Liability as a catalyst for product stewardship

1. Introduction

Under product stewardship companies assist their suppliers and customers in improving Environmental, Health and Safety (EH&s) practices. Product stewardship entails a variety of techniques. Reducing the use of potentially hazardous materials in process design, supplying customers with improved information and training, and tracking and recovering hazardous materials. Together these techniques provide for decreased hazards and wastes and increased use of recyclable materials, leading to both decreased operating costs and reduced liability. The question this paper addresses is the extent to which reduced liabilities can or should motivate investments in product stewardship.

To judge from recent trends, product stewardship has become an essential factor in supply chain design and coordination in environmentally intensive industries. For example, in the chemical industry it is the sixth and final code of Responsible Care, implemented by all members of Chemical Manufacturers Association (CMA) and a large portion of the members of the Synthetic Organic Chemical Manufacturers Association (SOcMA) in 1999. Together member companies of these organizations make up more than 90% of the productive capacity of basic industrial chemical manufacturing and of the chemicals produced in the United States. Similar endorsements of product stewardship are evident in Asia and Europe.

The rationale for product stewardship investments is twofold. First, product stewardship activities can be used to assure regulatory compliance and technical support for customers in the proper use of a company's products (see, e.g., Kleindorfer and Snir 2001 for a discussion). Second, and of key interest in this paper, under joint and several liability laws large companies may be held liable for environmental damage, even when damages are not a direct consequence of their actions. Identifying and mitigating these liabilities is therefore important. This paper investigates stewardship activities by large companies that have a distribution system in place characterized by supply chain partners with limited assets.

The collection, treatment, and disposal of hazardous wastes is the most important area where supply chain liabilities have been a centerpiece of industry concern. The legal framework in the United States is that resulting from the 1976 RCRA (Resource Conservation and Recovery Act), the 1980 CERCLA (Comprehensive Environmental Response, Compensation and Liability Act) and its 1986 revision under SARA (Superfund Amendments and Reauthorization Act), commonly referred to as Superfund. This legislation and its implementing regulations clearly recognize and promote product stewardship in wastes generation and disposal. Under RcRA/Superfund, any company that generates, transports, treats, or disposes of hazardous wastes may be held jointly and severally liable for damages caused. In these cases it becomes beneficial to assist others in the supply chain to reduce the risks associated with product manufacture and use. This regulation is not limited to the actual disposal process of wastes, but includes onsite contamination from insufficient attention to possible environmental damage. Companies with potential contaminants are lead to invest in mitigating action to reduce EH&s risks from improper disposal throughout the supply chain. This is especially true for companies with `deep pockets.'

A few recent examples illustrate that vicarious liability, where one firm is held responsible for the damage caused by a supply chain partner, is also of great concern for large chemical manufacturers. Dow-Coming has been found liable for damage caused by its silicone breast implants, which caused the company to file for bankruptcy with 170,000 compensation claims against the implant manufacturers and resulted in a $3.2 billion settlement (Washington Post, Jul. 9, 1998, Dec. 1, 1999). Following Dow-Coming's bankruptcy, Dow Chemical, which owns 50% of Dow-Coming, was also sued for its part in testing silicone liquid. Although it never manufactured implants, its R&D role proved sufficient to warrant an initial $14 million verdict against the company by a single plaintiff, in October 1995, and it faces another 13,000 lawsuits (International Commercial Litigation, Feb. 1996), whose status is unclear following the settlement (New York Times, Dec. 23, 1999). Although scientific proof of the connection between cause and effect is weak (e.g., Silverman et al. 1996), the jury here, as is often the case, was sympathetic to the plaintiffs' suffering (Chemical Engineering, Sept. 1996).

In Europe, the eco-logistics and EMAS programs are extremely important in promoting product stewardship (see Kleindorfer and Orts 1998). By requiring companies to initiate backward logistics for reclaiming package wastes, companies have an incentive to invest in assuring that packaging material are reusable or recyclable.

Concerning the legal foundations of product stewardship, these would seem to run counter to one of the basic axioms of legal thought of independent contractors: a company should only be held liable for actions taken by its employees, and not those of others. In the area of EH&s activities, a different set of legal precedents has evolved, however. These precedents recognize that limiting liabilities to a company's organizational boundaries would lead to possible moral hazard. Companies with harmful processes would outsource these to others with limited assets, who would file for bankruptcy once injured parties demand compensation (Goldfarb 1978; Ringleb and Wiggins 1990). It would then be left to juries and judges to `pierce the corporate veil' to identify which arrangements were put in place only to limit liability. Recognizing the transactions costs and the social losses associated with this process, legal thought and precedent has moved to different liability mechanisms. These essentially hold that for abnormally dangerous activities companies may be held vicariously liable for actions of supply chain partners (see, e.g., Evans 1994, for a discussion of the independent contractor concept and vicarious liability). Allowing for supply chain liability clearly reduces the moral hazard of under-investment in care and promotes product stewardship.

