In defense of big caps (less volatility in returns; better for buy-and-hold)

[raylopez99]

I was looking at a paper by the management firm Tweedy Browne found
here (the 56 page whitepaper): http://www.tweedy.com/content.asp?pageref=reports
("What Has Worked In Investing").

What I noticed is that while it's true that the stocks that got the
best return were the stocks that had a small market cap, and had a low
price/book value, it's also true, as can be seen by inspecting Table 6
in the above report, that the largest caps had very little volatility
in returns *regardless* of the price per book.

Looking at the chart, for the annual return (1963-1990, not the best
time for stocks, which is good for today's bear market climate):

Small caps: (going from lowest to highest price per book): 8.4% to
23%
Large caps: (" "): 11.2% to 14.2%

So, if you want the greatest return then by all means go with small
caps--but, and this is critical--you have to "time" the market and get
in when the Price / Book ratio is favorable to maximize your return.
Who has time for that?

By contrast, if you are a 'buy and hold' investor who wants to 'invest
and forget', the large caps, while giving a lesser return, are the
most steady. You can buy a large cap at any time, regardless of the
price per book, and I suspect the PE, and it will yield between 11% to
14%.

So you might say: large caps are a no-brainer.

RL

[JOSE BAILEN]

With data from 1927 to 2006, the Sharpe ratio of the typical small cap
value portfolio was 0.408008075. By contrast, the ratio for the
benchmark large cap portfolio was 0.320480566
and the Sharpe ratio for the typical large cap value portfolio was
0.340756585. This means that the risk-adjusted return of small cap
value stocks was higher than the return of ANY alternative portfolio.
In other words, while it is true that small cap value stocks are
riskier than large cap stocks, it is also true that the
overperformance of these stocks more than offset this "extra"
riskiness.

Moreover, when looking at 20-yr periods, a portfolio of small cap
value stocks beat the returns of any other portfolio 88.52% of the
time. When looking at 30-yr periods, small cap value stocks beat any
other investment style 100% of the time:

[raylopez99]

Very interesting, but your Excel spreadsheet makes my case.

First, there's a typo: some columns for High Book to Market and Low
BtM are switched around for the "6 portfolios" tab--clearly a stock
group with a high book to market will outperform a stock group with a
low BtM. But aside from that typo, the 10 year, 15 year and 20 year
columns make a strong case that large caps are sometimes better than
small cap. Note for example in the 6 porfolios worksheet: columns T
vs. W for the 10 year average; columns AA vs AD for the 15 year
average, and columns AH versus AK for the 20 year average. Sometimes
large cap beats small cap, even over a span of 20 years. I agree that
over 30 years small cap always has beat large cap, but think of the
heartaches you will have over that 30 years, as your portfolio yo-yos,
while meanwhile your large cap investor is sleeping soundly. And
sometimes the difference that small caps beat large caps is small:
for the 30 years ending 1974, small caps gained 14.95% (if bought at
the highest BtM) while large caps gained 14.29%, a difference of less
than 1%, hardly anything to brag about, even if risk adjusted.

Like those stats showing greater GDP per capita for the late 20th
century versus the late 19th century, when government was smaller, the
difference at least for the USA is not dramatic--less than 1% greater
with 'big government', with a whole lot of headaches more now
IM(politicial)O. Similarly, small caps will indeed yield better but
you have to be prepared for increased volatility, which my 'buy-and-
hold-and-forget' strategy with big caps avoids.

One final note, that's important: over time there is compression
between the returns in buying at high BtM vs low BtM (meaning, over
10, 15, 20 or 30 years it doesn't seem to matter as much whether you
buy a stock group at a high BtM or a low BtM) but this is true *much*
more for large caps than small caps--exactly what I initially stated.
For example, look at row 80 of the first worksheet "6 portfolios" and
compare column "Y" vs "AA" then column "AB" vs "AD". For the 15 years
ending in 2001, buying small caps at the worse book-to-market would
yield 3.87% while at the best BtM would yield 13.54%, a huge
difference of nearly 10%. By contrast, for the same 15 years ending
in 2001, buying large caps would give a pair of 11.04% vs 10.34%,
which is less than1% difference (note further that the large caps
don't follow the worst/best BtM rule for this time span either,
another benefit of buying a large cap in that, as I initially stated,
you don't have to 'time' the market and worry about getting the best
price, as you do with small caps; with large caps you can buy anytime
and get pretty much the same return, which is great for those
investors who want to be fully invested and/or invest in stocks at any
time, such as with every paycheck). So, if you bought small caps at
the wrong time in the 15 years ending in 2001, you would get 4%, which
is less than the 10% to 11% you would have gotten had you bought large
caps, at any time, regardless of the BtM.

Sounds like a case for buying large caps for those investors who
rather "buy and hold" and "pick and forget" their portfolio, aka the
'lazy man's portfolio' as columnist Paul Ferrell of MarketWatch likes
to call it. By contrast, with small caps you have to constantly
manage your stocks and wait for the right time to buy (and probably
the right time to sell). Unless you Jose start a microcap mutual fund
open to the public, it's too much bother for me to monitor the small
caps for the right time to buy (based on BtM), which can take one or
more years to work out, though of course I do have some mostly
financial small caps that you recommended, and they're holding up
nicely. Also I own Acorn mutual fund which has returned in the high
teens, though they also buy larger stocks not just microcaps.

RL

[JOSE BAILEN]

Actually, you don't need 30-yr periods to make sure that small cap
value portfolios always beat alternative investment styles (given
empirical evidence). If you just take 25-yr periods (I made the
exercise pretty quicly, see attached file) , in every 25-yr period
since 1927 small cap values beat the rest of investment styles. This
means that, if you are -for instance- 40, you have some capital to
invest in the stock market, and you don't plan/don't need to use it
till retirement at 65, you will always get a higher return if you
invest in a typical small cap value portfolio than in the rest of
stocks.

[raylopez99]

Thanks for this; however, the question still remains: it appears you
must time the market with small caps to beat large caps. For
instance, for the 25 years ending in 1997, if you picked small caps
when the BtM was unfavorable, your geometric yield was 9.48%, which is
lower than the 10.46% yield for large caps, with the same unfavorable
BtM.

RL

On Feb 23, 12:14 pm, "JOSE BAILEN" <jose.bai...@gmail.com> wrote:
> Actually, you don't need 30-yr periods to make sure that small cap
> value portfolios always beat alternative investment styles (given
> empirical evidence). If you just take 25-yr periods (I made the
> exercise pretty quicly, see attached file) , in every 25-yr period
> since 1927 small cap values beat the rest of investment styles. This
> means that, if you are -for instance- 40, you have some capital to
> invest in the stock market, and you don't plan/don't need to use it
> till retirement at 65, you will always get a higher return if you
> invest in a typical small cap value portfolio than in the rest of
> stocks.
>

>  Fama and French.xls
> 1831KDownload