Good money after bad: why do we become overly committed to losing projects?
Bary M. Staw
GOOD MONEY AFTER BAD
JACK WANTS TO BUY A USED CAR. After checking out several vehicles listed in want ads, he invests $3,000 in a five-year-old Chevy. Four days later the car’s transmission falls apart, and the garage mechanic tells him that some other parts on the car also look pretty worn. Jack doesn’t know what to do. Should he spend $700 on a new transmission or sell the car at a big loss and start again?
Jack faces a classic escalation dilemma. He must decide whether to quit an unrewarding course of action or commit more effort and resources to making it pay off. Similar dilemmas occur when one must decide whether to sell a stock that has gone down in price, to quit a job as it becomes increasingly frustrating or to persist in a personal or romantic relationship that is dissatisfying.
Research shows that people often become overly committed to losing courses of action. They throw good money after bad and expend resources beyond the point that would be considered reasonable by more objective outsiders. Organizations, as well as individuals, face escalation dilemmas. When a company finds itself with a money-losing project on its hands, it must decide whether to invest further, hoping for a turnaround, or to cut its losses.
Because some of the worst losses suffered by corporations result from escalation situations, a number of organizational researchers, as well as social psychologists, have been studying the problem. The research has shown how entire businesses can get locked into seemingly hopeless projects. In our view, the factors contributing to escalation in organizations can be broken down into four general categories: project, psychological, social and structural.
The first set of factors involves the project’s objective features, namely rewards and costs. Some projects pay off their initial costs quickly. But many projects, such as long-term development and construction projects, have substantial up-front costs and delayed benefits. These sorts of projects are likely to foster persistence. Managers in such situations often view early losses as investments and are likely to wait patiently before even thinking about withdrawal.
A project’s salvage value and closing costs can also impede withdrawal. If a venture can be closed without large costs and if the owners can recoup some benefits, then the decision to exit is not difficult. Unfortunately, many major projects have all-or-nothing outcomes. Half an airplane or half a building is worth little. And in some cases, after a company considers all the costs of withdrawal, such as payments to terminated employees, penalties for breached contracts and losses from the physical closing of facilities, it may find that it is more expensive to close down a financially troubled project than to continue its support.
While the project factors that discourage quick withdrawal from a losing venture often are fairly evident, the psychological factors that impede withdrawal are less obvious. These more subtle elements influence how information about a course of action is gathered, interpreted and acted upon.
The simplest psychological cause of persistence is reinforcement. Reinforcement theory suggests that managers should withdraw when faced with negative results. But often managers have succeeded in the past precisely because they ignored short-run, negative feedback. Most escalation situations involve intermittent reinforcement schedules. Sales may fall slowly and by fits and starts, typically providing the ray of hope that things will eventually return to “normal.” Thus, a previously successful manager may not pay attention to early-warning signals, getting locked into a losing course of action.
Another psychological reason decision-makers may remain committed to a failing project is that when people believe a project will succeed they often slant estimates of sales and costs in that direction. If the data are ambiguous, opinionated managers may seize upon the facts that support their position.
Even when decision-makers recognize that they have suffered losses they may invest further resources in a project rather than accept what they regard as a personal failure. To protect their self-esteem, managers may self-justify their initial decisions to start a project, spending even more time and money to “prove” it a success. The worse the news is about a project, the greater the desire to justify it. An early experiment by Staw, as well as later replications by others, has shown that those responsible for prior losses tend to persist longer than those without such responsibility.
Social influences provide other reasons for investing more deeply in a losing course of action. One such factor, the desire not to lose face with others, is closely related to self-justification. Managers may persist with failing projects not only because they do not want to admit to themselves that they made a mistake but because they are even more reluctant to admit it to others. In many cases accepting failure means loss of power or even being fired. Research by Staw and organizational consultant Frederick Fox of Minneapolis has shown that job insecurity and lack of managerial support only heighten needs for external approval. When managers become closely identified with a project (“That’s Jim’s baby”) and their fate is inextricably tied to it they can be forced to defend the venture despite mounting losses and even personal doubts.
Commitment to a losing venture can also be fostered by our social perception of leadership. Culturally we tend to associate persistence with strong leadership, as evidenced by popular phrases such as “staying the course,” “sticking to your guns” and “weathering the storm.”