Current tort reform initiatives would curtail liabilities, especially for small businesses. Supporters would reduce punitive damages for all defendants by requiring plaintiffs to provide clear evidence of negligence, and capping negligence awards to twice the economic loss. In addition, proponents would limit payment by certain businesses even further. For those businesses with less than 25 employees or $5 million in annual revenue, punitive damages could not exceed $250,000, and liability for retailers and implant components suppliers would be limited (Washington Post, July 7, 1998). Such changes in tort rulings may provide additional incentives for product stewardship. Companies whose supply chain partners may be indemnified from certain payments have greater incentives to verify that sufficient care is invested in the development, manufacturing, and sale of products.

Against this background, this paper examines the strategic importance of product stewardship, including process and product design changes, under a legal system of strict liability for harm arising from the activities of supply chain partners. The proposed game-theoretic model builds on the earlier work of Shavell (1984) in considering a specific case of joint liability in which suppliers are held liable only when customers' assets are insufficient to compensate for damages. The model extends Shavell's work by analyzing strategic behavior among supply chain partners, where the downstream partner has limited assets and thus is not able to compensate for all liabilities. It complements Watabe (1999) by introducing the possibility that both parties reduce liabilities in a setting where customer investments are not contractible.

The key result from the game-theoretic model is that product stewardship is a strategic response to limited liability in the presence of insufficient information regarding supply chain partners' actions and assets. When suppliers can perfectly discern downstream partners' risk-reducing activity, suppliers would either curtail their own investment or replace partners' investments. If, however, suppliers are uncertain about customers' investments, suppliers are apt to introduce stewardship measures that reduce liabilities throughout the supply chain. Following the supplier's investments, only large partners make additional investments in care, while smaller customers, with limited assets (and liabilities), choose not to invest additionally.

The remainder of this paper is organized as follows. Section 2 analyzes the baseline problem of joint investment in care under perfect information. For the supplier, customer assets drive the investment decision. If the customer has sufficient assets to compensate for possible damage, the customer invests in reducing accident probability, while the supplier would not additionally invest. In this case the customer's assets insure the supplier from environmental liabilities. When, however, faced with a potentially insolvent customer, the supplier has incentives to provide product stewardship, and the customer would not additionally invest. In Section 3, where asymmetric information on customer assets is introduced, uncertainty regarding customer assets leads to product stewardship actions by the supplier, and additional investments in care by large customers. Section 4 analyzes the advantages to the supplier of third party audits to verify customer asset levels and to monitor protective measures. It is shown that, when potential liabilities are large, suppliers have an incentive to require customer auditing by third parties. Section 5 provides generalizations of the proposed model, while Section 6 discusses implications of the analytical framework. Section 7 concludes the paper.

2. The Model Under Perfect Information

Based on the discussion in Section 1, it appears that joint and several liabilities are of great importance in situations where accidents may cause smaller companies to become insolvent. This induces companies with `deep pockets' to invest in mitigating accidents, even if they play only a small role in the product life cycle. This analysis emphasizes investments in care, by large companies, to reduce the probability of environmental accidents throughout the supply chain, assuming other mitigation actions have been optimally incurred. Such actions could include locating facilities in unpopulated areas or investments in legal teams to reduce the possibility of being held liable for adverse outcomes.

Investments in reducing hazards take on many forms in the production process. During the R&D phase, multiple tests are conducted to assure product safety under reasonable use. Investments in capital are essential to assure production processes that strictly adhere to design specification. Other, less visible investments, including employee training and monitoring and internal quality control reduce the possibility of adverse changes during production. While some of these actions are easily observed by outside parties including regulators, customers, and supply chain partners, other investments, especially those in ongoing quality control are more difficult to verify (Maxwell and Decker 2000).

An underlying assumption is that downstream supply chain partners have comprehensive information regarding investment in care by upstream partners. While some of these investments are difficult to verify by outside parties, it is reasonable to assume that customers have good information regarding suppliers' actions. As part of an acceptance policy of hazardous material, it is crucial for customers to know the quality and attributes of products received. This is often accomplished by physical evaluation of received inventory as well as quality inspections throughout the upstream production process. Through comprehensive quality assurance of production inputs, clients learn about suppliers' investments in reducing potential liabilities.

