: on one end, the inter- ests of the firms represented in these negotiations, and on the other end, the allegiances to criteria of professionalism, technical competence, and impartiality typical of a professional community. The tension between professionalism and the firms' interest may lead to a temporary situation in which the environmental professionals develop a common understanding of what constitutes the proper implementation of the code that differs from the position taken by some of the participating companies. These conflicts may be resolved in several ways. One possible way to resolve these disputes is for the professional communi- ty to simply ignore this disagreement. This is not likely to be a durable solution because as it creates a dangerous precedent. The conflict is also likely to resur- face at a later time in a more virulent form. A second option is to lower expecta- tions to avoid possible conflicts. This possibility often is identified as one of the main reasons for distrusting private standards (Cheit, 1990~. However, the evi- dence in support of this claim is not overwhelming. The third and final option is to reallocate the responsibility for resolving this matter from the community of environmental managers to the trade association (to be discussed later). An agreement on what constitutes a proper implementation of a voluntary code does allow environmental and code managers to determine whether a firm is "complying" with the code's requirements, but it doesn't necessarily make the use of extralegal sanctions more effective. A firm may have implemented a code poorly for several reasons. For example, a firm may lack scientific and technical expertise. A recent joint effort by the EPA and the American Chemistry Council to determine the root causes of regulatory noncompliance concluded that "hu- man error" and "procedures" (operating procedures not followed) are two of the main causes of noncompliance (U.S. Environmental Protection Agency, 1999~. Implementing the code also may create complex organizational problems (Fryx- ell and Vryza, 1999~. Finally, financial resources to properly implement the code may be in short supply. Thus, one should distinguish between voluntary and involuntary noncompliance. Involuntary noncompliance demonstrates the limits of assessing the credi- bility of voluntary codes from a narrowly defined economic perspective. If in- voluntary noncompliance is a significant aspect of a poorly implemented code- and there is significant evidence that implementing these codes can be
FRANCO FURGER 295 overwhelming for many small and medium-size firms applying sanctions to noncompliant firms may be wholly ineffectual. If noncompliance can be attribut- ed to limited technical and scientific expertise or to organizational barriers, the effective implementation of a voluntary code may depend on the availability of training and technical assistance and on processes of information sharing and mutual learning, rather than on sanctions. A collaborative approach to the code implementation has a dramatic impact on the code manager' s ability to gather accurate information about member firms. It also enables environmental managers to freely exchange with each other even sensitive bits of information about implementation problems. Conversely, legal- istic and confrontational relations are likely to undermine the free flow of knowl- edge and information among environmental professionals, and between these professionals and their trade group.20 The preceding discussion shouldn't be interpreted as evidence that collabo- rative and confrontational approaches are mutually exclusive. Rather, it suggests that sanctioning noncompliance is itself a choice that must be based on carefully assessing its costs and benefits. Those who favor the imposition of strict sanc- tions all too often are oblivious to the possibility that these sanctions legal or otherwise may be associated with considerable costs. By sanctioning noncom- pliant behavior in an inappropriate or excessive way, a trade group may compro- mise its future ability to shape the implementation process. This means that the role of code managers is subtler than usually assumed. Code managers are neither "private cops" nor mere "cheerleaders." Rather, they may be characterized as mediators. Their central position in the industry network confers on them the power to address disputes among member firms over implementation issues, between industry interests and the public sector, and between environmental groups and member firms. In short, they become what some scholars have dubbed "crosscutting ties."2i CONCLUSION In this chapter I have shown that the two dominant approaches to evaluating voluntary codes determining the impact of a code on environmental perfor- mance and assessing the code requirements