|New Threads Only:|
|New Threads & Replies:|
Forum List » Guru News and Commentaries|
Guru News, Stock picks and commentaries
Seth Klarman’s Two-Brain Framework
Posted by: Dr. Brian Zen (IP Logged)
Date: April 10, 2012 12:16PM
Analysis argues with intuition:
Two brains tracking fluctuation.
Analysis calls for caution;
Intuition calls for action.
Analysis causes paralysis;
Intuition causes emotions.
Use both intuition and analysis;
You’ll find great businesses.
Intuition leads with idea generation;
Analysis follows with deep calculation.
High probability insight occurs
when analysis agrees with intuition.
Seth Klarman: How Brains Could Fool Us
By MATTHEW INDYKE and BRIAN ZEN
Why is our own brain often our biggest enemy with investment decision making? What causes our errors in judgment? If you have played chess, you would know that the hardest part is choosing between what quick intuition tells you and what deep analysis tells you. Holding unpopular stocks like Hewlett Packard (HPQ), Seth Klarman, in his 2011 letter to Baupost Group partners, quoted psychologist Daniel Kahneman’s the two-brain-framework for explaining how the brain works. The hard part is how to hone both our intuitive insights and our analytical skills to make better decisions.
Do We Have Two Brains?
Kahneman visualizes the human brain as two separate systems. They are called system one and system two. System one is the intuitive brain. It thinks on the spot. It responds to commands without hesitation. It makes decisions quickly and unconsciously. We use system one when doing basic elementary math, such as 1 + 1 = 2.
System two is slower and more analytical. We have to think hard and long to find out that 62 x 78 = 4836. We need to remember what we were taught. We need to recall any past experiences that might influence how we approach a situation. System two is activated when doing complex calculations like those involved in financial statement analysis.
Understanding how our brains work is important to successful investing, says Seth Klarman. That includes knowing our mental limitations and internal biases. Know the maximum loss you are willing to accept and whether your perception of a company can one day be understood by others.
Should We Listen to Intuition?
System one helps us differentiate the good from the bad and the right from the wrong. We need it to control who we are and how we go about our daily routines. The people we interact with and the food we eat are processed by system one. The laws we follow and our ethical judgment are processed by system one. It keeps us sane and in touch with society. But system one is vulnerable when jumping to conclusions on a complex idea that seems so basic from the naked eye.
As an investor, you get up in the morning and you go to work. The process was completed using system one. But then you get to work and check the financial markets. You observe that the market is going up. Immediately, you get into buying mode and aggressively pursue stocks to buy. A novice investor likely uses intuition alone to guide its decisions. More experienced investors will tell you that doing this is not enough.
Seth Klarman notes that the errors associated with intuition alone can especially be avoided when operating as an organization rather than an individual. More investing minds create more ideas and help identify the ones that are insightful. Your intuition is valuable in terms of helping you sense the steps to get there and what to look for. But digging deeper into those details is another step required by the analytical part of the brain.
How Far Does Rationalization Go?
System two avoids following intuition only. Thorough analysis is necessary to solve a problem and answer a question. Knowledge will not be second nature to us. We have to think hard using a previously taught process ingrained in our brains. We have to be taught how to get there. We have to know how to get there. We have to remember how to get there the right way. And most of all, we have to think before we react. The educational knowledge and real world experience that we gain over the years expects to become the basis of our decisions. A seasoned investor like Klarman stresses that a company’s underlying value varies greatly from its actual price. Price is determined by supply and demand but value looks at factors like the amount of cash flow and assets a company possesses. A novice investor most likely makes an investment decision immediately after seeing favorable statistics. But system two cannot be ignored as it is necessary to further analyze the statistics. It requires more focus. And it requires deep analysis.
Why Do Smart People Sometimes Do Crazy Things?
System one or intuition could misguide us. Common sense makes us believe we can never make irrational decisions. But Kahneman suggests that most of the time, we later regret most of our decisions if we go only by intuition. So we need to be conscious of how our minds process information. A common mistake is to react quickly on what appears to be favorable circumstances. Ask any group of experienced investors and they most likely agree with Kahneman. The average investor will tell you there is a greater likelihood we don’t see high returns on our investments if we don’t spend sufficient time analyzing beforehand.
Seth Klarman knows that we need to train our brains to think a specific way in order to develop our investing skills. He goes on to mention that it is easier to predict how investors will react in a particular situation than it is to know a company’s bottom line. In other words, value investors may have better odds predicting behavioral shifts in investors’ mentality.
Intuition vs. Analysis: The Two-Brain Framework
Is system one’s intuition always reliable and accurate? Not entirely. All investors, novices or experts, are prone to following their intuition at any given time. Novices suffer from lack of experience. They haven’t seen the movie before. Experts may become arrogant and believe they are making a good investment because every past investment has worked.
Smart investors recognize that every good idea needs evaluation. That is where analysis comes in. It is rare for novices to be successful using system two. They generally require the experience of having done it before. Expert investors will presumably be effective with analysis. Their effectiveness is hindered when pride sets in or they become lazy. It is always necessary to think hard about the problem and how the idea applies in the real world.
Intuition is the beginning. It creates our raw thoughts. The thoughts become ideas. Analysis expands on the ideas. It brings out the details. It applies the ideas and details together. It thinks logically. Intuition and analysis, smartly used, will translate to high probability insights and profits.
Re Seth Klarman s Two-Brain Framework
Posted by: onthefringe (IP Logged)
Date: April 18, 2012 09:45AM
I just had a short discussion about math and business with another gurufocus regular, and I just needed somewhere to comment on it. This topic has been a concern of mine for years. This seemed like the most appropriate place to comment on the subject in a recent gurufocus article. I just typed this out off the top of my head without editing. And I type about as fast as I think. So this is going to be lengthy. And Dr. Zen, this is not a direct response to your article, nor the gurufocus member I discussed this with, so please do not take it in any particular way.....
