Students of Roger Montgomery's (Buffett's) intrinsic valuation method

[DeCal]

My 3 questions:

1) Can one use Commsec or other sites to filter price information, I.e. "companies with ROE >30%"

2) Does anyone know what goes into determining if a business is A1 or A2 etc?

3) What is the logic behind ROE/RR ^ 1.8 for Roger's formula.


If anyone can answer any of these questions, it'd be greatly appreciated.

DeCal

 

[Intrinsic Value]

My 3 questions:

1) Can one use Commsec or other sites to filter price information, I.e. "companies with ROE >30%"

2) Does anyone know what goes into determining if a business is A1 or A2 etc?

3) What is the logic behind ROE/RR ^ 1.8 for Roger's formula.


If anyone can answer any of these questions, it'd be greatly appreciated.

DeCal

I suggest you look at Rogers blog and/or buy his book. All the information you need is there.

 

[hrundi999]

My 3 questions:

1) Can one use Commsec or other sites to filter price information, I.e. "companies with ROE >30%"

2) Does anyone know what goes into determining if a business is A1 or A2 etc?

3) What is the logic behind ROE/RR ^ 1.8 for Roger's formula.


If anyone can answer any of these questions, it'd be greatly appreciated.

DeCal

Hi DeCal

To answer your first question, go to Commsec -> "News & Research" -> "Company Research" -> "Advanced Search Tool". Once there you have a very basic filtering tool. In the first list box select "Performance Measures", second list box "Return on Equity", third "Greater than", fourth type in '.3' (ie 30% in decimal). Then hit the 'add query' button and then the 'run query'. A list of companies with greater that 30% is displayed.

Regards.

 

[DeCal]

H999 much appreciated mate! Thanks for the time taken to write that.

Intrinsic Value - I own his book, I've devoured his blog. Firstly he has never told anyone the formula ^1.8, it was figured out in this very thread. It is important that the logic (math) to come up with the method is understood before it is used, hence my question. Roger unfortunately thinks it's not necessary to discuss - which is just him trying to keep him formula propreitary which is understandable, but annyoying in a "reveal all" book.

Secondly, I cannot find anywhere discussing what goes into how he "grades" companies (I.e. A1 A2...), it'd be great to know how he grades them.

-DeCal

 

[pedalofogus]

My 3 questions:

1) Can one use Commsec or other sites to filter price information, I.e. "companies with ROE >30%"

2) Does anyone know what goes into determining if a business is A1 or A2 etc?

3) What is the logic behind ROE/RR ^ 1.8 for Roger's formula.


If anyone can answer any of these questions, it'd be greatly appreciated.

DeCal

Hi DeCal,

Although I can't 'exactly' answer the last 2 questions that you have asked, I can say that in relation to the A1 and A2 ratings, I heard Roger on one of those TV programs one night (Switzer i think it was) and he was talking about his criteria for A1 and A2 companies. The factors he considers are Management, Competitive Advantage, Earnings Growth History, Number of Competitors, Future Validity of the Industry, etc. But the way he was talking it was giving me the impression that it is very much a subjective classification, and it is really just his way of determining the businesses that are worth watching, and that he considers to be 'Extra-Ordinary Businesses'.

Although I think his method for calculating IV seems very good, and his factors that he discusses in his book for an 'Extra-Ordinary Business' make a lot of sense, i don't use his Montgomery Quality Ratings (or whatever they are called) as a guide to my investments at all. In fact, when i dissected his list of A1's and A2's, i feel there were a lot of companies in there that were definitely not 'Extra-Ordinary Businesses'.

In relation to the use of the '^1.8' in his formula, I am not entirely sure of the validity of this. As mentioned in this thread earlier, he has stolen that from a previous author and just adjusted it to his individual needs. I have spent a bit of time trying to prove/disprove the effectiveness of it in giving a value for companies that retain all profits, but as yet i have been unable to do either.

