[This is a duplicate of a post made on the 'IV Value Central' board.]

Some time ago (May 2003 to be more precise) I tried to apply Ben Graham's principles to create a couple of stock portfolios, one for the “conservative” investor and one for the “aggressive or enterprising” investor. [Note that the term “investor” implies that the “aggressive investor” was still considered by Ben Graham an investor and not a speculator.]

Over a series of several posts I assessed results of stock screens to identify qualifying stocks to create two portfolios of 5 stocks each. Regrettably at that time I could not find more than that number of stocks without duplicating industries within a portfolio (or stocks across the two portfolios). This is substantially less than the 20 stocks that Ben Graham recommended for a diversified portfolio. Nevertheless fools do rush in….which is exactly what Ben was trying to prevent I guess.

The following post was the conclusion of the series:

http://boards.fool.com/Message.asp?mid=19395220

But just to save you the effort of wading through those posts, I'll summarize below:

Conservative Portfolio

Ben Graham's criteria for the conservative portfolio are:

1. An Adequate Size
For Industrials - more than $500 million in sales
For Utilities - more than $250 million in assets

2. A Strong Financial Condition
For Industrials - Current Assets more than 2 X Current Liabilities
Long Term Debt less than net working capital (i.e. Current Assets less Current Liabilities)
For Utilities - Total Debt less than 2 X the Book Value of TANGIBLE Equity

3. Good Earnings - positive in each of the last 10 years

4. Dividends - a continuous record of paying dividends in each of the last 20 years

5. Growth - an increase of at least one-third (+33%) of the most recent three year average compared to the three year average 10 years ago (i.e. the earnings for 1999-2002 are 1/3 more than the earnings for 1989-1992)

6. Good Value in terms of earnings yield - the current price is less 15 times the AVERAGE EPS for the LAST THREE YEARS (i.e. Price-earnings or PE ratio < 15). There is a potential modification to this as in one corner of his book Ben Graham inverts the PE to produce the earnings yield in percentage and compares it to the long term US Bond yields. So to allow some leeway here I relaxed this criteria so that the average earnings yield would be less than the average long term bond yield, which results in a modified criterion of 17 instead of 15.

7. Good Value in terms of book value - the current price is less than 1.5 times the current tangible book value (i.e. equity less goodwill and other intangibles). (i.e. the price-book or PB ratio < 1.5). This criterion can be modified if the PE ratio is less than 15 (i.e. the PB can exceed 1.5 if the product of the PE X PB is less than 22.5, which is 15 X 1.5).

The portfolio that was created comprised the following stocks with their purchase prices as of 1 May 2003 and number of stocks purchased.

Hawaiian Electric (HE) [Electric Utility] / 150 shares @ $40.73
Haverty Furniture (HVT) [Home Furnishing Stores] / 450 shares @ $13.53
Reliance Steel & Aluminum (RS) [Metal Fabrication] / 350 shares @ $17.56
Pulte Homes (PHM) [Residential Construction] / 100 shares @ $57.69
Kellwood Co. (KWD) [Textiles – Apparel/Clothing] / 200 shares @ $29.99

The total cost of this portfolio was $30,286.00 including a cost of $35.00 per transaction. [Discounted cash flow analysis was used to confirm that these stocks did represent value for cost.]

Since that time both HE and PHM shares have split.

The holdings, share prices as of 30 November 2004 were:

HE / 300 shares @ $28.25
HVT / 450 shares @ $20.15
RS / 350 shares @ $39.91
PHM / 200 shares @ $55.26
KWD / 200 shares @ $34.81

The market value as of 30 November was $49,525.00 with accumulated dividends of $1,213.55. This represents a total return of 67.5% over a year and a half.

In comparison, an index tracking mutual fund (TD US Index) has increased in value from $7.62 per unit to $9.78 per unit, with a dividend of $0.11 per unit. This is a total return of 29.8%.

Aggressive Portfolio

A second portfolio was created according to Ben Graham's aggressive criteria, which are:

1) Size - no limitations
2) Financial Condition - Current Assets > 1.5 times Current Liabilities
3) Earnings - no deficit in each of the last 5 years
4) Dividends - currently a dividend is paid
5) Growth - Last year's earnings greater than earnings 5 years previous
6) Value (PE) - PE < 10 (modified to 13 based on current bond yields)
7) Value (PB) - PB < 1.2

The stocks selected with shares purchased and prices were:

Skywest (SKYW) [Regional Airline] / 450 shares @ $13.61
Standard Pacific Corp (SPF) [Residential Construction] /200 shares @ $30.76
Building Materials Holdings (BMHC) [Home Improvement Store]/ 500 shares @ $12.46
Burlington Coat Factory (BCF) [Apparel Store] / 350 shares @ $17.60
Commercial Metals (CMC) [Basic Materials Wholesaler] / 375 shares @ $16.08

Total cost including$35.00 per transaction was $30,871.50.