against formal criteria may not provide satisfactory results. Accurate, public sources of information on environ- mental performance are rarely available. Furthermore, there is no consensus among scholars and practitioners on how to define measures of environmental performance that would allow meaningful interfirm comparisons. On the other end, evaluating voluntary codes on formal criteria alone may be misleading. As this chapter has shown, firms may undertake considerable efforts to implement a voluntary code, even though the code itself is not fully specified and/or doesn't include detailed reporting requirements and specific mechanisms of compliance assurance. This chapter suggests that a better way to assess voluntary codes is to
296 ASSESSING THE CREDIBILITY OF VOLUNTARY CODES focus on processes and structures that may facilitate or impede a credible code implementation. This includes industry networks, information diffusion and in- formation exchange, extralegal sanctioning mechanisms, and interorganizational learning, among others. The concept of credibility advanced in this chapter provides several lessons for the design of voluntary agreements. First, a contractual relationship between a regulatory agency and a trade association is likely to transform the relationship of the trade association to its members from collaborative to confrontational. In other words, it may crowd out collaborative dispositions and undermine infor- mation exchange and mutual learning. In addition, it may impair the trade asso- ciation's ability to mediate among opposing interests. Whether regulators could avoid these outcomes by providing significant benefits to participating firms remains to be seen. Second, most industrial codes may be characterized as an architecture of environmental management. Current voluntary codes are not designed to directly address highly specific environmental concerns. The incomplete, open-ended nature of many voluntary codes is typical of social obligations that are impervi- ous to formalization. As a result, they are ill suited to become part and parcel of voluntary agreements. Third, developing an in-depth knowledge of the implementation processes should enable decision makers in the public sector to focus on improving the self-regulatory and learning capacity of a trade association and its members, rather than concentrating on specific elements of the code. In less abstract terms, voluntary agreements should be designed to improve the problem solving capac- ity of the parties involved (Lindblom and Cohen, 1979~. NOTES 1 The term "voluntary initiative" is not particularly accurate, as some trade groups have made their codes a condition of membership. However, for the sake of clarity I will use this term through- out the chapter. Voluntary initiatives must be distinguished from "voluntary agreements." The latter usually refers to a formalized contractual arrangement between a private party, often a trade associa- tion, and a regulatory agency, while the former is defined as an institutional arrangement developed by a trade association for the benefit of its members. 2 In the United States, Standard Developing Organizations have been involved in setting health and safety standards in many industries and for many consumer products (Yilmaz, 1998; Krislov, 1997; Cheit, 1990; Hamilton, 1978; Hemenway, 1978). 3 One may attribute this result to fears of product liabilities. However, these fears may be compensated by concerns about antitrust laws. 4 See, for example, National Academy of Engineering (1999), Natan and Miller (1998), Ranga- nathan (1998), and Ditz and Ranganathan (1997). 5 Consider, for example, Responsible Care. Even though this code has existed for well over a decade, only one comprehensive evaluation has been conducted so far (King and Lenox, 2000). Their statistical sophistication and thoroughness notwithstanding, the authors of this study were unable to draw firm conclusions about the effectiveness of Responsible Care. Other efforts at evalu-
FRANCO FURGER 297 ating voluntary codes include Coglianese and Nash (2001), Cowton and Thompson (2000), Prakash (2000), and Nash and Ehrenfeld (1997). 6 This is so because by definition, nobody can be excluded from enjoying the benefits of a public good. 7 The term "asymmetrical information" identifies a situation in which access to critical infor- mation is uneven. Consider, for example, the market for used cars. Prospective buyers are usually unable to determine whether they are getting a "lemon," only the salesperson has that information. This is a classic case of asymmetrical information (Akerlof, 1970). 