Even God Would Get Fired as an Active Investor

Even God Would Get Fired as an Active Investor

February 2, 2016 Architect Academic Insights, Key Research, Tactical Asset Allocation Research, White Papers
Print Friendly

Empirical asset pricing research can sometimes get monotonous because you end up circling back relentlessly to the same conclusions: value works, momentum works, and yet, markets are remarkably efficient. But, sometimes, research uncovers absolutely stunning and counter-intuitive results–and this is where things get truly exciting. The study below is what we consider “exciting” research because the results are so profound (at least to us).

Our bottom line result is that perfect foresight has great returns, but gut-wrenching drawdowns. In other words, an active manager who was clairvoyant, and knew ahead of time exactly which stocks were going to be long-term winners and long-term losers, would likely get fired many times over if they were managing other people’s money.

Question: if God is omnipotent, could he create a hedge fund that was so good that he could never get fired? No. It turns out even God would most likely get fired as an active investor.

Strategy Background

We compute the 5-year “look ahead” return for all common stocks for the 500 largest NYSE/NASDAQ/AMEX firms. For simplicity, we eliminate any firms that do not have returns for a full 60 months. We look at gross returns and all returns are total returns including dividends. Next we create decile portfolios based on the forward five-year compound annual growth rate (CAGR).

We rebalance the portfolio on July 1st every fifth year. The first portfolio formation is July 1, 1926 and is held until June 30, 1931. The second portfolio is formed on July 1, 1931 and held until June 30, 1936. This pattern repeats every fifth year. To be clear, we know with 100% certainty the performance of the top 500 stocks over the next 5 years.

We are explicitly engaging in look-ahead bias.

Returns are analyzed from 1/1/1927 to 12/31/2009. Portfolios are value-weighted returns for month t are weighted using the market capitalization at the end of month t-1. All returns are gross of transaction costs, taxes, and fees.

Decile Portfolios

We first look at the decile portfolios rebalanced every 5 years. These portfolios highlight what perfect foresight can achieve. The Decile 10 portfolios represent value-weighted portfolios sorted on future top 5-year performers and the Decile 1 represent value-weighted portfolios sorted on future bottom 5-year performers. The compound annual growth rates for the 10 decile look-ahead portfolios are mapped below:

look ahead bias portfolios deciles

As expected, a portfolio formed on the names that have the best 5 -year performance, have the best 5-year performance. Duh. But the details are interesting…

Summary Statistics

Here we investigate some statistics and charts on the performance of the 5-year look ahead portfolio.

First, the raw summary statistics (labeling Decile 10 above as High MOM VW):

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.

The 29% CAGR is obviously awesome for the look-ahead portfolio. Expected.

But how about the drawdown associated with a perfect foresight portfolio? Can’t be that bad if you know the future, right?

Wrong! The worst drawdown for the look-ahead portfolio is devastating: -76%! (Aug 1929 to May 1932). But the pain doesn’t end there…here is a table of the top 10 drawdowns:

long dd
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.

Clearly, even a “perfect” long portfolio can cause a long-only investor pain.

How About We Create the Ultimate Hedge Fund?

In the analysis above we highlight that the perfect long portfolio still can create massive pain. But perhaps we can create the ultimate hedge fund portfolio: 1) long the names we know will perform the best over the next 5 years and 2) short the portfolio of names that we know will perform the worst over the next 5 years. Surely, this would be a god-like hedge fund?

The long/short portfolio is constructed as follows:

  • Fully funded long book and a short book that earns the risk-free rate on short proceeds.
  • The long/short weights are rebalanced monthly.

The following portfolios are examined:

  • 5 Year High MOM VW L/S = Long 5-Year winners; short 5-year losers
  • SP500 = S&P 500 Total Return

Summary Statistics

Here are the high-level stats:

long short god
The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained. In fact, these returns are EXPLICITLY IMPOSSIBLE TO ACHIEVE. 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. Note: these results were updated on 2/17/2016. The CAGR is higher on the “God hedge fund” than originally posted.

Yowza! Clearly, the ultimate hedge fund does amazingly well — 49% CAGRs would have you owning the world’s stock market in short order. Obviously, this sort of return is not possible over a long period — even if someone had perfect “Biff-like” foresight.

Yet check out the worst drawdown on the PERFECT hedge fund — 60%+. Incredible. And it gets better…

Here are the top 10 drawdowns for the ultimate hedge fund and the associated returns on the stock market:

god excel drawdowns
The results are hypothetical results and are NOT an indicator of future results and do NOT represent returns that any investor actually attained. In fact, these returns are EXPLICITLY IMPOSSIBLE TO ACHIEVE. 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. Note: these results were updated on 2/17/2016. The CAGR is higher on the “God hedge fund” than originally posted.