[Note that stocks included in the conservative portfolio were excluded from this portfolio to allow the creation of an amalgamated portfolio.]

As of 30 November share prices were:

SKYW / 450 shares @ $19.03
SPF / 200 shares @ $56.01
BMHC / 500 shares @ $36.45
BCF / 350 shares @ $23.31
CMC / 375 shares @ $45.34

Total market value as of this date was $63,151.50 with accumulated dividends of $566.50. This is a total return of 106%.

Well that's it you think, this is a winning stock selection strategy. Not so fast….there's a big BUT and that “BUT” has got to do with a thing called “risk”. Simply showing that one group of investments had a greater return over another group is meaningless if risk is ignored. Would you be impressed if I told you that my corporate bonds had a higher return than my treasury bills? You shouldn't because the corporate bonds represented higher risk and for assuming that higher risk I should expect a higher return. So a comparison must consider risk, but what is “risk”? How is it defined? Having decided to go off the deep end (or am I rushing in where angels fear to tread again), I will leave that for the next post.

Cheers

SNS

SNS

Nice returns.....hope you had your own money in these ports!

Efficient market, eh?

PosFCF

...er, I was saving money to build this house thing and the wise saying was don't invest money you'll need in the short term.

[pause]

darn.

Some comments on the categories:


1. An Adequate Size
For Industrials - more than $500 million in sales
For Utilities - more than $250 million in assets


That seems a trifle small these days. Did you apply an inflation factor to scale them to current dollars?


6. Good Value in terms of earnings yield - the current price is less 15 times the AVERAGE EPS for the LAST THREE YEARS (i.e. Price-earnings or PE ratio < 15). There is a potential modification to this as in one corner of his book Ben Graham inverts the PE to produce the earnings yield in percentage and compares it to the long term US Bond yields. So to allow some leeway here I relaxed this criteria so that the average earnings yield would be less than the average long term bond yield, which results in a modified criterion of 17 instead of 15.


Again...depends on when you grabbed Graham's books. Did you account for the gradual rise of P/E over time? When he wrote these criteria, the average P/E may have been only 15-20, whereas it is 23-25 today (over the long term...not talking about the short term P/E effect where it is closer to 28).


7. Good Value in terms of book value - the current price is less than 1.5 times the current tangible book value (i.e. equity less goodwill and other intangibles). (i.e. the price-book or PB ratio < 1.5). This criterion can be modified if the PE ratio is less than 15 (i.e. the PB can exceed 1.5 if the product of the PE X PB is less than 22.5, which is 15 X 1.5).


I know this is opening up a can of worms, but I'll do it anyway. Does book value still have much relevancy today? There are numerous arguments to the contrary.


Still all-in-all, a decent list of criteria if you are into ratio analysis.

Hi PaulEngr,

Interesting questions.

The criteria are not mine, but those of Ben Graham. I take no credit or blame.

To answer your specific questions:

1) I'm in the midst of having just moved houses so I can't place my hands on BG's book to check, but I did correct for inflation. I don't know with certainly if the numbers quoted or in the BG book or the inflation adjusted ones. Although they do look like original ones.

6)The only adjustment I made is the one described for bond yields. PE's have risen but they've fallen too. The past 100 years has seen dramatic 30 year market cycles. If we are repeating this type of cycle it's possible we haven't seen the bottom of market PE's yet.

7)It's possible to adjust book value to include things like R&D to reflect a more "knowledge-based" (terrible term BTW) economy. Some form of adjusted book value must have some relevance otherwise it would imply that returns do not depend upon the need for investment. If that's the case why do companies retain any earnings, borrow funds, raise cash from the markets, etc. Consider this, studies that have demonstrated that low book-value portfolios outperform high book-value portfolios are some of the main bits of evidence against efficient markets. So I wouldn't discount book value yet - and, yes, I know the arguments. It may just be that the market has a very hard time valuing knowledge and overvalues those more dependent upon "knowledge" than "bricks and mortar".

Cheers

SNS

It may just be that the market has a very hard time valuing knowledge and overvalues those more dependent upon "knowledge" than "bricks and mortar".

I'm not pointing to propping up the MSFT's of the world. On the contrary, It's not just an MSFT problem.

