The Cost of Being Wrong Is Less Than the Cost of Doing Nothing

In 1975 an engineer at Eastman Kodak named Steve Sasson built the first digital camera, a clumsy device the size of a toaster that recorded black and white images on a cassette tape. Kodak's managers, who earned most of their profit from film, are said to have told him to keep the invention quiet. The company held on to its film business for another two decades, and in January 2012 it filed for bankruptcy protection. Kodak did not fail because it tried something and was wrong. It failed because it chose to do nothing, and the topic claims that this is the commoner and costlier error.

The reasoning begins with an asymmetry in how we see mistakes. When a decision turns out badly, the loss is visible and the person who made it can be blamed. When an opportunity is missed, no one sees the loss, since the world in which the chance was taken never existed. Psychologists call this omission bias, the tendency to judge harmful actions as worse than equally harmful inactions. Because of it, individuals and organisations drift toward doing nothing, and they pay a hidden price that never appears on a ledger.

Business supplies many stories. Blockbuster, the video rental chain, is reported to have been offered the chance to buy the young Netflix for fifty million dollars around 2000 and to have declined. A decade later Blockbuster was bankrupt and Netflix was worth many times the sum. Clayton Christensen showed in The Innovator's Dilemma that well run firms often fail precisely because they listen to their present customers and avoid risky new markets, and so are overtaken by entrants whom they judged too small to matter. Caution turns out to be the most dangerous strategy.

Public life offers the same lesson. When the Covid pandemic began in 2020, countries that acted early on testing and isolation generally fared better than those that waited for certainty. In the United States, Operation Warp Speed committed billions of dollars to vaccine manufacture before the trials had finished, accepting that some of the money might be wasted, and vaccines were available within a year, far faster than in any earlier case. The money wasted on candidates that failed was small beside the lives saved by speed. Waiting for perfect information would have been the expensive choice.

Waiting for perfect information would have been the expensive choice.

There is also a case from learning itself. People who try things and fail learn more than those who never try. The story that Thomas Edison tested thousands of materials for a lightbulb filament, though doubtful in its details, expresses a truth about experimentation. Each failed trial narrowed the search. A firm that runs many small experiments and abandons the ones that fail gains information at a low price, while a firm that debates for years pays the price of delay and learns nothing. Jeff Bezos has described the habit as making decisions quickly with about seventy per cent of the information one would like.

Personal lives follow the same logic. A student who delays applying to a course until he is sure, a writer who never shows a draft and a citizen who does not vote because no single vote decides anything all pay a cost that cannot be seen. Years later each finds that the waiting has set into a habit. Research on regret bears this out, since studies by Thomas Gilovich and Victoria Medvec in the 1990s found that over the long run people regret what they did not do more than what they did. Mistakes of action can be explained, repaired and forgotten, while omissions keep presenting themselves as possibilities.

Yet the topic cannot be accepted without limit, because some wrong decisions cannot be undone. On the morning of 28 January 1986, the Space Shuttle Challenger was launched in freezing weather despite warnings from engineers at the contractor Morton Thiokol that the rubber seals in its boosters might fail in the cold. Seven astronauts died. The pressure to keep to the schedule, and an unwillingness to accept a delay, were the causes of the disaster, and the cost of the delay would have been a few days. Here the cost of being wrong was incomparably greater than the cost of waiting.

Medicine knows this well. In the late 1950s the drug thalidomide was sold in dozens of countries as a mild sedative and treatment for morning sickness, and it caused severe birth defects in thousands of children. The United States was spared largely because a regulator, Frances Kelsey at the Food and Drug Administration, refused to approve it without more data despite pressure from the manufacturer. Her inaction was the correct decision and her critics were wrong. The precautionary principle, which urges care where harm may be severe and irreversible, rests on the same understanding.

The useful distinction was set out by Bezos in a letter to shareholders in 2016, when he separated decisions that are one way doors from those that are two way doors. A two way door can be walked back through if the room behind it proves unwelcome, and it should be opened quickly and without a committee. A one way door cannot, and it deserves slow and careful thought. Most decisions are of the first kind, though organisations treat them as the second, and so lose time. The skill lies in knowing the door before one opens it.

Even this advice meets a difficulty. It is not always clear which kind of door one faces. A new technology may seem reversible at the start and prove irreversible once society has reorganised around it, as happened with leaded petrol, with the dumping of plastics and perhaps with social media. The costs of being wrong are often discovered long after the decision, when they can no longer be reduced, and the costs of doing nothing may be just as slow to appear. The topic holds in most of the ordinary run of choices, but those cases where it fails are rare and severe, and no rule can tell us in advance whether the choice before us is one of them.