Exchanges between healthcare providers and insurers: a case study.

Inter-organizational transactions can be discrete or relational (Mohr and Nevin, 1900). Two organizations can interact frequently but as a series of discrete transactions rather than a cooperative channel relationship. A distribution channel exists when the transactions between organizations are relational; the transactions are part of a social system that is ongoing, integrated, and cooperative. It requires the two organizations to coordinate efforts to work jointly towards improving the revenue stream and lowering the costs for both participants. Two organizations that are part of the same supply chain can only be classified as channel partners if they bring specialized expertise or resources together to achieve a mutual benefit rather than interact in a series of independent transactions. The physician-insurance provider dyad represents an ongoing relationship where efforts must be coordinated to provide healthcare to patients. In this context, the relationship is typical of a channel partnership rather than a simple supply chain transaction.

A distribution channel can exist in the sales function as well as the support functions within organizations. The sales channel involves buying, selling, and transferring title, whereas in the facilitating channel the actions of insurance, transportation, financing, market research and other activities necessary to support the sales function are coordinated (Rosenbloom, 2000). Correspondent activities occur within the physician-insurance provider relationship such as the marketing function that occurs when the insurance company promotes its preferred providers to policyholders, and the facilitating functions that consolidate billing, assist in practice management, and provide access to national statistics.

Within the channel framework, several dimensions are pertinent to the structure of the channel dyad. Legal contracts, power, conflict, and inter-firm communication are primary elements inherent in the channel relationship. We rely on interviews with healthcare administrators and prior channel research to demonstrate how these variables correspond to the physician-insurance provider alliance.

The Relevance of Transaction Cost Analysis

The channel relationship succeeds because the benefits exceed the costs. Legal contracts, monitoring effort, and relationship-specific investments are the primary costs inherent in the relationship. Transaction cost analysis (Williamson, 1975, 1985) has been adapted as a framework for explaining the environmental characteristics in which the channel partnership thrives (c.f. Rindfleisch and Heide, 1997; Frazier, 1999; Cannon et al., 2000). Theory underlying transaction cost analysis assumes that the channel relationship persists because the costs are lower than they would be if the manufacturer performed all of the distribution services internally. Transaction cost analysis is not unique to distribution channel research but it is particularly applicable given the context in which the channel dyad operates.

In the traditional channel setting, a manufacturer chooses to continue with the distributor arrangement because it is more efficient than performing these services internally. The distributor often has economies of scale and retailer alliances that can move the manufacturer's products to the end consumer efficiently from both a time and money perspective. Similarly, in the healthcare funding system, the insurer arrangement persists because the insurer offers the physician practice one central billing point where the cost of collection is expected to be less than billing each patient for the entire cost of the medical procedures. In addition, the insurer provides access to patients, without which the physicians would need a marketing or outreach program to attract patients to the physician practice. To the extent that the transaction costs exceed the perceived benefits, there are incentives to change the nature of the relationship. At the extreme, channel collapse occurs where one exchange partner leaves the relationship and structures the transactions internally. Dell Computer switched from selling their products in retail stores to selling them direct to consumers. This is one example of channel collapse where the costs of the distributor-retailer network exceeded the benefits. Likewise, the growing trend whereby smaller physician practices have restructured their organization to become fee-for-service is motivated largely because the transaction costs, tangible and intangible, inherent in the insurer relationship outweigh its benefits. The physicians choose to accept fewer patients, seek out patients using their own resources, and accept greater fee collection risks in return for less bureaucracy and more decision-making autonomy.

Transaction costs include both the ex ante contract negotiation costs and ex post coordination, monitoring and governance procedures. According to Williamson, "... any problem that can be formulated, directly or indirectly, as a contracting problem can be investigated to advantage in transaction cost terms" (1985: ix). The transaction cost analysis begins by examining the context of the relationship. The transaction cost framework identifies frequency, asset specificity and uncertainty as three fundamental environmental conditions that influence an exchange relationship (c.f. Williamson, 1985; Cannon et al., 2000). This framework serves as the starting point to investigate how the relationship structure drives the cost structure.

Frequency refers to transaction repetition and assumes that high repetition will motivate organizations to manage them internally (Williamson, 1985). This hypothesis has not been empirically supported in the channel literature (Rindfleisch and Heide, 1997) and is rarely a component in the channel transaction cost model.

Asset specificity or relationship-specific adaptations tend to create dependency and build switching costs (Cannon et al., 2000). When one exchange partner must acquire assets (human or physical) primarily to adapt to the needs of the other exchange partner, these relationship-specific assets are defined as transaction costs. For example, in healthcare when a physician practice must acquire software tailored to interact with one insurer or hire personnel with the skills to interact with one insurer, then relationship-specific assets have been acquired and they are included as part of the total transaction costs of the relationship.

Environmental uncertainty exists when conditions that affect the transaction cannot be known in advance (Cannon et al., 2000). This makes it difficult to create ex ante contracts that define the relationship. The exchange partners must then create extensive contracts to cover a range of eventualities or rely upon ex post governance to interpret and enforce the spirit of the contract. In the healthcare industry, contracts between physicians and insurers cannot anticipate the needs of every patient; hence, the conditions for payment of services are based on the average patient and diagnosis. Therefore, when patients do not fall into the average range for a particular diagnosis, negotiation and monitoring procedures are necessary to insure adherence with the contract.

We employ the assertions from transaction cost analysis to provide a framework for assessing the environmental characteristics in which the healthcare-insurer relationship thrives. Channel researchers have identified asset specificity and uncertainty as transaction cost drivers in typical channel partnerships (Rindfleisch and Heide, 1997). We expect then to find these transactions costs in the physician practice as one type of evidence that the channel framework can be applied to the physician practice--insurance company arrangement. In this way, transaction cost analysis provides an ex post identification of those environmental conditions that characterize surviving physician-insurer partnerships.

Research Method

Our primary goal is to assess the applicability of a proven theoretical framework to a new domain, If deemed applicable, this framework will serve as a rich foundation upon which optimal solutions to the healthcare funding problems can be formulated. In such circumstances, it is appropriate to use a single case approach to serve as the basis for demonstrating the correspondence (Aherns and Dent, 1998; Atkinson and Shaffir, 1998). Accordingly, our investigation utilizes one physician-owned practice with 8.5 FTE physicians (full time equivalents) who each see approximately 18-22 patients daily. They are one of the largest physician groups in the community. Six insurance providers--half are preferred provider organizations (PPO's) and the other half are health maintenance organizations (HMO's)--represent the majority of their patient base that utilizes employer-provided insurance. Over a two-year period, three in-depth and numerous informal interviews were conducted with the practice manager and Chief Financial Officer. The first round of interviews defined the environment in which the practice operates and assessed the correspondence between public discussion related to problems in healthcare funding and the practice's experiences. Further in-depth interviews elicited discussion surrounding the key channel variables and their importance in the physician-insurance provider relationship. Data from individual patient billing records were collected and several interviews were conducted to insure feasibility of our collection methods and calibrate the conclusions drawn from the data.