 

[DeCal]

Hi DeCal,

Although I can't 'exactly' answer the last 2 questions that you have asked, I can say that in relation to the A1 and A2 ratings, I heard Roger on one of those TV programs one night (Switzer i think it was) and he was talking about his criteria for A1 and A2 companies. The factors he considers are Management, Competitive Advantage, Earnings Growth History, Number of Competitors, Future Validity of the Industry, etc. But the way he was talking it was giving me the impression that it is very much a subjective classification, and it is really just his way of determining the businesses that are worth watching, and that he considers to be 'Extra-Ordinary Businesses'.

Although I think his method for calculating IV seems very good, and his factors that he discusses in his book for an 'Extra-Ordinary Business' make a lot of sense, i don't use his Montgomery Quality Ratings (or whatever they are called) as a guide to my investments at all. In fact, when i dissected his list of A1's and A2's, i feel there were a lot of companies in there that were definitely not 'Extra-Ordinary Businesses'.

In relation to the use of the '^1.8' in his formula, I am not entirely sure of the validity of this. As mentioned in this thread earlier, he has stolen that from a previous author and just adjusted it to his individual needs. I have spent a bit of time trying to prove/disprove the effectiveness of it in giving a value for companies that retain all profits, but as yet i have been unable to do either.

Cheers buddy, appreciate it.

I've found the math behind ROE/RR ^ 2

It appears that Rogers 1.8 is purely nothing more then a subjective drop, probably to make the formula "his own". There is no mathematical justification for the decline.

Cheers
DeCal

 

[pedalofogus]

Cheers buddy, appreciate it.

I've found the math behind ROE/RR ^ 2

It appears that Rogers 1.8 is purely nothing more then a subjective drop, probably to make the formula "his own". There is no mathematical justification for the decline.

Cheers
DeCal

Yeah, aparently he just bumped it down to 1.8 to be a bit more conservative.

If you don't mind me asking, what is the 'math' behind using the ^2 (or ^1.8 in Roger's case). It is obviously an allowance for the additional capital being employed due to lack of dividend payment, but I'm not sure why ^2 is the figure that was decided on.

 

[rx2]

Yeah, aparently he just bumped it down to 1.8 to be a bit more conservative.

If you don't mind me asking, what is the 'math' behind using the ^2 (or ^1.8 in Roger's case). It is obviously an allowance for the additional capital being employed due to lack of dividend payment, but I'm not sure why ^2 is the figure that was decided on.

I don't believe the formula (ROE/RR)^1.8 is 100% correct. While the values it produces correlates with a majority of Roger's Table 11.2, it does not correlate 100%. For example the first two rows (5% ROE and 7.5% ROE) in Roger's Table do not correlate with the output of the formula i.e. 5%/8%^1.8 = 0.429 whereas Roger's Table is 0.356.

Rogers exact formula still remains a mystery. I was personally interested in the formula for two-reasons a) Save time typing the table into excel b) Understand the mechanics and reasoning behind the numbers.

For now - I will use Roger's magic table and look forward to when the exact formula and reasoning is revealed.

Cheers

Rob

 

[DeCal]

I will give you the credit of assuming you could figure this one out, I'll start you off...

ROE/RR ^ 2 ...

If MV/BV = ROE/RR ^ 2

therefore
= ROE/RR * ROE/RR

Try to simplify the problem perhaps - then logically thing why you might times the multiple again by ROE/RR.

DeCal

 

[hrundi999]

I don't believe the formula (ROE/RR)^1.8 is 100% correct. While the values it produces correlates with a majority of Roger's Table 11.2, it does not correlate 100%. For example the first two rows (5% ROE and 7.5% ROE) in Roger's Table do not correlate with the output of the formula i.e. 5%/8%^1.8 = 0.429 whereas Roger's Table is 0.356.

Rogers exact formula still remains a mystery. I was personally interested in the formula for two-reasons a) Save time typing the table into excel b) Understand the mechanics and reasoning behind the numbers.

