The Incredible Shrinking Alpha is a good book on trading and investing. Its message has broad applications for investors and traders:
As more and more investors and traders discover strategies that outperform buy and hold, their usage of these strategies causes the strategies to converge with buy and hold over time.
In effect, it becomes harder and harder to outperform buy and hold as time goes on.
The Incredible Shrinking Alpha: beta, factors, and alpha
Investment researchers break down an investor/trader’s performance into 3 parts:
Beta measures the investor’s exposure to the market. E.g. if the S&P 500 goes up 8% and you are long $SPY, you earn 8% from “beta” (i.e. being exposed to the market).
Why is beta important?
- There is no point in paying high fees to a fund manager whose strategy is the same thing as being 100% long the S&P.
- You can easily replicate this beta by buying $SPY (a S&P 500 ETF). This is a “strategy” that you can easily copy.
“Factor” is a relatively new concept, first being accepted in the 1990s.
A “factor” is a characteristic of a stock that causes it to outperform in the long run. There are several popular factors:
- Size: historically, a portfolio consisting of small cap stocks tends to outperfrom a portfolio consisting of large cap stocks. Hence, an investor who buys a small cap index will outperform an investor who buys the S&P 500 in the long run. From 1927 – 2014, small cap stocks outperformed large cap stocks by an average of 3.4% a year.
- Value: historically, a portfolio consisting of “cheap” value stocks tends to outperform a portfolio of “expensive” growth stocks. Hence, an investor who buys a value stock index will outperform an investor who buys the S&P 500 in the long run. From 1927 – 2014, value stocks outperformed growth stocks by an average of 5% a year.
- Momentum: stocks that have outperformed in the “recent past” (e.g. past 12 months) tend to outperform in the near future (e.g. next few months). Strength begets strength. Hence, an investor who buys a momenum stock cap index will outperform an investor who buys the S&P 500 in the long run.
- Quality and profitability: stocks with low debt, stable earnings, sound management tend to outperform stocks with high debt, volatile earnings, and weak corporate governance. Hence, an investor who buys a quality stock index will outperform an investor who buys the S&P 500 in the long run.
- Volatility (betting against beta): low volatility stocks tend to outperform high volatility stocks when adjusting for risk. Hence, a trader could profit by going leverage-long on low volatility stocks and shorting high volatility stocks
“Factors” are important because an investor or trader can easily and cheaply replicate these sources of outperformance without paying high fees.
- In the past, good investors and fund managers would often tout their outperformance as “alpha” (i.e. their own skill) when in reality, all they did was implement some of these factors. Uninformed investors would pay these fund managers high management fees because they thought the fund’s outperformance was a result of the manager’s skill (alpha).
- When outperformance could be explained by quantitative “factors”, investors and traders could easily replicate this outperformance for themselves instead of paying such high fees by implementing these factors on their own.
Here is an example.
Fund Manager XYZ benchmarks himself against the S&P 500.
- From 1997 – present, the S&P 500 delivers an average annual return of 7.2%
- From 1997 – present, the fund manager earns an average of 8.1% a year
The fund manager boasts about his investing prowess. In reality, all the fund manager did was tilt his portfolio toward small cap, value stocks. This permanent tilt accounted for his entire “outperformance”. (Historically, small cap value stocks tend to outperform).
Instead of buying into how “smart” the fund manager is and paying him outrageous fees, you can easily learn about the factors that the fund manager is using and replicate his outperformance.
“Alpha” is true skill. Returns generated by “alpha” can not be replicated by a low-fee ETF because these returns are generated by the skill of the fund manager. Alpha cannot be replicated by a factor.
Over the past 20 years, more and more of what was once thought to be “alpha” has been discovered to be “factors”. For example, even the great Warren Buffett’s outperformance can almost entirely be attributed to the “size, value, and quality” factors.
- People commonly assume that Warren Buffett’s stock-picking ability is what makes him a good investor.
- But you could have easily replicated Warren Buffett’s performance just by buying a basket of “small, cheap, and high quality” stocks. (This can be done nowadays via factor-related ETFs and mutual funds.)
