During my preparations for the talks at the end of the month ( which is rapidly selling out ! ), I would go to Lee Kong Chian Library and take down some interesting observations on the market from the Bloomberg terminals.
Today's observation goes beyond the scope of my presentation in July so it's better to share this on my blog for intermediate investors. It will also give attendees and idea of what my presentation would be like at the end of the month.
An important question investors are curious to know is whether deep value strategies outperform a well-crafted dividends strategy. I have always maintained in my talks that a dividend strategy is, in many cases, inferior to the deep value strategy.
( Information on my dividends strategy will be shared on my talk itself but let's look at simple deep value strategy I back-tested on Bloomberg.)
Suppose we take the bottom 20% P/E ratios from a set of local SGX stocks. And then within this set, we find the bottom 20% P/B ratio. We will end up with a list of deep value stocks that belongs to the province of deep value investors.
Backtested 10 years and annually rebalanced, the performance is abysmal :Returns are 3.07% and the portfolio is highly volatile with a semivariance of 23.40%.
Investing in the STI ETF would have gotten better results.
As I could not believe my eyes, I then tried to observe the stocks selected by this approach. The market returned mostly Chinese companies.
So the next step would be to limit the scope of this strategy to locally domiciled companies.
The improvement is very dramatic - Once you filter out the Chinese companies, the returns become 52.46% annually ! The semivariance becomes higher at 34.69%.
This is superior to any dividends strategy I will be presenting end of this month and possibly superior to any performance of active managers covering the Singapore market !
Some preliminary suggestions for investors :
a) If you go for deep value, filter out the China stocks in SGX.
b) While your returns will improve, your risks will also be magnified.
c) The backtest does not account for the lack of liquidity of good deep value stock counters. So you might even be able to buy the stocks suggested by the screen becuase no one is selling them.
d) While it's tempting to leverage such counters, brokers are unlikely to give you decent lending rates for deep value stocks.
e) Unlike dividend stocks, deep value counters may not occasionally give you a reward you for holding counters over time.
f) No release of dopamine in your brain when you get rewarded with dividends. ( Which is what got me hooked more than a decade ago ! Dividends are the opiate for the investing masses ! )
Hi Chris,
ReplyDeleteThe problem with deep value stocks is there is no timeline for its value to be realized. It could be within 1 year, 3 years or even more than 5 years.
In the meantime, you might have little or no reward when holding these stocks. Of course, you can narrow down your search further by inputting low PE, low PB and high dividend yield. That might actually be a compromise if you are seeking deep value stocks coupled with decent yield for you to receive some income while waiting.
It need not be a direct Dividends strategy vs Deep Value Investing. Why not a Dividends plus Deep Value Investing strategy?
ghchua,
ReplyDeleteYou are absolutely right. There is academic evidence published in the FAJ this year that simply adding a yields criteria to a deep value strategy can indeed enhance returns.
Thanks for the reminder !