While clients often observe suppliers' actions, the opposite is rarely true. Under common practices, clients receive production inputs with the supplier having only limited information regarding actions taken to reduce potential liabilities. This limits the ability of bilateral contractual arrangements to reduce a supplier's liability. Without verifiable information regarding client behavior, suppliers cannot reasonably require customers to invest in care to reduce the supplier's expected liability. The value of monitoring client action is examined further in Section 4.

The model proposed in this paper extends Shavell (1984) to an interorganizational (von-Stackelberg) game in which investment in care is a strategic response to liability. Consider two firms, a monopolist supplier with a single customer who jointly produce and distribute a product to end-users. The product has intrinsic risks and may cause environmental damage. Denote L > 0 as the magnitude of loss caused if damage occurs. The joint liability framework is assumed to be such that the customer is held liable for all damage done, providing he has sufficient assets, with any residual liabilities borne by the supplier. In essence, the customer's assets (partially) insure the supplier from liability. To emphasize the importance of environmental liabilities, we ignore all other aspects of the bilateral relationship, including the production process and the market for produced goods.

5. Generalizations

An additional method to explain product stewardship is to differentiate between suppliers' and customers' impact on loss reduction. If the supplier and customer each affect a different loss, with a different risk reduction mechanism, then we should find each investing in his respective technologies. Consider the case when the supplier can influence only large-scale environmental damage, for example, through product design, reduced B-P-TS (Biocides, Pesticides, and Toxics), and improved monitoring of his facilities, and he will be held solely liable for damage that occurs from his range of responsibilities. On the other hand the customer may influence only local damage, for example: on-site accidents and emissions. If only the customer will be held liable for his range of responsibilities, we will find that each party invests in their respective actions, dependent on the magnitude of loss and the probability of being held liable.

Equilibrium behavior will also be influenced by the relative efficiency of each party's investments. The model analyzed here allowed only for equal efficiency between parties, but in cases where one party may have an advantage in reducing loss probability, due to superior knowledge or capital, we should expect threshold levels for investments to change. Namely, if a supplier may reduce risk using cheaper methods than customers, we should find these methods being implemented before customers' investment. Suppliers provide safety and use information to customers through Material Safety Data Sheets (MSDS), warning labels, training, and customer support help desks.

When a single supplier has a number of customers, who vary in future asset levels, the supplier should invest in care, dependent on customer insolvency probabilities, customer assets relative to loss potential and average asset level, of those customers with limited assets. Customers, who have below threshold assets, choose to not additionally invest, since the supplier's investments suffice given their liability level. Others, who have above-threshold assets, find that supplier initiatives do not suffice, and they are motivated to additionally invest in care. In many industries, product stewardship takes on this form. Suppliers choose their optimal level of care, given customer heterogeneity. In an industry characterized by large customers, average levels of care across the supply chain can be expected to increase.

6. Discussion

This paper develops an analytical framework to evaluate the importance of joint and several liability as motivating current practices of product stewardship by the largest chemical companies. Limited liability is an important driver of private investments to reduce the probability of adverse events and leads to underinvestment by small customers. Realizing that they will be held liable for residual liabilities, large suppliers invest in care. When suppliers have information regarding customers' compensation possibilities, either the supplier or the customer will invest in care, but not both. Under the more reasonable assumption that information regarding future assets is imperfect, we should find product stewardship initiatives where suppliers invest, with additional investment limited to large customers. The need for investments in care when customer assets are uncertain may justify monitoring customer action and asset levels. As potential losses increase, the economic viability of monitoring increases, and suppliers may take a more active role in evaluating their customers before supplying hazardous material.

6. 1. Product Stewardship and Monitoring

Based on the analytical framework proposed, the potential for product stewardship may be formulated. Companies who have relative advantages in certain risk reduction factors should implement these to reduce the liability of the entire supply chain. Such factors could include on-site activities such as improved design, and also off-site activities such as distributing information, accident and `near-accident' investigation and customer training. For example, Ashland Chemical, a chemical manufacturer and distributor, supports some of its clients' needs by selling chemicals on a `turn-key' basis, taking on all the responsibilities of providing and disposing of chemicals (Chemical Week, Feb. 2, 1994). When there are economies of scale, such actions should be even more pronounced. For companies with similar clients, a single program, of improved design or recycling, may affect damage probabilities throughout the product life cycle. These appear to be extremely beneficial for companies to pursue, in their attempt to promote stewardship programs. It is evident that such programs will be less common in intermediate process companies, with a limited client base.