Actually, the problem creeps up with tangibles too. The problem I've had with book values is that depreciation schedules are at best, laughable. And that's just on the tangible assets. An old circa 1930's coal fired power plant is very valuable these days due to restrictions on new construction (environmental permitting). And when's the last time you saw a Fournier machine in a paper mill considered "scrap", even though the value is zero on the books? On the other hand, computers are depreciated at 5-7 years last I knew. Do you actually get that much mileage out of your office PC's?

At a plant I worked at, due to restrictions on financing in Europe (Europeans do not recognize the concept of "good will"), they revalued all the assets in several plants. After resetting the depreciation schedules, a 30 year old plant I worked at consistently ran "in the red" simply because depreciation ate up over half the costs. And the plant really, truly could have desperately used some capital to straighten out some of the sins of the past. So, is the super-low P/BV deserving in this case, when you have a plant with a $2MM/month operating budget (before depreciation) bringing in $3MM/month in revenues?

When you get to intangibles (people and "intellectual property"), book values are pretty much out the window.

Hi PaulEngr,

I agree that there are significant accounting rule problems associated with book value. That's why an adjusted book value should be used, but that's not so dissimilar to the problems with reported earnings, is it.

Since book value is very much related to invested capital / equity you'd have to throw "return on investment" and "return on equity" out; leaving you only with returns. You'd be left with hunches for growth rates too, since these are dependent for the most part dependent upon the ability of a company to make good investments. And without that knowledge you'd not be able to distinguish between companies that destroy value and those that create it. Why bother worrying about capital expenditures either since this part of what leads to increased book value.

It's true that the Ben Graham criteria will not select tech companies. So what? There are other ways to select tech companies that are probably more useful then BG's criteria. For one, as I wrote earlier, I do believe that R&D and similar expenditures would be better represented as "invested capital". This would definitely increase the book value of such companies. I do treat such expenditures as investments in my own analysis. It's useful as it clearly indicates things such as when a company like Intel isn't investing as much in R&D as its R&D is "amortising". It also shows the difficult that a company like Microsoft has with any kind of investment (i.e. growth opportunities).

Ben Graham had a lot more years of investing experience than me. He experienced the great declines in both the thirties and the late sixties/early seventies. He gave a lot of thought to his criteria and he discounted expectations of growth heavily. Everyone is free to heed his advice or not. But it should be noted that there is at least one resident of Nebraska that has.

In short I can't tell you that you must use book values even if you are applying the remainder of BG's criteria. That's your decision. However, I continue to see a role for book value in stock screening.

Cheers

SNS

I agree that there are significant accounting rule problems associated with book value. That's why an adjusted book value should be used, but that's not so dissimilar to the problems with reported earnings, is it.

You and SNS might enjoy, if you haven't already, the book "Value Investing from Graham to Buffett and Beyond" http://www.amazon.com/exec/obidos/tg/detail/-/0471381985/102...

They go through numerous scenarios in which you start with book and make adjustments to come up with about 3 different estimates of the value of a company.

Book explicitly DOES NOT include Intellectual Property, which is almost certainly the most important asset of most of the modern high growth companies. But this is my bugaboo, not the books, they talk about lots of different adjustments.

R:)ph

Ralph,

Thanks for the recommendation.

You're right book value doesn't include intellectual property; and it's difficult to show the relationship between amount of $ invested to get the "bright idea" and the value of the "bright idea". If only it were simply a case of pouring more money into R&D. Of course, it's much easier to value intellectual property of proven value than that which is still only promising. "If only", the drug companies say.

It's an interesting topic, but I don't think there's a solution that'll satisfy most investors - never mind becoming accounting practice.

SNS

It's true that the Ben Graham criteria will not select tech companies. So what?

Actually, I hate tech companies. I don't even try to analyze them. They bob in the winds on nothing at all that makes any kind of business sense. And that's coming from a guy that used to work on computers with a soldering iron!

I'm more interested in a different angle with regards to book value (and good will for that matter).

What interests me is the companies where the issue of "the whole is worth more than the sum of the parts" or it's converse is true. Those that have the converse case are often ripe for takeover attempts...the old leveraged buyout: buy the company, sell off the parts that are wealth consuming instead of creating, and then sell off the rest for a profit.

The "normal cases" (sum of the whole is worth more than the parts) is interesting in that I can often identify companies that are much stronger financially from a value perspective than from a ratio analysis perspective, which often causes them to have artificially low prices (and profit making opportunities).