Let those numbers soak in a bit.

What the chart highlights is that even GOD HIMSELF would get fired multiple times over. The performance on the perfect hedge fund would get crushed many times over by the passive index.

These results highlight the fickle nature of assessing relative performance over short horizons. We’ve shown this quantitatively, but Ben Carlson talks about the challenge of short horizon thinking here, and Meb Faber recently highlighted that investors are terrible at timing active investments.

Takeaways:

  1. Keynes was right: Markets can remain irrational longer than you can remain solvent
  2. Active investors MUST have a long-horizon…and few investors actually have horizon.

Good luck out there…

h.t., Arturo B. . An old Chicago PhD (1980) we met at the Nantucket Project, who suggested we explore this research question…

***

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.

***

If you liked this post, don't forget to subscribe to our blog.




About the Author

Wesley R. Gray, Ph.D.

After serving as a Captain in the United States Marine Corps, Dr. Gray received a PhD, and was a finance professor at Drexel University. Dr. Gray’s interest in entrepreneurship and behavioral finance led him to found Alpha Architect. Dr. Gray has published three books: EMBEDDED: A Marine Corps Adviser Inside the Iraqi Army, QUANTITATIVE VALUE: A Practitioner’s Guide to Automating Intelligent Investment and Eliminating Behavioral Errors, and DIY FINANCIAL ADVISOR: A Simple Solution to Build and Protect Your Wealth. His numerous published works has been highlighted on CBNC, CNN, NPR, Motley Fool, WSJ Market Watch, CFA Institute, Institutional Investor, and CBS News. 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.


  • ptuomov

    If God would use a sensible risk model, an optimizer, and some common sense portfolio constraints, God would not get fired in this experiment.

    Keeping the portfolio small/mid/large cap balanced and industry balanced (relative to whatever benchmark chosen) would probably bring in the drawdowns a lot.

    If the point is that poor risk control can squander away even great insights, I agree. The same goes for trading costs, by the way.

  • That is a hypothesis that can be tested. Please explain the exact rules you believe would prevent God from getting fired and we can empirically test this proposition.

    In general, owning the stocks that will have the best 5 year returns and shorting the stocks that will have the worst 5 year returns is about the best strategy I can ever imagine.

  • ptuomov

    In a long-short context, here’s one thing that you could try to see whether risk control is important or not to the outcome of this experiment.

    After you have sorted stocks into ten perfect foresight portfolios and assigned each of them a decile assignment (as you did above), form ten portfolios from your overall top decile stocks by creating sector portfolios (say the ten GICS 2 sectors) out of the top ranked stocks. Then form your final long portfolio as the equal weighted portfolio of these ten sector-specific top deciles (averaging ten portfolios, one per sector). Similarly, for the short side.

    Another way to say this is that I’d try reweighting the stocks in your top decile not to be equal weighted but such that each sector has the same amount of weight (and equal weighted within sector.)

    My guess is that your average return will go down only a little but your standard deviation and drawdown will both go down a lot. But it’s just a guess.
    This is not the only way to use the perfect foresight information to get less volatile results, but it’s the simplest that I can come up with.

  • MrLovingKindness

    An omniscient being would short and long the best stocks every day (maybe every hour, minute or second, depending on the trading drag), and thus, avoid the draw downs completely (assuming the being knows the highest and lowest price points each day).

  • Now that would be great, but he’s need to convince the broker to give him free trading costs — which I’m sure he could do with a little negotiating. ha!

  • Steve

    I’m with you – this was an exciting one…thanks for sharing! I would not have predicted that result. Now matter how good your system, you’ve got to be doing this long-term.

    Also interesting to see that the strategies we talk about sit around the 7-9 decile mark.

  • Why would God disregard utility?

    “When early mathematicians first formulated principles of behavior in chance situations, they assumed the proper objective of the individual was to maximize expected money return. It was later found that this objective can be incredibly bad.

    The expected utility rule was proposed as a substitute for the expected return rule.”

    Pg. 207 here: http://cowles.yale.edu/sites/default/files/files/pub/mon/m16-all.pdf

  • Ashley

    My other takeaway from this is — If you’re reaching for the biggest returns, you’re going to be taking the biggest risks. I wonder how “God” would do if he factored in some measure of relative volatility while building the portfolio?

  • JS

    The fourth worst long-only portfolio (03/31/2000 to 03/31/2001) had a portfolio return of -34.03%. This is the exact same portfolio return for the 8th worst long/short portfolio (which happens to be the exact same time period). While it is certainly theoretically possible, it seems highly unlikely to the second decimal. Is it possible there is an error? It is also possible I missed something in the methodology.

  • Hey JS,

    Nice catch. Updated the L/S table. We have to handjam these things out of our systems to get them in to the blog.

