Annual US expenditures on business entertainment likely exceed $40 billion. Such “wining and dining” is often viewed with suspicion, as a way for one entity to influence another’s decision makers improperly. Indeed, such concerns often lead governments and other organizations to limit what kinds of meals and other gifts employees can receive (consider, e.g., 5 Code of Federal Regulations Part 2635). Yet, at the same time many businesses encourage and permit their employees both to wine and dine and to be wined and dined. Why?
The rationale might, at first, seem obvious: firms expect to generate more business, better cooperation, or other profitable quid pro quo from those being wined and dined. But this begs two questions: first, why do firms let their own managers be the beneficiary of such largesse if its purpose is to induce their managers to pursue actions desired by other firms? And, second, if firms benefit from better cooperation and otherwise smoother business relations, why do they not directly provide their own managers with the necessary incentives to cooperate?
A rationalization for wining and dining can be offered in an environment in which there are (i) no tax advantages to the practice; (ii) firms, if they wished, could directly provide incentives to their managers that would induce the same level of inter-firm cooperation as wining and dining; and (iii) wining and dining may be subject to abuse (e.g., expense-account fraud). All that notwithstanding, wining and dining nevertheless proves a low-cost way for firms to provide incentives for inter-firm cooperation. Indeed, if there is no potential for abuse, firms can achieve a first-best outcome (i.e., maximum efficiency with the firms’ shareholders capturing all the benefits that come from doing so). Even when abuses can occur, firms still do better when they facilitate wining and dining than they would if they didn’t.
Why is this? The key insight is that although shareholders (the firms’ owners more generally) can provide direct incentives, those incentives are expensive because shareholders cannot directly observe whether their managers are behaving cooperatively. This asymmetry of information would permit managers to capture so-called information rents (remuneration strictly above that necessary to compensate them for the actions they take). In contrast, a manager directly observes whether her counter-party is being cooperative. Consequently, in the quid pro quo implicit contract between themselves, the managers can avoid paying information rents. In other words, the shareholders can take advantage of the managers’ ability to monitor each other to get around their inability to monitor the managers directly.
Does this mean we should never be concerned about wining and dining? Of course not
What about potential abuse? In the real world, there are at least two dangers: one, the firms’ managers could collude between themselves to wine and dine each other regardless of whether cooperation is immediately needed; two, managers could simply pocket funds allocated for wining and dining other firms’ managers (this is no ideal concern, 14.5% of embezzlement cases involve fraudulent expense reimbursement). If the first danger is present, but not the second, then shareholders remain better off permitting wining and dining if they value cross-firm cooperation highly enough (specifically, would wish to induce cooperation via direct incentives in the absence of wining and dining). Intuitively, even if the managers collude to wine and dine regardless of the need to facilitate cooperation at a given point in time, over time they will still find themselves needing to cooperate. The fact that there is wining and dining means the information rent attached to such cooperation is less than it would be absent wining and dining, which means the shareholders are better off.
Likewise if the second danger, possible embezzlement, is present, but not the first, then shareholders are also better off permitting wining and dining (despite the potential for embezzlement) if they value cross-firm cooperation highly enough. Basically, the shareholders of one firm can utilize the manager of the other to help them police against embezzlement: the other firm’s manager, because she would not be wined and dined when she should based on her cooperation, has an incentive effectively to blow the whistle on embezzlement by withholding future cooperation, to the detriment of the embezzling manager.
What if both dangers are present? That is, what if managers are freely able to collude and there is zero direct deterrence of embezzlement (firms can never catch embezzlers)? In that case, wining and dining ceases to be of benefit. Yet, it doesn’t cost firms either: in this scenario, funds for wining and dining effectively represent an increase in managers’ base salary, so firms can reduce the base salaries by that amount. In other words, wining and dining simply becomes an alternative form of compensation (moreover, if tax advantaged, it could be a cheaper form of compensation).
Does this mean we should never be concerned about wining and dining? Of course not, but what this article reveals is that suspicion need not be the only reaction nor that letting one’s employees be wined and dined is without benefit.
Additionally, just as ‘The Market for “Lemons”’ by George Akerlof isn’t simply about used cars, this analysis isn’t solely about wining and dining. Rather, it has a more general message about how principals (e.g., shareholders) induce their agents (e.g., managers) to cooperate when such cooperation is personally costly for agents to provide, but beneficial to the principals. The key insight is the principals can leverage the agents’ better ability to monitor each other to obtain outcomes that are more profitable to them.