Sunday, October 28, 2012

Short selling and an Index of Crap

How would you judge an investing strategy that had the following fundamental economic characteristics: 1) Limited potential returns, but unlimited potential losses; 2) Skyrocketing competition; 3)Tax inefficiency; 4) Aggregate net losses over its history; 5) The elimination of a significant source of income in recent years; 6) Risk of asset repossession at creditors’ whim.
Having spent 15 years of my career doing nothing but short selling – including periods of great prosperity and other periods of fast, painful losses – I can argue with some authority that, as an investment strategy, shorting suffers from each one of these characteristics of a bad business.
- Joe Feshbach, noted short seller

There are some good reasons not to bother shorting stocks (and many specious ones, in my opinion.) That is a debate for another time, however. Assuming that one decides to construct a short portfolio, whether for the purposes of outright idiosyncratic profit, creating leverage for the longs, specific or general hedging, creating an insurance pool that behaves better than cash (to deploy on the long side during times of turmoil), or for the  social and financial watch-dogging that one might rationalize, there are a number of ways to structure a collection of short positions.

While incredibly annoying to hear repeatedly, the bromidic axiom that "unlike longs, shorts become a bigger problem as they move against you, and a smaller exposure as they work," the implications of this characteristic combined with the path dependency of certain types of shorts becomes important. Decisions around trimming, pressing, and exiting a short are often more complicated than initializing a short position.

Short-sellers have historically approached portfolio construction in various ways, specifically with regard to concentration and underlying characteristic. I will discuss the taxonomy of short-styles and thesis-labels for later, but broadly there are those that prefer concentration and those that value diversification. In the 80s and 90s a number of short sellers practiced a style of concentrated "catalyst-driven" shorting, where a deeply researched position would reach an inflection point and a moment of reckoning was said to arrive (missed earnings, profit warning, key metric degradation, lowered guidance, drug trial results, investigative journalism report, covenant breach, etc) upon which the stock would be expected to react quickly and acutely, allowing the investor to book profits and move on. Often such positions would be sized up to 10-15% of the portfolio running up to such an event, and the total book contained single-digit number of positions. A number of noted short-sellers had years of very positive returns with this event-driven strategy, before they blew up. Some post-mortems suggest that this catalyst-driven style became less reliable as investors "looked through" negative results and the presumed price-ceilings became illusory. It is easy to see how mishaps can occur with such a portfolio when one's thesis is incorrect, but even when one is "correct" it might be hard to behave around truant positions. Over time there appears to have been a shift to more diversification and an acknowledgement of the decreased dependability of "catalysts"; however, recently some short-sellers have begun exploiting their reputational equity and the power of the release of skeptical reports on stocks which allows for sizing up shorts (or adding leverage) before this new "event" (Chinese RTOs/IPOs, Citronification, Einhorning, etc.)

Even with a move away from catalyst-dependent shorting, some highly regarded investors maintain a reasonably concentrated book of 6-10 medium sized (and often "terminal") shorts with a long time horizon. The philosophy underlying this decision is reasonable, and often invoked on the long side and it goes like this- one should concentrate capital in ideas with the conviction, and where there is an over-whelming probability of making absolute returns over time (even on the short side.) There is a shortage of great ideas, and this approach allows one to avoid the mediocrity of diworsification. I think this is an excellent framework of short-selling and approaches what might be considered optimal given low constraints of time horizon, asset-liability matching, and time consumption on research and maintenance. If one has long-duration capital that allows one to withstand (and even exploit) volatility, and the research bandwidth that allows for deep research, high conviction, constant monitoring, and un-anchored resizing of positions, this might be the best approach.

However, many investors do not have these two elements of time on their side (i.e. infinite time horizon, and large amounts of time to spend on the shorts), and thus may have to trade optimal for robust. Specifically, the trade-off is lowered total returns in exchange for simplicity, time-commitment and arguably lowered brain-damage and psychological trauma. What I am proposing is original and field-tested (although the original parts are not field-tested, and the field-tested parts are not original.) Kidding aside, I am proposing the construction of an Index of Crap- a diverse collection of appropriately sized short positions that is in aggregate both of poor quality and overpriced (on an absolute and relative basis.) As a whole, the components of the Index of Crap (IOC, henceforth) will have low return on invested capital, poor earnings quality, deteriorating fundamentals, low barriers to entry, poor management with adverse incentives, solvency and liquidity issues, inflated asset values and low levels of free cash flow to the enterprise. Sometimes, these stocks will consume and burn cash, have over-earned recently relative to their normalized earnings power, insiders may be selling out, and of course are trading well above a reasonable estimate of fair value. However, it is important to note that singular positions will NOT share all these characteristics (if only) as can be seen in the components of the IOC.  As such, such a "low conviction" portfolio should yield absolute (and relative) returns over time, require relatively less maintenance, and be more valuable than cash for long-term value investing on the long side. Despite the immediately apparent flaws inherent in this approach, it will be useful for some as-is, and perhaps for others with a modified approach. Different categories of short behave differently in terms of their correlation to broad markets, the stage in the cycle that they typically "work," and the volatility of the price. For eg. a highly levered capital markets dependent business may go to zero but go up 3x on its way there, a fraud is generally uncorrelated to the market, and less financially levered secular decliners are less less volatile and may be late or counter-cyclical. Having a balance of such traits in the IOC is worth noting.