- In other words, it wasn’t Warren Buffett’s “small, cheap, and high quality” stocks that did well. It was that ALL “small, cheap, and high quality” stocks outperform on aggregate. Warren Buffett’s outperformance came from his investing philosophy (small, cheap, and high quality stocks) instead of from his stock picking skill.
Of course, this doesn’t diminish Warren Buffett’s achievement. He sensed these factors DECADES before academics and researchers discovered these.
The problem with Factors
Factor investing has a problem. All factors can go through very long periods of underperformance. In fact, this is true of anything in investing and trading.
- Good trading strategies that outperform in the long run can underperform in the short run.
- Factors that outperform in the long run can underperform in the short run.
The problem is that the “short run” can last 10 years. Here’s an example.
- “Value” stocks tend to outperform in the long run (i.e. 40+ years).
- However, value stocks can go through stretches of 10-15 years in which they underperform. The most recent example is from 2009-2018, in which a basket of value stocks would have underperformed the broad S&P 500
In other words, factors are inconsistent.
Here is an example from Jeremy Siegel. From 1927 – 2014, small cap stocks (the “size” factor) outperformed by an average of almost 2% a year.
However, this outperformance ENTIRELY came from 1975-1983. Without that period of outperformance, investing in small cap stocks (the “size factor”) would have been exactly the same as investing in investing in the broad S&P 500
The problem for investors and traders is that is it impossible to consistently and accurately predict which factor will outperform in the future. E.g.
- You don’t know if “small cap” will outperform in the 2020s.
- You don’t know if “growth” will outperform in the 2030s
- You don’t know if “value” will outperform in the 2040s
Hence, an investor or trader who relies on factors to outperform will need to sit through long periods of underperformance and accept those periods as a part of the game. In order to outperform in the long run, you must accept that there will be long stretches in which you underperform.
My biggest concern with factors
Factors sound appealing. In order to outperform, all you need to do is trade a portfolio that is tilted in favor of outperforming factors.
E.g. Instead of trading/investing in the S&P 500, trade/invest in the S&P 600 (a small cap index). Historically, the small cap “size” factor tends to outperform.
This sounds like a good idea on paper. But I have 1 big concern.
The more something becomes popularized in the markets, the less well it tends to work (Ray Dalio echoes this point in his book Principles). These factors are being understood and disseminated at a faster and faster rate.
- In the past, factor investing worked well (outperformance) because these factors were relatively unknown. Individuals like Warren Buffett were able to pick the low hanging fruit.
- With factor investing well popularized nowadays among institutional investors, it is possible that this won’t work as well in the future.
Once again, let me use Warren Buffett as an example. Warren Buffett has always favored “value”. However, Warren Buffett’s performance has deteriorated significantly over time. At the start of his career, the amount of competition in the financial industry was much less. There were fewer investors hunting for value stocks, which meant that Buffett could easily find dirt-cheap deals.
Today, so many investors are looking for “value” stocks that inevitably, all value stocks become more expensive. This results in “value” stocks performing less admirably than they did in the past.
Warren Buffett himself alluded to this in his interview with CNBC. At the 1:52 mark, Buffett mentions (I paraphrase):
The game [investing] has become much harder than it used to be. There are many more players, which makes the competition harder and harder.
Moreover, information is so much more readily available than it used to be. This results in your “edge” being eroded. By the time you’ve spotted a “good” deal, so have 10 other people. They will bid up the price, which makes the deal not as good.
In the past, there simply was not a lot of data available. (This is why Warren Buffett could capitalize on the “size” and “value” factors for so long. The data was not available, so many others didn’t know that these factors cause outperformanced. Buffett sensed that these factors would work, even though he didn’t have the data at the time. Others only started to capitalize on these factors in the 1980s and 1990s, when the data started to become available).
This is the biggest problem with the finance industry. In order to succeed in the finance industry, you need opacity – a lack of transparency. This means that “outperformance” is a zero sum game. In order for someone to outperform, someone else must underperform.
