General ThoughtsInvestment PhilosophyJoel GreenblattSuperinvestors

Some Thoughts on Joel Greenblatt’s Magic Formula and its YTD Results

“Value investing is simply figuring out what something is worth and paying a lot less for it” – Joel Greenblatt

I often describe my investment philosophy as a synthesis of ideas from Ben Graham, Walter Schloss, Warren Buffett, and Joel Greenblatt. At the core of my strategy is Graham and Schloss’ quantitative methods for valuing stocks. It’s far more difficult to make mistakes when you simply make obvious, simple decisions based on valuation. In each investment, I want to ensure I’m not taking on valuation risk. Many investors (including most value investors) overly complicate things and this can often lead to counterproductive results. That’s why many smart guys get mediocre results. They try too hard.

One of the things I occasionally like to do is look at simple, diversified value portfolios I track to see how they are performing. I think of these like “value indices”. Rather than track the S&P, which I hope to surpass by large margins over time, I also want to track the results of simple value indices which are very easy to replicate in practice if one wanted a passive approach that would likely beat the S&P over time. I’m talking about tracking things like the 50 S&P stocks with the lowest P/B or P/E ratios, and other simple value portfolios.

Please remember that these are hypothetical portfolios and past performance doesn’t guarantee future results. 

Understand Quant, But Don’t Let the Computer Do 100% of the Thinking

Again, just to be clear: I don’t invest in these quant strategies, I simply track a group of 8 or 10 of these portfolios… more just for fun than anything else. Most of these hypothetical portfolios are up a 20-35% this year, which is remarkable, but not surprising given that the market has performed very well YTD.

I don’t invest using these strategies for two reasons:

  • One, I hope to do significantly better over time (3-5 year periods-any 1 year period could go either way).
  • Two, although I admire a few investors who follow value quant strategies, I have never been a fan of completely systematic strategies with my own capital.

Some quant guys have very convincingly shown that quant strategies outperform even the smartest discretionary value investors. Their argument against making discretionary decisions is that it’s too easy to have lapses in judgement and make poor decisions. This is likely true, but it’s the reason value investing works. Discipline and common sense are required to have a sustaining career as a value investor. And it’s those traits that many lack. For those who are not disciplined, it’s likely better to use a computer to tell them what stocks to choose. For these folks, I truly believe that value quant strategies will likely work very well over time relative to their own results.

But the one thing that isn’t covered as much in a general sense is risk management… i.e. protecting the downside.

The Best Risk Managers Are Humans

The above subtitle might be the exception rather than the norm. For many investors, a quant value strategy (or an index fund) might be less risky than managing your own portfolio. But the best investors in history have generated far better results with far less risk of permanent loss (not volatility) than the best computer systems.

My first priority in investing is protecting capital from permanent loss. In order to properly do that (in my opinion), I need to be able to know something about what I own. I don’t want to invest my hard earned capital (nor my clients’ hard earned capital) into stocks that I’m not familiar with simply because they show up on some valuation screen that has historically said that these stocks do well.

I’m not suggesting quant strategies don’t work… it’s clear that a variety of quantitative value strategies have done well over time (Graham’s quant ideas, Greenblatt’s ideas, net-nets, etc…), and will probably continue to do well in the future, but I’m just more comfortable understanding something about the investments I make. This is not to say that I understand everything about each business, but I want to be able to understand why the investment makes sense. Simply allocating across 30-50 value stocks without any reason (other than the computer said to) is a difficult thing to do, and even more difficult once the strategy begins to underperform for a period of time-an inevitable occurance in just about all value strategies-they almost always work over time, but not always each month, quarter, year, etc….

Magic Formula is My Favorite “Quant” Strategy

I’ve talked about Joel Greenblatt’s Magic Formula before… it’s a value investing strategy that uses two simple inputs (return on capital and earnings yield) to pick a group of “cheap and good” stocks. Basically, Greenblatt wanted to test the ideas he used at Gotham Capital, where he was able to make 40% annual returns for 20 years using a combination of simple value investing ideas and special situation techniques.

I believe that Greenblatt, like Ben Graham, started the project as a way to teach individual investors more than as a way to use for his proprietary funds (although he does use those ideas). He wanted to provide a simple to follow strategy for everyday investors to use that would be superior to an index approach.

The results of this test were published in his book The Little Book That Beats the Market. These results, needless to say, were astounding and surprising, even to Greenblatt himself:

Magic Formula Results vs SP 500

Please remember that these are backtested results, and not guaranteed to continue.

The above is a chart I found on Jae Jun’s excellent site, which was reproduced from Greenblatt’s second edition of his book. The Magic Formula method of picking stocks averaged 23.8% per year vs a market average of 9.6% during the period from 1988-2009.

Magic Formula YTD Hypothetical Results

So far in 2013, the Magic Formula is the best performing strategy of all the hypothetical value strategies I track. I track a list of 30 “Magic Formula” stocks with market caps over $250 million, and also a list of the top 30 over $2 billion. These lists can easily be generated at Greenblatt’s Magic Formula Investing site. The smaller cap group is up about 42% year to date, and the larger group is up just under 40%. Here is a list of the top 30 Magic Formula stocks that were in the group at the beginning of the year (this is not a real money portfolio; just a hypothetical portfolio I track for fun each year): 

Magic Formula YTD Results

Please remember that these are hypothetical results, without consideration for commissions, taxes, slippage, etc…, and not guaranteed to continue.

