Whether you’re making a decision that has apparent, immediate consequences that could affect the rest of your life, like deciding to quit your job and open a business, or making a purchasing decision big or small, it is worthwhile to gather information and think about the future. When you gather information, you have to be careful, because people, businesses, and even statistics that provide that information might contain inherent biases.
Survivorship bias is often hidden in plain sight, but recognizing the bias when it exists could be the difference between building wealth over the long term and going broke. If a logical argument neglects to recognize that failure is a possible outcome, the argument may be subject to survivorship bias. If a statistic reflects the outcome of only the successful set, ignoring the set of those who fail, survivorship bias is at play, and can give someone a false understanding of the data.
David McRaney, the author of You Are Not So Smart explains survivorship bias:
Survivorship bias […] flash-freezes your brain into a state of ignorance from which you believe success is more common than it truly is and therefore you leap to the conclusion that it also must be easier to obtain. You develop a completely inaccurate assessment of reality thanks to a prejudice that grants the tiny number of survivors the privilege of representing the much larger group to which they originally belonged.
It’s probably easier to illustrate survivorship bias than to explain.
Survivorship bias in the stock market.
Mutual fund managers rely on survivorship bias for selling their products. Overall, and over the long term, actively managed mutual funds cannot beat the relevant market benchmark. Considering actively managed mutual funds are more expensive to own than their index counterparts, index-based, non-managed mutual funds are better choices. But mutual fund managers continue to attract investors because they’re able to advertise higher returns.
Money managers can advertise these higher returns overall because poor-performing actively-managed mutual funds are eliminated or merged into other mutual funds. This hides poor performance. As a mutual fund manager, kill the low performers, and when you report performance results for the funds that survive, your track record looks better than it is.
The danger in ignoring survivorship bias in the stock market. Well, it’s all about making the best decisions with your investments. If you are taken in by promises of big financial returns, you may not be putting your future self in the best position possible. When an investment manager says 90 percent of his or her funds beat the market, he or she isn’t including those that no longer exist. And your money could end up in an investment that, someday in the future, will no longer exist.
Survivorship bias in stock recommendation scams.
Step one: build a sufficiently large mailing list. Ten thousand names will do, if the recipients are already inclined towards buying.
Step two: Pick a stock, and send a letter to half the recipients on the list predicting the stock price will go up in the next week. Send a letter to the second half predicting the opposite.
Step three: In the following week, determine which half received letters that correctly predicted the stock’s performance. Delete the other names from the list of recipients.
Repeat steps two and three, each time eliminating the half of the mailing list that saw an incorrect prediction. After a few weeks, a couple thousand people will consider you a stock market prognosticating genius, and each week, you have a better chance of selling something to those who were lucky enough to be on the winning side of each prediction.
This approach isn’t limited to bottom-feeding stock scams. Hedge funds catering to the wealthy often employ a similar strategy by initiating many funds, waiting for performance results, and marketing and reporting to indexes only the successful hedge funds.
The danger in ignoring survivorship bias in stock scams. This should be apparent. If something sounds too good to be true, it is. Don’t invest in stocks recommended to you by a stranger. Or by a friend who thinks he or she has insider information. Or in individual stocks at all unless you are Warren Buffett, can negotiate for a discount, have a say in management decisions, or are conducting an investment experiment.
Survivorship bias in entrepreneurship.
Do you ever wonder why all business owners seem to be successful? In the classic book, The Millionaire Next Door, the authors point out that entrepreneurship is the path to success. This message permeates, and many people believe that owning a business is the best way to grow wealth. While investing as much as possible in broad stock-based index mutual funds can get you to financial independence over a long period of time, barring bad timing as many people who were in retirement during the latest recession might have realized, owning a business is said to be the key to speeding up wealth accumulation.
Of course business owners are better off financially than equivalent employees. If they weren’t successful, they would — and do — stop their pursuit of owning a business. In other words, those who don’t succeed drop out of the statistics.
The danger in ignoring survivorship bias in entrepreneurship. It’s easy to be seduced by the promise of riches that being a business owner seems to hold. Encouraging books and gurus often ignore the risk of leaving a job with a salary and benefits behind in favor of getting a project off the ground. At the same time, they emphasize the risk of staying in a job, subject to random layoffs and other stresses of answering to a boss. According to Bloomberg, eight out of ten new businesses fail within the first 18 months.
The remaining two become part of the statistics that show how worthwhile owning your own business could be, but chances are more likely any new business owner will eventually give up and no longer be counted among the “entrepreneur class.”
Survivorship bias in other aspects of life.
I go the gym three times a week to work out with a personal trainer. And after going this for about a year, not including time away for travel, I’ve made some progress. But I still look around the gym and see that I’m not nearly as in shape as just about everyone else. That’s to be expected, but not because of survivorship bias. Unless you work out constantly, every time you go to the gym, most of the people will be in better shape. The reason is that any person you see is a statistical representation of someone who is most likely to be at the gym at any particular time. And the people most likely to be at the gym at any particular time are the people who go to the gym the most often.
But survivorship bias manifests within fitness philosophies. An intense program can report high success rates for two reasons. First, the people who are most likely to fail don’t start in the first place. Second, people who being the plan but are destined to fail leave, and they are no longer included in the group of followers of the fitness plan.
Diets follow the same pattern. This phenomenon could also be considered selection bias, but selection bias is also important to marketers. If they can stop the “wrong” customer — someone not likely to succeed — from buying their self-help products initially, the seller can continue to report good success ratios with a clear conscience.
The key is to always ask questions.
I tend to be more skeptical than people I know. I don’t trust statistics immediately. I look to the source of information, and I try to think about logical holes, fallacies, and biases before making decisions. I may not always be perfect and may not always be the smartest person in the room, but recognizing biases has helped me make better decisions with my life.
If you’re only looking at successes for modeling your behavior, you’re missing an important piece of information. It’s quite possible that those who failed employed the same strategies as those who succeed. It’s also possible that failures vastly outnumber the successes, and are just not reported as such. So if you want to avoid failure, pay attention to potential survivorship bias.
Photo: Universitat Stuttgart (a tardigrade, Earth’s survivor)
Published or updated April 16, 2014.