8 The clustering coefficient is a local property: If a node has n immediate neighbors, then this neighborhood defines a subgraph that has at most n(n-1)12 edges, if the neighborhood is fully con- nected. The clustering coefficient of this subgraph is the fraction of this maximum that is actually realized. The clustering coefficient of the entire graph is the average of these fractions calculated for every node (Watts, l999b:498). 9 The characteristic path length is defined as the "average number of edges that must be traversed in the shortest path between any two pairs of vertices in the graph" (Watts, l999b:498). It is a global property of the network. 10 In this context, the literature on interlocking directorates is of limited help. In this chapter, I am concerned with networks of environmental professionals and top executives, and not with rela- tionships among boards of directors. 11 Consultants and other providers of professional services may be described as "weak ties" (Granovetter, 1973). As I will show, individuals and organizations with weak ties to otherwise disjointed networks are in an excellent position to mediate between conflicting interests and to influence the outcome of these conflicts. An indirect illustration of this argument is provided by Dietz and Rycroft (1987). 12 This is not surprising because acceptance of these codes by member firms would be seriously damaged by any attempt to incorporate such mechanisms from the outset. Accordingly, trade associ- ation representatives carefully avoid any talk of "enforcement action." 13 See Canan and Reichman (1993) for a discussion of the role of mid-level managers and engineers in promoting the implementation of the Montreal protocol. Also see the discussion in Brown and Duguid (1991). 14 It is fashionable to discuss the role of organizations located in the institutional environment of a firm in terms of first-, second-, third-, and fourth-party inspections and certification. Although this terminology occasionally may be useful, it tends to obscure important differences within each cate- gory. For example, a well-reputed industry insider and an international consulting firm both can be described as "third parties." However, a small firm may find it wholly unacceptable to be surveyed and certified by an international consulting firm, but most likely would agree to be audited by an independent industry insider. An NGO would take the opposite view. 15 The details of this argument are more involved. The groundwork was laid by several econo- mists in the 1980s (Mailath and Samuelson, 1998; Kreps et al., 1982; Kreps and Wilson, 1982). Economic historians have provided vivid illustrations of the role of reputational incentives in sus- taining trade (Greif, 1989, 1991, 1993; Milgrom et al., 1990). During the l990s, many legal scholars recognized the relevance of reputational incentives in sustaining social norms (Schwarcz, in press; Coffee, 2001; Posner, 1996, 2000; Choi, 1998; Bernstein, 1992; Ellickson, 1989, 1991; Charny, 1990). Political scientists have studied the role of reputation, guilt, and shame as a means of informal policing (Braithwaite, 1989, 1993). Surprisingly, sociologists with few exceptions (Etzioni, 2000; Raub and Weesie, 1990) have been slow to recognize the importance of this topic. 16 An increasing number of large industrial firms are requiring suppliers to achieve some degree of sustainability by adopting environmental management systems such as ISO 14000. The automo- bile industry is a case in point. 17 For example, the American Chemistry Council (ACC, formerly the Chemical Manufacturers Association) has negotiated better insurance rates for Responsible Care companies. Unfortunately,
298 ASSESSING THE CREDIBILITY OF VOLUNTARY CODES this approach removes selective incentives for ACC firms to improve their environmental perfor- mance. 18 In the case of the diamond trade, this has occurred, as documented in Bernstein (1992:121). A diamond trader whose reputation had been damaged by baseless gossip was able to restore his good name by posting a rebuttal on the DDC's bulletin board. 19 Consider, for example, the following requirement, which is part of the NPCA Coatings Care program: "I.III.1 Occupational Safety and Health: 1.2 Plan and carry out periodic targeted inspec- tions for conformity with site policies and practices." Obviously, a small firm will implement this requirement rather differently than a large industrial company. These differences, however, don't necessarily constitute evidence of inadequate code implementation. 20 A growing body of literature in economics demonstrates that confrontational interactions among the members of a group hamper the group's ability to cooperate. See, for example, Fehr and Gachter (2001), Falk et al. (2002), and Frey (1997). This literature also illustrates the relevance of processes of face-to-face communication in overcoming problems of collective action. 