Update on the Valuation Metric Horserace: 2011-2015

Update on the Valuation Metric Horserace: 2011-2015

April 6, 2016 Research Insights, Value Investing Research
Print Friendly
(Last Updated On: April 6, 2016)

Jack and I published, “Analyzing Valuation Measures: A Performance Horse-Race Over the Past 40 Years,” in the 2012 Journal of Portfolio Management.

horse race

Here is a summary of the research paper on our own blog.

The paper asked a simple question: “Which valuation metric has historically performed the best?”

Here were the participants in this horse-race:

  1. Earnings to Market Capitalization (E/M).
  2. EBITDA to Total Enterprise Value (EBITDA/TEV).
  3. Free Cash Flow to Total Enterprise Value (FCF/TEV).
  4. Gross-Profits to Total Enterprise Value (GP/TEV).
  5. Book Value of Equity to Market Capitalization (B/M)

Below are the summary results from the paper (July 1971 through December 2010). Note that these results include a range of companies from the largest to even very small companies, which may be uninvestable:

The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained. Indexes are unmanaged, do not reflect management or trading fees, and one cannot invest directly in an index. Additional information regarding the construction of these results is available upon request.
The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained. Indexes are unmanaged, do not reflect management or trading fees, and one cannot invest directly in an index. Additional information regarding the construction of these results is available upon request.

Here are the same results, but looking at only mid and large cap companies, which are investable. We also expand the sample from January 1964 to December 2010.

value 64 to 2010
The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained. Indexes are unmanaged, do not reflect management or trading fees, and one cannot invest directly in an index. Additional information regarding the construction of these results is available upon request.

Value investing has worked, regardless of how one identifies “cheap.” Enterprise values beat the competition after running the various metrics through a variety of robustness tests (results not published in the original paper confirm this finding.) Gross profits to TEV also does quite well, but it is also highly related to enterprise multiples. The only real difference between EBITDA and gross profits is operating expenses. Results for EBIT/TEV aren’t included to stay in line with the original paper, but the EBIT figure only differs from EBITDA by subtracting a GAAP estimate of “capex” requirements in the form of depreciation and amortization expenses.

And we aren’t the only ones to identify this performance phenomenon related to enterprise multiples. Other researchers have found the same thing in domestic and international markets. We expand on this concept in our book, Quantitative Value (summarized here). In short, enterprise multiples are pretty decent when it comes to sorting stocks on “cheapness.” We have done extensive internal research on understanding “why” this is the case and we will share these findings in the coming months.

Nonetheless, now that some time has passed since our 2012 publication, we can assess the horse race results over the past 5 years. Here are the updated results from January 1 2011 through December 31, 2015. Results are gross of transaction costs and management fees. All returns are total returns and include the reinvestment of distributions (e.g., dividends). The empirical test specifics are similar to the original paper.

Here are the results on the mid/large cap universe:

val 2010 to 2015
The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained. Indexes are unmanaged, do not reflect management or trading fees, and one cannot invest directly in an index. Additional information regarding the construction of these results is available upon request.

Yuck. Value as a whole didn’t work that well based on compound annual growth rates (shown above. Sharpe ratios have similar takeaways).

Here are the metric summaries:

  • B/M REALLY stunk.
  • Enterprise multiples stunk (although variations such as EBIT/TEV were a bit better).
  • E/M (which represents P/E) were lame.
  • Gross profits / TEV results were pretty solid.
  • FCF/EV was great!
  • Note: the results for some metrics are especially bad due to the last half of 2015

Really tough to assess much of anything over a 5 year window — especially on portfolio spreads — but a good learning point. Just because something works, on average and over a long period of time, doesn’t mean it will work all the time. This rule applies on two-levels: value as a whole, and the enterprise value in particular. On net, value hasn’t shined the past 5 years. Also, enterprise multiples, the historical horse-race victor, came in dead last.

Stepping back, the results over the full data sample (1/1/1964 to 12/31/2015) still support the broader conclusion that we should “buy cheap” — regardless of the metric chosen. And enterprise values — along with their close brother the GP/TEV ratio — seem to be the most effective valuation metrics, historically.

val 1964 to 2015
The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained. Indexes are unmanaged, do not reflect management or trading fees, and one cannot invest directly in an index. Additional information regarding the construction of these results is available upon request.