For now - I will use Roger's magic table and look forward to when the exact formula and reasoning is revealed.

Cheers

Rob

Hi Rob

I just checked Table 11.2 in my copy of Value.able. In the book I see:
RR 8%, ROE 5%, Table 0.429
RR 8%, ROE 7.5%, Table 0.890
Does your book have different values to the above? I calculated a few entries for ROE 5% and 7.5%. The values in the table agreed with the (ROE/RR)^1.8 formula.

Regards.

 

[rx2]

Hi Rob

I just checked Table 11.2 in my copy of Value.able. In the book I see:
RR 8%, ROE 5%, Table 0.429
RR 8%, ROE 7.5%, Table 0.890
Does your book have different values to the above? I calculated a few entries for ROE 5% and 7.5%. The values in the table agreed with the (ROE/RR)^1.8 formula.

Regards.

Hi,

Just checked Table 11.2 in my book and I have different values. In my book I see:
RR 8% ROE 5%, Table 0.356
RR 8% ROE 7.5%, Table 0.868

When I ran the ^1.8 formula it appears the first two rows in my book are completely different to what the formula produces, and there are some variations in other places too.

I have the 2nd edition of the book. What edition do you have? I am wondering if Roger updated the tables between editions, and if so, why?

Cheers

Rob

 

[hrundi999]

Hi,

Just checked Table 11.2 in my book and I have different values. In my book I see:
RR 8% ROE 5%, Table 0.356
RR 8% ROE 7.5%, Table 0.868

When I ran the ^1.8 formula it appears the first two rows in my book are completely different to what the formula produces, and there are some variations in other places too.

I have the 2nd edition of the book. What edition do you have? I am wondering if Roger updated the tables between editions, and if so, why?

Cheers

Rob

Hi Rob

That's very interesting. I have the very first edition. I'm also wondering why the table was updated.

Regards.

 

[rx2]

Hi GaryS,

Speaking of modifying, since I posted my last IV of MCE I have changed the way I calculate EqPS (see my post a couple of posts back). Based on this, my current 2011 forecast is now at $8.89 (with 14% RR).

As you mentioned Roger has posted on his blog that he considers MCE about 46% undervalued. Considering the EPS forecasts are for a near 50% increase in profit I'm not surprised. With a current price of $5.00 a 46% increase gets you a 2011 IV forecast of $7.32.

That puts my valuation about 20% off of his. Not sure why but there are many variables so who knows. For any Roger student interested here are the current numbers I am using for 2011:

BOYE: 59.893 m
EOYE: 76.871 m
# of shares: 72.964 m
DPS (f): 0.114 c
EPS (f): 0.486 c
RR: 14%
NPAT: 34.15 m
EqPS: $1.066
POR: 24.36%
ROE: 49.62%
RR: 14%
Step 1: 3.570
Step 2: 9.888

2011 IV = $8.896

My EPS/ DPS forecast numbers are from Comsec.

If anyone got closer to Roger's IV it would be great if you could compare and contrast your variables and methods to help others... and me

I am just curious how you managed to come up with a forcasted EOYE of 76.871 m?

The figure I come up with is 85.77m based on no additional capital raised for the forecast -

(Current EQPS .82 +
Forecast EPS 0.4686 -
Forecast DPS 0.114) *
Shares Issues 72.964 = $85.77

Regards

 

[rogerd]

Hi,

Just checked Table 11.2 in my book and I have different values. In my book I see:
RR 8% ROE 5%, Table 0.356
RR 8% ROE 7.5%, Table 0.868

When I ran the ^1.8 formula it appears the first two rows in my book are completely different to what the formula produces, and there are some variations in other places too.

I have the 2nd edition of the book. What edition do you have? I am wondering if Roger updated the tables between editions, and if so, why?

