In 1989 the investment banking giant First Boston reeled after the junk bond market collapsed, rendering its overleveraged positions in those bonds worthless. Parent company Credit Suisse injected new capital and by the early 1990s First Boston was regaining some of its lost ground.
Most of Wall Street had recovered well enough from the recession of the late 1980s, and bonus compensation at most firms remained robust in the new decade. But neither First Boston nor Credit Suisse seemed willing to justify paying outsized bonuses in 1991, when performance was improving, but not dramatically.
By industry standards, First Boston bonuses that year were small, and executives quieted rumbles of dissatisfaction by assuring its bankers that the bank was on track to return to the flush compensation of years past. But at the end of 1992, First Boston announced a second consecutive year of comparatively modest bonuses, and scores of its top bankers left.
Industry observers and insiders attributed the mass departure to First Boston’s reneging on a promise — no one had signed any papers, but it appeared that the bankers nevertheless felt they had been guaranteed higher bonuses. (This kind of scenario is what Wall Street points to today when asserting that it can’t afford not to pay large bonuses to top talent even in the face of broad outcry against such rewards.)
Informal contracts like the one First Boston’s employees believed they had entered into — stick with us and you’ll be compensated handsomely — are as integral to conducting business as formal contracts are, says Professor Marina Halac, who includes organizational economics among her research interests.
“It isn’t possible to draft a formal contract that can cover all relevant aspects of performance or expected obligations for parties,” she says. “Performance might depend on broad measures that can’t be written into a contract: how do you reward an employee for leadership, innovation, cooperation or initiative? Those qualities are difficult to measure objectively. So it’s not unusual for a formal agreement to be augmented by an informal one.”
Informal contracts are not necessarily difficult to enforce: incentives and threats to the relationship (and its financial value) provide leverage between principals and agents that can ensure a contract will be honored. In fact, if a relationship is particularly valuable, one party can make substantial promises to the other — since walking away from the contract and ending the relationship is costly, this party will have no incentives to renege.
But how can an agent (in the case of First Boston, bankers working on behalf of a principal — the bank) know whether a principal places a high or low value on the relationship when the principal has private information or an outside option that the agent is unaware of? Halac set out to establish a method for determining the value of such relational contracts, analyzing the dynamics of how parties to informal contracts induce each other to reveal clues to private information.
To do so, Halac created a model that reflects the three main aspects of the principal-agent setting. The first aspect is moral hazard, or the fact that the principal cannot observe the agent’s effort, so the principal offers motivation and incentives in the form of rewards that are contingent on performance. The second aspect is that not all performance or output can be measured, so some of these rewards are discretionary and informally enforced (in the case of First Boston, the promised bonuses).
The third aspect, and the innovation on which Halac’s paper rests, is that the principal may have private information or outside options that the agent isn’t aware of. Can a firm close up shop and take its operations elsewhere, or can it replace current employees with others who have similar expertise? The inability to know whether each party places a high or low value on the relationship makes it difficult to gauge whether the promises made through an informal contract are trustworthy. It also makes it difficult to determine what contracts a party will be willing to accept and which ones it will prefer to reject.
Even if counterparties can’t identify the details of such outside options and private information, Halac’s research suggests that it is possible to size up whether a relationship has high or low value. Since informal contracts are usually implemented over time, an agent may have the opportunity to learn the private information of the principal based on the way the principal behaves (or vice versa), akin to how poker players glean information about their competitors’ hands and adjust their subsequent decisions with their competitor’s choices in mind.
Principals and agents usually reveal information through their decision to honor or renege on a contract, or through their decision to accept or reject a proposed contract. For example, a principal might make a promise to pay an agent or to promote him to a higher position in the event that performance is high. If the principal does not renege on his promises, the agent can be reasonably certain that he is facing a principal that places a high value on the relationship. If the principal reneges on some or all such promises, the agent can reasonably assume the principal doesn’t place a high value on the relationship, and there’s a high likelihood that the agent will end the relationship.
One of the biggest factors to consider is how the principal wants to misrepresent its private information to shape the way in which it is perceived by the agent (or vice versa). “A party may have incentives to overstate or understate its private information,” Halac says. “The location of bargaining power plays a key role in this respect: If one party has power, it will probably want to convey that it places a high value on an informal contract. If its power is not that strong, it will probably want to convey that it places a low value on an informal contract.”
For example, a firm may have an incentive to make its employees (or contractors or vendors) believe that it values its relationship with them more than it actually does to provide an incentive for them to work hard or otherwise produce value for the firm — in which case the counterparty may believe that ending the relationship would be more costly to the firm than it really is.
Or, if employees organize and demand higher pay, the firm may want to take the opposite approach: It may not want to reveal that employees are highly productive lest they ask for a larger share of profits through wages. (There is evidence that this happens: research shows that in years when companies are in negotiations with unions, reported revenue decreases, suggesting that firms engage in efforts to make profits look smaller than they are.)
“The model,” Halac says, “can help predict the outcome of informal credit contracts, agreements between firms, supply-chain contracts — wherever informal contracts arise and parties don’t have access to the same information.”
Marina Halac is assistant professor of finance and economics at Columbia Business School.
Professor Halac's research studies dynamic incentive provision in settings with private information, contracting constraints, and learning. Halac teaches an elective course on Game Theory and Business.
Read the Research
"Relational Contracts and the Value of Relationships"