What's Working
Bear-Market Bargain Hunting

Suffering losses when the market tumbles is bad enough. Adding to the irritation are commentators who whoop with joy about how bear markets afford great opportunities to buy good stocks at bargain prices. Frankly, I'd rather pay full price for what I buy and enjoy a much higher brokerage balance. So I'm not going to wax poetic about how much I love these buying opportunities. But . . . that won't stop me from take advantage of them while they're here.

Screens for bargain hunters

Here are some stock screens created to identify potential bargains in the current down market. All are available as P123 screens.

• Book Value Bargains
  Identifies stocks trading below per-share book value
  
  
• Cash Bargains
  Identifies situations where net cash per share is high relative to the share price
  
  
• Debt-free Bargains
  Identifies debt-free companies with bargain-priced stocks
  
  
• Income Bargains
  Identifies good-yielding stocks with only moderate risk
  
  
• Performance Bargains
  Identifies companies with long-term records of financial performance whose stocks seem uncharacteristically cheap
  
  

In all cases, the fundamental factors were run against a universe defined as follows:

• No OTC stocks
• Market cap greater than or equal to $250 million
• Share price greater than or equal to 5
• No ADRs
• No Finance stocks

Since these are public screens, those factors are expressed as screening rules, rather than a custom universe. The first three rules are standard for me. The no-finance rule is geared for our current situation. I believe it is presently impossible to many finance stocks. To tackle that task, I'd want to have expertise in political science, since developments in that area are likely to trump anything we can glean from financial statements. As to the no-ADR rule, I'm on the fence. I put it in mainly as a comfort factor; to stick with more familiar U.S. stocks. Actually, though, I can't objectively say my quest for comfort will enhance probabilities of success. Before actually investing money, I expect to take a look at what shows up if I disable the no-ADR rules.

All of these screens will be discussed individually in subsequent blogs. But if you click on the links above, you can get access to all of them immediately.

Caveats and context

Before jumping head first into any of these screens, there are some important issues that need to be addressed.

1. Not necessarily good for all time
   
   These screens were created in order to take advantage of today's (hopefully) rare market environment. And if/when in the future we experience another period like this, perhaps they can be dusted off and used again. But if you're thinking in terms of a normal rules-based Portfolio123 portfolio, these are not likely to be ideal choices. Often, these screens will feature few or even zero passing companies; the fact that we can even consider getting viable lists (something that can still vary from day to day) says much about where the market is right now. And even where normal-sized lists can be obtained at other times, healthy markets may be less likely to respond to the factors used here.
   
   
2. A different approach to backtesting
   
   As usual, I look at the maximum 3/31/01-present period. I'm happy if it looks good, but I don't put too much emphasis on it since, as noted, these are not intended to be good-for-all-seasons screens. I look at the past year, but I don't expect superior performance (I am, after all, looking for stocks that are dirt cheap and stocks don't usually get that way unless they performed badly in the past). Instead, I'm satisfied if the screen's performance is not significantly worse than the S&P 500. In other words, I'm content to offset potentially superior upside performance with more-or-less market-matching downside prospects. I also run backtests on the 6/30/02-12/31/03 period. I do look to see some upside potential in that early-recovery period. (Note, though, that the prior period was different, with the worst of the collapse having been experienced in tech and telecomm. Many fundamentally sound companies in other sectors held up fairly well. The upshot: the 2002-03 screens produce much fewer passing companies than some screens give today.)
   
   
3. Supporting rules
   
   Often, cheap stocks are cheap for good reason; i.e., because they deserve to be cheap. These screens assume the present is an unusual time, when equity liquidation based on cash-raising or asset allocation needs makes for some notable exceptions. But that shouldn't blind us to the fact that some now-cheap stocks still reflect trouble and would likely have been cheap even if the market hadn't tumbled. Therefore, the screens make some effort to include at least one supporting rule to try to weed out these really-bad cases. (Hopefully, a lot of potential trouble can be avoided with the Sector!=FINANC rule.) Beyond that. . . .
   