So I'm looking for exceptions to the book value rule on a regular basis, but I steer clear of tech stocks entirely. Unless you can call material science companies "tech stocks".

What interests me is the companies where the issue of "the whole is worth more than the sum of the parts" or it's converse is true.

That is an interesting and potentially profitable approach...even more profitable if you have the funds to actually do a buyout (heh). I guess we all can't WEB.

In any event it's way beyond the simple BG criteria, which was intended not as a market beating strategy but as one by which an average investor could sleep at night without to much worry. I was interested in the experiment just to see it the BG criteria still had any legs at all.

SNS:

Thanks for the great post. I am looking for a screen w/ an annual hold and this may be it. A couple of questions: Have you backtested it in SIPRO? And, can I replicate it there? The eye popping monthly screens are great, but I want one w/ less trading involved, even if the returns are lower. I've yet to buy SIPRO, but am investigating it.

Thanks again,

Kevin

Hi Kevin,

Glad you enjoyed the post.

I haven't backtested the criteria. Ben Graham formed his views on the basis of substantial experience; and there have been numerous studies that have shown that stocks selected using value criteria like BG's outperform the market.

However, the view over the last few years is that Ben Graham's criteria are outdated. That was the question in my mind...."are Ben Graham's criteria still valid?" So backtesting would actually muddy the waters.

You should note that BG suggested his criteria, not as a way to beat the market, but as a way to sleep comfortably at night without worrying about investments.

SNS

I don't think Graham should be taken too literally, if only because of the time shift factor. Because Zweig does a superb job of updating him IMO, you pretty quick catch on what he's after. While he never specifically embraces the "random walk on Wall Street" school, facts are most investors would be better off buying the S&P 500 than doing what they do. So, with a lot of well-based and repeated argument, I think he counsels humility.

And Grahams hadn't TIPS, spiders, and index funds at his disposal, and did not need to contend with wholesale abuse of stock options as compensation, stagflation, and corporate management being press corps. Buffet & Munger have addressed these in some of their interviews.

Irrespective of what you might think of Graham, his basic philosophy today would say that if you can't consistently beat the market averages in all circumstances, your investment algorithm is broke.

I'm surprised there isn't more discussion of his formula than his "values investing philosophy". That's his scheme whereby you take your investable assets, divide in two halves, one in stocks or equivalents, one in bonds or equivalents. If the stock side goes up to 55% or better, sell off 5% and invest on the bond side. When the stock side drops to 45% or less, sell enough of the bond side to bring the stock side back to 50%. Always invest in value companies or their equivalent.

Comments?

I'm surprised there isn't more discussion of his formula than his "values investing philosophy". That's his scheme whereby you take your investable assets, divide in two halves, one in stocks or equivalents, one in bonds or equivalents. If the stock side goes up to 55% or better, sell off 5% and invest on the bond side. When the stock side drops to 45% or less, sell enough of the bond side to bring the stock side back to 50%. Always invest in value companies or their equivalent.

I agree with the value investing part. So far, it seems that the only viable "growth" based strategy I've seen out there that is actually working is the Rule Breakers strategy that has been promoted via TMF. I don't use it only because I can't wrap my brain around the concept of investing in stocks which are as big of a crap shoot as those are, without picking away at it via valuation. However, there are great value stocks that also happen to be growth stocks, contrary to popular opinion. This is not an A vs. B decision; growth stock strategies simply choose to focus on one part of value investing (longer term future returns only).

I agree partially with the rest of what you laid out as well. Graham's strategy is literally a 2-part portfolio with rebalancing. Portfolio development and rebalancing makes a huge difference in your returns. Almost any basic investment book or guide that isn't fodder for the camp fire should talk about portfolio mixes and rebalancing.

Where I specifically disagree is in the amounts. The percentage that should be in equities vs. debt instruments (bonds) vs. perhaps other asset classes is not a fixed number for every situation. It is dependent on how soon you are actually going to need the assets in your portfolio. A 50/50 mix is appropriate for someone midway in their retirement with regards to their retirement assets. It is a rotten mix for someone just beginning retirement or anyone who isn't yet retired, or someone approaching the end of their assets (principal is zero within a couple years).

And the reason that I said individual situation is that although there is a mathematically perfect portfolio balance, you can always adjust based on risk taking nature of the individual. For instance, what percentage of people saving for retirement under the age of say 45 have 100% investment in equities? Of the remainder, what percentage have more than say 25% in various fixed income instruments? Both groups need to learn more about portfolio mixes.

It's simply amazing how conservative most people are, to the point of unnecessary paranoia, or the opposite extreme of undue risk taking.