Cheers

Rob

Looks like RM changed the factor to 2.2 for the first two rows of those tables. My guess is it artificially boosts the safety margin needed to acquire these type of stocks with very low ROEs by reducing their IV, as they have slow IV growth (maybe even less than term deposit rate) thus a larger purchase discount is probably required to partially offset this risk, in case your estimates are off.

5%/8%^2.2 = 0.356

From experience the formula works alright for companies with moderately high ROEs of around 20-25% (Generally can get very close numbers to RM under this condition).

But tends to substantially overestimate companies with large ROEs (eg 35%+) with low payout ratios. I think mathematically it tends to slightly overestimate the power of compounding (retained profits). If we can somehow factor in logistic growth (s-curve) instead, it would be a little more consistent. RM mentions that due to our small economy, even the best businesses with strong competitive advantages will eventually see returns diminish.

The formula I think is very useful as a screener but nothing much more in my opinion. Its easy to calculate but definitely flawed in a lot of areas, but I think will be sufficient when incorporated with a margin of safety to avoid a lot of dud stocks/companies.

Cheers
-R

 

[rogerd]

H999 much appreciated mate! Thanks for the time taken to write that.

Intrinsic Value - I own his book, I've devoured his blog. Firstly he has never told anyone the formula ^1.8, it was figured out in this very thread. It is important that the logic (math) to come up with the method is understood before it is used, hence my question. Roger unfortunately thinks it's not necessary to discuss - which is just him trying to keep him formula propreitary which is understandable, but annyoying in a "reveal all" book.

Secondly, I cannot find anywhere discussing what goes into how he "grades" companies (I.e. A1 A2...), it'd be great to know how he grades them.

-DeCal

The way he grades his companies is a big mystery.
He has stated however that his grades are based solely on their risk of a 'liquidity event' such as a need to raise capital, liquidate, raise debt, inability to pay off interest. But I personally think he is hiding a 'lot' more.

I remember that he also mentioned that the digits 1-5 are based on discount to IV on a video on the sky business channel about a year ago (cant find it anymore unfortunately) where 1 represents the largest discounts to IV and an A symbolises little to no risk of a liquidity event (which can of course change IV), thus an A1 stock is worth the most consideration. But that is definitely contradictory, following his blog you will notice a lot of A1 companies are not trading at large discounts to IV, some might even be at a premium. So I dont really know what to believe anymore.

A1-A2 stocks dont always have the lowest RRs either and its quite hard to find a lot of common ground. I do know that A1-A2 MQR companies usually have shown consistently high financial performance (ROEs of 25% or more), have very little debt or very strong cashflow and have generally sound competitive advantages (not all, for example NCK I dont see a strong competitive advantage). Which quite matches the risk of liquidity event that he mentions. I think its a combination of the above.
Hope that helps.

 

[rogerd]

Forgot to also mention that morningstar is a nice place to filter companies:
http://www.morningstar.com.au/
Just scroll to the bottom of the page.

 

[Intrinsic Value]

If you don't want to go thru all the hooplah of trying to work out the formulas and trying to assess all the unknowns then i think if you pay attention to Rogers blog and his tv appearances you will get his top tips anyway.

The best three of the year in my opinion and he hinted very strongly at these stocks at the time were MCE Matrix, sitting at 6.90 today and he strongly recommended them at three dollars something , FGE Forge he strongly recommended at around 2.50 and they are 4.90 roughly today and the recent float of MLD Maca. I got in on the first two but didn't have any spare cash left for the Maca float which floated at a dollar and is now 1.70 odd.

As he himself has stated there might be only a few opportunities a year to crop up but when they do you need to take these opportunities.

 

[Trog Dog]

Looks like RM changed the factor to 2.2 for the first two rows of those tables. My guess is it artificially boosts the safety margin needed to acquire these type of stocks with very low ROEs by reducing their IV, as they have slow IV growth (maybe even less than term deposit rate) thus a larger purchase discount is probably required to partially offset this risk, in case your estimates are off.

