Are there any Graham stocks on the ASX?

[systematic]

IMO, the Defensive criteria is supposed to be defensive. For me this means 10+ years without a loss or a missed dividend. You could probably remove the current ratio criteria as it is specific to the business/industry

Hey odds-on, i'm cool to use 10 years for the Graham defensive. started with 5 years to see if we got any at all.
Do you think you'd prefer to remove the current ratio for now, to apply to industry separately?

What I'm interested in here is: I used the Graham Defensive criteria straight from the serenity site. Are you saying it'd be better to modify the criteria somewhat rather than as is? I'm just seeking your opinion on Graham's restrictive criteria is all. What are you looking for (compared to Graham's suggestion I mean). Hope that makes sense. Just elaborate a bit, if you will, on why tinker with Graham's rules? (I totally agree mind you, but I like to get others views too - best way to learn).

The safety criteria would have to be done on a case by case basis.

...how about Z score for the safety? Or?

 

[systematic]

Check out this link where they refine some of the Graham criteria

http://www.oldschoolvalue.com/blog/investing-strategy/graham-guru-stock-value-screen/

They are plenty of other screener backtests on this site - check out the CROIC screener $$$$$$$$

This site rung a bell; after I checked out your link realised it had been a while. Total respect to what they are doing (I love the stuff they are exploring), but I remember having a reservation or two about a couple things. Let me see if I can find something...
...Yeah, take the Negative EV - I thought that sounded really cool, but asked a quant I know (in US) who'd tested it a while back as to what the go was. Exceptional results, until you remove the lowest market caps. It's the old micro-cap dilemma in back testing. My rule of thumb (personally): ignore any tests that require the micro of micro caps to be profitable. Take out the bottom stocks (size wise) and the negative EV effect goes away. Sad, I agree! But true.

Anyway, never mind that! Good site with lots of ideas!

 

[systematic]

Ok I was being a little silly here but it would be interesting to see the performance of stock selection if you went completely against what made logical sense. Search for expensive and highly leveraged.

Totally agree with checking stuff out, it's all good. But no, I wouldn't go anywhere near expensive and highly leveraged unless looking for shorts. It just goes against the way the market is. Just my opinion!

 

[RandR]

Hey odds-on, i'm cool to use 10 years for the Graham defensive. started with 5 years to see if we got any at all.
Do you think you'd prefer to remove the current ratio for now, to apply to industry separately?

What I'm interested in here is: I used the Graham Defensive criteria straight from the serenity site. Are you saying it'd be better to modify the criteria somewhat rather than as is? I'm just seeking your opinion on Graham's restrictive criteria is all. What are you looking for (compared to Graham's suggestion I mean). Hope that makes sense. Just elaborate a bit, if you will, on why tinker with Graham's rules? (I totally agree mind you, but I like to get others views too - best way to learn).




...how about Z score for the safety? Or?

I honestly dont understand why a value investor following the principles of ben graham would scan for dividend yields... let alone use it as a deciding factor to eliminate investment options. seems crazy to me.

 

[systematic]

I honestly dont understand why a value investor following the principles of ben graham would scan for dividend yields... let alone use it as a deciding factor to eliminate investment options. seems crazy to me.

...Because that was in at least two of Graham's methods?

 

[craft]

Worth discussion on another thread.

If this thread is about Graham’s systematic scans then that is one thing but if it is about what Graham would likely look for today on the ASX – then the postscript is very telling.

I’m guessing now it’s just about the scans – so I’ll leave you to it, sorry for the interuption.

ps

Did you see the James Montier – Value Investing link in the PV thread?

 

[systematic]

If this thread is about Graham’s systematic scans then that is one thing but if it is about what Graham would likely look for today on the ASX – then the postscript is very telling.

...What do you mean? From what I've read of Graham (not much, so I could be way off!)...he was more a fan of using "scans" (investing in groups of stocks with desirable characteristics rather than individual securities) in later times?

Incidentally, James Montier like a good PE ratio and the like!

 

[RandR]

...Because that was in at least two of Graham's methods?

