The Magic Formula Experiment

[Totoro]

Intend to pick this thread up every so often in the coming years for myself and others to enjoy, so best to start afresh. I posted bits and pieces already around the forum earlier.

After reading the Little Book That Beats the Market and several other books, thinking a.o. about the soundness of the approach (vs. the off-putting name of "magic formula" ..) and checking up on Greenblatt and doing some desktop analysis, I decided to launch an experiment with real money: my own.

For those who don't know: The approach is basically selecting value companies with high returns on capital.

I started in february this year (2014) as follows:

• Ran the on-line screener for 50 companies larger than 300M market cap
• Selected 10 stocks from the 20 with the smallest market cap that I liked the most. This after reading their annual reports and doing some basic research. I put the 20 companies in three buckets ("1", "2", "3" and "out"). The way it worked out I took the full "1" bucket and a few "2" companies.
• In each of those 10, I put about 2.500 euro (+/- 3.2k USD at the time) in. So that's 32k USD at stake (4% of my NW).
• Then, I waited ..

First results after 9 months (Nov. 17 2014). Note that this is without dividends, but the general picture is the same (did some spot checks).

• S&P 500 gained about 13%
• The entire 50 company list gained about 13% too
• The smallest 20 companies gained about 17%
• My selection gained about 4% (oops)

Observations:

• There is a huge spread in returns between individual companies within the list. I'll post a histogram later on.
• Given that, 10 stocks is not nearly enough to diversify. The book mentions this, now I understand better why.
• The seven companies I liked the most (with a "1") had a 19% return in aggregate. That went well.
• The 3 companies I took from the "2" bucket ruined my performance. Two them tanked with 47%.
• The 10 companies I didn't select had wildly varying returns. e.g. two of them more than doubled, one halved.

 

[Totoro]

The return histogram. These are the 50 companies in the original screener. You'll see only 49 since one seems to have been taken over (didn't check up yet) - can't get a quote for it.

 

[Totoro]

My conclusions and way forward

• As a first step (just 9 months) this doesn't look bad. I am going to continue the experiment.
• I need to extend the number of companies I select to dampen volatility and/or only select larger cap companies.
• Since I screened out the biggest gainers and managed to include some big losers (although in bucket #2) I am going to trust the process completely this year. Scary as that might feel.
• In the mean time I found out about the 'january effect', since I only started in february so I might have missed some upside.

So this is what I'll do around the 15th of december this year

• Run the screener for 50 companies
• Select the 20 with the smallest market cap. Allocate again 2.500 euro to it (now worth 3k USD or so). Only check I'll do is if there are strange things like reverse takeover chinese companies in there.
• Buy them, and wait until next year december comes around to examine the results

Note that I am fully prepared to lose my full investment in this. It would be a setback obviously with it being 8% of my net worth. The money though is coming from my individual investment account.


Index tracker account (my benchmark) and cash/bonds remain untouched.

 

[Huston55]

Interesting experiment. I read that book several years ago. As an experiment, I made a 'watch list' and tracked it. I stopped tracking some time ago and frankly don't remember how well it performed. But, looking at my old "Magic Formula" watch list, I see that 13 of 26 companies I used to track no longer exist. Seems like pretty high attrition, even after several years.

 

[Totoro]

Would be interesting to see if it is because of merger/take-over, name change or just plain bankruptcy.

 

[LOL!]

This is a great idea for a thread. Thanks! Good luck!

 

[sengsational]

I have a plan to do this too, but haven't gotten around to tranfering the money to a broker with so many free trades per year. But I'm going to do zero research, and only look at it a few times a year, when I buy/sell. I don't even plan on knowing what my gains/losses are until I've been at it 2 years. T hats the plan, anyway.

 

[Huston55]

Would be interesting to see if it is because of merger/take-over, name change or just plain bankruptcy.

I wish I knew but, I don't know where to look up old tickers of now former companies.

 

[CaliforniaMan]

Now let me see if I understand the plan. It is to use screens that basically everyone else has access to as well as you, and screen for attributes that you think will beat the market.