Hi ZLegLogos,

I agree with you that there should be more discussion of the entire BG investment package, but what he said in the version of the "Intelligent Investor" was that you shouldn't have less than 25% in either stocks or bonds and that he suggested buying which ever seemed cheapest (i.e. highest yield) at the time. If you want the references I can dig them out this evening.

SNS

So my question is simply this:

What are you using to compare all of the stocks out there to the criteria outlined in the formula? I haven't found a solid or flexible enough stock evaluator yet...

Thanks!

Yates

Hi Yates,

At the time I set up the portfolios (May 03), neither did I find a stock screener in which I could directly implement the criteria. Instead I used the closed approximations I could and then delved into financial statements for the final filtering. Does that mean I found all stocks that met the criteria? No. But the ones I found did.

I haven't revisted the question of stock screeners since then for a number of personal reasons, but it would be interesting if stock screeners and databases have evolved significantly from then. I might add that the most difficult criterion to verify was the length of time a company has paid dividends....easy in some cases and darned hard in others.

SNS

Hello,
I just tried running the numbers for Graham's NCAV strategy which has the following criteria

Last Price >= 2	
Price/Book Value <= 0.8	
Price/Cash Flow Ratio >= 0.1	
Price/Sales Ratio <= 0.3	
Debt to Equity Ratio <= 0.1						

and ended up with the following stocks

Symbol	Company Name	Rank	Last Price	Price/Book Value	Price/Cash Flow Ratio	Price/Sales Ratio	Debt to Equity Ratio		NCAV	67% NAV	Qualifies
CLST	CellStar Corporation	1	4.62	0.56	79.2	0.06	0.07		8.416	5.638	y
ELXS	ELXSI Corporation	2	3	0.28	33.5	0.14	0.07		12.05	8.0735	y
GIII	G-III Apparel Group, Ltd.	3	6.64	0.68	29.8	0.23	0		9.59	6.4253	n
JLMC	JLM Couture, Inc.	4	3.38	0.77	16.2	0.26	0		4.421	2.962	n
PATK	Patrick Industries, Inc.	5	10.15	0.77	6.7	0.16	0.08		12.9361	8.66	n
KEQU	Kewaunee Scientific Corporation	6	8.66	0.78	7.2	0.25	0.01		10.68	7.1556	n
INMD	IntegraMed America, Inc.	7	7.23	0.73	4.8	0.24	0.1		9.444	6.32748	n
	

As the chart shows only two stocks(CLST and ELXS) qualified where the current price < 67% NCAV.

But the other indicators for these two stocks are negative. Is there something missing here or Iam just doing it all wrong. Please comment.

Thanks!!

Hi

I don't understand what you mean by the other indicators being negative. Please explain what you meant since the indicators for these two companies do meet the criteria.

For example, for CLST you have provided this information.

Symbol = CLST
Name = Cellstar
Rank = 1
Last Price = 4.62, which is > 2.00
P/B = 0.56, which is <= 0.80
P/CF = 789.2, which is >= 0.1
P/S = 0.06, which is <= 0.3
Debt:Equity = 0.07, which is <= 0.1
NCAV = 8.46
67% NCAV = 5.638

All of these indicators meet the criteria. It's the same for all the other stocks with the except of meeting the NCAV criteria.

SNS

Sorry, I meant the indicators other than the ones mentioned above. Like future earnings quotes from analysts are negative etc.

Hi

The following link is to an excerpt from a book called Exposing the Myths of "Can't Miss" Investment Strategies, by Aswath Damodaran. The excerpt explains why some or even many low PE / PB stocks deseve those low price - i.e. they have low growth potential or are high risk.

http://www.informit.com/articles/article.asp?p=170894&seqNum...

Clearly this appears to be the view of analysts for the two companies you mentioned. Unless you are fairly confident that the analysts are wrong these stocks would not be bargains. This is why many people combine fundamental analysis with things like Ben Graham criteria.

SNS

1 May 2003

dow 8454
nas 1473



30 Nov 2004

dow 10,428 +23%
nas 2097 +42%


Fine job applying Graham's principles to select your portfolios. Also very fine job timing the market (!)

Hi dcookie,

Thanks.

I tried my best in applying the principles. The timing was a pure accident as the motivation was the chance event of reading somewhere that Ben Graham's criteria didn't result in any good companies. I actually would have preferred a mixed market - for the experiment only!!!!

SNS

Mixed market might be the toughest to predict. To state the obvious, beta does it's best job for us when the market is rising. Outperforming a rising market is proof enough.

Cheers!