Great ... but the investing Ben Graham actually did was mostly arbitrage/liquidations/hedges and simply buying stocks at less then their book value on net current assets. I couldnt possibly see Ben Graham making investment decisions himself on either his own funds or those of his clients based upon dividend yields, nor any of his many students ...I know Ben compiled lists of things an investor should look for in stock selection (both 'defensive' and 'enterprising') in the intelligent investor. But do you think thats what he 'actually' used? As craft has pointed out with the postscript which is well worth a read and really shouldnt be disregarded as it provides a great insight into his actual investing thinking.

Ben also said the majority of investors should simply put their money in index funds and cash and re weight them periodically as the markets go up and down, which would probably have a better result I suspect then blindly purchasing whatever stocks meet a selection of scan criteria that Ben recommended for people who simply 'have' to pick stocks.

 

[systematic]

A modified defensive criteria :

-10 years of earnings
-10 years of dividends
-10 year average ROE of at least 15%
-Revenue and earnings growth

It would be interesting to see how many companies pass the above test. The safety criteria would have to be done on a case by case basis.

As above.
10 years of consistent earnings and dividends.
Resulted in 101 companies

Revenue & Earnings growth (I thought I might as well add in Dividend growth while I was at it)...of at least 3% per year (to put it in line with Graham's one/third over ten years).
Leaves 51 companies.

10 year ave ROE at least 15%
Man, you're putting me to work, thanks odds-on, (lol). I had to create a new category here (closest I already had was 5 year average ROE). Might as well go all in - all done now.
Left 32 final companies.


Nearly all of some sort of decent market cap/liquidity (from a cursory glance).


Added the PGP column. For those who haven't read the earlier post, what this is is simply a ratio of current price, to the Graham Price. Much like a PE or PB ratio. Lower is better (less than 1.00 means the current price is under the Graham Price).
And, the Graham Price is the square root of (Earnings per share * Book Value Per Share * 22.5)
So the PGP just lets you know where the current price is to the Graham Price. 1.00 means the current price is exactly the Graham Price. 0.50 means the current price is half the Graham Price (value). 2.00 means the current price is twice the Graham Price (expensive).


Here's the table....

​

Basically picks up more stocks than before because we're not requiring rules 2A and 2B (current assets twice current liabilities, and, long term debt less than net current assets). That makes quite the difference and loosens it up quite a bit - even when requiring a min. 15% 10 year average ROE.

Feel free to comment, offer thoughts - which are your fave (if any) stocks in the list etc?

 

[systematic]

Ben also said the majority of investors should simply put their money in index funds and cash and re weight them periodically as the markets go up and down, which would probably have a better result I suspect then blindly purchasing whatever stocks meet a selection of scan criteria that Ben recommended for people who simply 'have' to pick stocks.

Incorrect.


I appreciate your different view however, and will re-read the post-script. I intend on reading the book also - as soon as I can (which will probably be a long while away. Don't get to read as many books as I used to, sigh...it's not right I tell you...)

 

[odds-on]

Hey odds-on, i'm cool to use 10 years for the Graham defensive. started with 5 years to see if we got any at all.
Do you think you'd prefer to remove the current ratio for now, to apply to industry separately?

What I'm interested in here is: I used the Graham Defensive criteria straight from the serenity site. Are you saying it'd be better to modify the criteria somewhat rather than as is? I'm just seeking your opinion on Graham's restrictive criteria is all. What are you looking for (compared to Graham's suggestion I mean). Hope that makes sense. Just elaborate a bit, if you will, on why tinker with Graham's rules? (I totally agree mind you, but I like to get others views too - best way to learn).




...how about Z score for the safety? Or?

From my reading on the Altman Z Score, the key factors to determine whether a company will go bankrupt are cumulative profitability and earnings stability over ten years. Profitability and stability...........sounds like predictability to me!

http://www.gurufocus.com/news/36158...08-part-i-introduction-of-predictability-rank

For a defensive investment i am looking for predictability first and foremost. I would be intending to hold for a year or more so I (and the "market") will need confidence that the company is still going to be making cash over the next few years. Once I have found companies that meet the predictability criteria then I would check the capital structure is appropriate for the business. I would not use the current ratio and debt as a screener at this stage.

Now for value I would look be tempted to use historical price to NTA. Do not set a limit just look at the historical premium to NTA the company has traded at.

In summary....

Defensive Screening criteria

Predictable10 years of earnings and dividends
Predictable10 years of earnings stability
Value30% (?) discount to historical price to NTA

Once I have found companies that meet the above criteria I would check the capital structure is appropriate for the type of business. There is plenty of material on the web on what to check for industrials, banks, retailers....

