Written by Survivorship bias is a cognitive shortcut that takes place when the successful or surviving part of a group is mistaken for the whole group, due to the invisibility of the group’s failures. A common example of survivorship bias is the assumption that older buildings and architecture were much more durable and stable as suggested by the common saying “they don´t make them like they used to”. This assumption fails to acknowledge that only the sturdier buildings have survived into the present while the rest, the majority, have been destroyed or replaced. A contributing factor to this erroneous belief is the fact that we haven’t experienced the durability or, in this case, the survivability of modern buildings: presently, we are surrounded by both good and bad quality construction; the former will be preserved in time while the latter might not, just like with the older ones.
When we “miss what we’re missing” is how author David McRaney describes the survivorship bias. Indeed, if failures are invisible, successes are in the spotlight, and we not only fail to acknowledge that the failures might have held useful information, but we fail to acknowledge their existence altogether.
This bias is harmful due to how common it is and how easily it affects decision making. Furthermore, it affects a myriad of sectors ranging from business and finance to science, and even medicine. During the Covid 19 pandemic for instance, healthcare systems struggled to keep up with testing which might have skewed survival and death rates. Medical studies are also more often performed on stronger and younger patients who survive initial diagnoses, as weaker patients are less likely to survive long enough to participate in them, leading to overestimations of successful outcomes.
When thinking about research, it is obvious how error-inducing this bias is. Indeed, to be effective, research must be thorough and take into account as many variables as possible. More specifically, statistics rely on surveys and analysis of populations and, to be accurate, they have to put together groups that fully represent them. The survivorship bias skews researchers into only looking at a subset of those populations, leading to incomplete research. Similarly, when making decisions without analyzing all the available data, individuals will automatically not be making the best choices for themselves.
Background
A study that took place during WW2 has become the prototype of survivorship bias.
For context, Abraham Wald was born in 1902 in what was then the city of Klausenburg in the Austro-Hungarian Empire, today’s Cluj-Napoca in Romania. He developed an interest and talent for mathematics and went on to study the subject at the University of Vienna. He later moved to the United States to work at the Austrian Institute for Economic Research. Then, during World War II, Wald joined a classified program that assembled statisticians to focus on military research and strategy to help in the war, the Statistical Research Group (SRG).
At the time, the military came to the SRG with data on the placement of enemy bullet holes on planes that had come back from battle, represented by the red dots in the image below. The first conclusion reached was to install more armor in the areas where the planes were getting hit the most. However, Wald pushed the group to do the opposite: since the planes being analyzed were the ones that had come back from battle, the areas in more need of protection were the ones without apparent bullet holes, the ones where the planes that did crash must have been shot. Thus, the missing bullet holes were on the missing planes. This is where the notion of survivorship bias was first coined. In fact, the decision to reinforce the areas of the planes ridden with bullets failed to consider that the planes being looked at were the ones that made it back safely, the ones that survived. While the others’ perceptions had been distorted by the survivorship bias, Wald overlooked it and was instrumental in the reinforcement of the aircraft.
Had it not been for him, the group would have made a major mistake despite the stakes being so high, which illustrates how much bias affects decision making.
Survivorship bias in the business world
Survivorship bias has also crept into the business and finance environment and is apparent in various situations.
The first instance is the glorification of successful businesses and people. Every now and then, we hear an inspiring story about how some college dropouts became millionaires. Concrete examples are Steve Jobs, Mark Zuckerberg, Bill Gates, all of whom quit university and went on to become part of the richest people on the planet. Their fame has made them into inspirations and examples to follow. However, chances of becoming a millionaire after dropping out of college are rare. In fact, according to Ramsey Solutions’ National Study of (American) millionaires in 2024, 88% of millionaires graduated from college. Furthermore, the success of the examples above tends to be attributed solely to hard work, when in reality, for every successful college dropout, there are thousands who are not as lucky despite equivalent ambition. Moreover, variables such as luck, timing, networks and socioeconomic background also play a significant part in the path to success.
A similar example involves what are called “unicorn start-ups”. This term, coined by venture capitalist Aileen Lee in 2013, refers to a private startup company valued at over one billion dollars. Examples of unicorn start-ups are Uber Technologies Inc, Airbnb and Space X. People venturing into the business world often strive to one day find or create start-ups as the latter, in particular unicorns like the ones above, are viewed as the archetypes of success and entrepreneurship. At the same time, according to Forbes, 8 in 10 startups will fail within the first year of operation and unicorn start-ups got their name from their statistical rarity.
