If there was someone who inspired me to learn more about behavioral finance, it has to be Daniel Kahneman.
I first came across his work when I was taking a course in psychology in university, and the first chapter introduces “System 1 and System 2 Thinking”. And I was awed by how revolutionary it was.
Kahneman together with his friend, Amos Tversky have paved the way for behavioral science research and changed the way we analyze our judgement and decision-making process.
Here are my notes from reading Daniel Kahneman’s work and how it can be applied to investing.
Sections:
- I. Dual Process Model of Decision-making
- II. What is judgement
- III. How is judgement evaluated
- IV. Noise vs Bias
- V. Bias Observation Checklist
Dual Process Model of Decision-making
Let’s first begin with the fundamental concept of Kahneman’s work.
Even if you don’t know anything about behavioural science, you’ve probably heard of the phrase ‘System 1 and 2 thinking’.
This is arguably the famous theory in the behavioural science world which was popularized by Daniel Kahneman.
Kahneman’s model divides the mind’s processes into two distinct systems:
- System 1 “is the brain’s fast, automatic, intuitive approach”[1] which is unconscious and effortless
- System 2 is “the mind’s slower, analytical mode, where reason dominates” [2] which is conscious and effortful.
Understanding this concept is crucial in investing because it can be a costly mistake.
For example, only using System 1 thinking to make our investment decisions, such as:
- Panic buying or selling of stocks without logical reasoning
- Researching company without finding disconforming evidence
- Making judgements about a stock only based of reading news
In the later sections, we will explore how to improve our judgement and decision-making.
But first, what is judgement and good judgement?
What is judgement?
Here are the characteristics of judgement.
1. Predictive judgement is a judgement we make when we try to derive close to a true value.
For example, an analyst might forecast if the stock should be a given a buy/sell rating.
When a judgement aims at a true value, two different judgements cannot both be correct.
2. Judgement is not about expressing your own preferences, it’s about being accurate.
“The goal of judgment is accuracy, not individual expression.”
– Noise: A Flaw in Human Judgment
3. Good judgement depends on 3 things, what you know, how well you think and how you think.
Good judges are usually experienced, smart, open-minded and willing to learn new information.
How is judgement evaluated?
There are 2 ways to evaluate judgements, quality of output and process.
In Noise: A Flaw in Human Judgment, the authors outlined the questions you can ask yourself when evaluating your judgement.
- Output: What is the difference between the judgement and the outcome?
- It only works if the judgement is has a true outcome
- Process: What is the process of the judgement?
- How does the process perform when applied to a large number of cases? (Sample size must be large, statistics)
- Does it conform to the principles of logic or probability theory?
In investing, we are often told to focus on the analysis process instead of focusing on the outcome.
Because sometimes a good outcome (i.e. gains in the stock market) can be from a lousy process (i.e. are just lucky).
But this is not a standard practice in reality.
Most people tend to want to achieve a prediction (e.g. making a prediction of the intrinsic value) that matches the outcome.
Instead, we should be trying to achieve having a “judgement process that would produce the best judgement over an ensemble of similar cases”.
A good judgement process in investing includes having a stock research process that can be consistently used to generate the best judgment about the future potential of a company. The process has to be scaleable and work over a number of different companies.
How can we form accurate judgements? We have to reduce noise and bias.
Noise vs Bias
We have to first accept this:
“When there is judgment, there will be noise. And more of it than you think.”
Daniel Kahneman
We often hear things like “the data is noisy” or “the analysis is biased”, and think that they refer to the same thing.
But noise and bias are not interchangeable words.
What is the difference between noise and bias?
There is a useful diagram that illustrates the difference.
Imagine a team of friends playing a shooting arcade game. Each team uses a shared rifle and one person has one shot. These are the results:
- Team A – Ideal world where everyone hits the bulls-eye
- Team B – Biased. Shots are systematically off the target
- Team C – Noisy. Shots are scattered widely
- Team D – Noisy and Bias. Shots are systematically off the target and shots are scattered widely
Noise is about scatter and standard deviation (i.e. how far away the data is from the mean).
Bias is about being systematically off the target.
Some judgments are biased; they are systematically off target. Other judgments are noisy, as people who are expected to agree, end up at very different points around the target.
How can we reduce noise?
Noise is wrecking our judgement. How can we reduce it?
Noise can be reduced in two ways, by performing (1) Noise Audits and by applying the (2) Six Principles of Decision Hygiene.
Noise Audit
This is the process of a noise audit:
- People are presented with a problem that is realistic, and typical issues they could encounter on their job
- Employees are given the same question and are asked to put a dollar number or in some other way indicate what they expect to happen
- Anlayze the variability of the case. If the judgements are variable, then the errors are variable.
- Note: It’s not about having the correct answer (i.e. accuracy), we are interested in the variability of judgments.
Let’s look at a case study of a noise audit to better understand how it works.
Case study: Noise Audit in Asset Management Firm
What is the study about?: In a random selection of 42 experienced investors to run an exploratory noise audit and estimate the fair value of the stock.