Both hypothetical Magic Formula portfolios are trouncing the S&P so far in 2013. Of course, they are just as likely to trail the market average by a similar amount over such a short time frame, I just found it interesting to note the fantastic results so far YTD.

But short term results are trivial… the real takeaway is in the principles behind the strategy. These are simple and logical…

Magic Formula’s Secret is in its Logic and Simplicity

Greenblatt’s name for his strategy is tongue in cheek. There is nothing magic about it. It simply systematizes the principles that Greenblatt used to invest for 20 years. It uses one simple metric to determine valuation and one simple metric to determine quality.

At the core, this is all I do when I invest. I try to find good companies selling at cheap prices. Valuation is the most important. As Buffett said in his 1965 letter to partners when he described what he looks for in general investments:

“Their main qualification is a bargain price; that is, an overall valuation of the enterprise substantially below what careful analysis indicates its value to a private owner to be. Again, let me emphasize that while the quantitative comes first and is essential, the qualitative is important. We like good management, we like a decent industry, we like a certain amount of “ferment” in a previously dormant management or stockholder group. But we demand value.

In his early days, Buffett was looking for cheap and good stocks. Just like Greenblatt. I agree with Buffett that value takes precedence over quality. The magic formula weights both of them equally, which is one reason why I choose to make my own investment decisions.

I want to ensure that what I’m buying is undervalued. I want a margin of safety in each investment. The result is a huge margin of safety at the portfolio level which contains a nice group of individual low risk undervalued securities. Insisting on a margin of safety in each individual position and also diversifying adequately also allows the portfolio to easily weather the inevitable errors that one will make. On balance, over time, this strategy works very well.

Greenblatt-an Evolution of Strategy?

I find it very interesting to study Greenblatt’s career. He made 40% annual returns for 20 years using his own value investing strategy of buying cheap and good stocks along with numerous special situations. Then he came up with the systematic strategy he called the Magic Formula. But even after writing that book, he still invested using his old techniques of concentrated, disciplined value investing. As of late, he seems to have changed his ideas on this… in his Market Wizard interview he mentioned he is now managing money using automatic quantitative strategies. However, he did say if he was starting out, he’d do the same thing he did when he started Gotham.

One of the interesting things I heard Greenblatt say at a class in Columbia I happened to watch was that he liked the Magic Formula, but preferred to invest his capital using his own decisions and common sense. I think his thinking may have evolved on this, but I agree with his original idea-that the Magic Formula proves that these principles work (buying cheap and good stocks works well over time). It’s just that I want to decide for myself what stocks are cheap and good.

It’s more a matter of risk management than absolute performance. I may only keep up with the Magic Formula over time, and one might argue ‘why not use it in that case’? The answer for me would be risk. I want to understand what I own, so that I’m comfortable owning the shares of those businesses when the market takes a big drop. I want to know the values of my businesses, so I know how to respond to various market conditions.

After all, we own pieces of businesses, not pieces of paper.

It’s just a matter of preference. I’m a numbers guy and so I love reading about quant strategies and noticing their results. But I will never forget the first two rules of investing:

  • Rule #1: Don’t Lose Money
  • Rule #2: Don’t Forget Rule #1. 

The best way I know how to manage risk is to understand something about why I own the businesses I do. I want them to be good businesses, but even more important than that, I want them to be undervalued. To sum it up, I want to ensure that I understand why I own the stock and what my general estimate of the value is.

As usual, my post got somewhat longer than I expected (but this is an interesting topic to me so it was bound to happen!). But the easiest way to sum it up is the way I started the post… with one of my favorite quotes from Greenblatt:

“Value investing is simply figuring out what something is worth and paying a lot less for it.”

That’s the crux of what we’re trying to do. And at its foundation, it’s really just that simple…

34 thoughts on “Some Thoughts on Joel Greenblatt’s Magic Formula and its YTD Results

  1. On thing I don’t like magic formula is that the investor needs to re-balance the portfolio every year which increases the cost, decreases the return.

    1. Yeah, I think it makes it much more difficult in practice. Although with a decent size portfolio, buying and selling 30 stocks (especially the larger cap stocks) shouldn’t be too costly.

      The thing I like most about the “formula” is the principle behind it. I like the fact that two simple metrics (ROC and EBIT/EV) are extremely powerful. It’s really simple stuff.

      I try to keep those two principles in mind, and remember that I’m buying a piece of a business. What am I paying for it, and how good is the business at generating good returns on the capital that it has to invest?

      These two big picture items helps me quickly determine whether I’m interested or not.

    2. Remember, your trading costs (for me average $6.95/trade) are highest your first year. After that, you may see the same stock on the list, which Mr. Greenblatt explains you should decide if you want to keep it or sell it. I am willing to incur a cost of $6.95/trade on $1,000.00 invested per stock. That is a .70% cost in. I do plan to go with Merrill Edge soon which will allow for zero costs. FOR WHAT IT’S WORTH, I am up 5.59% on my Magic Formula portfolio since Jan 2, 2015. The DJIA is at about -2.9%. I am a blessed man. I feel that those who don’t succeed in a strategy like MFI, try to presume why it can’t work and try to outthink it. If you are mentally gifted (I’m an average Joe who’s succeeded greatly in life), you may try to out-smart or “out think” the quant strategy that is Don’t do that! It’s a numbers game, not an intellectual exercise. Just do what he says and you should do quite well in comparison to the general markets.