5%/8%^2.2 = 0.356

From experience the formula works alright for companies with moderately high ROEs of around 20-25% (Generally can get very close numbers to RM under this condition).

But tends to substantially overestimate companies with large ROEs (eg 35%+) with low payout ratios. I think mathematically it tends to slightly overestimate the power of compounding (retained profits). If we can somehow factor in logistic growth (s-curve) instead, it would be a little more consistent. RM mentions that due to our small economy, even the best businesses with strong competitive advantages will eventually see returns diminish.

The formula I think is very useful as a screener but nothing much more in my opinion. Its easy to calculate but definitely flawed in a lot of areas, but I think will be sufficient when incorporated with a margin of safety to avoid a lot of dud stocks/companies.

Cheers
-R

Ok now it makes sense - when ROE<RR the formula is (ROE/RR)^2.2 and when ROE>=RR the formula is (ROE/RR)^1.8 - given the original Simmons formula was (ROE/RR)^2 these modifications fit with RM adjusting it to be more conservative - the ^2.2 produces a lower multiplier. The ^1.8 actually produces a higher multiplier for those businesses with a higher ROE - which RM says you should pay more for when they retain all their earnings.

 

[resourceboom]

I think it is unrealstic to have a single formulae that can accurately provide an IV for many different companies. However as Roger says this is a great starting point and requires all the other aspects of the company to be looked at, and in my opinion the most important thing to determine is how sustainable is the ROE expected for this company, and determining the expected long term dividend payout as the 2 parts of the formulae are so different.

Regarding the capital re-invested portion of the equation, I believe this is the loosest part of the formulae, as it is very hard to be accurate, over the longer term.

I think an example may help show how compounding can make such a difference over time if it can be sustained.

eg 30% ROE for a company that doesn't pay out any dividends, vs one that pays out all:
Equity1 Equity2 Ratio Year
100M 100M 1:1 0
130M 130M 1:1 1
169M 160M 1.06:1 2
219M 190M 1.15:1 3
284M 210M 1.35:1 4
369M 240M 1.54:1 5
480M 270M 1.77:1 6
624M 300M 2.08:1 7
811M 330M 2.46:1 8
1054M 360M 2.92:1 9
1370M 390M 3.51:1 10

Ratio Times Book Value formulae suggests you should pay for 30% ROE at 10% RR:
7.225 3 2.4:1

Just some things to think about in my opinion.
Apologies in advance if some of my math or logic is wrong

 

[sultanmudo]

From my studies (university) and working in the finance industry:

RR = Required Rate of Return and is in relation to a per annum basis. So say the ASX has averaged 10.50% returns in the past 25 years, your Required Rate of Return might be 12% as you expect to do better than the market with your research and selections.

Margin of Safety (What i like to call 'Buffer') = As people have stated, the percentage difference between your intrinsic value calculation and the current share price.

I've expect an RR of between 10-12% is fair enough (i'm sure people would be reasonably happy making 10-12% on their funds per annum) and that the Buffer is more stock relative rather then providing a basis figure such as 20%. This can prove to be a good guideline, but I think some companies command and deserve a premium, so in some circumstances you might be willing to accept a 5-10% buffer as it may not be very often that the company trades under its IV anyway (i.e. a timely opportunity).

I'm continuing to put more and more companies through my value rating and evaluation spreadsheet. I've put approx 20 companies through at the moment, and the more I put through, the greater precision my rating system exhibits. The two companies that have stood out so far on my ratings system are ACR and NCK. I also did a valuation of Forge, can't remember the exact figures for my IV as i'm at work, but I know it was a reasonable way above the current SP.

PM me if you want to discuss my spreadsheet or thoughts further. Always looking for others input/ideas to help refine my own strategy and bounce ideas.

Would you please send me a copy of your value rating and evaluation spreadsheet. I have not yet rec'd my copy of the book. Please forward as an attachment to gdpeters@aapt.net.au