Mind Against Money – Randomness


The human mind likes stories, even craves them.  We tell our stories about all sorts of things, even when there is no story.  It’s part of some type of innate need to make sense of our environment, of the world we live in.  It’s clear that our mind is uncomfortable with randomness, and often tries to create cause and effect relationships where there are none.

This is an important topic to understand when it comes to investing in the stock market, because there is a lot of random noise in pricing investments and if you don’t realize how much of this is meaningless, you can drive yourself crazy trying to figure out why certain moves happen and predict short-term changes.

The other important reason for better understanding randomness in stock investing, is the related concept of mean reversion.  I will write a focused article on this in the future since it’s so important but I’ll touch on it in this once since randomness and mean reversion are linked.  In fact, mean reversion is not only a powerful concepts in investing, it’s critical to many other areas as well.  I first learned the concept as Le Chatelier’s Principle in chemistry, the “equilibrium principle”.  Physicists might think of Newton’s Third Law which states that equal and opposite forces occur in reaction to an applied force.  Economists use mean reversion as they describe market forces (e.g. competitors entering in response to unusually high profit margins, thereby eventually bringing profit margins back towards an equilibrium value).  There are many other areas where mean reversion is visible as a powerful force.

Before diving into randomness and it’s importance in stock market investing, first we need to understand what defines the price of a company’s stock in the first place.  We all know that a share of stock is a partial ownership in the underlying company.  The more shares you own, the larger a portion of the company that you own.  However, it’s not  straightforward to value that portion of ownership, even if you know the current financials of that business.

There are several ways to value a company, mostly based on a fundamental analyses and often several approaches are used to triangulate an estimate.  One of the top ways is to estimate the future cash flows of the business using past, current, and projected financial data including sales and profits, growth rates, etc.  Generally people spend money now to buy stocks with the expectation that they will get even more money over time from those investments.  This leads to values based on multiples of sales or EBITDA (earnings before interest, taxes, depreciation, and amortization).  You can also benchmark by how much similar businesses in the same industry have been purchased for.  For example, large established commodity businesses like oil companies typically have stock prices at a P/E ratio around 10 while a “specialty” company with less cyclical earnings growth might sell for a P/E twice that high.  So even the same $ earnings can be valued quite differently.

You can also take into account all the assets and liabilities to do a bottoms-up book value estimate.  A company with high debt will be discounted.  One with great real estate locations for physical stores, or great brand equity will have a premium.  Then you can factor in things like the quality of the management team, the R&D pipeline, etc.  You can also factor in estimates around competitive advantages and how sustainable these are (e.g. the “economic moat” term coined by Warren Buffett).

In the end, if you step back, you find there are a lot of assumptions, and that you could come up with a quite wide range of values based different, but equally reasonable assumptions.  One telling data point is that smart professionals who live and breathe this stuff, who visit specific companies asking probing questions of top management, getting tours of facilities, and understanding these companies to a level that individual investors aren’t able to do, still don’t do any better on average than blind individual investors.  That should give you some idea of how difficult it is to find undiscovered bargains through fundamental analysis.

On top of the variation and complexity of doing fundamental analysis, the “market” prices fluctuate even more on a day-to-day basis.  Even when the true value, defined as expected long-term cash flows, of the underlying business doesn’t change that quickly, the price of that business on a given day can change a lot.  Many of the value assumptions are psychological and depend greatly on the overall investment mood.  This is why behavioral economics is such a hot field right now…..the old view of the rational human and perfectly efficient markets has some serious flaws.

To further complicate it, we have business cycles.  Business demand and supply fluctuates over time, leading to sales fluctuations and even bigger profit fluctuations.  If you’re valuing a business on predicted cash flows and those cash flows go up and down large amounts on a routine basis, it’s difficult to estimate a true value to the business.  In addition, central banks make money cheaper or more expensive depending on how they are trying to manipulate investment and economic activity, which creates further swings in business and stock market values.

Tied to this are changes in alternative investment options.  Part of the value of business ownership (stocks) is relative to other options like fixed income investments such as bonds.  Historically, stocks have had a “risk premium” when looking at earnings growth about 3% above what bonds deliver.  It is closer to 6% right now because of extremely low bond yields set by central banks.  This makes stocks (earnings) more valuable and increases their prices.

Lastly, the value of investments (i.e. future cash flows), depends greatly on inflation.  In a high inflation environment, those future cash flows will be worth less, and the underlying investment is discounted.  This is why P/E multiple decline when inflation rises, and vice versa.  As an example, currently there is a large debate over the high P/E ratios in the US stock market.  A valid justification of the high multiples is that inflation is low and expected to remain low so a higher P/E is rational.  In addition, as mentioned above, bonds are paying nothing so that makes the earnings from stocks more valuable.