  2. “the best investors in history have generated far better results with far less risk of permanent loss (not volatility) than the best computer systems” – how do you know that?

    When keeping track of the quant portfolios, how do you measure their risk of permanent loss against yours?

    1. Hi Fabian,

      Let me first say that there are a lot of great quantitative value “systems” out there (Magic Formula being one of them) that I think will likely work out well for practitioners in the future. I suppose my statement should have been more “far better risk adjusted returns”. I don’t necessarily think that the best computer systems carry a significant risk of permanent capital loss. For instance, buying the low 50 P/E stocks in the S&P will (just a guess) probably yield 1-2% better than the S&P 500 over time, and those 50 stocks will probably carry on balance the same or maybe slightly less risk. So I think that quant systemts (assuming they are using simple, logical value principles) don’t carry a lot of risk at the portfolio level simply because there is a lot of diversification.

      This diversification will also limit the quant systems’ returns. You won’t find a quant system that could come close to Joel Greenblatt’s 20 year record at Gotham (40% average returns with -5% being his worst year). So I guess my point was that the best value investors will always be better than the best computer systems. Certainly from an absolute return standpoint, and (in my opinion) from a risk standpoint as well.

      Your question about risk is a good one…. there is really no quantitative way for me to measure risk of permanent capital loss vs a computer system. That type of risk is more intangible. But to me, I just think about it logically. For example, to me, it’s riskier to own a lot of stocks that I know nothing about (or maybe a little something about some, not much about others, etc…) than it is to know something about each investment I make. I don’t need to know everything about each stock (there are always unknowns and uncertainties), but I want to understand why I made the investment. Why is it cheap? Why does the investment make sense? What is the value I’m paying for?

      These simple questions are hard/impossible to answer about each position in a 30-50 stock quant portfolio. So to me that carries more risk. That’s really how I measure it… I know I wouldn’t be able to publish this opinion in the academic journals, but that’s just how I look at it. Doesn’t mean it’s the right way, just my thoughts.

      I think that making good investments means paying less than the true intrinsic worth of the asset you’re buying. And if you don’t know the intrinsic value, then you’re taking on a certain amount of risk.

      It’s an interesting topic to consider. I like reading about quant ideas and like using them for screens, but I prefer to know something about the business I own-not necessarily everything-just why it’s cheap and what my general estimate of value is…

  3. Subbed to your website.

    Seems like you think like me, though by the sounds of it, many years ahead of me yet – nice to know that I am on the right path. Joel Greenblatts book is the best book I have read on Value Investing, but like you I don’t really like to just go around buying companies I have no idea about.

    Have you read Seth Karlmans Margin of Safety? I am working my way though that now, it is a nice read actually. I’m sure you have managed to pick it up.

    1. Thanks for reading Benny. Yeah Greenblatt’s first book is in my top 5 of all time when it comes to investing books. Klarman’s is in there as well. Margin of safety is a great one. I have a bound copy of it.

      Keep it up… investing is a journey and all knowledge is cumulative. Thanks for checking out the blog.

    2. Well, I knew nothing about STRZA and made 68% with MFI so far this year.
      USNA 59%
      GME 40%
      LCI 16%
      Sure there are losers, but it’s the few huge gains that make the difference.

  4. Hey, John Huber:

    Your posts have given me a great deal of information. Like you I’m gaining human capital and constantly reading. I’d recommend F Wall Street by Joe Ponzio.

    What is your opinion on George Soro’s theory of reflexivity? I was reading Seth Klarman’s Margin of Safety, and it mentioned how in rare times price will actually affect value of the underlying company. This actually made me fearful since then value investing wouldn’t work. One of his examples was a company in financial distress. In order for it to live on, the market will decide whether it’s worthless or not. If it thinks it’s worthless the company will go insolvent because no one wants to buy the stock they’re selling or no one wants to give a loan to the company.

    Even if the company is valuable, if the market doesn’t recognize it: management may take the market’s perception to heart and do actions that will eventually which will bring the market’s perception to reality.

    Soro’s interests me since sometimes price does affect value, and his theory of reflexivity sort of correlates with value investing.

    What do you think?

    I’d also like to have your insights on this paper on his theory:

    1. Thanks for the nice words. I appreciate you reading… Joe’s site is a great site. I’ve read most of his posts.

      As for Soros, he built an incredible record. I’ve read his book but find his methods too abstract for my liking. Soros is a philosopher. I get the impression from reading his work that he desires more than anything else to be the smartest man in the room. I’m not trying to disparage him, this is just my opinion I got from reading his writing. I do think he is an incredibly smart guy, and he probably has difficulty explaining his ideas to the everyday person, and that’s probably why his writing comes off like that.

      I personally prefer much simpler ideas. Reflexivity is a way to think about the world. It has some merit I think… but it’s very difficult to replicate a method like Soros’. It’s not impossible, but very difficult. It’s an art form.