While one can improve on the construction of the IOC by ranking factors such as margin of safety, risk/reward, bottle-rocket-quotient (TM), short-interest/rebate, level of innate hedging to the longs and timeliness (you wish), that is antithetical to very reasons one might be interested in this. As such, I suggest an equal-weighted (or two-tiered, for the more engaged) sizing such that one can withstand the doubling and perhaps even tripling of individual positions as long as they are deserved members of the IOC.

With all that rambling, hand-waving and over-simplifying out of the way, I present some of the components of the index of trash i.e. the fun stuff. While I will build a representative IOC portfolio with specific companies in my next post, here are some broad component categories (not all of which are mutually exclusive- in fact, may good shorts share characteristics) with a couple of potential areas of interest in parenthesis:


  • Massively overvalued/priced for perfection: This group may include "good" companies. Whoa! That does not sound like the dark characteristics above and we've heard a number of reputable self-righteously indignant investors drone on about "not shorting simply on valuation" and how "overvalued stocks can get more overvalued." Blah-blah. The IOC will contain a few stocks where valuation math requires a suspension of disbelief- a world with no competitors who like money too, continued pricing power for a rather reproducible good, and the assumption that unending growth will not be costly. While these can stay expensive or get more expensive, over time such stocks are unlikely to provide total shareholder returns. [Growing restaurant franchises, retail clothing concepts, cloud software valued at 30-130x earnings]
  • Broken growth: Similar to the above category in terms of valuation, but where the implied growth has demonstrably stalled, often with an ownership base that is concerned with one specific metric. [Variable-cost software as a service, specialty retail, specialty bedding, organic food]
  • Other People Like Money Too: Generally first-movers in a commodity business with high initial paybacks and low barriers to entry, and typically some kind of volume tailwind. They are often priced on trailing earnings discounting no new entrants and stable pricing and returns on capital. [Pressure pumpers, fracking supply chain, industrial auctioneers.]
  • Old company in a new world: A former franchise or near-monopoly business might find itself in a new world with a different set of economics of customer appetites, but is currently valued in line with its former glory. Sometimes technological obsolescence is the driver, while other times it's simply shifting customer preferences.  [Postage processing, franchise microprocessors, former regulated near-monopolies, newspapers, wireline telephony, patent cliff and generic flood]
  • Over-earners or peak-cycle-extrapolated: A business that has recently earned more than it should on a normalized or mid-cycle basis, and valued as if those earnings are normal. Sometimes there are idiosyncratic reasons why the company has had excess returns recently or has pulled forward future earnings. [Handgun manufacturers, healthcare IT services benefiting from mandated one-time sales]
  • Frauds and fads: Companies with shady accounting, stated assets that might not exist, scientifically implausible technology, fad products that are unlikely to create recurring revenues, and richly valued companies that repeatedly fail to deliver financially [Chinese media companies, oil exploration with undelivered production, electric vehicle and battery complex, shake diets, natural gas vehicle outfitters, mismarked real estate]
  • Structurally disadvantaged businesses: A business may have part of its cost structure impair its profitability and competitiveness. While generally such factors are transient (input cost), sometime they are permanent (unions.) [Union-staffed supermarkets]
  • New regulation and competition: Often less-regulated or favourably regulated businesses earn above-normal returns until regulation, sometimes in combination with new competition entering. [For-profit education, sub-prime credit lenders, prepaid debit cards.]
  • Secular decliners and dying businesses: These are businesses with a diminishing revenue pie (customer shift) or declining share share of the pie (commoditization), often with high operating leverage left behind from their better days. If they have enough financial and operating leverage, these tend to be potential "zeroes." [Wireline telephony, newspapers, radio, mail, directories.]
  • Cyclical and macro inflection points: Sometimes supply/demand imbalances are undiscounted because of inertia, euphoria or insistence that "this time is different" despite market levels of new supply coming on line. [Iron ore, China-dependent consumption.]
  • Controlled-Flight-Into-Terrain: Borrowed from the aviation crash assessment industry, CFIT describes an otherwise functioning business that is being run into the ground by decidedly poor management actions such as misallocation of capital, poor product decisions, and misguided corporate actions. Sometimes even activist investors are unable to change the course of these impending wrecks. [Truck manufacturing, mailing equipment, empire-building technology]
  • Safe havens or areas of perceived security: During periods of instability, investors of seek safety and sometimes just the illusion of it. Pockets of the market receive inflows that significantly skew valuation as investors rush into purportedly "defensive" securities. Sometimes securities and entire asset classes that pay a dividend may attract the yield pigs. [Liquidating REITs financing payouts through asset sales, telecoms that pay large dividends, consumer staples, "safe" banking jurisdictions like Australia and Canada]
  • Reverse-Greenblatts: Businesses that are the opposite of Joel Greenblatt's formulaic casting net i.e. cash burning businesses with low returns on investing capital and low free-cash flow yields to the enterprise, and preferably with large fixed liabilities (LED suppliers, solar, network equipment]

These are just some categories of the types of shorts that the IOC will contain. In my next post I will start putting together bits and pieces of potential crap.