Article by Bivek Neupane
In the previous article, we saw how biases develop and take their final form. But since that article was getting too long, I had to hold my horses a bit. So here it is now.
In this article, we are going to explore what these biases are and how are they actually diverting us to take a much-needed rational decision. But before we go there, I just want to clarify something that I wrote in part 1 of this article. Later when the article was published, it felt as if I missed clarifying some vital points.
I happen to claim that “models beat experts in their own game”. Well, I actually did not claim but rather just pointed out that this is what past studies have shown us. However, this statement can easily be taken out of context. That is why I want to set the record straight. When I say models beat experts, I do not mean that experts are totally unnecessary and worthless. Absolutely not. In fact, they are very important but just not at every step of decision-making. From all the papers and books I have read on cognitive science, it seems like the decision-making process can be more or less divided into 3 broad categories:
- R & D
- Assessment of the outcomes
Where experts are really good and can add significant value is at the 1st and 3rd phases. However, during the implementation phases, human beings should leave things to some form of mechanical system or algorithm that can do the job way better. Now that the load is off my shoulder, let’s talk biases:
- Anchoring: This means we tend to rely massively on a certain anchor to base our final decisions and this anchor is almost all the time “completely irrelevant information”. An example would perhaps help you understand. Simonson and Drolet (2004) did a comprehensive study on this phenomenon. In their research, they constructed a short game in which people were asked their willingness to pay for a variety of products. One of those products was the toaster that you can see in the figure. The rules of the game are also explained there and are pretty simple. Results: People who had their SSNs (last two digits of course) above 50 tended to report a willingness to pay an average price of $32.50 whereas buyers with SSNs < 50 reports $25. Hence, the conclusion was that the last two digits of a person’s SSN actually influence his/her willingness to pay. This conclusion also perfectly applies to other products. Now, the question arises, but how does it affect an investment decision process? Again, let's go through an example.
Imagine a fund manager who is doing a DCF analysis that requires a 5-year revenue or cash flow forecast. Now, construct these two possibilities:
- While in the middle of it, one of his colleague comes into his office and tells that his daughter scored 92 % in school test.
- While in the middle of it, one of his colleague comes into his office and tells that his daughter scored 34 % in school test.
According to the anchoring bias study, this would mean that the manager is more likely to assume a higher growth rate in the first scenario and contrary on the second. This in turn affects the whole valuation process and the ultimate buy-sell decision. But you know what’s the dangerous part? It's that the manager has absolutely no idea that he is being biased. This is because these kinds of biases largely happen on a subconscious level.
I am sure you have heard this line before, “It's not what you said that hurts but how you said it.” Well, with framing, the concept is similar to what this statement is trying to convey. Behavioral psychologists have demonstrated that the way you present information or ask questions to other people actually matter. This trick is actually very widely adopted by marketing and salespeople to make and close the sale. Here’s an example: consider the image below.
Actually, this picture is kind of a well-known one, so I guess you have already seen this one. You are seeing two orange circles. I need you to look at these two circles and tell me which one is larger? If you are a normal person like me with full of behavioral biases, then chances are high that you will say “Well, obviously the circle on the right is larger!”. The correct answer is that both circles are exactly the same size. Because the circle on left is being surrounded by large circles, our brains perceive the circle in the middle to be smaller compared to the large ones on the boundary. The exact opposite happens in the right circle. This, my friends, is called the framing effect.
This effect is also quite prevalent in the investment management business. People like to talk about their returns and how many folds they increased their capital. An equity fund manager who is giving a presentation about why her potential clients should get into business with her would present the known information slightly differently. For instance, this statement is absolutely true: “Equities provide a significantly higher risk premium in the long run. In fact, U.S. markets have earned about 9.9% per annum over the sample period of 1927-2013.” However, her potential clients are more likely to be dazzled by this statement: “Equities provide higher risk premium in the long run In fact if you had invested $100 in U.S. markets in 1927, that $100 would be about $370,000 until 2013.”
Do you see my point? Both statements are actually conveying the same information. But having the dollar amount in the sentence captures the attention way faster than a mere 9% figure.
- Availability bias
Sometimes also known as recency bias, it is the tendency of human beings to overemphasize the events that have occurred recently. I personally think this is that one bias that we as investors are most prone to. For example, if a certain stock or an asset class is having a good run (both fundamentally and price-wise), then our mind starts to develop a narrative that this is going to go all the way and thereby leading us to ignore the historical risk premiums and empirical results. This kind of biases is actually one of the explanations behind why value investing strategies work. Usually, people tend to extrapolate the good earnings too far into the future and when the earnings result does not match the expectations, the stock experiences a huge sell-off from which value investors can capitalize. This was exactly what had happened in the massive tech bubble of the 2000s. The expectations from the investment public towards the internet stocks were so much that those firms just could not keep up with earnings. (Well, to be honest, not only those firms were not generating expectation matching earnings, they were generating negative earnings every quarter.)
- Physical state
“How were you feeling when you got out of bed thirteen years ago, when you’re looking at historical simulations? Did you like what the model said, or did you not like what the model said? It’s a hard thing to back-test.” --Jim Simons, CEO, Renaissance Technologies, LLC
This might be the most simple and intuitive out of all the biases. I guess we all have experienced this one many times. This is so simple that it is often overlooked in academic literature. When you are extremely hungry, how do you start to react? Do you think you can make a rational decision at that state? I don’t know about you, but I definitely cannot. This is just one example. There are hundreds of these scenarios where we might be more susceptible to make mistakes than others. Another such example that comes to mind is making some decisions immediately after you have woken up or before going to bed very late at night. I don’t think this requires much explanation since it is fairly intuitive.
So, this was a slight introduction to behavioral biases that are within us. Starting from the next article, I am going to write series of articles about how academic finance research is used by practitioners around the world and how it has helped in shaping an industry that is now worth more than USD 5 trillion. This would perfectly fit the research I am doing right now on risk premiums on Nepalese capital markets. Until then, let’s stay rational (at least, let’s try..!).
Bivek Neupane is a MSc. Finance and Economics student. He is also a CFA candidate. He is specializing in quantitative finance and research. His other interests include Factor modeling, Portfolio optimization, Behavioral finance, Alternative asset management, etc. If you have any questions, do not hesitate to contact him. You can connect with him via LinkedIn, his blog, or e-mail.
- Simonson, I., and A. Drolet, 2004, Anchoring Effects on Consumers’ Willingness-to-Pay and Willingness-to-Accept, Journal of Consumer Research 31, p. 681-690.