Portfolio review: A good four years, and a humbling quarter
Beginning in June of 2008, I set out to invest money in portfolio of value stocks that I would pick myself. I got started with Joel Greenblatt's The Little Book That Beats The Market, moved on to The Intelligent Investor and Security Analysis... and then got my hands dirty. I set up a portfolio called JEM (the name is a historical artifact I won't get into here) and began investing.
Almost from the beginning, I outperformed the market significantly. At the end of April of this year, I outperformed the fund's benchmarks (the Russell 3000 and the S&P 500) by 20% on a cumulative basis since inception. I was feeling good. May and June of this year were terribly unkind, and I gave back nearly every dollar of outperformance. This has led me to some introspection - was my underperformance a function of changes in the portfolio? Market shifts? Or other factors (explainable or not)?
Here are some brief thoughts on the portfolio's performance at its four-year anniversary.
Portfolio statistics.
My initial hypothesis was that the portfolio simply underperforms in bad market climates. I considered using market shorts (buying puts or shorting an index) to protect my downside. Shorting an index would only work in the event I could be sure I was generating true alpha. I did a series of cross-section analyses as well as running regressions against the Russell 3000 and the S&P 500 to determine if I was generating alpha.
Cross-Section Analyses
Not surprisingly, the portfolio's returns tend to cluster around those of the Russell 3000.
The months in which it underperforms the Russell 3000 (27) outnumber those in which it outperforms (22). But the average outperformance (3.9%) more than makes up for the average underperformance (2.6%). So from a cross-section perspective, the portfolio's performance is good, but not great. I would like to consistently outperform the benchmark index, although that might simply be too much to ask!
Summary statistics
In summary, this portfolio is more volatile than the market benchmarks (33% standard deviation vs. the Russell 3000's 22%). But the market benchmarks also are not highly correlated with this portfolio - R squared is 57% over the last four years. This suggests that either the JEM portfolio is not well diversified, or that the Russell 3000 is not an appropriate benchmark. It may be a little of both; in the future I may compare the portfolio's performance against a value index as well as the total market indices. Regardless, I computed the alpha of the portfolio and found that I have generated more than 4.5% alpha over the last four years against the Russell 3000 and S&P 500. The portfolio's beta is 1.12 against the Russell 3000, although it does suffer from low correlation as I mentioned above. (I want to caveat this by saying that a significant portion of the alpha is generated by outlier months, and I fully recognize that. Happy to engage in a more detailed discussion if you're interested.)
Portfolio concentration.
As I gained what I believed was confidence, I allowed the portfolio to become more concentrated. (I believe I might label it something else with the benefit of hindsight.) For most of the portfolio's life, I kept a cash position of 10% - 15% of the portfolio's value, and typically held between 15 and 20 equally-sized positions. In recent months as I have rolled out of older positions, I have double- (or in some cases triple-) downed on investments, allowing the number of positions to drift down. This represented a significant lack of discipline on my part, because it allowed the portfolio to become less diversified. In addition, I took a highly-leveraged position in RIMM by writing puts and purchasing calls, which has presented significant headwinds to the portfolio in recent weeks even as the market has recovered from its drop at the end of May. Although my investment in RIMM was consistent with my traditional investment thesis for the fund (high return on assets, low P/E ratio), doing so using a leveraged technique was a mistake. In the future I am going to "stick to my knitting" and make equity investments, perhaps purchasing long-dated call options if I believe an investment offers significant upside.
Failure to make new investments.
Over the last 12 months, I found a dearth of new opportunities, and thus did not liquidate positions as called for by the fund's investment policy. Although my typical hold period is around 12-15 months, many of my positions are 24 or more months old at this point. Although I have not explored whether momentum effects are responsible for any of the portfolio's returns, it seems possible since I enjoyed so much success with a very consistent hold period. If that is the case, I am now realizing the reversal of those momentum effects. Although outright losses were responsible for some of the decline in the portfolio's value over the last quarter, most of the losses came from the evaporation of prior gains which I failed to lock in by liquidating positions at the appropriate time. In combination with my above remarks on concentration, it is clear that I have made a series of humbling mistakes by departing from the fund's investment policy during the last 3 - 6 months.
The path forward.
"We should be careful to get out of an experience the wisdom that is in it - and stop there; lest we be like the cat that sits down on a hot stove lid. She will never sit down on a hot stove-lid again - and that is well; but she also will never sit down on a cold one anymore." -Mark Twain
I want to be very cautious about my next steps with the fund. As demonstrated above, I enjoyed substantial success with the fund over the past 48 months. My departure from the fund's investment policy is indicative of a seat on the hot stove-lid. The wisdom I gained is not to give up and begin indexing, but instead to return to my core values. I have confidence that my philosophy (deep hat to Joel Greenblatt, of course) is sound. In order for the performance of the portfolio to reflect that success, I must execute consistently. I plan to return to my core philosophy by resizing positions within the portfolio.
A curiosity.
One item I have noticed is that my stock screens tend to identify certain industries (for instance, semiconductor testing equipment or small pharmaceutical companies) which have a number of undervalued companies. In some cases, this is because of shifts in industry fundamentals. Although I do not plan to pursue the strategy actively, I am considering exploring a market-neutral approach to these investments in which I will short weak competitors against a long position in strong companies in an undervalued sector. It seems to me this may be a way to generate pure alpha. I have read a significant amount of Ray Dalio's work on risk parity, and while I am not sure I am a believer, I do find his concept of purchasing alpha and beta separately to be fascinating. Given my fund's performance, I believe that I can offer pure alpha and would like to explore mechanisms to do so in the future.
Please note that JEM is not currently open to new investors, and this does represent a solicitation of any kind. Individuals curious about my investing style, portfolio calculations, or just friendly banter can contact me by e-mail at jmatthewhouse at gmail.com or on Twitter at @jmatthewhouse.