Why Blending Value And Momentum Is A WISE Investment Strategy

5th Apr '13 by Edward Croft

7 comments 15534 reads

Last year I noted that blending a strategy designed to include cheap stocks with rising stocks could be the 'ultimate market strategy'. The evidence that Value & Momentum are excellent bedfellows is utterly compelling. The two strategies are complementary as value tends to prosper when momentum lags and vice versa. Given that the returns to value and momentum are fairly uncorrelated investors in both can reap additional returns for less volatility.

If you are a hardcore value investor it's worth seriously considering if you could really stomach the inevitable 3 or 4 year periods where you massively underperform the market. As mentioned in the article linked above the career of Tony Dye (a dedicated value investing fund manager) famously didn't survive the dotcom bubble, while others threw in the towel right at the wrong time.

While momentum investing is anaethma to most serious investors, there's a growing body of research that shows it should be taken just as seriously as growth investing and clearly a dose of it might just save a few careers. So if you wanted to build a systematic investment process around value and momentum how might you go about it?

Getting WISE

Societe Generale's quant team have been running a so-called 'WISE' investment strategy across various markets for over 8 years. At it's core the strategy blends a few value and momentum ratios through a very simple scoring system. The results historically have been quite impressive - especially given the large cap bias - with the original backtest showing a 12.6% annual performance on the long/short portfolio.

The team backtested all the traditional value ratios (P/E, P/B, Yield, P/S, P/CF) including the enterprise value ratios (EV/EBIT, EV/EBITDA, EV/Sales) and added in the 'growth relative ratios' for good measure (PEG and its enterprise ratio cousin the VEG).

They studied each company's ratio on an absolute basis against the entire universe (e.g. large caps in Europe), against their local market index (e.g. FTSE 100) and against their sector. They also tested each ratio relative to each stock and sector's historical average in what looks like a very comprehensive backtest.

On the momentum side they tested the classic pure price momentum ratios (like 6 month relative strength) but also heavily tested earnings surprise ratios - like earnings revisions, upgrades and downgrades.

To figure out which ratios to include in the end portfolio, they picked the ratios that…

Unlock with Email

Unlock with Google

Already have an account?
Login here

SevenPillars

5th Apr '13

I've felt for some time and it is one strategy that I've tried to adopt, that finding value plays which also have momentum behind them, preferably early stage momentum, is a good way to trade over the length of time periods mentioned in the article above.

To do this successfully you have to be able to mix both fundamental and technical analysis. It never ceases to amaze me those who simply dismiss one or the other, what's the point in buying what your fundamental analysis might suggest is a good undervalued company if the trend on the chart is clearly down? You might as well wait for some indication that the falling trend is over. The only way this can be done is by taking notice of the longer term charts, daily, weekly and monthly. You only have to look at the up and down movements on these longer term charts to see that it is surely better to be on the right side of the trend? That means you have to wait and be more patient when a downtrend is in place, but I've found it does save you from being in something that has further to fall, which eventually means that you can buy more as the price bottoms out and reverses.

Fundamentalists should embrace charts, not see them as voodoo or be afraid of them!

Edward Croft

The MomentumRank as it currently stands blends mostly short term relative strength / breakout indicators, but we've also factored in the 3 year performance as it's been shown that 3y losers outperform 3y winners. We will probably remove this latter rule from the MomentumRank in the interests of keeping things simple - but if you screen using it you'll find lots of the kinds of shares that you mention - value shares that have been long term losers which are now in recovery. If anyone wants to play around with these ranks - fork this screen http://www.stockopedia.com/screens/compositerank-7347/

littledavesab2

When I saw this atricle I was going to pop a link on your blog but I see you got here first !

Mark Simpson

Ed,

For the momentum rank you may want to consider including the 3 auto-correlation factors:

Short Term - 1m - Negative
Medium Term - 6m-1y - Positive - I seem to remember a paper suggesting 9m formation period had the best SR over a range of holding periods when we discussed this before.
Long Term - 3-5y - Negative

So Momentum Rank = RS9m - RS1m - RS3y would would one option.

However RS3y is a much weaker effect in most studies so as you suggest I'd exclude it. And also my personal feeling is that it is probably a function of the value effect - people underestimating the competitive pressure that good/bad companies face from competitors/shareholders - rather than returns to under/over-reaction (or time-varying risk premia if you are an EMH acamdemic :-) that you are capturing via auto-correlation.

If you don't have RS9m in the dataset maybe take the average of 6m & 1y as a estimated proxy. So

Momentum Rank = (RS1y+RS6m)/2 - RS1m would make the most sense to me.

As an aside I remember you sharing before a surface graph showing the interaction of value and momentum and my recollection is that combining the two within share selection had little benefit. It was better to run 2 distinct portfolios 1 value and 1 momentum? This would then preclude using the composite rank as a strict portfolio selection tool.

I still struggle to personally implement a pure momentum strategy - buying something that has already gone up loads feels too much like I'm the greater fool! Particularly because it feels to me like the current market (last two days excepted) has been overvaluing strong growth - ARM Holdings (LON:ARM) ASOS (LON:ASC).

So my personal strategy has been to buy on value, hold on momentum (even when no longer good value) and then sell on lack of momentum - therefore tracking this new momentum rank that you've created will be very useful way to make this more formal.

Love your speed of implementation on things like this btw.

Cheers,

Danger

manxman

11th Apr '13

Thanks Ed. I've been looking at the Value momentum screen, but from what i can see jack Hough is a writer and not an investor. So I'll look into the composite rank screen above.

Isn't this what omahas finest does by mixing growth and value strategy, or is their a slight differnce between momentum and growth. These terms are usually much of a muchness.

14th Apr '13

Momentum and growth are very different things. Momentum is to do with the correlation between past prices and future prices. Growth investing is trying to identify companies whose future potential earnings growth has not been fully appreciated by the market. One is concerned with prices, the other earnings. Growth companies can have momentum and vice versa but the two are certainly not synonymous.

Momentum is present in many asset classes such as commodities not just equities whereas growth investing obviously applies only to equities.

Despite being a fairly naive strategy momentum has shown a pretty robust long term cumulative outperformance, whereas the evidence for growth strategies is less clear - probably because it is almost impossible to build a consistent growth strategy. Simply picking high P/E etc. companies that the market expects to grow quickly is the opposite of a value strategy and will earn a negative premium on average. Relying on brokers forecasts is often shown to be little better than rolling a dice, and following fund managers relies on a very long record to show that it is any better than chance.

Buffett's strategy is is more akin to a quality and value strategy. Identifying companies that have good long term business fundamentals and economic moats so should earn a high long term returns on capital and then buying those businesses at a discount to a sober assessment of their true worth.

bargainvalue

7th Dec '15

I wonder, do you have any data on the PEG and VEG ratios? How they improve returns, as a standalone indicator used for creating a portfolio? I have recently tested EV/EBIT (P/S, P/E in the past); it will be interesting to compare these ratios with the two, which are adjusted by EPS growth.

Why blending value and momentum is a WISE investment strategy

Getting WISE

Unlock the rest of this article with a 14 day trial

About the Author

Edward Croft

7 comments

Unlock the metrics for all these stocks and more...

Unlock the metrics for all these stocks and more...