Math is important. But in terms of business and investing I get the impression that it is sometimes overemphasized for the wrong reasons. It's referred to as this very complex and precise thing that can only be understood by the very few brilliant, and is inaccessible to mere mortals. In essence it's become a phallus contest for nerds. And it's really not that precise in business because it requires assuming that a bunch of real life uncertainties are indefinitely certain. Maybe high level mathematics without context (or with limited context) is a much more reliable tool for short term traders.
In reality, a lot of high level math is just a very dense summation of a lot of layers of simple low level math (at least that's what I see, and at least as far as finance is concerned). We could say something like "the high level math would be summarized in the financial statements and the low level math are the details in the footnotes." Break down high level math into smaller parts and build it back up, and it becomes much easier to understand at a fundamental level. And if you refer to the math in context of something you already know very well, it allows you to compare theory with reality. That approach can give you an intuitive understanding of how drivers relate to value. The kicker is this "intuition" is really based off of fundamentals that are used (or should be used) in the math. The two are very much connected.
It may sound like I'm rambling but I think this is so important for a number of reasons.
Take revenue forecasting for example. Early on in my career I was in a few roles that required forecasting. Two stuck out in particular. One was for an international finance department of a healthcare organization. They used three month moving averages. I asked why. I was told that's the way the VP of finance did it. End of discussion. No talk about the underlying business.
When I moved on to corporate strategy at another company, we often used forecasts from management consultants, information firms, and market research/industry firms. I tracked down a couple of contracted consultants (yes, the consulting firms outsourced much of their forecasting to other third party consultants) who were responsible for the bulk of the work. They too extrapolated trends in past data into the future. It was basically a fancier version of the 3-month moving average. Neither approach really seemed tied to the fundamentals of the business. But they are taken at face value, and explained as something the common person can't quite grasp.
A much more accurate way of doing things would have been to gone through the client accounts (both were services companies with individualized accounting and customer databases), look at the pricing and terms of individual contracts, look at retention, look at revenue from new clients, new business from old clients, and sum the results together. Or outsource this task to someone else. It's closely tied to the business and is much more transparent (partly because it is based on reality). But still, you have to know any forecast has to be built on assumptions that are, with near certainty, going to be inaccurate.
I think this math first approach without really understanding what lies behind the numbers is potentially harmful both for businesses and investors. Part of the reason it exists is because we often take things at face value without demanding an explanation. Part of this is a little bit of intellectual elitism. For those who know better, it is just being plain lazy.
For professional investors, it seems forecasts in valuations often utilize a fancy version of the three-month moving average. And you end up with these wacky assumptions built into models. On Bloomberg TV the other day, an analyst explained that he thought Apple was cheap because it had a PE ratio of 13 (on record high earnings no less). He then goes on to say his valuation is based on a 20% growth rate for the next 5 years! And leaves it there. What?!?!?!
At least explain how you came to that conclusion. Is it because that's how fast the company grew last year (which is not enough for me)? Even if you end up being right, and Apple ends up being extremely undervalued based on realized future growth, you essentially based your assessment on a wild guess backed by numbers. That's a huge risk I'd pass up on any day, regardless of the potential payout.
There's nothing wrong with this method per say. You just have to be very honest with what you are or are not assuming and whether it is realistic.
Way too much faith is put into these equations that really make extremely broad assumptions. And that goes beyond forecasting revenues. It's kind of left up to this idea that the experts and economists say so, so it must be. I would have so much more faith in the approach if the assumptions, strong limitations, and inaccuracies behind such models were made crystal clear. It's not that they are wrong, per say. But they in no way can capture everything.
It pays to know what you are really communicating when you put together a valuation. Math is simply a communication tool for most of us (something all together different for engineers and scientists). Not much more than a blog post or article in numeric form.
At the business level, assuming math is inaccessable makes things so much less efficient than they could be. I see no reason why the CFO should know more about detailed sales and marketing metrics than marketing managers who spend time trying to manage for results. Same for operations, etc. It's their world. They should know it inside out. And if they don't know they should be taught how to look at things the right way. It would really allow them to make decisions independently and efficiently. These things are usually not that complicated. But they are often dismissed as such. And so you have the blind leading blind.
I think for those who are not confident in their ability to understand math, particularly in context of business, it pays just to go out and learn about what you don't know, but want to learn more about. And preferably to learn it in context of something you are familiar with. For me that's "Mathematics For Economists Made Simple" by Viatcheslav V. Vinogradov. I don't understand half the stuff in the book off the bat. But it gives me context about what's going on in the heads of the people who put these models together. You have to spend time with these things if you are not familiar. And if you are anything like me, you may come to a few very basic conclusions:
1) A lot the methods used in modern finance are simply estimates of estimates of estimates (of estimates of estimates). So it really pays to understand what you are implicitly assuming (or not assuming) when you put these estimates together.
2) A lot of higher math can be broken down into smaller parts digestable by the average joe
3) Taking the numbers at face value is risky. It generally helps to know what drives them. Earnings and even cash flow can change in an instant. Assets are much more secure, but they can be sold off, acquired, or damaged/destroyed. The business as a whole generally will not change nearly as rapidly.
4) With math, familiarity, context, and repetition helps.
OK, I'm done. Don't know if I actually made my point. :)
Re Seth Klarman s Two-Brain Framework
Posted by: batbeer2 (IP Logged)
Date: April 18, 2012 10:38AM
Beware of geeks bearing formulas. - Warren Buffett