Trending Value Metrics by O’Shaughnessy
James O’Shaughnessy is a well known “value quant” for his book What Works on Wall Street (4th Ed). He has a new column in Marketwatch discussing what he calls “the top stock-market strategy of the past 50 years.”
According to Jim, using a combination of value and momentum strategies — “Trending Value” — is the best performing strategy since 1963. To capture this, he ranks stocks based on:
• Price-to-Sales
• Price-to-Earnings
• Price-to-Book
• Price-to-Cash Flow
• EBITDA/Enterprise Value
• Shareholder yield (dividend yield + rate of share repurchases)
O’Shaughnessy ranks all of these on a 1-100 basis for his Trending Value portfolio. He works with the top 10% of those ranked stocks with the best composite score. He selects a concentrated portfolio of 25 stocks based on trailing six-month momentum, creating an extremely cheap group of stocks that are on the mend.
Its an interesting ideas, one worth exploring . . .
>
Source:
The top stock-market strategy of the past 50 years
James O’Shaughnessy and Patrick O’Shaughnessy
Market Watch, December 16, 2011
http://www.marketwatch.com/story/the-top-stock-market-strategy-of-the-past-50-years-2011-12-16
Tweet
Facebook
Reddit
Digg this!
December 19th, 2011 at 12:33 pm
The question I would have is how well has this strategy performed, particularly in the last couple of years with all the market volatility? I believe the original book was published in 1994, and so we should have a reasonable data set. Also, I think that Msrs. O’Shaughnessy have run a pair of funds using this strategy for some years now. What’s been the performance of those fund compared to the market in general and other funds?
December 19th, 2011 at 12:49 pm
I think my innernet connection must have crossed over into an alternate reality … Barry Ritholtz talking fundamentals …
December 19th, 2011 at 1:04 pm
Very interesting update from O’Shaughnessy; thanks. I read the original edition of his book in 1994.
While claiming an eye-popping return north of 20% for the strategy, O’shaughnessy also reveals that its volatility was 17.69%, only slightly below the market’s volatility of 18.26%.
What investors need to realize is that any strategy with volatility that high (for me, anything over 10% is ‘high’) is going to whack you with 30% to 50% losses, every decade or two. Returns that high are not available from fickle Ms. Market without some gut-wrenching floggings along the way.
Significantly, O’Shaughnessy does not disclose the strategy’s drawdown in 2007-9 (you’ll have to buy his book to find out). But in my experience, momentum strategies tend to get hit HARD when momentum goes south. I will guess that his Trending Value portfolio probably took close to a 50% hit during the period, if not higher.
O’Shaughnessy castigates market timers … but the fact is, highly volatile portfolios are exactly the vehicles which scare hapless punters out at the lows. Thus, his stock-picking strategy would benefit from being mixed with other asset classes to damp its high volatility, perhaps supplemented with mechanical timing techniques.
December 19th, 2011 at 1:15 pm
highest drawdown = 50.55%
http://www.whatworksonwallstreet.com/pdf/wwows_ex_28_table5.pdf
December 19th, 2011 at 1:16 pm
I don’t have all O’Shaughnessy’s metrics, but here is what a quick screen came up with:
https://docs.google.com/a/itaas.com/spreadsheet/pub?hl=en_US&hl=en_US&key=0AogBs5RKoOq9dE9QUmlOWDg4Q0ZGenN5U0dlVWVmNmc&output=html
December 19th, 2011 at 1:31 pm
@Da55id — Thanks for the reference!
Trending Value’s 50.55% drawdown was only a fractional percent less than the S&P 500′s drawdown.
It shows that while momentum strategies kick ass in bull markets, when the bear comes along you can hardly tell them apart from a plain-vanilla stock index.
December 19th, 2011 at 1:31 pm
@Da55id — Thanks for the reference!
Trending Value’s 50.55% drawdown was only a fractional percent less than the S&P 500′s drawdown.
It shows that while momentum strategies kick ass in bull markets, when the bear comes along you can hardly tell them apart from a plain-vanilla stock index.
December 19th, 2011 at 1:40 pm
So they’ve probably just figured it out before it begins to underperform. What is so magical about 50 years? Does that make it any more probable this strategy will outperform for the next 50?
December 19th, 2011 at 1:50 pm
Of course, as we are in the midst of a wee bit of a credit crisis, one might wish to take into account balance sheet elements such as current ratio, interest coverage, and total debt to equity. Think of these elements as measures to consider . . . you’re trying to buy the stock of a company which has fiscal airbags installed.
What I suspect most investors are hoping and wishing for is a Price-to-Cataclysm metric. “Golly! What would be a fair price on this stock if net sales fell to zero for a couple of quarters? Aaaaaaah!”
December 19th, 2011 at 3:14 pm
@discusCS — 1963 is the starting date of the Computstat, needed for some of O’Shaughnessy’s screens.
However, he also used CRSP data (start date: 1926) for some longer-term analyses.
December 19th, 2011 at 5:16 pm
Good grief, BR! Curve fitting again! He’s been touting this stuff for years, and if memory serves, even tried it in a mutual fund.
December 20th, 2011 at 1:52 pm
Value + Momentum does work, but work I did using data from AQR and Research Affiliates shows that, for the R1000 universe, at least, fundamental index plus momentum works even better when measured using rolling 5-year periods: fewer periods of underperformance, more periods of outperformance.