Now I would assume that since everyone else has access to these same screens, including pretty much every mutual fund manager in the world, and since they have probably have a lot more time to spend analyzing screens than you do, they pretty much have tried all of this in the past and are still trying now. And we all know how well it has worked out for the active managers of mutual funds.

So somehow, using the same information that everyone else has, and having less resources to analyze it, you will beat others trying to beat the market. Of course next year some stock screens will have worked better than others. We will certainly know next year which ones those were.

Instead of trying to pick the best screen, why not just pick some random ones. You've got about the same chance to beat the market. Maybe it might even work better, because since everyone reads the same news I would bet there are a lot of people picking the same screens as you.

Sorry if I sound a bit harsh, I just don't understand how people think this kind of thing can work better than chance.

 

[clifp]

Now let me see if I understand the plan. It is to use screens that basically everyone else has access to as well as you, and screen for attributes that you think will beat the market.

Now I would assume that since everyone else has access to these same screens, including pretty much every mutual fund manager in the world, and since they have probably have a lot more time to spend analyzing screens than you do, they pretty much have tried all of this in the past and are still trying now. And we all know how well it has worked out for the active managers of mutual funds.

So somehow, using the same information that everyone else has, and having less resources to analyze it, you will beat others trying to beat the market. Of course next year some stock screens will have worked better than others. We will certainly know next year which ones those were.
.

Have you actually tried purchasing a small cap stocks? Cause if you had you understand why this isn't a strategy that can be pursued by money manager and especially indexers.

The average market cap of VTI is $40 billion or more than 100x large than what size companies he is buying. VB Vanguard Small Cap the average market cap is just under $3 billion or 10x as big as he is buying. The average position size of VB is over $5 million. The liquidity of companies stock is so small that isn't not worth the time and effort for almost any institutional money managers to deal with them.

I actually own a couple of microcap stocks. One of the Seneca foods is just over $300 million market cap. You've never heard of Seneca but you probably have seen their products Libby frozen food (especially vegetables) in your grocery store. I bought 130 share for $25 back in 2007 and since climbed to $30-$31 hardly a great investment. I was using a system that was screening for stocks that were profitable and selling for below book value. It is hardly a tiny company 1.3 billion in sales and more than 3,000 employees, but barely profitable. I have been trying to sell the stock for the better part of 6 months. Right now the bid is $25 and the ask is $37 the spread will close to a $1-$2 during trading hours. Prior to last week heavy volume of 1,800 share being traded, there were 14 straight days of no shares being trade.

Now imagine you are a money manager how would you take even a $1 million position in a company like this without taking a beating when buying or selling? I decided that with my several million portfolio it wasn't worth my time, because I was basically limited to taking four figure position of the size Totoro is talking about. So even with 30 stocks were talking a few percent of my portfolio. But if had 500K and time and interest it would be worth pursuing.

Contrary to your assertion institutional investor are at huge disadvantage to individual investor when it comes to trading in microcap stocks. There is generally either 0 or 1 analyst following and I'm willing to bet I know more about the company than the analyst.

 

[Totoro]

I wish I knew but, I don't know where to look up old tickers of now former companies.

Did a quick google search, it seems that kind of service only exists at high cost. I just put it on my business idea list (not that anything will happen because of it).

 

[Huston55]

Did a quick google search, it seems that kind of service only exists at high cost. I just put it on my business idea list (not that anything will happen because of it).

I looked too but, couldn't find a site that lists them. Too bad because, now I'm curious about what happened to those old companies.

 

[Totoro]

It is to use screens that basically everyone else has access to as well as you, and screen for attributes that you think will beat the market.

Yes, everyone else has access to that same screener. I won't be doing any additional analysis though.

And we all know how well it has worked out for the active managers of mutual funds.

85% lag the market. I think it's 50% (almost by definition) once you remove the management fees.

Instead of trying to pick the best screen, why not just pick some random ones. You've got about the same chance to beat the market.