Cheers

odds-on

 

[odds-on]

Revenue & Earnings growth (I thought I might as well add in Dividend growth while I was at it)...of at least 3% per year (to put it in line with Graham's one/third over ten years).
Leaves 51 companies.

Not really a surprise there. Wonder if you compared that list of 51 against an "investment grade" list produced my favourite prudent investing poster over at HC (his posts are worth reading) and see how many matched.....I bet at least 60% would.

And, the Graham Price is the square root of (Earnings per share * Book Value Per Share * 22.5)
So the PGP just lets you know where the current price is to the Graham Price. 1.00 means the current price is exactly the Graham Price. 0.50 means the current price is half the Graham Price (value). 2.00 means the current price is twice the Graham Price (expensive).

In addition to the Graham price you could use
- discount to historical price to sales or
- discount to historical price to NTA.

The type of stock that this screening criteria produce will rarely be cheap, I think it would be more appropriate to use price criteria based on historical valuation.

Feel free to comment, offer thoughts - which are your fave (if any) stocks in the list etc?

ARB, CCP and MND are rated highly in the press and on forums.

Cheers

odds-on

 

[odds-on]

investors should simply put their money in index funds and cash and re weight them periodically as the markets go up and down.

RandR,

Do you have articles/experiences/links to where investors have successfully implemented this type of strategy for a period of a few years as I would be interested to read about. I suspect it could get a return of around 9% or so with little work. Some questions:

-Would you value average or just average?
-Would you rebalance every 3 months? 6 months? 12 months?
-Do you add some sort of weighting system i.e. the larger the spread between term deposit yield and the median dividend yield the more you put into the index tracker?

Cheers

odds-on

 

[odds-on]

If this thread is about Graham’s systematic scans then that is one thing but if it is about what Graham would likely look for today on the ASX – then the postscript is very telling.

I’m guessing now it’s just about the scans – so I’ll leave you to it, sorry for the interuption.

ps

Did you see the James Montier – Value Investing link in the PV thread?

I did but forgot to say thank you. Have not had a chance to read it properly yet as am working my way through some other investment books first.

 

[systematic]

I'm not really sure there's very much interest in Graham's actual methods, but anyway...


So, just briefly, I had a look at the 51 stocks mentioned above, produced by:
earnings and dividend payments over 10 years.
earnings, revenue and dividend growth over 10 years.

As per some of odds-on suggestions, I added in the Altman Z score (safe zone only).
That narrowed those 51, consistently earning companies down to only 32.

Adding the ROE (10 year average) of minimum 15% now leaves only 20 companies.


Similar to some of odds-on's suggestions, instead of Graham price - here we will use a relative PE ratio. Let me take a second to explain that. It's basically what odds-on is talking about with relative premium to sales or NTA ratios.
To explain further:

The relative PE is simply current PE divided by 5 year low PE.

(I'm using 5 years because that's what I have already and don't have the time at the moment to create another 10 year ratio).

So, a score of 1.00 means the company's PE is at the 5 year low PE. If it's 1.45, that means the current PE is 45% greater than the low. So - the 5 year low PE is 10, and the current PE is 14.5, that gives a relative PE of 1.45 A current PE of 40 would mean a relative PE of 4.0

Small differences in numbers is irrelevant. Just remember it's a percentage off the 5 year low (PE). And, that you are looking for companies that are trading near their lows, as opposed to highs (presumably).



Here are the 20 companies that meet all the criteria as before + the Z score safe zone...sorted by the extra column which is the relative PE ratio, describe above.

​

Feedback please:
So now we have determined the inputs, what do we make of the outputs? If we've done well with the input criteria, we should have our desirable stocks.

So...are GUD, MND, OKN & BHP (followed perhaps by REH & FWD) excellent (consistent, quality & safe) companies - at a 'cheap' price? In other words, does our criteria 'work'? Why / why not?

 

[Julia]

When you are ascribing value to companies do you take into account the environment in which they operate?
Just as one example, MND, should the so called mining boom falter or even fall over.

 

[systematic]

When you are ascribing value to companies do you take into account the environment in which they operate?
Just as one example, MND, should the so called mining boom falter or even fall over.

