What Is Survivorship Bias?

By: Rupesh Oli

Introduction

Survivorship bias is a form of selection bias that influences people to evaluate the performance of securities as a comprehensive sample without taking into consideration the securities that have exited from the market. Winners are only considered not the losers, thereby leading an investor to fall prey to survivorship bias when taking certain investment decisions. If we try to look at it from the perspective of mutual fund performance, in the majority of cases, the merged mutual funds or the funds that are no longer functioning are discarded when evaluating the mutual fund performance. The same goes with the market performance index, where the stocks that dropped from the whole index for whatever reason are abandoned when evaluating the market performance index. All these instances reflect the survivorship bias where the data set that is currently existing or surviving is only taken into consideration, and the ones which disappeared are overlooked.

Survivorship bias leads to survivorship bias risk, whereby the investors take a misguided investment decision solely based on the published investment fund return data that reflects only the successful funds instead of all the funds. As a result, the failed funds are discarded from the available data set resulting in reported returns being overly optimistic. When researchers fail to consider the complete data set for evaluation, the occurrence of logical error ultimately leads to the wrong conclusion. Hence, it is vital to be aware of the fact that when a certain investor only focuses on the positive aspects neglecting the negative ones, it engenders screwed information.

You May Also Like:

What is Gambler’s Fallacy? How it Impacts Decision Making while Trading?

6 Behavioral Biases In Investing: How To Minimize Their Influence?

Examples

Example 1 – Mutual Fund

Fund	Return	Contemporary Status
A	10%	Active
B	11%	Active
C	-6%	Closed due to Acquisition
D	-8%	Closed due to Poor Performance
E	8%	Active
F	9%	Active

To illustrate survivorship bias more clearly, let’s consider 6 mutual fund returns as of now.

Scenario I: Consider Active Funds only

As depicted in the table above, we have a total of four funds that are still active: A, B, E, and F. If we calculate the average return for these four funds, it would be 9.5% [(10+11+8+9)/4].

Scenario II: Consider All Six Mutual Funds Stated Above

Considering all the six mutual funds, the average calculated return would contrast if compared to the 9.5% we just obtained above in scenario I. As a result, the new average return would be 4% [(10+11-6-8+8+9)/6].

Result: Scenario I Vs Scenario II

In Scenario I, we ignored the survivorship bias as we only considered the mutual fund that is still active. However, in Scenario II, we took all of the mutual funds including the ones closed due to poor performance or acquisitions. So, what we found out is, the average return under survivorship bias is more than half (4%) lesser than the overall average return (9.5%).

This single instance is more than enough to justify the biasness that survivorship bias carries along with it, impacting our financial decisions by quite a hefty margin.

Example 2 – Non-Financial Sector

To justify survivorship bias, let’s consider the impact it had on the medical research sector. A trial was conducted at Harvard Medical School and Beth Israel Medical Deaconess Medical Center in 2010 to improve the rate of survival of patients who were suffering from trauma. In this particular study, they targeted the patients who received 4-8 blood transfusions within 12 hours of their initial injury. The trial aimed to recruit 1502 patients, however, they failed to recruit the target they set and only 573 patients were recruited. It is because of this the trial was abandoned.

The major reason behind the failure of this study was the survivorship bias. The trial only included those patients who survived their initial injury, who then obtained care from the emergency department before they received 4-8 blood transfusions. However, they did not include the ones who died from the initial injury, which acted as the major obstacle to finding suitable patients for the trial, ultimately lowering the target number of patients they previously set for the recruitment.

Therefore, it is crucial to take into account the impact of survivorship bias, whether it be on the financial or non-financial sector.

Impact of Survivorship Bias in Investment Decision Making

As survivorship bias excludes failed companies as they no longer exist, it often leads to higher skewness as only the companies successful to survive till the end of the period are included.

Investors tend to focus on winners omitting the concealed information of losers, the common trait shown by almost every one of us. As a result, investors/traders only consider the investment options that are doing better in the contemporary context perceiving them as the attractive option to invest. Hence, they divert from the businesses/stocks that failed and consider only who survived. This has a major impact when investors try to invest considering the performance of the whole index, as the indices incorporate and depict the overall market performance rather than a particular stock.

Survivorship bias, in general, tends to provide overly optimistic conclusions that might lead to entirely wrong analysis. It might even distort the calculation of returns on stock markets.

It profoundly impacts our perceptions and judgment when it comes to financial decision-making as we, most of the time, will not be able to make fruitful decisions for ourselves as we lack the overall data to make rational decisions.