      Plain vanilla value investing has some art as well, but it’s less abstract. It’s far easier to figure out a future estimated stream of cash flow on a stable business, or better yet, figure out how much the assets are worth that the company currently has than trying to figure out the opinions of value that the majority of market participants are going to place on a particular asset class or group of stocks.

      Keynes talked about the market being a beauty contest where the winner was determined not on beauty alone, but by who the majority of voters thought was most beautiful. This might be true in the short term, but in the long term, it’s not a voting machine, it’s a weighing machine.

      Your comments about companies caring about the way people think about them has merit when it comes to leveraged institutions such as financials. These companies could go bankrupt if there was a colossal run on the system (like 2008 or worse), but these events are rare and the way to defend against them is to not take on leverage risk. Companies without debt (or low amounts of debt) don’t have to worry about how the market thinks about them.

      So I respect Soros’ results, certainly they are tough to beat… but even tougher to replicate. Buying cheap stocks in the tradition of Schloss or Graham is much easier for an investor like me to replicate. I have the discipline and the emotional mindset to do so, and it doesn’t take a genius to do it.

      I usually ask myself these questions: Do I understand how the business makes money? Is it cheap? What are the insiders doing? What are the risks? What’s the upside?

      I try to find easy to understand situations where I can buy assets or earnings cheap. Doing that over and over again over time tends to work very well.

      I haven’t read the paper but if I do I’ll let you know.

      Thanks again for your thoughts. Always fun to consider different ideas. Thanks for reading…

  5. Thanks for your comment, John.

    Soros interests me greatly, and I attempted to crack into understanding what Soros does and how he does. I was thinking of trying to master reflexivity even though it’s so difficult. What I got out of it is that his theory states that our perceptions distort reality (ex: treating drug addicts like criminals will likely create criminal behavior) The we view reality is perception, and what Soros does is identify a bubble bursting or a bubble forming.

    Bubbles are formed through positive feedback which is a loop that supports the bubble. An example of it would be in the paper: when equity increases -> book value increases -> which increases earnings -> which increases demand for REITs (this is the bias/perception) -> which increases the price -> which increases the equity (not sure how it does this)

    The bubble supports itself through this loop, until negative feedback is introduced which disrupts the bubble. Which means that something came along to disrupt the bubble, which then causes a downward spiral with it’s own loop.

    As you have said Soros is an amazing philosopher and a great intellectual. Explaining his ideas is difficult (even he has admitted this in a lecture) I’d agree with you: value investing is much more easier to replicate (but of course harder to stick with due to the psychological aspects of it.)

    Identifying what majority of the market thinks is no easy feat and that’s what makes Soros: Soros. I was just thinking of somehow implementing reflexivity into a reflexive-value investing hybrid.

    Your risk post, on leverage, valuation, and business risk struck me as gold. When I struck upon your blog, it really resonated with me (investing in compounders, basket of quantitative cheap value stocks, special situations.. etc) I still want to understand Soro’s reflexivity and maybe incorporate it within my own strategy. If that’s even possible. Many people who work alongside with his Quantum fund, sure as hell have done it.

    Hard to replicate.. care to help me? :p

    1. Yeah people have done it… so it’s certainly not impossible. I think any time you want to make 30-40% annual returns you have to do things very differently than most. And Soros certainly did that. So did Greenblatt, Pabrai, and others who achieved those types of results. Schloss and Graham made 20% returns for decades just doing the same thing over and over again. Base hits… I like their strategy, but certainly it’s possible to improve them. But the key for me is always focusing on the downside. I’m not interested in generating higher returns at the expense of downside risk management. I’m interested in taking very small incremental risks for very large asymmetric upside… almost like a business would think of operational leverage.

      I think that focusing on getting on base, so to speak, will often lead you to enough fat pitches over the course of your career to hit enough home runs. The key is stringing together a lot of low risk results. Grind out consistent profits and at some point, maybe the GEICO type investment comes along.

      Good luck and glad you’re motivated. Lots of opportunity for individual thinkers in markets…

  6. Thanks John for your interesting insights. I too am a fan of Joel’s value investing strategy. I feel it could be improved upon or at least give more comfort to the investor is by looking at each stock more carefully like you do. One method that could be employed would to use Weiss Ratings(he is very conservative) to choose only “A” or better stocks coupled with 3 month and 1 yr. track records(choosing the best after looking at their charts for volatility and increasing value) leads to the best 5 picked quarterly. I like the 250 mil. min. cap range and higher since it removes some volatility.
    Also I like to stay out of the mkt. when it closes more than 3% below the 250 day simple m.a.(S&P500 daily chart). This will keep me out of bad situations like we weathered in 2001-03 and 2008. (back in the mkt. when the reverse occurs, 3% above the m.a.) At my age now 70 this yr. I cannot withstand big drawdowns and feel this value system coupled with some technical analysis can give one a good night of sleep.
    Thanks again, Let me know how you feel about all this?
    Much appreciated, Roger

    1. Hi Roger,

      Thanks for the comment, and I appreciate you reading… I’ll chime in briefly on the technical analysis… I don’t use it and I don’t believe it adds any value long term. I think those simple moving average techniques can get you out before a market crash, but on balance, they’ll give you too many whipsaws and cause lots of false signals for your account. In hindsight, it makes a lot of sense. The problem is we don’t know when the next 2008 is coming, so it’s very difficult to see in real time. I’ve never seen any reliable track record that was using real money that consistently resulted in significant outperformance over a full cycle (say 15-20% per year over 10 years or more).