Those who persist and succeed in the face of failure are especially rewarded. Two major heroes in business and politics, Lee Iacocca and Winston Churchill, persevered and ultimately succeeded in apparently doomed situations. Our research shows that perceptions of leadership are enhanced by persistence and that managers who turn losing situations around are held in high esteem.
Project, psychological and social factors often combine to push managers into staying with a decision longer than they should. This combination of forces does not, however, account for some of the most glaring organizational debacles. In many cases, factors within the organizational structure itself are to blame.
Probably the most basic structural factor is administrative inertia. Just as individual attitudes will not always change undesirable forms of behavior (for example, pledges to stop smoking don’t always lead to quitting), organizational preferences don’t always result in action. All of the rules, procedures and routines of an organization, as well as the sheer trouble it takes for administrators to drop day-to-day activities in favor of a major operational disruption, can prevent withdrawal from losing projects.
Withdrawal may also be hampered by the politics of the situation. A recent and highly visible example of this was British Columbia’s decision to stage the world’s fair, EXPO 86, in Vancouver despite huge predicted losses. Originally, EXPO was expected to operate close to the financial break-even point. But as plans for the fair progressed, the projected losses grew dramatically. At first, organizers attempted to minimize the financial problems by providing overly optimistic estimates of revenues and costs. But even when dire financial projections were accepted and the fair’s director recommended cancellation, plans for EXPO did not change. Politically it was too late to call it off. The fortunes of too many businesses in the province were tied to EXPO, it was popular with the voters and the future of the Premier and his political party depended on it. Thus, the province initiated a lottery to cover the expected deficit of $300 million and the fair was held as scheduled.
Once a project is identified with the core values of an organization, support for it can run even deeper than that provided by administrative inertia or politics. Take, for example, the Pan Am Corporation, which, more than most airlines, suffered major losses after deregulation of the industry.
Although the prospects for significant profits in the airline industry were not terribly bright, Pan Am chose to remain in its core business, air travel, and instead sold off most of its other profitable ventures. First the corporation sold the Pan Am building in New York to meet increasing losses and then relinquished its Intercontinental Hotels chain. Finally Pan Am sold its valuable Pacific routes to United Airlines. Following these divestitures, the company was left with only U.S. and international routes in corridors that are highly competitive and often unprofitable. The possibility of selling or closing the airline and keeping most of the other profitable subsidiaries was apparently not considered since, after all, Pan Am is known as an airline, not a real estate or hotel business.
A wide range of forces can lead managers and organizations to persist in a failing course of action. Not all of these forces are relevant to every case, and not all influence the situation equally. In many instances commitment to a course of action builds slowly, fostered first by psychological and social forces and only later influenced by structural determinants. As forces for commitment accumulate, however, it becomes increasingly difficult for an organization to cut its losses.
Holding to original standards can be painful, but it is not impossible. One way to keep the goal in sight is to schedule regular intervals for stepping back and looking at a project from an outsider’s perspective. Managers can encourage their subordinates to reevaluate projects by establishing a climate in which accurate information is conveyed, regardless of whether it is supportive or critical. One forum for increasing candid feedback is a variant of the popular “quality circle.” Key staff members could regularly convene to evaluate the hurdles a project faces and its prospects for success. To encourage objectivity, a manager from another department could be invited to attend or even chair such sessions.
A radical way to correct a losing course of action is to replace those who allowed the original policy or project to go astray. Obviously this tactic can be disruptive and costly. And it may be difficult to draw the line for such purges, since even those indirectly associated with a losing project may be committed to it. A less drastic and probably more beneficial tactic is to assign initial and subsequent decision making to separate groups, as some commercial banks now do. When a loan starts to sour it is handled by a new group of employees, rather than by those who originally funded and serviced the account. Industrial companies could similarly turn failing projects over to a new team.
There are, no doubt, other organizational tactics for reducing excess commitment. But the most general strategy is to move toward an “experimenting organization,” in which every program is reconsidered regularly and every line of business is for sale at the right price. Managers in an experimenting organization would be judged not on their success or failure rates but on how they recognize and cope with the problems facing their units. Movement toward such an experimenting organization would no doubt call for some major shifts in business practices and even in the culture of entire organizations, but it offers the most complete solution to the escalation problems facing industry today.
COPYRIGHT 1988 Sussex Publishers, Inc.
COPYRIGHT 2004 Gale Group