|
|
 | | From: | Darrin | | Subject: | Book to Market (BtM) vs Price Earnings (PE) | | Date: | Sun, 16 Jan 2005 10:46:38 +0800 |
|
|
 | Does anyone have any sources that provides a comparison of the two
measurements?
I remember reading somewhere that Book-to-Market (BtM) was a more
reliable ratio to look at for long term investment because it was less
volatile to short-term flucuations on the balance sheet in comparison
to Price Earnings. Does this make sense?
Also, the media tend to frequent provide charts that shows the ASX to
the median PE to reflect if the market is over or under valued. Is
there a similar chart that compares the ASX to BtM?
Thanks in advance...
|
|
| | From: | Travis Morien | | Subject: | Re: Book to Market (BtM) vs Price Earnings (PE) | | Date: | 15 Jan 2005 22:39:00 -0800 |
|
|
|
Fitzroy wrote:
> BtM and P/E are linked.
>
> In simple terms :
>
> Book Value reflects Shareholder Equity
> If a company paid no dividends, Shareholder Equity would increase
> by the amount of annual earnings.
Unless the company was unable to reinvest those dividends properly or
overpaid for assets and acquisitions, which means that retained
earnings do not contribute to book value growth.
I've been using a measure of long term book value growth vs. retained
earnings to assess the quality of a company's past record of
reinvestment.
>
> Return on equity links these two numbers, the ratio
> of (Earnings) / (Equity).
>
> Using BtM as the sole arbiter of value is therefore
> misleading. A company with high ROE would command
> lower BtM. For insrance, Woolworths generate their
> profits with very few assets, their ROE is an amazing 50%.
> They make 700 million on 1,300 million of Equity.
> They are a high turnover business.
Nobody is saying that BtM is a sole arbiter of value. What they are
saying is that price to [some accounting variable] is a useful
quantitative way to rank stocks according to cheapness or expensiveness
(I think "value" is somewhat misleading when used in this sense) and
then researchers can study whether cheap stocks outperform expensive or
vice versa.
> Proper valuation though does not stop there.
> (Accounting) Earnings themselves can be misleading,
> they need to be reconciled to cash earnings.
These academic "value" studies are not about proper valuation, they're
about systematic trends across the whole market.
David Dreman's "Contrarian Investment Strategies: The Next Generation
(http://www.travismorien.com/FAQ/shares/contrariantng.htm) makes a
strong case that investor psychology, a general trend toward people
paying a lot of money for stocks with interesting stories and too
little money for other stocks where the future isn't so promising
(inept valuation in other words) leads to a systematic tendency for
cheap stocks to outperform expensive stocks.
This is an exploitable investment phenomenon of more than just academic
interest.
>
> Growing companies tend to require more capital in
> the early stages, and therefore incur high depreciation
> charges, which reduce earnings but not operating cash
> flows.
>
> To cut a long story short, valuation is a company
> specific exercise which requires in depth analysis of :
>
> * Sales and Sales Growth
> * Assets and Asset Growth
> * Margins and Margin Growth
All true. Proper Graham and Dodd style value investing is very
different to "value indexing" type approaches. I find it unfortunate
that the word "value" gets used in such confusingly different contexts.
> Finally, regarding the numbers Travis put up from
> William Bernstein's site.
>
> There have been many Fama / French type studies all
> showing similar results.
> Using a (market value) / (some value parameter) they
> have classified stocks into 'value' and 'growth'.
> Other researchers have used Price / Sales, Price/ Earnings,
> Price / FreeCashFlow rather than FF's Price / Book.
> I believe that the highest correlation to returns
> was achieved by Price / FreeCashFlow.
>
> Regardless, the main point is that the high
> price / value stocks have built into them implied
> growth rates which have been not attained in practice.
Yep.
> That of course is not the fault of the stocks themselves
> but of the investors who have bid them up to those
> levels. Bear in mind that by virtue of the portfolio
> formation methods employed by the researchers, some
> stocks can be classified as 'growth' when bid to
> stratospheric levels, but a year earlier they may well
> been classfied into the 'value' percentiles, prior
> to the price hike.
Yep.
> It all points to one thing. Valuation is a company
> specific exercise. You dont have to buy stocks after
> they have become expensive.
>
> Is there any practical use for the Fama / French type
> cross-sectional exercises ? Yes, there is. It is a
The Australian value index calculated by Dimensional Fund Advisors from
Fama French data has beaten the ASX300 Accumulation index by about 7%pa
from January 1980 to November 2004. (I'll get the December 04 data in
the next week or so).
