Ben Graham Scan

[mpauls]

Biaggio, I don't mind one bit that you posted those scans, that's why they are there after all. 

 

I read (briefly) some of the comments here on Ben Graham's interview.  What he is saying is partly true.  In B.G.'s early years, finding FCC filings that show a company holds $xx in treasuries wouldn't have been disclosed nor would these assets have showed up in the price.  In the interview, he was implying that most people would not be able to do as will as he did by just digging through the basic numbers within financial statements.  He wasn't saying that the playing field is level.  Also, as suggested by someone already, Graham was bored by investing and wasn't interested in it late in life. 

 

General comments are almost always taken out of context and I am certain that this is the case with this particular question.  If they would have continued in greater detail we would not be having this conversation.  Moreover, does anyone need proof that systematically high returns are impossible to come by or otherwise the consequence of pure luck?   

 

 

[Kraven]

Dear Kraven, I think we have reached a point where we have to agree to disagree. These are more or less the conclussions.

 

Let's state again the plain facts:

 

- Graham said that he had an experience of 30-odd years buying net-nets and that he estimated that they yielded 20% during this period.

 

- He was so convinced of the formulaic approach working better than classical Graham and Dodd that he actually made it his main strategy.

 

Words and actions match, they are consistent. He literally put his money where he put his mouth. So my interpretation is to take them at face value, and assume that they were true.

 

Correct me if I am wrong, but you think that he was tired and burn out of investing and that he was just rationalizing his laziness, because you implicitly assume that Graham and Dodd investing must necessarily produce better results that applying a mechanical formula. 

 

We would really need an audit of Graham's returns to know who is right.

 

Txitxo, I think we will have to agree to disagree.  That's fine.  It was an interesting debate and discussion.

 

Correct me if I am wrong, but you think that he was tired and burn out of investing and that he was just rationalizing his laziness, because you implicitly assume that Graham and Dodd investing must necessarily produce better results that applying a mechanical formula.

 

No, not at all.  I have tried to make this point repeatedly.  I BELIEVE Graham meant what he said when he said it.  That is, that late in life he determined that a mechanical approach was superior.  What I don't think is that OTHERS who are trying to find support for that premise (i.e. that mechanical investing is superior) have on their side significant evidence by referring to Graham.  Why do I think this?  He had spent a lifetime, approximately 40+ years to be exact (from around 1917 - late 1950s), preaching the gospel of individual security analysis.  He wrote the bible, Security Analysis, and put out 3 versions during that time.  He wrote countless articles earlier in his career on it. 

 

He then retires and more or less stops participating in the investing world.  Then years later he all of a sudden re-emerges.  He updates The Intelligent Investor, etc.  His assertions that mechanical investing are superior are primarily (solely?) found in interviews included in the Institutional Investor and the oddly named Journal of Medical Economics.  Perhaps a Financial Analysts Journal as well.  So the basis by which you have determined that he essentially found Jesus late in life is due to around 15-20 pages of interviews in 2-3 magazines vs the hundreds or thousands of pages from earlier.  While volume means nothing, I point it out simply to demonstrate that it isn't so cut and dried.

 

I leave it at that.  I do believe Graham's mechanical strategies are workable and can be successful, particularly with respect to net nets.  I don't believe though that they are wonderful support for a purely mechanical approach.

[west]

It was the turnkey analyst blog that tried to recreate the magic formula results and couldn't: http://turnkeyanalyst.com/2011/06/909/

 

If I recall the guy is a finance professor and researcher.  It's slightly troubling that he couldn't replicate the results.  If anyone's interested in the magic formula there's a yahoo newsgroup with members who have been trying to replicate Greenblatt's numbers since 2007 if not earlier.  I say possibly earlier because I joined in 2007 and they've been discussing it ever since, I never searched the archives.

 

The magic formula yahoo group is worth a read if anyone is interested in quant strategies.  For how simple the magic formula is in real life most of these people got absolutely slaughtered in 2008, and they haven't come back yet.  The ones who've done well have integrated technical analysis, or simple scoring systems.  The ones who followed exactly what Greenblatt mentioned have done the worst.

 

Forewarning: I'm jumping into this thread without having read the full thing...

 

I have been forward testing the data from the Magic Formula website for about two years now.  I wanted to see that it actually worked in real life before throwing any money at it.

 

I'd say, for large caps, the screen does pretty well vs. the market for *almost* all periods I've tested it for.  The thing that's scary to me is watching how the construction of the lowest market cap limit dataset has changed over the years.  For example, in October 2010 the dataset had lots of Chinese fraud companies in it.  The results from that dataset were *terrible*.  In October 2011 that same dataset was full of pristine businesses and did pretty well over the next year.  Did the MF actually adjust so the bad businesses "magically" disappeared?  Or does the dataset get edited?

 

At this point, I would feel very safe investing in the $5b floor, 50 company dataset screen (which, oddly, has been one of the most consistent and better performers of the datasets I've collected).  It does about 2%-6% better than the market on average.  I'm not 100% sure if this would be worth it outside of a tax free account though.

 

On the other hand, I would not touch the $50m/30 company dataset unless I went over the companies myself.  Too much potential for bad companies to show up again.

[twacowfca]

I don't think most posters have studied Graham's life thoroughly.  His pre 1929 period used a lot of leverage which was gradually pared down as the depression ground everybody down.  His arbitrage, long/short and special situation plays worked OK during that time, but the others didn't. By 1936 his funds were back to even, and he made up the losses of those few of his investors who weren't able to stay with him through the cycle.  He quotes very specific results for his unleveraged funds from 1936 through 1946 of 17%+ net to his investors or 20%+ gross returns to Graham and Newman.  However, if one reads carefully what Graham states in his memoirs, it is evident that his gross returns were significantly higher than that because those returns were after large salaries had been paid to the principals.  I guesstimate that Graham's gross returns were 22% or higher during that time when the DJIA returned @ 10%.

 

Graham testified before the US Senate in 1955 that the returns of his fund had been about 20% since his business had been reorganized in 1936.  Therefore, it seems that his average return to his investors had by then increased from about 17% from 1936 to 1946 to 20% for the entire 19year period, apparently net to his investors.  His gross returns may have averaged something above 24% including salaries for the entire 19 year period and above 27% for the 1946 to 1955 period.  This is all taken from his own writings and congressional testimony.  He testified to Congress that investors occasionally sold their interests in his fund at a substantial premium to other investors.  :)

 

He testified before congress that almost all his operations consisted of reorganizations, recapitalizations and special situations that weren't readily available to the public or to other funds.  Apparently the vaunted net nets were a small, but profitable part of this, during a time when good companies sometimes sold for net net prices.

 

That was the heart of his career, careful analysis of special situations for the most part.

[Valuebo]

I don't think it was simply advancing age. How old is WEB? Rather Graham simply realized something that's quite true, with a more competitive market with investors who have more and more information available to them, markets do become more efficient, especially compared to Graham's heyday. I wonder what he'd think about current markets.

 

Markets became more efficient but mispricing will always exist due to our emotions, biases and specific misunderstandings.

 

Fun fact: As late as 1975, WEB's Berkshire was trading for a record 60% discount to book and 3 years later you could still buy it for a discount to book. Even after BRK good in the large cap area, we saw two or three very obvious mispricings for the stock in the last 15 years. The same goes for a lot of other companies with great management. I think we are far away from a truly efficient market, but I'm sure you agree! :)