21 The relevance of crosscutting ties hardly can be overemphasized. For example, Lipset (1959) has shown that a key requirement of a functioning democracy is to have crosscutting ties among the various social and political groups. In his now classic The Strength of Weak Ties, sociologist Granovet- ter suggests that missing crosscutting ties among ethnic Italian families in a Boston neighborhood account for the inability of this neighborhood to organize against "urban renewal" (Granovetter, 1973:1373-1376; see also Granovetter, 2000:1079). Locke notes that missing crosscutting ties between business and labor in Turin, Italy, explain the very confrontational and ultimately self-defeating path of industrial restructuring at FIAT, the largest Italian carmaker. By contrast, in the case of Alfa Romeo, the existence of numerous crosscutting ties among the parties involved led to a cooperative restructur- ing process, with positive economic fallouts for the entire Milan area (Locke, 1995). REFERENCES Akerlof, G.A. 1970 The market for 'lemons': Quality uncertainity and the market mechanism. Quarterly Journal of Economics 84(3):488-500. Arora, S., and T.N. Cason 1999 Do community characteristics influence environmental outcomes? Evidence from the Toxics Release Inventory. Southern Economic Journal 65(4):691-716. Bendell, J., ed. Bernstein, L. 2000 Terms for Endearment: Business, NGOs and Sustainable Development. Sheffield, Eng.: Greenleaf Publishing. 1992 Opting out of the legal system: Extralegal contractual relations in the diamond industry. Journal of Legal Studies 21(21): 115-157. Blomquist, W.A. 1992 Dividing the Waters: Governing Groundwater in Southern California. San Francisco: Institute of Contemporary Studies Press. Braithwaite, J. 1989 Crime, Shame and Reintegration. Cambridge, Eng.: Cambridge University Press. 1993 Shame and modernity. British Journal of Criminology 33(1):1-18. Brown, J.S., and P. Duguid 1991 Organizational learning and communities of practice: Toward a unified view of work- ing, learning and innovation. Organization Science 2(1):58-82. Canan, P., and N. Reichman 1993 Ozone partnerships, the construction of regulatory communities, and the future of glo- bal regulatory power. Law & Policy 15(1):61-74.
FRANCO FURGER 299 Charny, D. 1990 Nonlegal sanctions in commercial relationships. Harvard Law Review 104:375-467. Cheit, R.E. 1990 Setting Safety Standards: Regulation in the Public and Private Sectors. Berkeley: Uni- versity of California Press. Choi, S. 1998 Market lessons for gatekeepers. Northwestern University Law Review 92:916-966. Coffee, J.C.J. 2001 The Acquiescent Gatekeeper: Reputational Intermediaries, Auditor Independence and the Governance of Accounting. Unpublished manuscript, Columbia Law School, Center for Law and Economic Studies, New York. Coglianese, C., and J. Nash 2001 Bolstering Private-Sector Environmental Management. Issues in Science and Technolo- gy. [Online Edition]. Available: http://www.nap.edu/issues/17.3/ [Accessed September 28, 2001]. Cohen, M.A. 1999 Monitoring and enforcement of environmental policy. Pp. 44-106 in International Year- book of Environmental and Resource Economics, Vol. III, T. Tietenberg and JH. Folm- er, eds. Northampton, MA: Edward Elgar. Coleman, J.S. 1988 Social capital in the creation of human capital. American Journal of Sociology 94:S95- S120. Cooter, R.D. 1996 Decentralized law for a complex economy: The structural approach to adjudicating the new law merchant. University of Pennsylvania Law Review 144(5):1643-1696. Cowton, C.J., and P. Thompson 2000 Do codes make a difference? The case of bank lending and the environment. Journal of Business Ethics 24(2, Part 2):165-178 I. Dietz, T.M., and R.W. Rycroft 1987 The Risk Professionals. New York: Russell Sage Foundation. Ditz, D.W., and J. Ranganathan 1997 Measuring Up: Toward a Common Framework for Tracking Corporate Environmental Performance. Washington, DC: World Resources Institute. Ellickson, R.C. 1989 A hypothesis of wealth maximizing norms: Evidence from the whaling industry. Jour- nal of Law, Economics and Organization 5(83):83-97. 1991 Order Without Law: How Neighbors Settle Disputes. Cambridge, MA: Harvard Univer- sity Press. Etzioni, A. 2000 Social norms: Internalization, persuasion, and history. Law and Society Review 34(1): 157-178. Falk, A., E. Fehr, and U. Fischbacher 2002 Appropriating the commons: A theoretical explanation. Pp. 157-191 in National Re- search Council, The Drama of the Commons. Committee on the Human Dimensions of Global Change, E. Ostrom, T. Dietz, N. Dolsak, P.C. Stern, S. Stonich, and E. Weber, eds. Washington, DC: National Academy Press. Fehr, E., and S. Gachter 2001 Do Incentive Contracts Crowd Out Voluntary Cooperation? University of Southern California Center for Law, Economics & Organization Research Paper No. C01-3. [On- line]. Available: http://papers.ssrn.com/abstract_id=229047 [Accessed February 15, 2001].