Looking up to and trying to emulate success stories is an example of survivorship bias and its consequences. Firstly, it drastically limits the knowledge and awareness needed to have a chance of actually succeeding by leaving out important voices, the voices of failures which are vital in understanding successes. To quote author David McRaney again, “The advice business is a monopoly run by survivors”, only their advice and stories are deemed relevant. Secondly, it leads to overly high degrees of optimism which can influence risk-prone decisions. Finally, it suggests causation from correlation by creating the illusion of certain patterns: dropping out of college does not necessarily put you on the path to becoming a millionaire even though a few millionaires did so.
Studies on mutual funds are perhaps the most famous example of survivorship bias in the business world. A mutual fund is an investment fund that pools money from investors to purchase stocks, bonds, and other assets and securities. When looking at mutual funds, studies tend to only include ones that currently exist and fail to show data on funds that no longer do. Funds cease to exist in the case of mergers and acquisitions but also during restructuring and poor performance. This failure to count lost funds leads to misleading positively biased results that do not actually depict the returns realized by all mutual funds, since funds that close cause a negative return that is not considered.
Finally, marketing campaigns can also transmit biased information. Indeed, many rely on attractive figures in terms of client satisfaction and durability of the product: “90% of people loved the product!”. These figures are not necessarily biased or false, but it is important to look at their sources and the factors they consider: these include the sample size and composition and for how long the product was used. For instance, the study might have been set up for success by only using the testimonies of regular and loyal customers.
How to avoid the bias
Knowing about its existence and understanding how it can influence and impact our judgement is already a huge step in trying to avoid bias. Being selective of data sources, always striving to see the bigger picture and practicing critical thinking are other ways of fighting against it. Since it is present in so many different situations, awareness of the bias can already lead to better and more informed decisions, from financial investments and ventures to medical and scientific conclusions, but also common opinions and values.
Conclusion
Survivorship bias is omnipresent in our everyday life, impacting our decisions and opinions.
However, it is not the only bias and many others like the anchoring, availability and confirmation biases also guide our conduct every day. Although it is impossible to be immune to them altogether as they are unavoidable cognitive occurrences, being aware of them and their significance is enough for a more informed point of view in a variety of subjects and, in particular, decisions as an economic agent.
Marta Nascimento
Sources:
“Survivorship Bias – the Decision Lab.” n.d. The Decision Lab. https://thedecisionlab.com/biases/survivorship-bias.
Team, Cfi. 2024. “Survivorship Bias.” Corporate Finance Institute. May 24, 2024. https://corporatefinanceinstitute.com/resources/career-map/sell-side/capital-markets/survivorship-bias/.
Penguin Press. 2018. “Abraham Wald and the Missing Bullet Holes – Penguin Press – Medium.” Medium, June 17, 2018. https://medium.com/@penguinpress/an-excerpt-from-how-not-to-be-wrong-by-jordan-ellenberg-664e708cfc3d.
Solutions, Ramsey. 2024. “The National Study of Millionaires.” Ramsey Solutions. October 3, 2024. https://www.ramseysolutions.com/retirement/the-national-study-of-millionaires-research#:~:text=Eighty%2Deight%20percent%20of%20millionaires,38%25%20of%20the%20general%20population.&text=And%20over%20half%20(52%25),13%25%20of%20the%20general%20population.
TEDx Talks. 2015. “Missing What’s Missing: How Survivorship Bias Skews Our Perception | David McRaney | TEDxJackson.” https://www.youtube.com/watch?v=NtUCxKsK4xg.
Gratton, Peter. 2024. “Survivor Bias Risk: What It Is, How It Works.” Investopedia. September 18, 2024. https://www.investopedia.com/terms/s/survivorship-bias-risk.asp.
Peachman, Rachel Rabkin. 2024. “America’s Best Startup Employers 2024 Methodology.” Forbes, March 8, 2024. https://www.forbes.com/sites/rachelpeachman/2024/02/21/americas-best-startup-employers-2024-methodology/#:~:text=Anyone%20who%20has%20worked%20at,fail%20in%20the%20long%20run.
theawarenessnews.com 