This week’s “chart to make you say hmmm” …
[logarithmically-scaled]
Here I’m trying to understand what has driven market gains during different periods from mid-1999 to the present, growth or value stocks.
I’ve assumed that two traders began with $10,000. The first trader (in Ferrari red) was only long those days when the S&P 500/Barra Growth Index outperformed the Barra Value Index (the same day). The second (in stodgy banker grey) was only long when value outperformed growth.
This would of course require each trader to peek into the future and know where each index would close at the end of the day, but the point of this exercise isn’t to actually trade the relationship (yet), just to better understand it.
Unfamiliar with the S&P 500/Barra indices? They are created by splitting the S&P 500 into two equal halves based on each stock’s price-to-book ratio; an expensive half (growth) and a cheap half (value).
What the data is telling us…
From mid-1999 until mid-2001, the market performed significantly better on days when growth outperformed value (i.e. red line going up, grey line going down). Things became a bit muddled after that with the market not particularly driven by either.
But look at what happened just prior to our current bear market…everything flipped. The market now performs significantly better on days when value outperforms growth.
The results suggest that this growth vs value vs broad market relationship has a significant and consistent impact on the market over long periods. But the fact that it has evolved so radically suggest that how different markets are driven by that relationship is very different.
Two outstanding questions:
(1) Why the difference between the two bear markets? (my non-fundamentals-oriented brain isn’t equipped to answer this one)
(2) How do we capitalize on this relationship? Not sure yet – I haven’t figured out a consistent way to predict whether growth or value will outperform the next day (and without that ability, this observation is only an interesting anecdote). More to follow.
Happy Trading,
ms
To stay up to date with what’s happening at the MarketSci Blog, we recommend subscribing to our RSS Feed or Email Feed.
Filed under: Evolving Markets, Stock Market Mechanics | 8 Comments
-
Subscribe to RSS Feed
Subscribe via Email
My email: michael @ marketsci.com (remove spaces)
Nice analysis, but you should compare against the S&P equal-weighted, otherwise you are comparing Growth and Value against a S&P bloated with growth stocks until 2001, and then bloated with value/banking stocks in 2006-2007.
RE to Andre: good point…if (for example) the cap-weighted S&P 500 was overwhelming weighted towards growth until 2001, it doesn’t mean the market liked growth at that moment in history, it means the market WAS GROWTH at that moment in history. Will run the numbers. michael
RE to Andre: [update] ran the numbers on the equal-weight index…same conclusion. But thanks for the comment – it was a very good thought. michael
“I haven’t figured out a consistent way to predict whether growth or value will outperform the next day (and without that ability, this observation is only an interesting anecdote)”
What happens if you trade ‘a day late’?
But at tonight’s close when you would have bought at yesterday’s close – had you known the future.
Is that worth a quick graph?
RE to Mark: good question, but no edge whatsoever. michael
One difference in fundamentals: while both stock market crashes were large, the recessions in the broader economy were quite different. The recession of 2000 was quite small, while the recession of 2008 has been enormous.
An interesting feature of this study — assuming that Andre’s concerns make no significant difference in the result — is that the market actually went up(!) during overall bear markets when value outperformed growth. So the days that value outperformed growth were overall up-days in a bear market. It’s not just that on those days the market went down less (which would make sense since value is presumably lower beta), but it actually went up. (I know I keep repeating myself, but it seems quite amazing.)
One possible interpretation is that days on which value outperformed growth were bottom-fishing days. On those days people are looking for solid stocks with low PE ratios, i.e., value.
What about days when growth outperformed value in a bear market? Not clear. They must overall be down days since that’s what the data shows. But if they were significant down days, then one would expect value to outperform growth a a result of its smaller beta. Perhaps they were mixed down days, days in which economic fear predominated, leading people to sell everything — and especially companies (mainly value companies) that are tied to the fundamental core economy.
All that is just speculation, but it was fun to make up a story.
You might find this research from cxoag very interesting: http://www.cxoadvisory.com/blog/external/blog10-06-08/
“investors may be able to exploit time variation of the value premium based on the state of the economy, moving out of (into) value as recessions approach (end).”