300 Frey, B.S. ASSESSING THE CREDIBILITY OF VOLUNTARY CODES 1997 Not Just for the Money: An Economic Theory of Personal Motivation. Cheltenham, Eng.: Edward Elgar. Fryxell, G.E., and M. Vryza 1999 Managing environmental issues across multiple functions: An empirical study of corpo- rate environmental departments and functional co-ordination. Journal of Environmental Management 55(1):39-56. Furger, F.L. 1997 Accountability and self-governance systems: The case of the maritime industry. Law and Policy 19(4):445-476. 2001 Global markets, new games, new rules: The challenge of international private gover- nance. Pp. 201-245 in Ruler and Networks: The Legal Culture of Global Business Transactions, R. Applebaum, W.F. Felstiner, and V. Gessner, eds. Oxford, Eng.: Hart Publishing. Gherardi, S., D. Nicolini, and F. Odella 1998 Toward a social understanding of how people learn in organizations. Management Learning 29(3):273-297. Granovetter, M.S. 1973 The strength of weak ties. American Journal of Sociology 78(6):1360-1380. 2000 A Theoretical Agenda for Economic Sociology. Unpublished manuscript. Palo Alto, CA: Stanford University, Department of Sociology. Greer, J., and K. Bruno 1996 Greenwash: The Reality Behind Corporate Environmentalism. Penang, Malay.: Third World Network. Greif, A. 1989 Reputation and coalitions in medieval trade: Evidence on the Maghribi traders. Journal of Economic History 49(4):857-882. 1991 The organization of long-distance trade: Reputation and coalitions in the Geniza docu- ments and Genoa during the eleventh and twelfth centuries. American Economic Review 83(3):525-548. 1993 Contract enforceability and economic institutions in early trade: the Maghribi Traders' Coalition. American Economic Review (83)3:525-548. Hemenway, D. 1978 Industrywide Voluntary Product Standards. Cambridge, MA: Ballinger. Howard, J., J. Nash, and J. Ehrenfeld 2000 Standard or smokescreen? Implementation of a voluntary environmental code. Califor- nia Management Review 42(2):63-82 King, A.A., and M.J. Lenox 2000 Industry self-regulation without sanctions: The Chemical Industries Responsible Care Program. Academy of Management Journal 43(4):698-716. Kreps, D.M., P. Milgrom, J. Roberts, and R. Wilson 1982 Rational cooperation in the finitely repeated prisoners' dilemma. Journal of Economic Theory 27(2):245-252. Kreps, D.M., and R. Wilson 1982 Reputation and imperfect information. Journal of Economic Theory 27:253-279. Krislov, S. 1997 How Nations Choose Product Standards and Standards Change Nations. Pittsburgh: University of Pittsburgh Press. Light, P.C. 1995 Thickening Government: Federal Hierarchy and the Diffusion of Accountability. Wash- ington, DC: Brookings Institution.