This is an incredible amount of “macro” factors that you have to not only understand, but also predict.  On top of all these, there are company and industry specific factors if you’re trying to time sectors or choose individual stocks. The combination of fluctuating numbers (day-to-day sales, unusual expenses, lucky profits, new regulations, safety incidents, legal actions, unexpected success from a product developed 10 years ago, the loss of business as a customer goes bankrupt, a new positive or negative media article that you didn’t know was coming, etc, etc), along with the even more important but even more unpredictable changing moods of people who are pricing these companies minute by minute, and you essentially get nothing more than noise.

This has been studied to death and the conclusion of all the studies is the following……you can’t really predict this stuff.  Over the short (less than a year) and even medium (3-5 year) timeframe, no one can predict how all of these factors will change, particularly how fast they will change.  Timing the market is extremely difficult if not impossible. It’s important to note that it does become possible to predict, with some (but not certain) accuracy, what will happen over a longer timeframe of 5-10 years or more.  Over longer time, the market, and associated stock prices, tend to drift closer to an equilibrium value based on a smoothed value of profits, sales, costs, etc over the shorter term volatile values.  In addition, people continue to learn and get better at things (I.e. productivity gains) so businesses continue to become more valuable over time, driving prices up faster than inflation.  This is a powerful force for long-term investing.  Those overly conservative savers who think owning a business (stocks) is just like gambling, forget that in this analogy you are the house, not the individual gambler.  You may lose many hands for a while, just like a gambler can be on a hot streak of wins for a while, but over time the house always wins.

By the way, it is the long-term aspect of investing where individual investors have a big advantage over the professionals.  The pros need to deliver results on an annual or even quarterly basis.  As we just discussed, it is nearly impossible to beat the market short-term with any consistency.  Even if they identify bargains, they will be out of a job if it takes too long for the market to realize it.  This makes their jobs much more difficult.  It’s much, much easier to get long-term gains from the stock market than it is to make/predict short term gains.

When track records of individual managers are analyzed, what you find is that the “outperformance” in a particular year seems due to luck, nothing more.  If you look at longer-term track records, active managers have results that converge to the market average (or worse).  Add in their high fees (and often higher volatility), and their results consistently underperform the market.  You can do an exercise with coin flipping to help understand this.  Imagine you take 1000 coins and only select the heads as the winners and start flipping.  In round 1, purely based on luck (random probability), you get about 500 “winners”.  In the next round, there will be about 250 winners.  Then 125, 62, 31, 15, 8,4,2,1.  Out of the thousand investment managers, many bets are made on investment guesses.  Some play out simply due to luck.  The few that “manage” to flip heads 9 times in a row make it onto TV as an expert that accurately predicted the latest stock market crash/boom, next big start-up sensation, new stock picking method, oil price, gold boom, etc.  When you look at the same person’s record the next year or year after it, they are usually completely off.  Their earlier result was simply luck, despite the fact that we want their result to have an explanation.

Given the clear data, it’s pretty amazing that so much money is spent (wasted) on active management funds that clearly don’t deliver any value.   This is a multi-billion dollar industry that clearly only takes money from investors.  It’s quite incredible.  The good news is that individual investors are getting more educated quickly (I think the internet is helping spread this information and fight the Wall Street marketing machine to an amazing degree that wouldn’t have been possible in the past).  The flow of money out of actively managed funds and into low-cost index funds has been amazing over the last decade.

A lot of money is also spent on “chartists”; the men and women who look at stock market price charts and try to predict what will happen next.  This “head and shoulders” pattern is a classic sign that the stock will break out to X price soon.  Or hedged predictions based on charts such as “a key support level is at $20 so if the stock price goes below that, it will go down further, but if it holds, it will go higher”.  What a great job…….it’s like weather forecasting.  If the stock drops, I told you so, if it goes up, I told you so and either way I’m right!  Unfortunately, it’s not really clear if you should buy or sell based on some of these recommendations.  Either way, these types of chart reading technical analysis (and there are many variations), have been studied and none of them have proven to be helpful.  It’s because there is no underlying pattern in the first place.

There are similar examples of data-driven methods, where new combinations of more fundamental indicators are used to predict future returns.  These don’t have a much better track record or predictability than the chart analyses described above.

The best book I’ve read on randomness and it’s role on wall street is by Burton Malkiel.  I highly recommend this book if you want to learn more.  In addition, I’ve seen good reviews  on several of Nassim Taleb’s books like the Black Swan and Antifragile but I have not read those myself and can’t give a personal recommendation.

A good example of the folly in not understanding randomness is chasing momentum, particularly in actively managed funds.  The hot stocks from the prior year almost always underperform the next year.  But inflows to these funds happen after they have already done well and made it to the magazines or top funds list. Once they have done unusually well, they are more expensive and the probability of a lower future return is now higher.  Yet this is when the investors decide to put money in.  It’s madness!   I am shocked at the common financial articles on investing recommendations where writers show investments that performed poorly and they then recommend NOT investing in them.  Presumably this “proves” that this is a bad investment and will do poorly in the future.  They then recommend the investments that did the best relative to others in the comparison group.  This must “prove” that these are the strongest investments and will continue doing better in the future.  In reality, it’s often the opposite that occurs because the probability of better future returns goes up as the price goes down.