Suppliers facing customer heterogeneity should have incentives to promote third party monitoring, especially when customer asset levels are diverse and loss potential is large. Such monitoring will provide suppliers with insight into customer care choice, a proxy for asset level. Given this information, suppliers may tailor their investments in care, to be congruent with customer asset levels. For those customers with few assets product stewardship initiatives should increase, due to customers abstaining from investment, while for large customers, risk reduction investment may decrease, since those partners have an independent interest to invest in care. iso 14000, as an international standard of environmental action, provides one such mechanism of assurance. Customers who become certified undergo evaluation that they adhere to their own Environmental Management System (EMs). If suppliers believe that customers' EMS prove sufficient investment in care, they may relax their own monitoring of practices (e.g., Kleindorfer 1997).

6.2. Supply Chain Design and Care

In a bilateral relationship, with a single supplier and a single customer the customer has no incentive to reveal his future asset level, inducing product stewardship activities by the supplier. These stewardship initiatives are most pronounced when the supplier is uncertain about the customer's future assets available for compensating potential damage. Uncertainty regarding the level of insurance provided by his customer and the resulting residual liability induce the supplier to invest in reducing liability throughout the supply chain.

In a competitive setting, with multiple customers, this result may be reversed. Clients with sufficient assets would prefer to credibly indicate their assets to win distribution contracts from monopolist suppliers. With assets shielding suppliers from potential liabilities, these become a competitive advantage. To credibly indicate their ability to compensate for future losses, large distributors would be willing to undergo comprehensive financial scrutiny both by suppliers and by third parties. Smaller customers, without the ability to compensate future losses, could do little to mimic their competitors' asset base.

This is most evident in hazardous waste disposal. Facilities with wastes that may be hazardous should prefer disposal companies with sufficient assets to be held liable for any damage incurred. Such large waste management companies should have a distinct competitive advantage due to their ability to limit others' liability. This is especially true in client industries where facilities producing hazardous waste do not have the capability to properly dispose of it, as in the medical or film industries. The Council of Logistics Management finds that the threat of liability, specifically improper disposal by waste contractors, is a key factor in DuPont's success in the film recovery business, which extracts silver and 'unzips' polymer material from film used in medical services, offset printing and electronics (Council of Logistics Management 1993). For similar reasons transportation of potentially hazardous material should be contracted out to large carriers, who have the financial capability to cover potential liabilities.

6.3. Directions for Future Research

Investments in environmental protection often require both easily observable actions such as capital improvements and actions that are more difficult to ascertain including ongoing employee training and process control. The private nature of some of these investments, especially by customers, makes contractual solutions difficult to implement. A necessary condition for such contracts includes easily observable and verifiable actions, which form the basis of the contract. Under the less restrictive assumptions of private investments, a strategic game between the supplier and customer emerges. Each party makes their investment considering the other party's actions, and responding to these. Future research should complement this research by examining possible contractual remedies to increase mitigation efforts, and the information requirement to implement such contracts. Watabe (1999) examines one such contract in a competitive setting with clients solely investing in care.

The analysis herein evaluates these investments in a supply chain setting where supplier investment both precedes customer investment and is observed by the customer before he makes his investment decision. This leads to a sequential game where the supplier's actions consider the client's best response before undertaking product stewardship initiatives. Other strategic interactions along the supply chain should be considered where the client does not observe supplier investment. This would lead to a Nash game between both parties with simultaneous investments by both parties. While the precise outcomes of this type of interaction are beyond the scope of this paper, simultaneous investments would increase investments by customers and lead to less investment in care by suppliers. As a first-mover, suppliers are disadvantaged because customers can react to observed investments. Without the benefits of observing supplier action, customers would have to increase investment relative to the results of the von-Stackelberg game (Maxwell and Decker 2000).

The analytical framework examines the implications for product stewardship for large suppliers but can easily be extended to address other issues regarding supply chain partnerships. In industries with suppliers with limited assets, upstream partners have few incentives to privately invest in reducing downstream liabilities. In this context we would find `deep pocket' participants as sole investors, regardless of their position along the supply chain. Alternatively, large partners could contractually require suppliers to invest in care. By examining production processes and delivered products, large customers could verify sufficient investments by suppliers, reducing the problems associated with underinvestment induced by limited liability.

Conclusion

The model proposed in this paper allows insight into the developing initiatives of product stewardship. These initiatives, it is hypothesized, should be beneficial in reducing liability, in instances where companies may find themselves liable for environmental damage caused by supply chain partners, when those become insolvent. Under current legislation, companies may find themselves compensating for independent contractors actions, either in cleaning hazardous waste sites, or when handling abnormally dangerous material (Evans 1994). Product stewardship and other mechanisms of investing in care should become more common. These efforts are extremely effective when companies sell to heterogeneous customers, or have specific advantages in providing customer support. On the other hand, customer monitoring may be an effective tool in reducing the investments in care taken on by suppliers, by tailoring investment based on customer assets available for compensation.'