But there is an interesting puzzle here. For things to work over the long-haul, they can’t work all the time, or everyone “arbitrages” the easy money opportunity. So in a sick twisted way, believers in value, and the enterprise value metric in particular, can be comforted by the recent poor performance. This poor performance is laying the ground work for sustainable expected performance over the long-haul.

That’s the theory, at least 🙂

Note: Thanks to the illustrious Yang Xu for putting these numbers together!


Note: This site provides no information on our value investing ETFs or our momentum investing ETFs. Please refer to this site.


Join thousands of other readers and subscribe to our blog.


Please remember that past performance is not an indicator of future results. Please read our full disclaimer. The views and opinions expressed herein are those of the author and do not necessarily reflect the views of Alpha Architect, its affiliates or its employees. This material has been provided to you solely for information and educational purposes and does not constitute an offer or solicitation of an offer or any advice or recommendation to purchase any securities or other financial instruments and may not be construed as such. The factual information set forth herein has been obtained or derived from sources believed by the author and Alpha Architect to be reliable but it is not necessarily all-inclusive and is not guaranteed as to its accuracy and is not to be regarded as a representation or warranty, express or implied, as to the information’s accuracy or completeness, nor should the attached information serve as the basis of any investment decision. No part of this material may be reproduced in any form, or referred to in any other publication, without express written permission from Alpha Architect.


Definitions of common statistics used in our analysis are available here (towards the bottom)




About the Author

Wesley R. Gray, Ph.D.

After serving as a Captain in the United States Marine Corps, Dr. Gray earned a PhD, and worked as a finance professor at Drexel University. Dr. Gray’s interest in bridging the research gap between academia and industry led him to found Alpha Architect, an asset management that delivers affordable active exposures for tax-sensitive investors. Dr. Gray has published four books and a number of academic articles. Wes is a regular contributor to multiple industry outlets, to include the following: Wall Street Journal, Forbes, ETF.com, and the CFA Institute. Dr. Gray earned an MBA and a PhD in finance from the University of Chicago and graduated magna cum laude with a BS from The Wharton School of the University of Pennsylvania.


  • Thomas Musselman

    Don’t know how possible or difficult (nor if you’ve done it elsewhere) but if not would a combo a la O’Shaughnessy’s “Trending Value” do better than ebitda-type quintiles under the argument he raises about different value criteria themselves trending or going in and out of favor? Just a thought.

  • Sure. There are lots of possibilities to refine value. One could also look at quality metrics (which we do). As we have highlighted in the past, value –>momentum and value –>quality, are arguably the same thing: http://blog.alphaarchitect.com/2015/03/26/the-best-way-to-combine-value-and-momentum-investing-strategies/.
    All of that is interesting and important analysis, but is beyond the scope of the “horse race,” which is meant to be based on pure “cheapness” metrics. A good project for a future research endeavor

  • Steve

    Thanks for the update, Wes – and Yang – I always love your work!

    Some thoughts:

    I am still a fan of the composite. Of course there’s many ways to go about that (the kitchen sink approach, or a more thoughtful approach toward categories as well as equal weighting vs some other kind of weight, winsorising, z-scores etc). My motivation is simply the empirical evidence that shows that compositing (or even just holding portfolios of cheap, based on individual measures) smooths the ebbs and flows. These measures of value (I believe) capture ‘angles of value’ – they are proxies…I do not believe any of them (including Enterprise Multiples) capture the value of a firm ‘accurately’ or ‘truly’. I’d love to find the ‘best measure’ (my personality is bent that way)…but I remain agnostic. Even David Dreman (who loved the PE) wrote that the other value measures will put you on to other good stocks. Just for fun, if I had to pick a single measure it would probably be OCF/EV. Maybe. But I might change my mind 🙂

    If you are not a US investor you should be aware that not every single country will follow the US (or the global) data. e.g. In some countries the B/M works way better than the US data shows. Possibly partly a function of the sector make up of various markets. That’s not a comment about international investing. It’s just to say I think even the US data gives a general idea and that we (and I do mean myself as well) can want it to tell us more than it is.

    That also goes to how much we split up the data. Following my idea that these financial ratios can only put us in the ballpark, I sometimes wonder if we should be splitting much further than a median. For example, I’ve never seen a study anywhere that showed expensive median B/M beats cheap median B/M. I’m only posing that as a thought I’ve sometimes had. I still rank stocks from first to last – have done for a long time and will continue to do so, But I have wondered. Dreman used to go as broad as the top 2 quintiles to look for value ideas.