Sorry Julia, I'm not sure I understand the question: as in, do you mean me personally? Ben Graham's method? Or was it a "thinking out loud" question (in which case, apologies for misunderstanding - and it's a good question!).

As for me personally (which isn't really relevant): No to the second part of the question (I don't at all take into account the environment a company operates in), because I don't do the first bit of the question anyway (i.e. I don't ascribe value to companies. I'm anti that approach).


But, relevant to this thread - good question. Oddson was talking about applying debt criteria and the like individually.

 

[Julia]

Sorry Julia, I'm not sure I understand the question: as in, do you mean me personally? Ben Graham's method? Or was it a "thinking out loud" question (in which case, apologies for misunderstanding - and it's a good question!).

Yes, I meant you personally.

Thanks for your response. This is, I guess, where Graham devotees are different from eg trend followers.
Personally, I'd never buy anything without considering the company's prospects in a given physical and economic environment. e.g. using MND as an instance, if mining really falls over, MND is going to be disadvantaged compared to when mining boom is at full swing.

PS Absolutely not disposed to turn this into yet another FA v TA discussion. I was just curious about this point.

 

[odds-on]

I'm not really sure there's very much interest in Graham's actual methods, but anyway...


So, just briefly, I had a look at the 51 stocks mentioned above, produced by:
earnings and dividend payments over 10 years.
earnings, revenue and dividend growth over 10 years.

As per some of odds-on suggestions, I added in the Altman Z score (safe zone only).
That narrowed those 51, consistently earning companies down to only 32.

Adding the ROE (10 year average) of minimum 15% now leaves only 20 companies.


Similar to some of odds-on's suggestions, instead of Graham price - here we will use a relative PE ratio. Let me take a second to explain that. It's basically what odds-on is talking about with relative premium to sales or NTA ratios.
To explain further:

The relative PE is simply current PE divided by 5 year low PE.

(I'm using 5 years because that's what I have already and don't have the time at the moment to create another 10 year ratio).

So, a score of 1.00 means the company's PE is at the 5 year low PE. If it's 1.45, that means the current PE is 45% greater than the low. So - the 5 year low PE is 10, and the current PE is 14.5, that gives a relative PE of 1.45 A current PE of 40 would mean a relative PE of 4.0

Small differences in numbers is irrelevant. Just remember it's a percentage off the 5 year low (PE). And, that you are looking for companies that are trading near their lows, as opposed to highs (presumably).



Here are the 20 companies that meet all the criteria as before + the Z score safe zone...sorted by the extra column which is the relative PE ratio, describe above.

View attachment 49953​

Feedback please:
So now we have determined the inputs, what do we make of the outputs? If we've done well with the input criteria, we should have our desirable stocks.

So...are GUD, MND, OKN & BHP (followed perhaps by REH & FWD) excellent (consistent, quality & safe) companies - at a 'cheap' price? In other words, does our criteria 'work'? Why / why not?

A few questions...

1. Do you intend to start performing business analysis on each business in the list? That is a lot of work that requires many years of practice and will create human judgement risk!
2. Why not just use the above 20 to create a portfolio? Most of those companies seem from memory to be ok.
3. Why not use a value weighting system to create the portfolio? The stocks at the top of the list get more cash and so on down the list.

As far as I am concerned the purpose of the Defensive graham criteria is NOT to start doing in depth business analysis. Just buy a portfolio that meet the Defensive criteria and move on with life. Obviously the cheaper are going to be out of the favour with the market e.g.mining industry.

Cheers

Oddson

 

[odds-on]

Systematic,

Thank you for all your effort, I am enjoying this thread. I remember an article in the AFR Smart Investor about a retail investor who used a quant system on the ASX and made 25% per annum. In the article the investor states he does not ever know what the business is behind the three letter symbol he just knows that it is a good business if it meets his 25 (i think) criteria. The retail only uses information from the AFR share tables. He has a keen interest in statistics and spend some time going through old share table data looking for patterns eventually creating his criteria.

IMO, the key to make this quant systems working is sticking to the criteria with discipline. Some research has already been done on Joel Greenblatt's magic formula fund, there is an option which allows the investor to pick from the magic formula list performs unsurprisingly it performs worse than just following the magic formula!!!

Cheers

Oddson