6.2 Foundations: Grossman and Hart on insurance companies

In their foundation paper on this subject, Grossman and Hart (1986) used the example of the incomplete contracts for profit-enhancing investments by participants in the insurance industry to explain the allocation of ownership rights to clients. Grossman and Hart consider the choice between ownership of the client list by the insurance company, and ownership of the client list by the insurance sales agent. In their model (and in practice) the insurance company determines the type of insurance policies available for sale, and their retail price. Substantial cost is incurred in the identification of clients and in selling them policies. Insurance sales agents are paid via an up-front commission at the point of sale, and a (smaller) renewal commission when the customer renews their policy in subsequent years. The up-front commission rewards the effort expended on identifying and selling to customers. The presence of the renewal commission incentivizes the agent to select clients who will be persistent (that is, to avoid forced selling) and to provide post-sale support to the client so that the company can cover the substantial fixed costs of making the sale.

In general (motor vehicle, fire, and accident) insurance, policies are usually written for one year, whereas most forms of life insurance involve a long-term contract. This means that general insurance customers have a much greater tendency to switch companies than do life insurance companies. It also means that for life insurance the critical sales effort is in initial selection of the client and matching of the appropriate policy, whereas in general insurance post-sale support to the client is much more important. It also means that for life insurance changes to the competitiveness of the company will not affect retention of existing clients, whereas changes in the competitiveness of the prices of a general insurance company will affect the retention of existing clients. In other words, complete contracts on agent effort and on insurance company competitiveness cannot be written.

Grossman and Hart (1986) show that in the portion of the insurance industry where agent effort in post-sale service to the client is most important, independent agents (agents who sell for a range of different insurance companies and own their client base independent of the insurance firm) dominate sales. In other words, the more important is post-sale effort from the agent, the more likely are the agents to own their client lists. In contrast, in life insurance where post-sale service to the client is least valuable, exclusive agents of individual insurance companies dominate sales, and this in turn means that the client list is owned by the firm. Thus, the structure of insurance firms, and in this case the ownership residual rights of control over the client list, is determined by whether the ownership by the insurance company or the sales agent delivers the largest ex post surplus from the production and sale of insurance.

Rebitzer and Taylor (2007) provide an analysis that is similar to that provided by Grossman and Hart (1986), but their example is law firms, and in particular, explaining why law firms are structured as partnerships with “up or out” rules for promotion from associate to partner. Their explanation is based around the idea that partners of a law firm control critical assets of the firm-specific knowledge about the business needs and interests of long-term customers. The specificity of this knowledge gives individual lawyers an incentive to try to increase their share of firm profits by threatening to leave and take their clients with them. Law firms respond to this threat in two ways. First, they fire lawyers who are not promoted to partner within a specified period of time to ensure that staff below the level of partner do not have the ability to develop long-term specific relationships with key clients of the firm. Second, the partnership structure provides incentives for each partner to refer the client to other specialist partners when the needs of the client require that, and reduces the incentive for any individual partner to leave and take with them the clients with whom they have built specific relationships, by creating an ownership structure in which all partners (and only partners) share in the profits of the firm.

Other professional services firms rely on long-term customer relationships, but as Rebitzer and Taylor (2007:225) acknowledge, not all of them use the same “up or out” tenure and partnership structure used by large law firms. The variation can be accounted for by a number of factors, including the ability of the firm to protect aspects of the relationship with the client through patents on key intellectual property, and the effectiveness of non-compete clauses in the contracts of senior staff. The importance of the reputation of the firm vis-à-vis the reputation of the individual partner or client-relationship manager, and the ease with which clients distinguish the reputation of the individual from the reputation of the firm, are also likely to be critical.

A related literature considers the relationship between delegation, information, and the contribution of different parties to the generation of total surplus in any activity. Short of allocation of ownership to a particular party, delegation of decision rights and compensation packages linked closely to the outcomes of those decisions may provide mechanisms that reduce the inefficiency created by incomplete contracting. When information is publicly available or otherwise easily measured, it is efficient to centralise decision-making. But in the presence of factors that will make performance contracts incomplete, such as the need to adapt to a local environment or employees with valuable private information, delegation of decision-making together with pay for performance may represent an optimal solution (Aghion and Tirole, 1997). Among empirical investigations of this issue, Wulf (2007) finds that the compensation of managers who have broader authority is more sensitive to firm-level performance, and in a survey of sales force personnel, Ghosh et al (2011) find a positive relationship between performance pay and the level of delegation.