Not only does it impact on an individual basis, but it also possesses systematic effects as well. It affects high-level decision making resulting in systematic challenges across multiple disciplines. It is because survivorship is one of the common biases and interpretation of data is required in every sector.

It becomes extremely misleading when an investor expects a similar return on their investment based on the company’s past performance, suppose 4 years as of now. There are very minimal chances that a company would perform exactly the same as what is performed in the last 4 years. Let’s connect it to a mutual fund. Suppose a company produced a figure on its average return of last 4 years constituting only 80 funds as they dropped 20 funds because of a merger or poor performance. In this scenario, it would be unwise for any investor to invest based on the last 4 year’s performance as some of the funds were already excluded. In the future as well, the company might launch funds that fail to deliver the investor’s expectation. Hence, expecting the future return solely on past performance is not acceptable.

You May Be Interested In:

What is Sunk Cost Fallacy? How it Affects Investing Decisions?

What Is Prospect Theory?

How to Avert Survivorship Bias?

Embrace survivorship bias by accepting the fact it exists as one of the common biases impacting our decision-making process.

Before delving down into any sort of investment, make sure you collect and examine data and information from only credible sources and try to integrate both qualitative and quantitative data in your analysis.

Make sure you do not become biased by including only the positive aspects of the information you derived from various sources. Hence, consider both the positive and negative sides of the sector you are going to invest in. For instance, when investing in mutual funds, make sure you consider not only the surviving funds, however, also the defunct funds while calculating the average return so that you see the overall scenario more clearly before investing.

Never forget to ask yourself what’s missing from the data set you are analyzing. It might help you figure out the data set that didn’t survive. Once you figure out the loophole of the missing data point, invest your time to research it.

The critical examination of data sources is very necessary in order to derive an accurate conclusion refining your investment decision. Further, ensuring the data set reflects the overall observation and does not miss out on the critical ones leads you to further refinement.

Exploiting Survivorship Bias

I would like to point out one of the perfect instances showcased by Wayne Duggan, one of the reputed writers in the world of investment. However, I will present here with a slight modification. He states, “Scammers have even taken the advantage of survivorship bias to convince people that they’re the skilled stock pickers through the use of simple probability”.

Let’s suppose, on any given day, a person mails about 15,000 different people saying that a particular stock they selected will gain at least 9% in the following week. If a scammer picks about 10,000 different stocks and if only one of every 25 is able to gain 9% that week, it implies 600 people received the accurate predictions.

When the scammer repeats the same process the next week by sending 600 picks to those remaining 600 people, it means 24 people would receive two different stocks that gained 9% in back-to-back weeks.

Ultimately, the scammer will be sending the follow-up message claiming if these 24 people just pay $1000 each week for a stock-picking newsletter subscription, they can become rich in no time. This is how the majority of people fall into a scam. We, therefore, ignore the fact that the majority of people who took the same approach failed and consider only the small percentage of those who took such a risky approach and ended up getting lucky.

What is Recency Bias? How it Impacts the Decision-Making Process?

Does Survivorship Bias Really Matter?

Evan Gilbert and Dave Strugnell in their research paper entitled “Does Survivorship Bias really matter? An Empirical Investigation into its Effects on the Mean Reversion of Share Returns on the JSE Securities Exchange (1984-2006)” came with some meaningful insights regarding survivorship bias.

The presence and significance of survivorship bias were clearly reflected based on the difference of returns between the complete and listed groups of firms. The average returns of currently listed portfolios considering both high P/E and low P/E are significantly higher than those from the portfolios of the complete data set. It is obvious as delisted firms are excluded when we consider only the currently listed portfolios. Further, they mentioned that data cleaning is vital if we want to reach a meaningful conclusion with minimal biases in the output.

Further, any research that excludes delisted shares in their analysis is likely to be subject to the survivorship bias, and the inclusion of data of delisted shares is likely to possess a significant impact on the results obtained. Hence, researchers need to be aware of such biases and include the inclusive data, if possible, when conducting the empirical analysis.

The Bottom Line

To conclude, survivorship bias is one of the common biases that not only an investor, but people of every background possess. Hence, the proper knowledge of survivorship bias and the potential impact it carries is vital so that we could take our decision rationally. Along with it, insights on some of the methods to avert it would be much more beneficial. Predominantly observed in the financial sector, particularly, in the mutual fund, every investor needs to consider survivorship bias from the next time onwards when analyzing the data set leading them to more refined and precise conclusions.

From The Author:

Hot Hand Fallacy and Its Impact on Decision Making Process

What Is Loss Aversion? How It Affects Decision Making Process?

(Liked this article??? If you are also interested in publishing your articles related to business, finance, and economics, then mail us your article at Investopaper@gmail.com.)