      So I’m not a big fan of the types of strategies that you are talking about. Doesn’t mean it won’t work, I just don’t believe it will work over time… just my opinion.

      Plus, there is a philosophical difference between those systems and how I invest. When I buy a stock, I think of it as a piece of a business, and not a trading vehicle that I need to dispose of after it drops below some statistical average. In fact, when a stock I own goes down, I often want to buy more assuming the conditions and the economics of the business haven’t changed.

      Again, I don’t want to discourage you, but I would provide a word of warning that based on my own experience and empirical observation, I have never come across solid evidence that those moving average systems work consistently. I think the easiest and most reliable way to allocate capital is to buy good businesses at cheap prices.

      It’s a great discussion, and I’d be happy to talk more about it… might be worthy of a post at some point.

      Thanks again for reading, and thanks for the comment.

    1. Hi Mike,

      There are a lot of different value quant strategies that have done well over time… low P/E and low P/B are probably the simplest and most commonly cited. A professor at Chicago named Joseph Piotroski did a study that combined low P/B stocks with some quality metrics. His study is worth reading. There are lots of studies done on Ben Graham’s net-net strategy and that has worked for the past 75 years or so (over time). Another strategy that has done well in various tests are negative enterprise stocks (basically, stocks with net cash in excess of market value). Toby Carlisle and Wes Gray wrote a book called Quantitative Value (so they literally wrote the book on Quant Value 🙂

      That book is excellent…

      There are others as well but that gives you a few ideas…

  7. Very informative posts here.
    Last time I invested was about 20 years ago, and it feels like starting over now.
    I may inherit some big money in the next 5 years, but for now I’m just learning and “warming up”.
    Which software (or web-based) tools that allow investors to assess and track investments do you recommend ?
    Does Sean Hyman’s strategies and reports have merit, and are worth the $50 or so dollars per year ?

    1. Hi Martin, welcome back to the game. I am not familiar with Sean Hyman. My recommendation would be to save the $50, spend $10 or $15 on a used copy of Intelligent Investor, and download and start reading through Buffett’s shareholder letters (Free, and you can find them on this site). That’s all you really need to get started. Take an accounting course if you are not familiar with financial statements.

      I don’t use much software. I use Google spreadsheets to keep track of my ideas, and I use Value Line and Morningstar. that’s about it.

      Feel free to reach out with other questions: john at

  8. Peoples’ desire to be above average is a delusion which leads to a gambling instinct. That causes people to deviate from the rational and optimal path. You should check out Larry Swedroe’s articles on Seekingalpha, where he debunks most of these value investors, and especially active managers who take in fees. They nearly all underperform their relevant benchmarks, including the “Superinvestors” that Warren Buffett picked out. The problem in my opinion with value investing is it is overly popular, so there aren’t as many irrational counter-parties to exploit any more (and on average it’s cheaper and easier to exploit them passively). Also there isn’t any strong reason that one should be able to out-perform professionals who are highly paid and work hard at value investing all day long (and who also underperform their benchmarks).

    1. Hi Connelly,

      Your comment touches on an interesting age-old debate that will always be a part of markets. I’ve always been interested to read things like this, where the logic is basically: No one can beat the market consistently or predictably over time… basically, it’s luck, and any attempt is inevitably doomed to be average.

      But I think it’s really a simple concept. In any endeavor that requires certain combinations of skill, talent, brainpower, etc… there will always be some that excel, some that are average, and some that do poorly.

      I don’t think it’s delusional to attempt to become great, as some have achieved it. Most of the things that are referenced about “see what this guy says, he proves that attempting to beat the market is futile” is usually contrived by someone who has either been unable or unwilling to beat the market themselves.

      The interesting thing is that I probably share a lot of thoughts in common with you, as I believe that most (in fact, the vast majority) will not beat the market over time. However, there will always be some that do, and those ones will continue to be able to do it consistently.

      Also, keep in mind that the edge is not in superior brainpower (although that helps). Nor is it in large resources, huge bankrolls, and lots of staff. The professionals actually have a lower probability than individual investors of beating the market. This is a topic for its own post, but two large reasons are capital constraints (harder to beat the market with $1 billion or more when you have to diversify), and the institutional imperative. The latter is a broad term, but it is very powerful, and it causes even smart value investors to revert to mediocre results over time if they don’t understand the basic framework of what cause superior results. Basically, too many positions, too much pressure to keep pace with the market, and too much focus on short term results (all of this because of fear of losing AUM).

      There are many ancillary points here that I’d love to discuss.

      I’m genuinely interested in this topic, and your comment, as I think a lot of people feel this way. But it’s unfortunate because individual investors have the best chance of incredible results.