Unless the approach is sound. The arguments for that are:

• These are low P/E companies - i.e. cheap. Buying cheap can't be wrong.
• They have a high return capital (a proxy for good). Buffet uses the same criterion (net earnings on tangible capital in 'normal operation', buffet prefers >20%). As a private owner it is pretty much the only metric that matters at the end of the day (apart from expansion possibilities).
• It has worked in the past
• It is not accessible for large investors if you go small and micro cap
• There are lumpy results. Some years not so good, other years good. This disqualifies the approach for most money management firms and impatient or nervous investors (speculators?)
• The screener updates daily (almost). If others believe in the approach prices will rise and the relevant stocks will drop out of the screener pretty fast.

Now, the arguments against are why I am running the experiment and not go full in:

• The screener works on the last full year reported earnings. That's a poor proxy in my view for true earnings power. There is no correction to normalize earnings, adjust for one-offs, strategic considerations, you name it.
• Out of a 1000 strategies there are always some that work well out of pure luck. No way to know for sure.
• The book was published in 2006, the advantage may have diminished now. Then again, it still worked up to 2014, certainly in small cap.
• There are some corrections going on behind the scenes that do not seem to be fully disclosed. I don't like hand waving.

Sorry if I sound a bit harsh, I just don't understand how people think this kind of thing can work better than chance.

Not harsh, valid points.

 

[Huston55]

Now let me see if I understand the plan. It is to use screens that basically everyone else has access to as well as you, and screen for attributes that you think will beat the market.

Now I would assume that since everyone else has access to these same screens, including pretty much every mutual fund manager in the world, and since they have probably have a lot more time to spend analyzing screens than you do, they pretty much have tried all of this in the past and are still trying now. And we all know how well it has worked out for the active managers of mutual funds.

After 30+ years of index fund availability (admittedly more mainstream in the past 20+ yrs), and data which clearly demonstrates that passive index investing outperforms active fund investing (After 5 yrs ~75% of active funds lag their index, and only ~50% of those active funds remain within their style criteria or even exist at all), STILL only ~25% of investments go into index funds.

Clearly, availability of information doesn't translate into action by all.

BTW, even though I invest in index funds, I still don't eat enough vegetables.

 

[ERD50]

...
Unless the approach is sound. The arguments for that are:

• These are low P/E companies - i.e. cheap. Buying cheap can't be wrong.
  ...

...

Yes, it can be wrong (I've got the capital losses to prove it!). Sometimes (often? usually? almost always?), there is a reason for the low P/E. I purchased a few low P/E companies with that reasoning, and they continued to go down, down, down. Their business was failing, and that's why they were 'cheap'. If the majority thought they were growing, they wouldn't be 'cheap'.

-ERD50

 

[CaliforniaMan]

Have you actually tried purchasing a small cap stocks? Cause if you had you understand why this isn't a strategy that can be pursued by money manager and especially indexers.

The average market cap of VTI is $40 billion or more than 100x large than what size companies he is buying. VB Vanguard Small Cap the average market cap is just under $3 billion or 10x as big as he is buying. The average position size of VB is over $5 million. The liquidity of companies stock is so small that isn't not worth the time and effort for almost any institutional money managers to deal with them.

I actually own a couple of microcap stocks. One of the Seneca foods is just over $300 million market cap. You've never heard of Seneca but you probably have seen their products Libby frozen food (especially vegetables) in your grocery store. I bought 130 share for $25 back in 2007 and since climbed to $30-$31 hardly a great investment. I was using a system that was screening for stocks that were profitable and selling for below book value. It is hardly a tiny company 1.3 billion in sales and more than 3,000 employees, but barely profitable. I have been trying to sell the stock for the better part of 6 months. Right now the bid is $25 and the ask is $37 the spread will close to a $1-$2 during trading hours. Prior to last week heavy volume of 1,800 share being traded, there were 14 straight days of no shares being trade.

Now imagine you are a money manager how would you take even a $1 million position in a company like this without taking a beating when buying or selling? I decided that with my several million portfolio it wasn't worth my time, because I was basically limited to taking four figure position of the size Totoro is talking about. So even with 30 stocks were talking a few percent of my portfolio. But if had 500K and time and interest it would be worth pursuing.