    Please note: a lot of the above are just thoughts I’ve had…I don’t necessarily agree with what I am saying(!)

  • Steve

    Also…

    Looking at the 5 year results…I just noticed that an average of the (equal weight) value quintile still did better than the expensive quintile (10.6% vs 9.2%)

  • Adam Kearny

    Hi Wes,
    One question: if you were (loosely) curve-fitting this five-year period of “value” underperformance to past historical periods, in what spots would the time machine put us right now, and what did that historically mean for going-forward “value” performance? In other words, is this sort of like 2000, 1972, 1987, etc? Which point-in-time comparables are closest (i.e. what were the other years where we had just finished a crappy 5-yr run for all the “value” strategies)? Rich Pzena often talks about these 5-7 yr periods of “growth” vs. “value” (the “value cycle”).
    I get the feeling we have a 2000-02-style cycle directly ahead, and the macro drivers and set-up of the past few years are similar to late 1990s–strong USD, foreign economies with big structural problems and their markets (and corp profits) peaked a couple years ago (1997 EM redux); tech/biotech running crazy; S&P mulitples expanding for years while profits flat-line.
    “Value” outperformed 2000-02 not because it went up (it didn’t–it went sideways) but because it topped back in 1997 (think 2011) and had already had its bear market, while both “growth” and S&P500 which had decoupled from “value” for three years, tanked and came back down to earth.

  • Fabian

    Hello Wes,

    very interesting article (again)!
    What is not so clear to me is, why is the value weighted selection under-performing the EQW?
    When we are looking at the quintiles we have with the EQW portfolio a quite good sloop from cheap to expensive.
    How can this get lost with a value weighted portfolio, if we expect that cheap beats expensive?

    It would be interesting to read your thoughts on this.

    Thank you!
    Best Regards
    Fabian

  • For the full period analysis, EW does beat VW. This is likely due to a small-size tilt and/or a better portfolio construction concept.

    Over short periods, lots of things can happen. VW portfolios are more robust to liquidity issues, but can also create oddball portfolios. For example, if Apple is in the cheap portfolio along side a bunch of smaller companies, the portfolio is effectively Apple. Also, from 2011 to 2015 size underperformed, so any exposure to “small” is going to do worse, all else equal. So a combination of possibly skewed portfolios and poor small-cap performance contribute to VW > EW.

  • Fabian

    Thank you! Maybe, I misunderstood the VW portfolio. I thought, one would weight the companies e.g. EBITDA/TEV ranking, which should not care about Apple’s market cap.

  • Sorry for the confusion. Value-weight = market-cap weight. In academic research that is the common label, but I will be the first to admit that it is confusing.

  • Fabian

    okey, this relaxes me a bit. Did you ever compare EQW, with VW and MOSW (margin of safety weight leike I would call it for now)?Thank you!

  • Hans

    Hi Wes. Could you post the annual performance for the whole period? Thank you for a great blog. Hans.

  • here are ew for e/m, ebitda/tex, fcf/tev, b/m since it is easy to spit out of system

  • Hans

    Great. Thank you. This is based on rebalancing only once a year, right?

  • JDM

    Wes, I really enjoyed Quantitative Value and appreciated the rigor of your approach. My questions:

    1. Have you updated the Graham Simple Value performance from 2011 to present?

    2. Do you have recommendations for research on a robust method for selling a stock? I was intrigued by Graham’s simple 2 years or 50% method as a protection against some behavioral bias, but one could miss out on some major hitters and/or unwillingly realize the full extent of a cyclical downturn, (such as oil stocks bought during the summer of ’14). I know buy and hold is preferable but at some point the gains will have to be realized, and I haven’t found very much information widely applicable on selecting which winners/losers to keep and which to move on from.

    Thanks,

  • Hi JDM,

    1. We have not done that one specifically, but you can be sure it is roughly in line with the other valuation metrics.

    2. For our systems the buy/sell is built in — if something is cheap/quality, buy, otherwise sell. The concept being that one should always be focused on the holding characteristics that statistically drive expected performance. We deal with tax problems via structures and frictional costs are fairly low these days. But depending on circumstances, buy/sell rules can change dramatically