25 Oct 2004  Research & Ideas

Planning for Surprises

A company doesn't need a crystal ball to see impending disasters. Harvard Business School professor Max H. Bazerman and INSEAD professor Michael D. Watkins explain how to foresee and avoid predictable surprises.


The train wreck that was Enron's collapse is only one big, blatant example of how some disasters catch us unawares—but shouldn't. In fact, according to Max H. Bazerman and Michael D. Watkins, many surprises in all types and sizes of organizations are predictable and avoidable. Predictable surprises, say Bazerman and Watkins, are a common form of leadership failure. "Predictable surprises happen when leaders had all the data and insight they needed to recognize the potential, even the inevitability, of major problems, but failed to respond with effective preventative action," they say. Here's the good news: There are reasons why leaders fail to prevent predictable surprises and there are ways to identify trouble while there is still time to stop it.

As authors of a new book from Harvard Business School Press, Predictable Surprises: The Disasters You Should Have Seen Coming and How to Prevent Them, Bazerman and Watkins recently collaborated on the following e-mail interview with HBS Working Knowledge.

Martha Lagace: What distinguishes a predictable surprise from any event seen with 20/20 hindsight?

Max Bazerman and Michael Watkins: Bad surprises really do happen to good leaders. But there must be a point where we hold leaders accountable for their failure to prevent predictable surprises. There must be a point at which we conclude that leaders have been misguided or negligent or both.

When should we hold our leaders accountable for failing to take actions to prevent predictable surprises and when should we give them a pass? If they have made reasonable efforts to scan the environment and tap into available information, yet still fail to see the piano dropping, we should let leaders off the hook. They likewise get a pass if they do what a reasonable person would do to make prevention of a potential disaster a priority in the face of competing priorities, and to mobilize a response in the face of concerted opposition for special interests.

Far too often, we only address problems after the predictable surprise has occurred.

Q: Why are predictable surprises so common?

A: Our research shows that there are psychological, organizational, and political factors that conspire to keep us from dealing with problems that are worthy of our attention.

Psychological vulnerabilities have to do with well-recognized biases in the way people think, such as self-serving illusions and overcommitment, as well as the tendency to stick with the status quo and to discount the future.

Organizational vulnerabilities arise because of structural barriers to the effective collection, processing, and dissemination of information, such as the division of organizations into independently operating silos and the filtering of information as it passes up through the hierarchies.

Political vulnerabilities contribute to predictable surprises when a small number of individuals and organizations are able to "capture" the political system or organization for their own benefit.

The area of decision bias has grown as an important lens of analysis in many areas of business, from finance to marketing to negotiations. We also believe that cognitive biases explain why we allow predictable surprises to occur. For example, people tend to exist in a state of denial that leads us to undervalue risks. In addition, people overly discount the future, reducing our willingness to invest in the present to prevent some disaster that may be quite distant.

People also try to maintain the status quo, creating a barrier to the dramatic changes that are needed to address predictable surprises. Finally, many are more willing to run the risk of incurring a large but small-probability loss in the future rather than accepting a smaller, yet certain loss now. We don't want to invest in preventing a problem that we have not experienced and can hardly bear to imagine. Thus, far too often, we only address problems after the surprise has occurred.

Q: Your book uses examples of widely known surprises in the public record, including the 9/11 attacks. As a business example, you use what you call the looming crisis of frequent-flyer programs. Why do you see frequent-flyer programs as a "house of cards"?

A: Predictable surprises loom in most organizations. Frequent-flyer programs are simply one example that affects a lot of people. Most of us collect our miles and even think of them as an asset that we can rely on using in the future. We think this is optimism. For many U.S. airlines, the debt owed to customers in the form of miles is a value significantly larger than the airlines' market capitalization. This is simply not sustainable. Airlines have already reduced the value of miles by making seats less and less available. But, a larger predictable surprise still awaits. We stick by our recommendation: Use your miles!

Q: What other looming, predictable surprises have you noticed since the book manuscript was completed?