      Just through writing this simple blog, I’ve had the good fortune of connecting with incredibly talented investors, some of whom run small private funds that you’ve never heard of, and others who are quiet individual investors who have been investing using Graham and Dodd principles since the 1960’s. I was introduced to a gentleman through a friend of the blog who I talked to last week who has managed his own money for a few decades and is now retired who has averaged better than 30% annually, building a small stake into a large retirement account in the process. This is an exceptional story, but as my friend Nate over at Oddball stocks says, there are “lots of Graham and Dodd investors out there quietly making 15-20% per year doing the same thing over and over again”.

      I’d add to that comment by saying that I know of numerous small value funds that manage small sums who are not part of the public domain that have achieved excellent results over the past decade. Some in excess of 30% annually.

      There are lots of people doing it. And lots of people saying it can’t be done. I think this will always be the case. Some will argue it can’t be done while a select few will happily let them think that while they continue to produce above average investment results.

      Buying a dollar for 50 cents seems to be logical and commonsensical in every aspect of life, except in the stock market. For some reason in this field, that logic doesn’t seem to take hold with many participants.

      I’d sum it up by saying that I think you’re right in that most people won’t beat the market. But I disagree that attempting to do it is futile. With a few key gifts (average to slightly above average intelligence is all that’s needed) and the right mindset, and the proper discipline, it can be done and will continue to be done.

      1. Hi John,

        Thanks for your detailed reply — it’s always good to have a nice debate!

        So we should distinguish between beating the market and beating the relevant benchmark. A lot of value investors beat the S&P 500 which is easy since the equity risk premium is higher for value. What they don’t realize is they should be comparing against a passive vehicle of their same asset class such as Vanguard Small-Cap Value, on an after-tax basis. Since within that asset class historical returns have been good, and active value investors are adding or removing economic value relative to that benchmark. However even better would be to weight on some of the quality factors such these Robo-Buffett factor published by AQR or Piotroski F-score. Since many of the techniques used by value investors have now been made passive, it’s appropriate (at least a couple years after the technique is published) to use the best possible passive benchmark as the hurdle.

        Now I agree that it’s going to be infeasible to get 20%+ consistently using passive techniques. And I agree that it’s worth doing for certain cases and for small investors who can get an edge, and if they have an edge I wouldn’t be surprised if they can maintain this return. (I do not think it’s at all worthwhile trying to pick mutual funds). However, I also think a lot of people who aim to be in that situation are basically delusional and are using generic value techniques with average ability, and have no real edge. My main challenge is to clearly identify who are the constrained counter-parties that one should exploit, and clearly identify what is the edge and how can it be maintained. By maintained I mean that the passive benchmarks keep improving so you have to consider the opportunity cost of active investing relative to those. (In other words, the argument for active value investing was much easier 20 years ago than today).

        I think that properly considered, this problem of active value investing to beat the proper passive benchmarks is extremely hard. So I’ve focused a bit more on the passive techniques recently. But also I continue to do value investing in the active part of my portfolio for only situations that are no-brainers in my strike zone, and have clear constrained counter-parties and/or catalysts.

        I actually have done value investing for the last couple years, or value/quant. What I noticed was my performance generally tracks these benchmarks but is less tax efficient. My annualized return since I am a newbie starting in 2012 is 22% vs 19% for Vanguard Small Cap Value. It is easy to get good performance in a rising market, and on an after-tax basis my performance was almost certainly worse. So I may be a bit jaded or perhaps also critical of value investors who keep saying “Superinvestors!” or “I beat the S&P 500!” Here are some representative articles of Larry Swedroe’s explaining what is wrong with such claims:

        In addition to Larry Swedroe, I also enjoy following the ideas of Chris DeMuth on Seekingalpha, and you can absorb a similar criticism of generic value by reading his articles, since he’s a specialist value arbitrager.

        Finally, I should add that when following the ideas of the best investors, one must also consider that they are likely in the top 5-10% in terms of the distribution of luck. For example Buffett has used significant leverage (Forbes claims an average of 1.6) which obviously increases the risk of bankruptcy — and no one would even be talking about him if he had gone bankrupt.

        In summary, these aren’t arguments against the worth of active investing in every case, but they are arguments that restrict the set of situations in which value investing can add any value.

        1. Hi Connelly,

          Excellent thoughtful comment. I happen to agree with most of your thoughts. One quick comment on Buffett though: The 1.6 leverage factor is there, but it incorporates everything (the entire capital structure). We (or most of us at least) don’t run mini-Berkshire Hathaways (conglomerates that use insurance float to invest into equities and other businesses), so it’s not really accurate to say “well Buffett uses leverage and that’s why his stock picking has been good” (not implying that’s what you meant, but I’ve heard that argument a lot).

          Buffett has used leverage, although very moderately for that business. And that has helped him increase the book value of Berkshire, but if you judge the stock investments on their own, he still has averaged over 20% per year, with a huge pile of money to move around, vs. an index that has averaged about 10. In fact, there is a study of Buffett’s results from 1976 to 2006 that basically say if you just copied Buffett’s moves (bought the stock he bought on the last day of the month that it was announced publicly and sold stocks that he sold on the last day of the month that it was announced), you would have beat the index by about 10% annually (20% per year) from 1976-2006. So his stock picking alone should be judged independently of his performance as CEO of Berkshire. And of course, we know about his 33% results of his partnership, when that was all he was doing is owning equities with no real leverage.

