How to satisfy your urges without losing your appetite

I refer, of course, to stock picking -- which, for the average private individual, is a huge waste of time and money. I'm not going to rehash the thousands of studies proving this, nor will I dwell on the fact that in most cases, the time spent doing due diligence on a stock, even if at all capable of adding edge, would have been more risk/reward profitable if spent on work (or flexible independent contracting if you work for a fixed salary w/little upside for performance).

Furthermore, if you are someone working in finance with a legal edge (informational, or wrt. investing skills), you're probably better off investing the time in work and trying to increase your bonus.

Nonetheless, people will still stockpick -- and don't get me wrong -- I enjoy stockpicking for my personal portfolio. Telling people not to stockpick and invest in virtual stockpicking contests instead, is like telling a non-compulsive but bad poker player to quit playing for money because the house/his opponents always win. There is a large component of inherent enjoyability in taking risks, but how can you do so in the stock market w/o having your face ripped off, as Michael Lewis put so piquantly? (Just as real mafiosi have been observed imitating the Sopranos, sometimes I wonder if the people in finance that I know who use that expression are quasi-ironic, perhaps indirectly imitating a relatively rarely used expression popularized by Liar's Poker, or dead serious...)

One way is to reduce the risk of bad investing. If you can restrict your attention to a set of stocks that will likely do better than others on average, you can have your fun (and your full wallet too). Of course, if markets were perfectly efficient, you wouldn't be able to do this regardless.

This is where Joel Greenblatt's recent book comes in. For a short and enjoyable read, check out the hokily titled The Little Book That Beats the Market (co-authored with Andrew Tobias).



Details below:

It essentially claims is that if you asset allocate to baskets of stocks with low P/E (a value stock trait) but hi ROE (calculated w/a little more care than I take here, so that it is actually return on capital, not equity), over the long run, you’ll eke out a decent edge over the market, based on statistical evidence over large groups of stocks over longish time periods. He basically says only intelligent, non-hyperbolic things throughout the book, and mentions topics like how to rebalance portfolios tax-efficiently. (I typically find that one of the top signs of investment crankdom is a writer who tells you to ignore the effects of capital gains/losses when trading. If in fact you have a collection of stocks w/high return and high risk, but little dividend generation, you can selectively sell big losers to offset capital gains, as well as $3,000 of regular income a year w/carryforward.)

Furthermore, Greenblatt makes a decent point about why this effect would exist in the first place and then continue to persist, even after discovery, by appealing to a broad set of behavioral biases that make it unlikely that the effect will disappear. I’m mangling the argument by compressing it, but roughly speaking, he argues that this effect kicks in over long periods, is smaller in large-cap stocks, reflects genuine and persistent risk-aversion preferences among most individuals, also appeals to investor tendency to rank welfare by reference points (investing in a bubble is better than not doing so, because not making money when others do is very painful, and losing money when everyone loses money isn't very painful), and such tendencies drive market movements in ways that are difficult to "arbitrage" away.

Similar arguments are used in mainline finance academia to explain the persistence of value beating growth, despite the disclosure of said effect. For the quant in me, he reduces the chance that this is data-miney by creating baskets of all stocks broken up by decile ranking (using his low P/E and high ROC metric), and showing that over many years, each decile does progressively better than the next by discernable amounts. He also acknowledges various potential problems upfront, such as whether most of this effect is found in small, untradeable penny stocks, survivorship bis, etc.

Performance of Magic Formula (universe of 3500 largest US stocks, choosing top 30, rebalancing once a year) vs S&P:
Over 1988-2004 (yes, this conveniently excludes 1987), all annualized:

• 30.8% w/formula12.3% over universe of equally weighted, 3500 stocks
• 12.4% S&P 500 (equal vs market weighted)
• 22.9% larger cap (top 1000 stocks by market cap) w/formula
• 11.7% larger cap (equally weighted index of top 1000 stocks by market cap)

Best of all, he offers an automatic screener for _utterly_ free on his website that allows you to choose a minimum market cap, and then picks out the top 25-100 stocks meeting his criteria. If nothing else, limiting your fundamental analysis efforts to that universe should prevent you from systematically losing to the market, unless you somehow have biases that always allow you to pick the underperforming stocks from the Greenblatt set -- which may still outperform the market. (It is also fairly difficult to argue that you can systematically bad stocks, as this implies that you can systematically refuse to buy stocks that you like, and instead only buy stocks you dislike, for a market-beating edge.) In essence, if you believe stock picking has alpha, and that you can try to grab some of that alpha, or even if you get utility from stock-picking, despite believing that the house always wins -- at least selecting from the Magic Formula's top picks, you make it less likely that you’ll tank your returns.

If I were writing an academic rebuttal, there are millions of things I could bring up, such as whether this strategy is properly risk-adjusted (he claims yes, and I roughly believe his arguments), whether 17 years is a large enough sample, etc -- and yes, these are all things that I worry about. Nonetheless, I am inclined to believe in this strategy.

I usually hate testimonials, but knowing that lots of people like something is more valuable than not knowing it (even if it is to avoid things that your dentist tell you to invest in, for instance). Hedgies love Greenblatt (they also love raising new money in a record-breakingly starved for alpha markets, so take this with a huge hunk of rock salt)– he’s one of the founders of Gotham Capital, which returned outside investor money on the order of 10 years ago after achieving 50% annualized returns, and since 1985, has had 40% returns. He is still very rich/not looking to maximize cash flow for time spent on writing this book. Specifically respectable people in the hedge fund industry seem to almost uniformly respect him (despite huge differences in investing approaches, etc), and many successful hedgies revere his first book as well (which is focused on special situations investing).

On a more personal note, once I have a Greenblatt set, I research and screen on some additional criteria. Meta-criterion: Select effects that you have a strong prior belief in, and at least some statistical/academic evidence for as well. For instance, in descending order:

1.Pick small caps, or even micro micro caps.
2.Look at insider holdings, and more importantly, sales and purchases -- Smartmoney has a free, clever tool that graphically depicts purchases and sales, and also gives insiders a letter grade.
3.Choose stocks w/minimal analyst coverage.
4.Choose stocks w/low $ price per share (a weak argument)
5.Choose stocks near all time lows -- as always, make sure that this is not caused by expectation that the firm is about to bankrupt, etc (weaker argument)
6.Pick up stocks opportunistically, say, by buying the biggest intraday downmoving of the Greenblatt screened list, assuming that a 10-K release didn't just give new E for the PE, and RO"C", both of which could screw up the screener (data is pulled once a day) . (Weakest...OK, starting to really get out there now...)
...and so on.

I should probably make it clear that this is a way to get your stockpicking jollies and probably not get shafted, not an argument for "this is how you can beat the market by 100% every year if you follow my whiz-bang patented pick tips!"