FRANCO FURGER 30 Lindblom, C.E., and D.K. Cohen 1979 Usable Knowledge: Social Science and Social Problem Solving. New Haven, CT: Yale University Press. Lipset, S.M. 1959 Some social requisites of democracy: Economic development and political legitimacy. American Political Science Review 53(1):69-105. Locke, R.M. 1995 Remaking the Italian Economy. Ithaca, NY: Cornell University Press. Mailath, G.J., and L. Samuelson. 1998 Your Reputation Is Who You're Not, Not Who You'd Like to Be. CARESS Unpub- lished working paper. Philadephia: University of Pennsylvania, Department of Eco- nomics. Miles, M.P., and J.G. Covin 2000 Environmental marketing: A source of reputational, competitive, and financial advan- tage. Journal of Business Ethics 23(3, Part 1):299-311. Milgram, S. 1967 The small world problem. Psychology Today 1(2):60-67. Milgrom, P.R., D.C. North, and B.R. Weingast 1990 The role of institutions in the revival of trade: The law merchant, private judges, and the champagne fairs. Economics and Politics 2(1):1-23. Nash, J., and J. Ehrenfeld 1997 Codes of environmental management practice: Assessing their potential as a tool for change. Annual Review of Energy and the Environment 22:487-535. Natan, T.E.J., and C.G. Miller 1998 Are tonics release inventory reductions real? Environmental Science and Technology 32(15):368A-374A. National Academy of Engineering 1999 Industrial Environmental Performance Metrics. Washington, DC: National Academy Press. National Research Council 2002 The Drama of the Commons. Committee on the Human Dimensions of Global Change, Ostrom, E., T. Dietz, N. Dolsak, P.C. Stern, S. Stonich, and E.U. Weber, eds. Division of Behaviorial and Social Sciences and Education, National Research Council. Wash- ington, DC: National Academy Press. Olson, M. 1965 The Logic of Collective Action. Cambridge, MA: Harvard University Press. Osborne, D.E. 1988 Laboratories of Democracy. Boston, MA: Harvard Business School Press. Ostrom, E. 1990 Governing the Commons: The Evolution of Institutions for Collective Action. Cam- bridge, Eng.: Cambridge University Press. Ostrom, E., J. Burger, C.B. Field, R.B. Norgaard, and D. Policansky 1999 Revisiting the commons: local lessons, global challenges. Science 284(5412):278-282. Ostrom, E., R. Gardner, and J. Walker 1994 Rules, Games, and Common-Pool Resources. Ann Arbor: University of Michigan Press. Pargal, S., and D. Wheeler 1995 Informal Regulation of Industrial Pollution in Developing Countries: Evidence from Indonesia. Washington DC: Policy Research Department, Environment, Infrastructure, and Agriculture Division, The World Bank. Posner, E.A. 1996 The regulation of groups: The influence of legal and nonlegal sanctions. University of Chicago Law Review 63(133-197).
302 ASSESSING THE CREDIBILITY OF VOLUNTARY CODES 2000 Law and Social Norms. Cambridge, MA: Harvard University Press. Prakash, A. 2000 Responsible Care: An Assessment. Business and Society 39(2):183-209. Ranganathan, J. 1998 Sustainability Rulers: Measuring Corporate Environmental and Social Performance. Washington, DC: World Resources Insititute. Raub, W., and J. Weesie 1990 Reputation and efficiency in social interactions: An example of network effects. Amer- ican Journal of Sociology 96(3): 626-654. Rees, J.V. 1994 Hostages of Each Other: The Transformation of Nuclear Safety Since Three Mile Is- land. Chicago: University of Chicago Press. 1997 Development of communitarian regulation in the chemical industry. Law and Policy 19(4):477-528. Schwarcz, S.L. in Private Ordering of Public Markets: The Rating Agency Paradox. University of Illinois press Law Review. Tyler, T.R. 1990 Why People Obey the Law. New Haven, CT: Yale University Press. Tyler, T.R., and J.M. Darley 2000 Building a law-abiding society: Taking public views about morality and the legitimacy of legal authorities into account when formulating substantive law. Hofstra Law Review 28(3):707-739. U.S. Environmental Protection Agency 1999 EPA/CMA Root Cause Analysis Pilot Project: An Industry Survey. EPA-305-R-99-001. Washington, DC: Office of Enforcement and Compliance Assurance. U.S. General Accounting Office 1990 Improvements Needed in Detecting and Preventing Violations. Washington, DC: U.S. General Accounting Office. Walton, S.V., R.B. Handfield, and S.A. Melnyk 1998 The green supply chain: Integrating suppliers into environmental management process- es. International Journal of Purchasing and Materials Management 34(2):2-11 Watts, D.J. Yilmaz, Y. 1998 Collective dynamics of 'small-world' networks. Nature 393(6684):440-442. 1999a The Dynamic of Networks Between Order and Randomness. Princeton, NJ: Princeton University Press. l999b Networks, dynamics, and the small-world phenomenon. The American Journal of Soci- ology 105(2):493-527. 1998 Private Regulation: A Real Alternative for Regulatory Reform. Cato Policy Analysis 303. Washington, DC: Cato Institute.