The same story plays out at a macro level as more individual investors pile into stock investments only after the market has done well for a while.  It’s starts to feel “safe”, popular press articles are more positive on investing, and everyone seem to be doing well.  This is when you should be scared and be expecting lower returns going forward, not getting optimistic!  When the markets go down a lot, you should be more and more optimistic about future returns (and should really celebrate if you are still in the accumulation phase) but this is usually the opposite of what most people “feel”.  If you do what feels right, you will severely underperform the market.

The stock market fluctuates.  A lot.  Clearly stocks are very volatile compared to stable investments like bonds or cash.  But whether the market goes up quickly or down quickly, over time the extremes normalize and the long-term path looks like a smooth trajectory upward.  You just need to understand that the short-term fluctuations don’t really have any meaning and need to be ignored.  It’s simply random noise as a complex set of constantly changing factors, feelings and predictions, are used every day to adjust the prices of stocks.

This randomness is very difficult for us to accept.  Everyone sort of understands this, but they don’t really accept this.  It seems that there must be some way to figure this out.  Some new algorithm or something than can help make meaning out of all this noise. Some new leading indicators, or prediction method.  It is a desperate human need to make patterns and sense out of noise.  We are very uncomfortable with the thought of randomness and lack of meaning and mentally fight to create cause and effect links where none exist.

One of my favorite story examples on randomness and mean-reversion is from Dan Kahnemans book Thinking Fast and Slow.

He describes the situations of flight instructors teaching new pilots how to improve.  They are convinced that their method of yelling at pilots when they have a poor flights is the best way to operate.  After all, they have decades of empirical data that show this works.  When pilots are yelled at following a poor flight, they tend to improve on the next flight.  At the same time, positive reinforcement doesn’t work.  Pilots praised for a really good flight, tend to do worse on the next flight.  Hence a punitive but seemingly effective training program is put in place.  However, if you understand randomness, you wonder is there is really a cause and effect relationship here.  Isn’t it simply likely that a pilot with a particularly good flight will tend to have a worse flight the next time, as his/her performance moves back to his/her average?  Similarly, an unusually bad flight will tend to be followed by a better flight.  Whether the pilot is yelled at or praised actually doesn’t play any role in changing the outcome.

As we described above, the same thing happens with investments as unusually good or bad returns tend to be followed by the opposite at some point.  By the way, this is the only reason rebalancing works, and it clearly does.  The challenge is that the timing cannot be predicted with accuracy.  Again, we are all desperate to construct stories to explain things. But there is no explanation for short-term stock market movements.  So the talking heads on TV have to come up with reasons for the audience, because people will stop watching if they tell the truth that it’s just random.  So instead of saying “the S&P 500 is up 0.5% today which seems strange because global stock markets declined 1% yesterday”, they say “the S&P 500 is up 0.5% as investors shake off the 1% global declines yesterday on optimism for XYZ”.  Or maybe the market drops on “profit-taking” or maybe it goes up on “bargain-hunting”.  What does that mean?  Nothing.  They are just making up reasonable-sounding explanations as they attempt to make sense of the data.

The same happens with existing companies.  Maybe the stock is up despite an earnings loss because it wasn’t as bad as some expected.  But if the stock goes down in the same example, it’s obviously because of the earnings loss.  There are endless speculations to be made and after all the effort, we’re back at the same point……unable to predict what will happen next.

Hopefully some of this article resonated with you and helped you understand that randomness is much more prevalent in our world, and certainly the stock market, than most people realize.  As a scientist, I struggle with this myself and it’s taken years of learning from others and my own experiments to better grasp this (my own attempts at short-term market timing and stock picking have gone as well as the research would predict……very poorly!).  I’m still uncomfortable with the lack of control this means, but I’ve come to accept that this is reality.

The benefit I hope IBFree readers really get out of this is to find acceptance and feel more comfortable with short-term stock market volatility and avoid doing dumb things that most individual investors do.  Repeat after me:  Because the stock market always goes up and down, I will 1. Not sell after a big decline 2. I will buy more if I can when prices are low and news is negative, and not worry if prices continue to decline after I buy.  3.  I will not avoid stocks altogether simply because they are volatile.

I assume you want/need money for many years and the biggest long-term risk to a portfolio is not volatility in the market, it’s a long-term low return, which you will get with “safe” non-volatile investments.  Which of course means the approach that feels “safe” short-term is actually the riskiest one in the long term.  Human progress (i.e. productivity, innovation) marches on and will continue in the future.  If you can drown out the short term fluctuations, and wait patiently during the longer term rise, you will be well-rewarded.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s