A: The flu vaccine crisis is a classic example of a predictable surprise. As The New York Times noted in the October 17, 2004 Health section, "The shortage caught many Americans by surprise, but it followed decades of warnings from health experts who said the nation's system for vaccine supply and distribution was growing increasingly fragile." There have been numerous disruptions in the supply of vaccines for flu and other diseases. In addition to disruptions with the flu supply in each of the past four years, the Times notes that there have been shortages in eight of the eleven vaccines for childhood diseases in the U.S.

Why were we vulnerable to this predictable surprise? Because the U.S. has become dependent on just two suppliers, while Great Britain has five suppliers to reduce the risk of supply disruptions. While fear of lawsuits has played a role, the fundamental problem is that the economics of vaccine production are unattractive for pharmaceutical companies in comparison to other opportunities. So the private sector has responded rationally, by exiting the business. Of the twenty-five companies that made vaccines in the U.S. thirty years ago, only five remain today. The Federal Government has, regrettably, permitted this to happen by not treating vaccines as a critical public good and providing appropriate subsidies for their production. One contributing problem is the absence of a single agency responsible for overseeing the U.S. vaccine supply.

Q: You write in the book, "By deliberately assuming a veil of ignorance, individuals can learn to see beyond themselves and more effectively ward off a predictable surprise." What's a veil of ignorance and how can it help executives avoid predictable surprises?

A: American philosopher John Rawls developed the concept of a veil of ignorance to encourage us to think about what the situation would look like without the partisan perceptions that we bring to most issues. We always have our own views. But the key issue here is the difference between hypotheses and assumptions.

In practice it is difficult to assume complete ignorance. But it is essential to try to be conscious of the assumptions you are making about what is possible and critically what is not possible. To the extent that you can treat these as hypotheses to be rigorously challenged and tested, rather than as assumptions that are taken for granted, you reduce the potential to be predictably surprised. People always learn with a point of view; the key is to be open to altering that point of view in the face of reality.

For example, we question whether the Bush administration objectively considered the issue of weapons of mass destruction (WMD) in Iraq from an objective standpoint (under a veil of ignorance), or whether the administration started with a partisan perception and then avoided any reasonable challenge to this view. Richard Clarke, the former counterterrorism czar under Former President Clinton and President Bush, made a compelling case that the administration never escaped their desire to see the data as they wanted to see it.

Q: At the organizational level, you make the point that special interest groups can really interfere with efforts to avert a predictable surprise. What is the power of special interest groups?

A: Any time reform would yield broad but modest gains and deep but narrow costs, you can expect the potential losers to become strongly mobilized to oppose any change. They can often do this successfully until the wall begins to collapse and a crisis allows leaders to overcome their resistance.

Far too often, a select group is successful in lobbying the U.S. Congress to the detriment of society. In our book, we document the dysfunctional role of the airlines in weakening airline security throughout the '90s and into the twenty-first century. We also show the dysfunctional role of the lead auditing firms in keeping auditor independence from becoming a reality. Our book is another call for meaningful campaign finance reform in this country. Until this occurs, special interest groups will continue to cost society in terms of the loss of jobs and the growth of the deficit, and by extension in the lives of citizens.

Waiting for Your Miles to Explode

by Max H. Bazerman and Michael D. Watkins

In the aftermath of September 11, 2001, U.S. airlines faced significant financial challenges. Many people were newly afraid to fly, and many remain so to this day. With the 1990s financial boom over, the number of business flyers fell dramatically. The airlines lost billions of dollars in equity after 9/11, United Airlines and US Airways filed for bankruptcy, and many other airlines continue to struggle for their economic life. We expect that most of the major U.S. airlines will survive in some adjusted form. But eventually, another crisis will hit: the billions of dollars of debt that airlines owe their customers in the form of frequent-flyer miles.

The majority of the examples in this book have concerned governments and very public corporate failures. Our selection was partially based on the availability of this data, and partially due to our desire to present vivid stories that readers already know something about. Early on, we argued that predictable surprises await most organizations. However, we do not have the right to publicly announce predictable surprises based on private discussions. Thus, future corporate predictable surprises have been underrepresented in our examples in relation to governmental crises. Yet, in some cases, information about future corporate predictable surprise can be found in public sources. The future explosion of the airlines' frequent flyer programs is one such example.