DFA went live with a value index fund in 1999. The results have been
in line with the historical magnitude of the "value premium".
http://www.dimensional.com.au/strategies/prices/performance_data/
If "practical" investment strategies are ones which make lots and lots
of money, value indexing looks pretty "practical" to me!
> guide to high-level asset allocation. There are some
> fundamental historical relationships between stockmarket
> multiples and fixed interest returns.
> You must not tempt fate by pouring money into stocks
> when market multiples are prohibitive.
I'm not all that worried about market multiples unless I've been
hearing a lot of talk lately about a "new era" and people insisting
once again that "this time it is different" and "traditional Graham and
Dodd techniques have no application here".
If you go to www.rba.gov.au and look up historical price earnings
multiples and chart them you'll see they were atrociously high in the
early 1990s. This was the result of low earnings in the recession no
doubt. Nevertheless, market PERs gave a very misleading impression
that the All Ords was very overvalued in the early 1990s.
Travis
www.travismorien.com
|
|
| | From: | ASX Shareholder | | Subject: | Re: Book to Market (BtM) vs Price Earnings (PE) | | Date: | Mon, 17 Jan 2005 14:21:08 GMT |
|
|
| "Travis Morien" wrote in message
news:1105857540.111880.269280@c13g2000cwb.googlegroups.com...
> >
> I've been using a measure of long term book value growth vs. retained
> earnings to assess the quality of a company's past record of
> reinvestment.
I believe that such re-investment needs to achieve positive
NPV discounted at WACC.
Buffet, as usual, cuts through all the crap and simply
requires each dollar of re-invested earnings to deliver
at least one dollar of increased market capitalization.
Same diff.
> Nobody is saying that BtM is a sole arbiter of value. What they are
> saying is that price to [some accounting variable] is a useful
> quantitative way to rank stocks according to cheapness or expensiveness
> (I think "value" is somewhat misleading when used in this sense) and
> then researchers can study whether cheap stocks outperform expensive or
> vice versa.
>
I am simply saying that BtM would not be my personal
choice for the value/growth proxy in a cross-sectional
F/F type exercise.
>
> I'm not all that worried about market multiples unless I've been
> hearing a lot of talk lately about a "new era" and people insisting
> once again that "this time it is different" and "traditional Graham and
> Dodd techniques have no application here".
>
William Bernstein calls this 'Dumb money' .
Dumb money watches the Price to Hype (TM) ratio.
|
|
| | From: | John Wright | | Subject: | Re: Book to Market (BtM) vs Price Earnings (PE) | | Date: | Sun, 16 Jan 2005 21:51:08 +1100 |
|
|
| "Darrin" wrote
> ... that Book-to-Market (BtM) ...was less
> volatile to short-term flucuations on the balance sheet in comparison
> to Price Earnings. Does this make sense?
>
It does.
The common item in both ratios is the market price - which is by its nature
volatile, being at the mercy of a whole lot of factors - market sentiments,
interest rate, factors like SARS outbreak, oil price, 9/11, war, [US
election year? ...]
With P/E, it is a ratio of two items BOTH of which are highly volatile. E -
the latest year's earning - can change drastically from year to year as
compared to a measure like say last 10 year's earnings. On the other hand,
in BtM ratio at least ONE - namely the book value - is a little more
stable - represents accumulated shareholder value over a number of years.
One year's figure will have relatively less impact on it. Hence the lower
volatility for BtM.
Regards - JW
|
|
| | From: | Travis Morien | | Subject: | Re: Book to Market (BtM) vs Price Earnings (PE) | | Date: | 17 Jan 2005 07:01:42 -0800 |
|
|
|
ASX Shareholder wrote:
> "Travis Morien" wrote in message
> news:1105857540.111880.269280@c13g2000cwb.googlegroups.com...
> > >
> > I've been using a measure of long term book value growth vs.
retained
> > earnings to assess the quality of a company's past record of
> > reinvestment.
>
>
> I believe that such re-investment needs to achieve positive
> NPV discounted at WACC.
>
> Buffet, as usual, cuts through all the crap and simply
> requires each dollar of re-invested earnings to deliver
> at least one dollar of increased market capitalization.
> Same diff.
I require each dollar of reinvested earnings to translate to an
increased dollar of book value (it is supposed to, I know, but if
you've ever charted the retained earnings vs. book value of, say,
Telstra and compared this with the same chart for Timbercorp you'll see
what I'm talking about).
And furthermore I require the return on equity over time not to be
sliding.
WACC is a tricky concept, I've never been too happy about most of the
formulae used to estimate cost of capital. There is a lot of BS,
particularly with cost of equity capital where cost is often estimated
via beta or some similar nonsense.