Contrary to your assertion institutional investor are at huge disadvantage to individual investor when it comes to trading in microcap stocks. There is generally either 0 or 1 analyst following and I'm willing to bet I know more about the company than the analyst.

Yes I have traded small caps, over the counter, and the prices were highly manipulated. They were all legitimate companies, but traded on very low volume. It is very easy to manipulate prices when only a few hundred shares are traded and that is exactly what happened. I mean how can it be otherwise? Sometimes they would rise 2 to 5% on only a couple hundred shares traded. With such small volumes traded so infrequently there is no way to match buy and sell orders by prices rising or falling until buy volume matches sell volume. So they basically guess at a price, someone wants to buy, go up a half a point, look for a seller. My investment did very very well, well, until it didn't. That was a couple of decades ago, and I haven't been back to play that game since.

 

[CaliforniaMan]

Now I don't mean to say that trading small caps cannot be fun. I had a friend who became technology director of a new medical device company. The company went on the market at $10 a share. He was obviously proud that as an insider he got to buy it at $8 a share and told me I should get some. A year or so later I bought at just above $2 a share. We both sold later at 16 when the company was bought out.

Another time I owned local company. I doubled or tripled my money (I forget) all during a down market. When the market turned around and started going up, I thought boy these stocks are going to do really well now. They went to zero.

"And she'll have fun, fun, fun
Till her daddy takes the t-bird away"

 

[Totoro]

Yes, it can be wrong (I've got the capital losses to prove it!). Sometimes (often? usually? almost always?), there is a reason for the low P/E. I purchased a few low P/E companies with that reasoning, and they continued to go down, down, down. Their business was failing, and that's why they were 'cheap'. If the majority thought they were growing, they wouldn't be 'cheap'.

-ERD50

If the business loses money and destroys capital you are right. If it still makes money there is always a price at which point it is cheap. Negative P/Es are filtered out via the high capital return requirement. The last part is quite crucial.

Few examples from the list of 20 companies I picked from last year
STRA Strayer Education Inc XNAS:STRA Stock Quote Price News
Declining revenues, stock priced doubled. I didn't actually include this one because of the declining revenues. I was wrong.

http://quotes.morningstar.com/stock/s?t=ETM®ion=usa&culture=en-US
Declining revenues for years, stock price from when it showed up in the list until today: +5%

http://quotes.morningstar.com/stock/s?t=EGL®ion=usa&culture=en-US
Revenues halved past few years, results for that one are +15%. Note the heavy volatility!

Now in the long run you are again obviously right. If the company keeps on declining, even with profit it eventually dies out and no value is left. That can take a long while though. If the owners gets paid in the mean time, the result can still be happy.

For an extreme example - just random find. Look at this company the past 5 years.
http://financials.morningstar.com/income-statement/is.html?t=RAI®ion=usa&culture=en-US

 

[gauss]

I looked too but, couldn't find a site that lists them. Too bad because, now I'm curious about what happened to those old companies.

If I were going to check into this I would start with the SEC Edgar site that provided access to SEC filings of public companies.

I know that it was free back in 2009, but there may be a bit of a learning curve to get proficient at it.

-gauss

 

[Huston55]

I looked too but, couldn't find a site that lists them. Too bad because, now I'm curious about what happened to those old companies.

If I were going to check into this I would start with the SEC Edgar site that provided access to SEC filings of public companies.

I know that it was free back in 2009, but there may be a bit of a learning curve to get proficient at it.

-gauss

Great suggestion Gauss...thanks.