In 1981, American Airlines introduced the most innovative marketing program in the history of the airline industry. The plan stated that flyers could earn miles for the flights they took and redeem these miles for travel awards. From a myopic perspective, this appeared to be a brilliant marketing strategy. The idea was to give flyers a unique reason to choose American. From a long-term financial perspective, it created a predictable surprise that is destined to explode.

Following American's lead, other major U.S. airlines hurried to create their own frequent-flyer programs. Soon they were offering double miles to their most frequent passengers, miles for hotel stays, miles for car rentals, miles for credit card use, etc. The expansion of benefits required to remain competitive in the industry resulted in significant liabilities. By 1987, various newspaper analyses estimated that the airlines owed their passengers between $1.5 and $3 billion dollars in free trips.

The programs' structures gave passengers an incentive to be frequent flyers of the airlines that flew to the largest number of cities—United and American. Despite the mounting long-term debt, American and United Airlines' frequent-flyer programs were paying off in the short-term with small gains in market share. As a result of losing ground to United and American, Delta Air Lines announced on December 15, 1987, that all passengers who charged their tickets to American Express would get triple miles for all of 1988. Delta assumed that passengers would switch their loyalty to Delta in exchange for triple miles. Instead, as Delta should have predicted, all of the major airlines announced that they too were offering triple miles, as well as a number of other new benefits.

The escalation of competition among frequent-flyer programs continued. Airline debt rose significantly, with some estimates placing it as high as $12 billion by 1990. 70 As early as 1988, Mark Lacek, then director of business-travel marketing at Northwest Airlines, lamented the escalation of mileage promotions: "It's suicide marketing. Insanity." 71 Around the same time, Fortune magazine concluded, "In the annals of marketing devices run amok, few can compare to the airlines' wildly popular frequent flyer plans." 72 Yet for the next fifteen years, the problem continued to grow.

Each airline has a marketing benefit from having an attractive frequent-flyer program. But when all of the airlines have such programs, more and more flyers redeem miles for their tickets rather than paying in cash. The systems have significant operating expenses, and, most importantly, the long-term debt to customers is staggering. Collectively, all airlines have been made worse off by the frequent-flyer war. Over the past quarter century, the air carriers escalated frequent-flyer competition and made matters worse for all companies involved.

Over the years, the airlines have succeeded in making your miles less and less valuable. They have increased the number of miles needed for many routes, decreased the number of seats available for miles redemption, created "blackout" periods during which miles cannot be redeemed, and invented a host of other methods for reducing the value of miles. Despite their efforts, the debt owed to consumers remains enormous. Estimating the total debt that the airlines currently owe their customers is a tricky accounting question. It is made even trickier by the airlines' massive campaign contributions to politicians (even during bankruptcy) to influence accounting laws that allow them to understate the true amount of debt. In its broadest sense, the timing of when a contingent liability (the technical accounting term that describes the mileage awards owed) should be recorded is associated with the matching principle. That is, the liability (and corresponding reduction in income) should be recorded in the same period that any associated revenue has been earned. But what amount should be recorded as a liability? This is perhaps the biggest debate surrounding the recording of a contingent liability.

Like many contingent liabilities, reporting of the airlines' frequent-flyer obligations is made difficult by subjectivity, the different ways that the contingent liability can be assessed, and the desire of the airlines to understate their debt. There is no sure way of knowing whether and when members will redeem their accumulated miles. In addition, for passengers who do end up using their miles, it is unclear how they will redeem them. Aside from traveling for free, travel awardees can opt to use the miles for first-class and business-class upgrades and for other benefits such as hotel, rental-car vouchers, and consumer products.

We can say for sure that there is a strong market for frequent-flyer miles. Airlines sell hotels, car-rental companies, and credit-card companies miles that they can distribute to their customers. Airlines also sell miles to corporations to be distributed as they see fit. As of 2003, the going rate per mile fell to between one and two cents. Using the conservative one-cent-per-mile figure, when a customer has 25,000 miles in her frequent-flyer account, an airline would owe her $250 of value. But because of the uncertainty described earlier, it could be argued that this figure should be adjusted downward: Not everyone will redeem his or her miles. In addition, more and more passengers are using their miles in ways that create more than one-cent-per-mile of benefit.