          Anyhow, I liked your comments. I haven’t read the links, but I’ll take a look.

          I happen to agree that for most investors, a value weighted or Magic Formula type approach to investing would be a very good alternative to other ideas. Or maybe just a value index. I personally think that the Magic Formula, or Quant Value, or any of the other approaches that focus on simple value metrics like low P/B, low P/E and some others will likely add noticeable value over time. And I think your criticism has some merit as there are a lot of value investors who underperform. But I think that’s just a product of the competitive nature of markets (there will always be some that do better than others, and of course many who are average). It’s certainly not easy, but the principles are simple.

          But I understand your skepticism. I’ve been skeptical myself when I hear some smart guys who are well respected, even in the value investing community, who seem to produce very average results over time. But I think it’s more a product of how they operate their businesses, the institutional imperative I discussed. One value guy that is well known will frequently discuss Buffett, Graham, and all of the rhetoric that makes sense, but then you look at his portfolio and he owns nearly 50 stocks, some long, some short, and some with P/E’s of 50+, and it leaves you scratching your head. I’ve written a few posts on thinking differently.

          Study those who have verifiable long term (20 years or more) results like Greenblatt (30% for 20 years), Buffett in the 1950’s and 60’s, Graham, Schloss for 50 years, etc… there are many others as well like Pabrai, Greenberg, Klarman and others.

          By the way, an investor’s track record really needs to be 10 years before you can get a decent gauge of your true ability. Anything shorter is subject to too much luck. 5 years is the absolute minimum that I would draw conclusions from, but when I study other investors to learn, I want to know how they’ve done over 10 years. No institution will allow you that much time, but individual investors can build a plan for the long term.

          And one last word about luck…. I would say the short term results are luck. They could be far above or below an average. But over time, I don’t ascribe luck to those with 30 year track records of trouncing the market. Other than the luck you would ascribe to the natural ability they were blessed with. From that perspective, they are lucky. Buffett calls himself lucky the day he was born. But ascribing their performance to luck is like saying Michael Jordan is lucky or Tiger Woods is lucky. They are lucky to have been given great talent, but their performance is skill based, and has nothing to do with luck (in other words, if they were lucky, there might be a chance that I could one day go out and shoot a 65 on the golf course (when my handicap is around 18)… it’s just not going to happen. Nor will I ever be able to beat Jordan in basketball). Similarly, Buffett, if he was given the chance to start over at age 21, would still trounce the market and just about everyone else. He’s just that good. I don’t know a single person that would honestly bet against Buffett if he were given a small portfolio of $1 million and given 20 years to compound it. If it were luck, or even a large portion due to luck, that wouldn’t be true.

          So in the end, I think the index approach has a lot of merit, especially for people who don’t have the time and energy to put into investing. And I think there are great investors, just like great athletes, who will always be better than the majority of us. But I also think investing is an endeavor that one can become proficient at over time…. like playing the piano. If one works hard enough, they will be able to play some nocturnes by Chopin, maybe even some Beethoven sonatas. With some talent, they’ll even play Liszt and Rachmaninoff, but most of us will never be able to be good enough to play at Carnegie Hall, regardless of our work ethic. But it doesn’t mean we can’t gain proficiency that allows us to play at an above average level.

          I think the same can be said for most individual investors. They may never be the next star fund manager, but they have a decent shot of gaining above average proficiency if they have the mindset and the work ethic to be able to invest 10,000+ hours into this.

          For the ones that don’t have 10,000 hours to spend practicing this, I think finding a value manager with the right mindset or just simply going the passive approach is the way to go. And I think you mention a lot of good points regarding the challenges that they’ll face with the latter approach.

          Great comment! Thanks for reading Connelly. Always interesting to discuss these types of ideas.

          1. Thanks for your thoughts John, and also for this blog. I’ve enjoyed reading now and then over the last couple years.

            The way I look at it is, there is a finite supply of alpha in the value space. The “basic edge” of the entire space is supplied by psychological errors of other constrained investors. But much of the value space can now be replicated passively. For example, Buffett in the 1990s had outperformed passive value, but in the 2000s underperformed passive.

            One interesting phenomenon is that even within the area of the “basic edge” of value investing, mutual funds and humans tend to underperform the passive value averages. For example, Greenblatt was discussing somewhere how the average person who chose particular Magic Formula stocks would actually greatly underperform the Magic Formula average performance. And many funds underperform their indices for institutional reasons, even before subtracting fees. But this implies that finding what the average value investor likes, and inverting it, should do better even within value. Cheap companies that people know and understand should on average underperform the Fukushima power plants and Greek mortgage issuers of the world (the ideal investment would probably emit both radioactivity and unstable Greek credit derivatives). Even in a ghetto, some apartments are cheaper than average. They aren’t pretty, or safe, but you get paid a sweet premium to live in them.

            As I said, my view of good active investments is heavily influenced by understanding (1) Can a passive strategy not do this well? (2) Are the natural buyers gone?, (3) Who are the constrained counter-parties to exploit? For example, derivatives, over-shorted issues with high borrow cost (which one can capture through IB or with a synthetic long if one’s broker does not share short interest), event driven, spinoffs, OTC issues especially foreign ones that aren’t penny stock scams (RHDGF?), imminent catalysts (AVG for mobile growth — short puts looks promising to me and Okumus has a large position), etc. Typical early Buffett and early Greenblatt.