    BTW, if you have the capabilities to replicate the study, let us know what you find. Happy to share details on methodology, etc. We know of no other investigations into this topic so we are “pioneering” the research effort.

    Would be great to see what others find.

  • Patrick Luby

    Thank you…fascinating article. Have you done a similar analysis on asset allocation?

  • Hi Patrick,

    We haven’t done a study like this in that context, but in general, strategies that work, tend to have major pain points at some point…that is a recurring theme in our research.

    When we get some R&D bandwidth we’ll explore some of the questions folks have posted on the blog…

  • JS

    Not exact, but Figure 1 in this firm’s blog does a good job of contextualizing the extreme nature of the “failure” of a diversified portfolio in 2015 (which is, I expect, the most imminent reason for asking). There are several exhibits that could be helpful.

    http://gestaltu.com/2016/02/navigating-active-asset-allocation-when-diversification-fails.html/

    I am not affiliated with the firm or article, just thought it might help.

  • love that piece. Thanks for sharing.!

  • Prateek Sharma

    The value-weighted portfolio holds a very special significance, if the two fund separation theorem holds, that is all risky investments are based on a unique risky portfolio, than value-weighted portfolio is the only portfolio that will clear the market. Under some (highly questionable) assumptions, the value-weighted portfolio is the optimum portfolio, one with maximum expected Sharpe ratio. However, as soon as you make a person clairvoyant, which means that he/she can predict, with certainty, which stocks will get what return in the future, the notion of risk is eliminated. In this certain world, value-weighting would not make any sense, and neither would any expectations about the Sharpe ratio. In fact, the optimum strategy would be to invest the entire capital in a single asset that “you know” would yield the highest returns for your investment horizon (5 years, 10 years, 20 years , whatever it may be). If shorting is allowed, the optimum strategy would be to short the single asset that is going to perform the worst, and go long the single asset that is going to perform the best.

  • Kevin Osborne

    Investors are terrible at timing active investments because they don’t have enough information or don’t know how to value that information. Yet small investors have a huge advantage over the big funds because of the nimble trading available to them as opposed to moving a large position that alone changes the value of a stock. If an investor learns how the big boys choose their stocks it is possible to piggy back on their moves, in and out, and make very good returns, certainly better than the average market, even if generally long. (although shorting when the clouds appear will help considerably)

  • Yo

    that’s called high frequency trading

  • Kevin Osborne

    Thank you.

  • Mark Dodson

    Wes – this is great stuff. Consulting services departments of wirehouses often use trailing 3yr (and 5yr) comparisons with the S&P 500 as one of their primary method for determining what managers get hired and fired from their platforms? Knowing your lookahead period was 5yrs, leaving this option out, I’m still curious how often this perfect 5yr foresight portfolio found itself lagging the S&P 500 after a random three year period.Would it be possible to see calendar year returns for these strategies as well?

  • had to post in 2 sections because it is too long.
    Also posting 1yr rolling cagr. most of the time it wins, but even God can lose.

  • the rolling cagr chart is kinda fascinating. In normal markets the thing kinda grinds, but when the S&P blows up you get massive spikes…so one hypothesis is you need chaos for active mgmt to really shine and as buffett says, “figure out who is swimming naked.” Who knows. Main point is market volatility and market psychology is incredibly complex.

  • Mark Dodson

    Wes – thanks for taking time to share all this data and the chart. I appreciate it!

  • Novice

    Regardless of the level of transaction cost, it does not make sense to make “value-weighted portfolios sorted on future top 5-year performers” as the best performing stock of the future would outperform this portfolio in every period (5-year periods with this rebalancing strategy).

  • Hannibal Smith

    Hate to be a party pooper, but there’s nothing surprising at all about these results. Being right is not the same thing as making money as Ned Davis rightly wrote an entire book about.

    The proper way to have constructed the deciles if you had perfect hindsight was on a risk/reward basis. Then “God” wouldn’t have been fired since he would have beat his tracking index. A better question is why the hell would “God” be a “professional” manager? No, he’d be an independent investor.

  • MAR

    I think what is interesting from this data is that someone like Buffett has close to a 20% CAGR return over a long period of time, which based on your data seems impossible. Assuming here that 30% top decile returns come down a few % due to real transaction costs etc. For someone like Julien Robertson who had a reported 32% CARG return over 20 years your data would imply statistically it was almost impossible to achieve. They’re not God but they’re pretty close.

  • Hannibal Smith

    Buffett ran a hedge fund for the first 20 years or so, so those returns are baked into the cake for later CAGR. Very misleading. And Robertson was a product of the pre-1990 era where data was not widely electronic and available. The trick is whether or not these two can still do their anarchronism in this day and age… they’ll have to adapt and so far Buffett seems to be failing at it.