So, how do the major airlines account for unredeemed miles? First, they ignore the miles until they reach blocks of 25,000 miles for a specific individual (they act as if 24,000 miles has no value to you). Then, for each 25,000-mile block of unredeemed miles, they enter a liability; that is, they recognize that they owe some amount for the future use of that block of 25,000 miles. As of 2003, for Delta, Northwest, and US Airways, the liability entered ranges between sixteen and twenty-five dollars per ticket, to account for the cost of writing the ticket and the meal you will eat on the plane. 73 These airlines totally ignore the possibility that you or your company might have paid for a ticket with real dollars if the free miles did not exist. Thus, they are selling 25,000 miles for an amount between $250 and $500 and listing the liability at an amount less than twenty-five dollars. Because the airlines do not provide their rationale for the value they put on these liabilities, it can only be inferred. We must conclude that the airlines are hiding frequent-flyer debt. When we use the conservative estimate of $250 per ticket, and accept the airlines' deceptive procedure of ignoring mileage that is not in 25,000-mile chunks, the debt that the five largest U.S. airlines (American, United, Delta, Northwest, and US Airways) owe their customers exceeds $10 billion. Obviously, if we use the figure of $500 per ticket, the total debt exceeds $20 billion. Less than $2.5 billion is recognized on the airlines' books. To put this $10 billion in perspective, as of August 12, 2003, the combined value of all of the stock for all five airlines fell well below $4 billion.

Eventually, most of us will decide to redeem our miles. When we do, a predictable surprise will emerge. The airlines will need to default on the debt they owe us and face legal action, or their revenue will crumble dramatically. Meanwhile, the airlines have been effective in lobbying to keep the real debt off of their books. If the real debt were listed, the predictable surprise would occur even faster.

We have spoken to many airline executives involved in the mileage programs. These executives are happy to explain the benefits created by the frequent-flyer programs, such as customer loyalty. They are also pleased with their ability to generate revenue by selling miles to hotels, car-rental companies, and so on. What they choose to ignore is that customers' loyalty to other airlines creates a real problem for them. They also prefer to ignore the huge debt that is mounting day by day, as customers rack up more and more miles. Anxious to justify their failed course of action, they ignore the strong evidence that frequent-flyer programs are a multi-billion-dollar mistake created by a series of irrational decisions. Like so many other predictable surprises, this motivated irrationality keeps executives from identifying their missteps and developing a more rational strategy for the future. Finally, the airlines have lost sight of their mission: To make a profit for their shareholders. Instead, they have falsely defined their goal as beating the other airlines. But being the best of a series of bankrupt firms is not a winning strategy.

Long ago, the airlines should have reduced the costs of their frequent-flyer programs. Promotions that are sure to be matched by competitors benefit only the customer. American should have realized when it created its program that other airlines were likely to follow suit. Delta should have considered the likely response of its competitors before announcing triple miles, and all airlines should have considered how to manage the decisions of the other parties in the industry. Instead, the airlines have chosen to bolster short-term financials and use deceptive accounting to hide their future problems. This strategy has worked for a while, but eventually the house of cards will collapse.

Our advice: Use your miles before they explode!

Reprinted with permission of Harvard Business School Press. Predictable Surprises: The Disasters You Should Have Seen Coming and How to Prevent Them by Max H. Bazerman and Michael D. Watkins. Copyright ©2004 Harvard Business School Publishing Corporation. All Rights Reserved.


70. The number of frequent-flyer miles had accumulated to $800 billion by 1990, according to Barron's. 20,000 miles are typically redeemable for a free flight within the continental United States. If we value a free flight anywhere in the United States at $300 dollars, the estimated cost to the airlines in 1990 stood at approximately at $12 billion. A. Zipser, "Sky's the Limit? Frequent-Flier Programs Are Ballooning out of Control," Barron's 70, no. 38, September 17, 1990. 16-17, 44.

71. "Free-for-All in the Skies," Time, March 7, 1988.

72. "Cutting Back on Flier's Freebies," Fortune, June 6, 1988, 149-152.

73. For comparison, the imputed average value for United Airlines is sixty-eight dollars per ticket and $118 dollars for American Airlines.

About the author

Michael D. Watkins is Professor of Management Practice at INSEAD, and founder of Genesis Advisers, a leadership and strategy consultancy.