Results: Median noise was 41%
Conclusion: Given that there is a huge difference among investors in the same firm, using the same valuation methods is not good news.
Six Principles of Decision Hygiene
There are six principles for organisations or individuals to take on if they want to minimise noise:
- Accept that decisions are about accuracy, not individual expression
- Think statistically, and take an outside view of the problem
- Structure judgement into independent tasks – prevents the problem of excessive coherence (i.e. when we distort information that doesn’t fit into an emerging story)
- Decision-makers should resist premature intuitions
- Take independent judgements from multiple judges and factor them in
- Favour relative judgements, which tend to be less noisy
Bias Observation Checklist
Recognizing the importance of System 2 thinking, having a good judgement and eliminating noise and bias.
There are also some common psychological biases to be aware of to reduce errors in our judgement and decision-making.
In “Noise”, Kahneman and the authors propose a “Bias Observation Checklist”:
1. APPROACH TO JUDGMENT
Substitution
- “Did the group’s choice of evidence and the focus of their discussion indicate substitution of an easier question for the difficult one they were assigned?”
- “Did the group neglect an important factor (or appear to give weight to an irrelevant one)?”
Inside View
- “Did the group adopt the outside view for part of its deliberations and seriously attempt to apply comparative rather than absolute judgment?”
Diversity of Views
- “Is there any reason to suspect that members of the group share biases, which could lead their errors to be correlated? Conversely, can you think of a relevant point of view or expertise that is not represented in this group?”
2. PREJUDGMENTS AND PREMATURE CLOSURE
Initial prejudgments
- “Do (any of) the decision-makers stand to gain more from one conclusion than another?”
- “Was anyone already committed to a conclusion? Is there any reason to suspect prejudice?”
- “Did dissenters express their views?”
- “Is there a risk of escalating commitment to a losing course of action?”
Premature closure; excessive coherence
- “Was there accidental bias in the choice of considerations that were discussed early?”
- “Were alternatives fully considered, and was evidence that would support them actively sought?”
- “Were uncomfortable data or opinions suppressed or neglected?”
3. INFORMATION PROCESSING
Availability and salience
- “Are the participants exaggerating the relevance of an event because of its recency, its dramatic quality, or its personal relevance, even if it is not diagnostic?”
Inattention to quality of information
- “Did the judgment rely heavily on anecdotes, stories, or analogies? Did the data confirm them?”
Anchoring
- “Did numbers of uncertain accuracy or relevance play an important role in the final judgment?”
Nonregressive prediction
- “Did the participants make non-regressive extrapolations, estimates, or forecasts?”
4. DECISION
Planning fallacy
- “When forecasts were used, did people question their sources and validity? Was the outside view used to challenge the forecasts?”
- “Were confidence intervals used for uncertain numbers? Are they wide enough?”
Loss aversion
- “Is the risk appetite of the decision-makers aligned with that of the organization? Is the decision team overly cautious?”
Present bias
- “Do the calculations (including the discount rate used) reflect the organization’s balance of short- and long-term priorities?”
Note that this checklist is not exhaustive and it serves as a starting point for you to design your own bias checklist.
Some of these biases are easier to spot, such as loss aversion whereby investors can be overly conservative with estimates.
While other biases are harder to recognise, such as anchoring bias because we are not aware that it has already happened to us.
Some examples of these happening in our investing research process:
- News & secondary information: If the Google search result is negative news, we are quick to associate the negative news with the company, and vice-versa if it’s positive news.
- Stock prices: We search the price of the company on Google, and our mind subconsciously tags this stock as “cheap” or “expensive” based on price.
To reduce anchoring bias, here are my two suggestions:
- Always read from the primary source: Read the primary information first (i.e. Annual report, Investor Relations) before forming too early of a judgement about the company
- Stock prices: Avoid searching the historic price of the stock when you analyze the company, and compare it to the current price. Price is seldom a reasonable factor of value.
Nobody is perfect
We are all prone to making errors in our judgment and decisions.
Even Daniel Kahneman himself admits that he is sometimes wrong but he is happy because he learned something. He says in an interview:
“Those moments of even finding that I’m wrong – there is some pleasure in that there is a feeling I’ve learned something. I used to think something and now I think something else and then solving it is even more fun. But even when I solve it, I know that it’s temporary and that I will change my mind again. For me, it’s one of the joys of work. It’s the only time that I feel quite certain that I’ve learned something is when I look at what I used to think, and I say, ‘What an idiot’ – how come I didn’t see this before? Which happens to me a lot.”
Often, there is a price to pay for your mistakes, and sometimes we can’t be sure if we can amend them.
But we can be certain that our mistakes are our best teachers.
We have to continue building up our latticework of mental models, and avoid making the same mistakes twice.
Books referenced in this article:
Noise: A Flaw in Human Judgment
If you’ve enjoyed this article, you’ll enjoy my weekly newsletter.
Enter your email below to join and receive a free copy of Charlie Munger’s Investing Principles Checklist.