            I think humans can also add economic value through position sizing. While we’re psychologically terrible at refusing to buy hairy companies that we ought to, we’re psychologically great at figuring out when something we’re familiar with is a steal and loading up. A computer would not be able to pinpoint that AAPL at precisely $400 or HPQ at precisely $12 are steals, so it has to diversify more, since the future price is more of a random walk to the computer. But I can easily take a large position without worrying since I understand those companies well. Then since I don’t believe there are many mispricings in large cap, I can just close those positions as they revert to cheap.

            Obviously at a high level I’m expressing opinions that look a bit subtle or even contradictory. But another way to look at this is as a scientist I have a strong belief that the factor analysis of French and Fama etc. explains most of stock price variance, and that markets are efficient up to risk premiums. So I believe most people should just passively harvest that “basic edge” and have that as a fraction of my portfolio. But I also have strong belief that arbitragers can add value in certain niches. So my investing philosophy is the intersection of those two beliefs.

          2. Interesting… thanks again for the comments Connelly. By the way, I love Charlottesville. Beautiful campus up there. If you’re ever in Raleigh, reach out and we’ll grab coffee sometime.

  9. Hi Mr. Huber,

    Really enjoyed you post! I am 18 and have also read Greenblatt’s book and have created a portfolio and utilized his strategies. The book is, without a doubt, clear and powerful.

    I was wondering, however, what exactly Greenblatt uses to generate the stocks he selects (on his website). I understand he seeks bargain companies that appear to be undervalued (low p/e ratio). But what exactly is the other aspect he looks for? The EBITA and so on?

    Thank you so much and I will certainly be checking in!

    1. Hi David,

      It’s great that you’re looking at this stuff at such an early age. If you start early and stay disciplined, you’ll do very well with such a long investment “runway” ahead of you.

      As for Greenblatt, he doesn’t disclose specifically how he chooses the stocks he buys for his fund, other than to say his “formula” is based on the same principles he talks about in his book: namely, return on capital and valuation. He uses EBIT/tangible capital for his ROIC calculation in his book, and I assume that’s what he uses to pick stocks, with minor adjustments here and there depending on the company or industry. Each situation is slightly different, and each business model is slightly different, so adjustments might be made on a case by case scenario.

      Thanks for reading.

  10. Just wondering why the screen chooses a one year holding period, it seems quite arbitrary to me. I feel that if a stock were to have large gains, say 30%, wouldn’t it be wise to lock in those gains and sell right away, instead of waiting out the full year…?

  11. Hi John,

    Question #1: Are there any stock funds that apply magic formula investing?
    Question #2: Is it better to buy small, middle or large cap enterprises with magic formula investing?

    Thank you for an answer!

    1. Hi Robert, yes Greenblatt runs a management firm that runs mutual funds based on the formula. However, the exact method of his operation he doesn’t disclose. It’s not exactly the formula described in the book (2 metrics, buy 30 stocks, etc…). His funds hold 300 stocks or more. As for market caps, I think the smaller caps open you up to a higher potential (but also more volatile) return. Large caps won’t produce the same results, but the formula still works (according to Greenblatt) on the large cap universe.

  12. John,
    I was interested to read your thoughts on MFI. I’m quite new to Greenblatt’s books. In deciding to put his strategy into play, i.e., pick 5 to 7 to start with, aiming towards 20 or so over the course of the year, I noticed a few things:

    1) He gives two “how-to” options, the second or DIY one being to use ROA = 25+ and P/E low but not below 5. The odd thing is that at his MFI site, among 30 or 50 companies with market caps $50m+ there are very few that meet this criterion. My count the other day was 6 of 50 and a few others that are close. That makes one wonder exactly what the correlation is between using his lists and using the two-part criteria. I ended up choosing primarily from those several that met the latter (interestingly, my first three choices are among the top four rated at the MagicDiligence site). I wish there was a way to ask him about this (see next point).

    2) In the appendix or notes of the second edition of his book, he discusses the numerous suggestions he got to also short the worst 30, but rejects it as not gaining anything and creating much greater risk of going broke. But when I look at the Gotham funds website, shorting too is what he’s doing now with all of his funds! I emailed to find out where there’s an explanation for this, but he’s yet to respond.

    3) In telling readers how long to hold, Greenblatt faced an obvious pickle, to which he took the understandably easy, yet arbitrary, way out. I read somewhere that in his study he turned a stock over when it doubled in price or after two years, whichever came first. At least that has some sense to it beyond taxes. To me, this is one place where individual choice most definitely ought to come into play. In choosing the first five, I rejected a number of companies, at least for now, because they were at or near highs (P/E’s also usually reflected this) or their quarterly briefings were predicting tough times ahead. While I’ve been around life and various companies long enough to have a decent sense of different industries and, thanks to Jim Stark, some idea of industry rotations in different periods of bull markets, I don’t usually feel knowledgeable enough to judge whether or not a stock might continue to run, let alone want to carry the extra risk of that uncertainty.

Leave a Reply

Your email address will not be published. Required fields are marked *