> I am simply saying that BtM would not be my personal
> choice for the value/growth proxy in a cross-sectional
> F/F type exercise.
Alright, then what would you use and why?
Remember, for the purposes of an F/F exercise you may only use simple
accounting variables, you do not have the liberty of being able to do a
full formal fundamental analysis and valuation for every stock.
> > I'm not all that worried about market multiples unless I've been
> > hearing a lot of talk lately about a "new era" and people insisting
> > once again that "this time it is different" and "traditional Graham
and
> > Dodd techniques have no application here".
> >
>
>
> William Bernstein calls this 'Dumb money' .
> Dumb money watches the Price to Hype (TM) ratio.
I've actually heard "the market has changed a lot since Graham and
Dodd" used a few times in aus.invest lately, while that is a worrying
sign at least it hasn't been used to justify buying a stock at 1,000
times sales, recently. It is used to promote some other form of
stupidity which I need not go into again right now.
Travis
www.travismorien.com
|
|
| | From: | Travis Morien | | Subject: | Re: Book to Market (BtM) vs Price Earnings (PE) | | Date: | 15 Jan 2005 21:26:46 -0800 |
|
|
|
Darrin wrote:
> Does anyone have any sources that provides a comparison of the two
> measurements?
>
> I remember reading somewhere that Book-to-Market (BtM) was a more
> reliable ratio to look at for long term investment because it was
less
> volatile to short-term flucuations on the balance sheet in comparison
> to Price Earnings. Does this make sense?
http://library.dimensional.com.au/articles/is_there_value_btm_ratio
>
> Also, the media tend to frequent provide charts that shows the ASX to
> the median PE to reflect if the market is over or under valued. Is
> there a similar chart that compares the ASX to BtM?
Probably but I don't have one.
There is a lot of good data in the F tables at www.rba.gov.au, you
might be able to create your own.
Travis
www.travismorien.com
|
|
| | From: | Fitzroy | | Subject: | Re: Book to Market (BtM) vs Price Earnings (PE) | | Date: | Sun, 16 Jan 2005 06:15:44 GMT |
|
|
| "Darrin" wrote in message
news:b4lju0pi2ftlm8nu9vc9bqp8h0br2069tq@4ax.com...
>
> I remember reading somewhere that Book-to-Market (BtM) was a more
> reliable ratio to look at for long term investment because it was less
> volatile to short-term flucuations on the balance sheet in comparison
> to Price Earnings. Does this make sense?
>
BtM and P/E are linked.
In simple terms :
Book Value reflects Shareholder Equity
If a company paid no dividends, Shareholder Equity would increase
by the amount of annual earnings.
Return on equity links these two numbers, the ratio
of (Earnings) / (Equity).
Using BtM as the sole arbiter of value is therefore
misleading. A company with high ROE would command
lower BtM. For insrance, Woolworths generate their
profits with very few assets, their ROE is an amazing 50%.
They make 700 million on 1,300 million of Equity.
They are a high turnover business.
Proper valuation though does not stop there.
(Accounting) Earnings themselves can be misleading,
they need to be reconciled to cash earnings.
Growing companies tend to require more capital in
the early stages, and therefore incur high depreciation
charges, which reduce earnings but not operating cash
flows.
To cut a long story short, valuation is a company
specific exercise which requires in depth analysis of :
* Sales and Sales Growth
* Assets and Asset Growth
* Margins and Margin Growth
Finally, regarding the numbers Travis put up from
William Bernstein's site.
There have been many Fama / French type studies all
showing similar results.
Using a (market value) / (some value parameter) they
have classified stocks into 'value' and 'growth'.
Other researchers have used Price / Sales, Price/ Earnings,
Price / FreeCashFlow rather than FF's Price / Book.
I believe that the highest correlation to returns
was achieved by Price / FreeCashFlow.
Regardless, the main point is that the high
price / value stocks have built into them implied
growth rates which have been not attained in practice.
That of course is not the fault of the stocks themselves
but of the investors who have bid them up to those
levels. Bear in mind that by virtue of the portfolio
formation methods employed by the researchers, some
stocks can be classified as 'growth' when bid to
stratospheric levels, but a year earlier they may well
been classfied into the 'value' percentiles, prior
to the price hike.
It all points to one thing. Valuation is a company
specific exercise. You dont have to buy stocks after
they have become expensive.
Is there any practical use for the Fama / French type
cross-sectional exercises ? Yes, there is. It is a
guide to high-level asset allocation. There are some
fundamental historical relationships between stockmarket
multiples and fixed interest returns.
You must not tempt fate by pouring money into stocks
when market multiples are prohibitive.
|
|
|