   
4. Rules-based investing versus fundamental analysis
   
   We at Portfolio123 exhibit a wide variety of investing approaches but one thing we all have in common is our use of rules-based protocols, as opposed to in-depth fundamental analysis of individual companies. I've done a lot of both over the years, but my preference is for the rules-based approaches we use here. The tests applied here give reason to believe that these screens can work in the usual way. Still, this is an atypical situation and as noted, we need to avoid being blas? about the possibility that some of these situations really do warrant the low prices we see. Hopefully, the supporting tests will help. But if one is tempted to supplement the rules-based approach with classic one-at-a-time fundamental analysis, today's market and the nature of these screens present appropriate opportunities, especially the Cash Bargains screen.
   
   
5. Measuring profitability
   
   Several of these screens include rules requiring companies to be profitable, which are expressed using our basic standby:

   EPSPExclXorTTM>0

   Understand, though that Thomson Reuters, our data provider defines this EPS number based on GAAP (Generally Accepted Accounting Principles) reporting (a standard practice among financial data providers). It excludes rarely-seen "extraordinary" items (such as the cumulative impact of changes in accounting policy). But nowadays, many non-recurring and often large items (both gains and losses) are included in the GAAP results we use.
   
   As is usually the case when times are tough, there's an army of reporters out there wagging their fingers at those who supposedly try to disguise how bad things are by looking instead at some species of operating earnings (case in point: the Ahead of the Tape column in today's Wall Street Journal). Actually, though, there's a lot more to this topic than some circa-1999 internet startup based in somebody's garage trying to raise billions in an IPO by persuading potential investors to evaluate it based on IBAE (income before all expenses), which is pretty much what many have in mind when they bash operating earnings. Actually, though, the notion of trying to measure a company's performance without the distortions caused by unusual developments unlikely to be repeated in the future comes straight out of Graham and Dodd.
   
   I don't really want to spark a full-fledged accounting debate here and now, but instead, to alert you to the fact that we're likely to see a ton of fourth-quarter write-offs that make it much harder than usual for normally-profitable companies to report positive EPSPExclXorTTM. As these write-off hit the database and the EPSPExclXorTTM figures plummet (starting, probably, in late January), we may want to consider substituting a rule like this:

   HistQ1EPSActual+HistQ2EPSActual+ HistQ3EPSActual+ HistQ4EPSActual >0

   These are historical quarterly EPS numbers calculated based on the sort of operating criteria used by Wall Street analysts as bases against which their estimates can be compared. (These aren't the shady IBAE metrics but are instead along the lines of what Graham and Dodd had in mind.) This isn't my first-choice since companies too small to warrant analyst coverage will necessarily vanish form your lists. (As of today, 4,012 firms can pass the EPS test, but only 2,947 make it when I switch to HistQ_EPS.) Use of EBIT, EBITDA or even Operating Earnings won't help much since many of these non-recurring items enter the income statement as "operating expenses." If you want to try to eliminate non-recurring items but don't want to limit yourself to stocks with analyst coverage, here are some other potential, although admittedly imperfect, workarounds:

   SalesTTM- COGSTTM- IntExpTTM- DepSCFTTM>0
   
        Sales - cost of goods sold - interest expense - depreciation
        (3,493 passing companies)
   
   ItemTTM(OTLO,0)-ItemTTM(SCEX,0)>0
   
        cash from operations - capital spending
        (4,531 passing companies)

   I'll discuss this topic further later on, as we get a better picture of how the write-offs impact results (this could turn into a very substantial issue for us since December-quarter write-offs will be in the TTM figures all through 2009). But for now, and for these screens, the simple EPSPExclXorTTM metric should suffice.

OK. Enough with the context. As noted, subsequent blogs will look at the screens individually, but if you want to get started right away, the screens can be accessed by clicking on the above links or through the P123 screens section. Note, too, that all of these screens are fairly simple, so it should be easy for you to come up with your own wrinkles. (Also, you could incorporate the first five lines into a custom universe, which is what I'd have done had I kept the screens private).

Book-Value Bargains »