درباره حاکمیت قراردادی همکاری های پژوهش: تخصیص کنترل و حقوق مالکیت معنوی در سایه فسخ بالقوه قرارداد

This paper investigates the governance design problem of a large company that wants to engage a small and cashless firm into a research collaboration. This analysis reflects the frequently observed collaborations between pharma companies and biotechs, and an actual research contract is assessed to link theory to practice. The governance form refers to the allocation of control rights over the research process and property rights over research output(s), as determined by the initial contract; yet this contract is incomplete. The parties negotiate at a later stage from bargaining positions that depend on the initial choice of the governance, but they also contemplate the possibility that the collaboration will be terminated. By means of a simple model that captures the core aspects of the contractual environment, I answer a key research question: How should governance be designed in the shadow of potential termination to provide the research firm with the incentives to invest in the collaboration? First, it is in the company’s interest to choose a governance form that eliminates the possibility of termination and stabilizes the collaboration whenever possible. Second, if the collaboration cannot be stabilized, the company faces a trade-off between reducing the probability of termination and providing incentives, which is ultimately resolved by making the collaboration highly unstable. Third, property rights and control are substitutes in the governance design: If the company collects more property rights, it must relinquish more control rights.

The formal contracts that underlie alliances and collaborations between firms produce a large variety of governance forms. In research collaborations, for example, in addition to spelling out the allocation of intellectual property rights over research outputs, contracts establish the configuration of control over the research process. Yet research contracts remain incomplete; they cannot establish clearly how the benefits from continued collaboration should be shared between the partner firms. Moreover many interfirm relationships, as highlighted by Kale and Singh (2009), present a paradox, in that they are sources of competitive advantage (Dyer and Singh, 1998), yet they are also widely exposed to failure and prone to disintegration (Kogut, 1989, Parkhe, 1993 and Doz, 1996). Research collaborations in the biopharmaceutical industry offer clear examples of the use of formal contracts, governance forms, and instability, and I shall discuss them in detail, using the contracts that link pharma companies and biotechs. These contracts have become increasingly complex, with clear allocations of intellectual property rights (IPRs) and control rights, but research collaborations remain very unstable. According to a Bio Partnering report from 2004:1 Large pharmaceutical companies and biotechs have achieved a level of deal-making sophistication. Still, less than half of respondents reported that their alliances were successful [...] better alliance governance practices could salvage 85% of the value now lost to failed partnerships – a potential sum of US$2.7 billion. In this article, I therefore develop a model that captures the core aspects of the contractual governance of research collaborations. On the one hand, I provide an incentive-based rationale for how control over the research process and property rights over the research outputs get allocated between the participant firms. On the other hand, instead of interpreting instability as a symptom of inaccurate governance practices or a natural consequence of the limitations to contracting, I take a different perspective and ask: How do accurate governance choices affect the instability of a collaboration? The model is designed to represent a research collaboration between a large pharma company and a small biotech. The company and the biotech sign an incomplete contract ex ante, that allocates control rights over research and property rights over the outputs. The contract is incomplete, and the parties anticipate that they will negotiate ex post to allocate the benefits from continued collaboration. Yet, the initial contract sets the boundaries for the negotiation, as the parties expect to obtain a larger negotiation payoff when they are assigned more control and property rights. They also contemplate the possibility that they might prefer to terminate the collaboration. Besides distinguishing between ownership and control, and showing that they are sensitive to different elements and have different incentive effects, I add two aspects that provide a more nuanced analysis of the role of governance. First, the probability of termination is endogenous to the model, as well as linked to the allocation of property rights. Second, the parties’ relative bargaining position shifts according to the allocation of control rights, which helps us explain how the value created gets allocated between the firms. From a modelling perspective, the pharma company acts as the principal that designs governance forms and provides funding to a cashless biotech, the agent, which must make a nonverifiable value-creating investment.2 The analysis highlights that a clear definition of ownership patterns and configurations of control is crucial to address moral hazard and potential termination simultaneously. From a positive standpoint, governance changes when the range of uncertainty over the benefits from continued collaboration varies and when the participant firms’ relative evaluation of property rights in case of termination shifts. From a normative standpoint, I offer managerial prescriptions regarding how much control over the research process and how many property rights a company with a strong bargaining position should allocate to its partner to motivate value-creating investments. The primary prescription is that when the principal (large company) retains more property rights than the agent (small firm), it also relinquishes more control to the agent to balance their relative bargaining positions and provide appropriate incentives to invest. In other words, when designing governance to provide incentives to invest, ownership and control should be considered substitutes. Moreover, an unstable relationship should not always be governed to minimize the probability of termination. Instead, this analysis illuminates an important trade-off for managing research collaborations, in which stabilization efforts conflict with the provision of incentives. A high probability of termination therefore may be effective to obtain larger value-creating investments by the agent.

In this paper, I investigate the research collaboration between two firms. The incomplete contract used to structure the relationship allocates ownership and control rights, and the governance produced affects the distribution of the value created by setting the boundaries of the negotiation. Overall, I show that there are more nuanced links among the initial contractual provisions, incentives, and stability of a relationship than usually studied in prior literature. This recognition leads to quite different prescriptions for contract design, illuminating how ownership and control rights get allocated in the shadow of potential termination and as tactics to obviate moral hazard. This framework could be extended to relationships that involve information- and knowledge-intensive activities, such as between a firm and a talented individual or a firm and newly formed spinoffs. For example, Panico (2009) focuses on the allocation of control and studies the relationship between an employer and a knowledge worker. Ownership does not enter into the picture, and termination is excluded from the analysis. Also, Arora and Merges (2004) stress the role of property rights in supporting the relationship between a parent firm and a spinoffs and provide a few examples of how patents can be used to structure the collaboration. Beyond the ownership of assets required for production and commercialization, patents can limit the ability of one partner to act opportunistically after the other has specialized its technology. But control, as opposed to property rights, is not addressed by Arora and Merges. The model herein could provide an interesting means to investigate the combined use of property and control rights by the firm in both kind of relationships.