The SEC Edgar DB had most of them. I'm guessing my old "Magic Formula" watch list is ~ a decade old. So, just for $hit$ & giggles, below is the list including those companies that no longer exist and those that still live

The Dead
KG: King Pharma
AEOS: ?
DRTE: Dendrite Intl (Stwe)
FTO: Frontier Oil
JH: Harlan John (Book binding)
HOC: ?
INSP: Blucora (Computer processing)
KSWS: K Swiss (Footwear)
KOSP: KOS Pharma
MVL: Marvel Entertainment (Patents)
MRX: Medicis Pharma
VCI: Valassis Communications (Advertising)
WON: ? (Perhaps the ticker should have been "LOST")

The Living
KND
HRB
CECO
CSGS
FCX
JAKK
KFY
MDLN
NSS
OVTI
PGI
UNTD
UST

 

[sengsational]

Now I would assume that since everyone else has access to these same screens, including pretty much every mutual fund manager in the world, and since they have probably have a lot more time to spend analyzing screens than you do, they pretty much have tried all of this in the past and are still trying now. And we all know how well it has worked out for the active managers of mutual funds.

According to the book, which I read, the companies in the screen are small enough that any mutual fund on the planet, even small ones, would bid the price up since there are not that many shares for sale. Thus, these companies are not analyzed by mutual fund managers. So your "competition" is primarily insiders and others using the magic formula.

If you look around on the internet for stock picking histories and portfolio simulations, you can see some folks that have been at it for a long time, with returns like those predicted in the books.

The book's 2001 title was originally "...that beats the market" and in 2006 it was released as "...that still beats the market", but the latter put more emphasis on the point that beating the market doesn't mean making money!

 

[Totoro]

The Dead

I think most are not quite dead but have morphed into something bigger (how poetic).

Checked four names, could find three of them:

FTO: Frontier Oil

Merged with hollyfrontier corp after some very profitable years.

KSWS: K Swiss (Footwear)

Got acquired. Suspect shareholders bled in years before though.

MVL: Marvel Entertainment (Patents)

Marvel got acquired by Disney. Not patents but movies and comic books I don't think shareholders "lost" per se on the acquisition deal.

Now what happened to the living in the past 10 years? Did some rough checks. Numbers are not really accurate, just to give a feel.

KND : yo-yo stock. Currently -30% or so.
HRB : +20% or so after long decline
CECO: -80%.
CSGS: yo-yo stock. Current on a roll + 100%
FCX : went nowhere. +20% or so
JAKK : crashed -75%
KFY : went down, recently recovered +30%
MDLN: couldn't find
NSS : no history
OVTI : yo-yo. Currently up 100%
PGI : sort of status quo
UNTD: down 10%
UST : seems to be T-Bills, not a company

At first glance it seems you are better off being dead

[Edit] At second glance these yo-yo effects offer huge arbitrage opportunities. As long as MF gives it the nod when it's on yo-down instead of yo-up

 

[CoolChange]

....
KND : yo-yo stock. Currently -30% or so.
HRB : +20% or so after long decline
CECO: -80%.
CSGS: yo-yo stock. Current on a roll + 100%
FCX : went nowhere. +20% or so
JAKK : crashed -75%
KFY : went down, recently recovered +30%
MDLN: couldn't find
NSS : no history
OVTI : yo-yo. Currently up 100%
PGI : sort of status quo
UNTD: down 10%
UST : seems to be T-Bills, not a company
....

I think I am missing something significant here. I checked a couple of these symbols that looked familiar; and, they seem to be much too large for your criteria, even a decade ago.

For example, FCX = Freeport-McMoRan to me which I believe has had a multimillion dollar market cap for well over a decade.

Please educate me.

 

[Totoro]

These weren't mine, but companies Huston55 bought ten years ago. Just curious as to what happened with them over time.

The MF screener itself gives you small and large cap companies, so apparently Huston55 also included large caps.

I'm only going for the smaller ones (around and below 300M, down to 50M) though. Basically use the standard screener (50 companies) and select the bottom 20 in market cap.

 

[Huston55]

These weren't mine, but companies Huston55 bought ten years ago. Just curious as to what happened with them over time.

The MF screener itself gives you small and large cap companies, so apparently Huston55 also included large caps.

I'm only going for the smaller ones (around and below 300M, down to 50M) though. Basically use the standard screener (50 companies) and select the bottom 20 in market cap.

It's been so long ago since I set up the screen that I don't remember the criteria but, it must have included a large cap choice. Just for clarity, I established the screen and shadowed these positions; I didn't buy them...thankfully.