- Outperforming is easy in an era (e.g. 30 years ago) when most people had limited access to financial information.
- Outperforming is hard today when most people have almost unlimited access to financial information.
In other words, the more something that “works” gets shared, the less useful it becomes over time. This problem arises in zero-sum games such as investing/trading and gambling.
Think about it this way.
- It’s easy for a professional poker player to win if he is surrounded by amateurs
- It’s hard for a professional poker player to win if he is surrounded by other professionals who are all armed with similar knowledge and tactics.
This book is called “The Incredible Shrinking Alpha” because the amount of alpha in the finance industry is shrinking.
- There are more and more professionals in the game. In the 1940s, 90% of U.S. stocks were owned by households. Today, more than 80% of U.S. stocks are owned by institutions. In other words, the amount of “dumb money” has decreased significantly while the amount of “smart money” has increased significantly.
- Thanks to the internet and its rapid dissemination of information, it is harder and harder to consistently outperform. Strategies that once worked are instantly shared, which makes these strategies work less well in the future.
The book The Incredible Shrinking Alpha points to a 2013 Credit Suisse report which illustrates this. The report plots the 5 year average standard deviaion of excess returns in U.S. large cap mutual funds from 1967 – 2013.
- If the standard deviation is high, then it means that some professional fund managers are performing exceptionally well, i.e. a wide dispersion around the mean “buy and hold”
- If the standard deviation is low, then it means that more and more professional fund managers’ performance is becoming closer and closer to buy and hold.
As you can see, professional investors and traders are becoming increasingly undifferentiated. Sure, some funds will outperform in certain years and others will outperform in other years. But on average, everyone is becoming closer and closer to buy and hold. That’s why even the greatest investors and traders like George Soros and Stan Druckenmiller have been reduced to being no different than buy and hold in recent years.
Over time, the premiums (outperformance) related to the size, value, and market beta factors have all decreased. “Small cap”, “cheap” stocks are becoming more and more like the broad index.
Things change over time. In the past, small cap stocks outperformed. But perhaps this won’t be true in the future.
Small cap stocks tend to outperform because the best companies were all small at one point in time. E.g. Apple was once a small company, Wal-Mart was once a small company, Amazon was once a small company.
Hence, a portfolio tilted towards small cap stocks can capitalize on the explosive growth of these future-superstar companies.
But times are changing. In the past, small companies turned to the public market to raise capital through IPOs. Now, small and promising companies are staying private for longer thanks to venture capital, private equity, and angel investing. Think Uber, Airbnb.
So in the future, perhaps the best companies won’t be “small” and available to the public. The best companies will be “small” and private, and only turn public once they’re big. This will hurt portfolios tilted towards small cap stocks, because these portfolios can’t invest in the best small cap stocks if these stocks aren’t on the public markets until they’re big.
The other problem
The other big problem with investing/trading a factor is that you need to use it for a very long period of time.
- All factors go through periods of underperformance. These periods can last very long (e.g. 10 years)
- All factors go through periods of outperformance.
If you give up on a factor after a period of underperformance, you might just be getting rid of the factor just before it is about to outperform. In other words, trading around a strategy can lead to “buying high, selling low” the strategy.
In the stock market, the “short term” can be 10-20 years. For example, stocks outperform bonds in the long term (i.e. 40+ years).
- Stocks underperformed bonds from 1998 – 2008
- Stocks underperfored bonds from 1968-1982
Don’t let the short term dissuade you regarding what works best in the long term.
So knowing this, perhaps you will probably decide to stick to a factor/strategy through thick and thin.
Herein lies another problem.
Some factors that outperformed in the past may permanently stop outperforming in the future.
How do you differentiate between a period of “temporary” underperformance (e.g. 10 years) and a permanent undperformance for factors?
You can’t. Not without 20/20 hindsight.
Hence, investing/trading with a factor relies on a leap of faith. Faith that what worked best in the past will continue to work best in the future. But then again, everything in investing/trading requires somewhat of a leap of faith because there is no such thing as 100% certainty in the markets.