Seeking Alpha
Value, special situations, research analyst, registered investment advisor
Profile| Send Message| ()  

I developed some models for use if, as many assume, we are and/or will soon be in a recession. The more I thought about that scenario, the more I fixated on historical precedent regarding success at such times comes coming from investment strategies designed for recovery. Committing funds this way is scary since every downturn looks, while we're in it, like it could be "the big one." So I will present the bear-market models I created. But first, here's a detour that puts some risk-controlled recovery strategies on the table.

A bottom?

I don't know if we are at, near, or just beyond a major market bottom. But I am confident that whenever the bottom occurs, there won't be widespread recognition until well after the fact and well after the "easiest money" has been made.

The present reminds me of my early Value Line days, specifically, an occasion when Arnold Bernhard, the company's late founder (who'd seen it all, having entered the business in the 1920s) invited several younger analysts to a luncheon. During the conversation, the research director asked "AB" whether he thought we'd ever again see a buying opportunity as spectacular as the epoch one that occurred in 1974-75 (when the country was mired in energy, financial, economic and political crises). He answered that he thought we were experiencing such an opportunity right then and there. Naturally, we up-and-comers knew better and dismissed his optimism as an idle, albeit well intentioned, pep talk.

That luncheon occurred in July 1982, when the U.S. again seemed mired in hopelessness (this was about three years after the infamous Business Week "Death of Equities" cover story). But in fact, we were just a few weeks shy of the start of one of the most powerful bull markets in history, far greater than the one launched in '75.

Anecdotes like that sound good. Most market veterans have one or more similar tales. But a case for recovery now need not rest on anything like that. This past weekend, Barron's argued effectively for a shorter, shallower downturn than many expect. It cited the drop in energy prices, the increase in exports relative to imports, and a prospective inventory re-build effort.

I'd like to add a fourth consideration: the likelihood that the financial restructuring-bailout will be far more successful than many now imagine. At this time, we have an entire generation of investors who take in stride the notions that free markets are best and that government intervention can only range between bad and very bad. Without attempting to launch a major political debate at this time, I suggest that government, even in the context of capitalism, has played constructive economic roles at times in the past and can do so again under the right settings, which may well be the case right now. This could later become a big surprise that lights a bullish fuse.

A question of timing

A best-case scenario would suggest we've already seen the bottom. But even bulls could easily envision at least one more downward leg. What we saw in recent months wasn't a conventional bear market, where some stocks did badly, others did very badly, others did more badly, and so forth based on different characteristics that repelled investors in different ways and to varying degrees. Instead, we got a widespread liquidation of equities as an asset class, with little or no regard to differences between one stock from another, and this liquidation had little, if anything, to do with whether equities in general were worth owning. The liquidation instead was a byproduct of something else, a widespread and misplaced belief that it was a good idea to lend a lot of money to people who had no chance to repay so they pay irrational prices to purchase homes there was no chance they could afford.

We don't know when all such equity liquidations will conclude, but we may see some increases in stocks that reflect little more than some re-adjustments of asset allocations, as the process zigs and zags its way to an end. If that's what's been happening lately, with the recent up days, we may still have to experience a more conventional recession-related bear market, one where stocks on the whole go down but this time, reflecting assessments of the stocks and companies.

Based on that prospect, I will post subsequent articles explaining the bear-market strategies I've been working on. On the other hand, recoveries tend to come suddenly and unfold rapidly when we least expect them. So even the most strident bear would do well to at the very least keep an on-paper recovery portfolio in place, and be able to switch into it in whole or in part at the drop of a hat. To accommodate that, I lead with a recovery strategy.

What businesses should be considered

When it comes to recessions, equity investors can quickly reel off a list of so-called "defensive" businesses that ought to be less impacted than most by downturn and tend to focus primarily or exclusively on these. (I'll have more to say about this next time.) But when it comes to recoveries, the approach is usually different.

One typical early-bull strategy involves going back into the hardest-hit areas, reasoning that these should snap back most quickly. Another approach involves abandonment of the concept of preferred businesses and returning instead into stocks across the board, regardless of industry.

I suggest a different approach: creating a list of preferred industries, just like the defensive ones that are created for bear markets, with this list being geared specifically for the type of upturn that is anticipated.

It sounds simple, but it can have a very substantial impact on the nature of one's recovery strategy. This can be seen by considering what's on my list of eligible sectors, and more importantly, what's omitted:

  • Services
  • Consumer Discretionary
  • Transportation
  • Technology

Finance is missing. So too are Housing and Energy-Resources.

Rather than going back into the hardest-hit areas, I'm continuing to avoid them. Business and market cycles are really a modern misnomer. When we look closely, we see that rarely does the market (or even the economy) have all it oars in or out of the water at the same time.

Consider energy. We all know what happened recently. But recall 1982 until the mid-2000s. That was the time of the epoch bull market . . . or was it? For much of that period, energy was in a dismal bear market: we didn't notice or care much because we had too many other fun things to absorb our attention. But the fact remains, that as the great 20th century bull run roared ahead, many passengers were left stranded at the station.

That's not the only example. But you get the point.

The above list allows me to start investing for recovery without having to tackle the most troublesome issues of the day: What will the financial restructuring mean for shareholders of those companies? Will energy and commodities get back onto bullish tracks? Are home prices near a bottom? How long will it take for capital spending to get going again (notice I also omit this late-stage sector)?

Who cares! There's no law that says a recovery strategy must be across the board or that it must include groups that led on the way down. We are all every bit as free to build a "play offense" eligibility list that serves the same function as the "defensive" eligibility lists so many are accustomed to using when things look bad.

I backtested my eligibility list on Portfolio123.com to get a sense of how these limitations might impact prospects. I didn't use a model portfolio approach. Instead, I used Portfolio123's advanced back-tester to create hypothetical portfolios at the start of each week between 3/31/01 and 9/13/08. All were then "held" for only four weeks. In other words, portfolio 1 ran from day 1 through day 28; portfolio 2 ran from day 8 through day 35, portfolio 3 ran from day 15 through day 42, and so forth. The results reflect the average performance of the 390 four-week portfolios thusly created.

Table 1 compares the performance of all stocks in the eligible sectors with the performance of all stocks in general

Table 1

 

Average "excess" (vs. S&P 500)
share price change when . . .

Market is Up

Market is Down

All Stocks

2.23%

-0.20%

All Eligible "Play Offense" Sectors

3.92%

-1.12%

Computations are for stocks with market capitalizations of at least $250 million

If past is prologue, I'm definitely well positioned to play offense. I add some more downside risk. But I'm prepping for an upturn and in this regard, prospects are clearly enhanced.

Note though that back-test results cannot be followed rigidly since past isn't always prologue. I keep that in mind here: housing and finance were strong through much of the test period and until the end, energy-resources generally ranged from neutral to positive. Going forward, I can easily envision prolonged concerns in those areas resulting in a less favorable up-market-down-market profile for the full universe than what is suggested above.

Modeling with multi-factor ranking systems

Next, I rank stocks in the eligible sectors according to multi-factor ranking systems I developed. Hopefully, these can improve the upside-downside profile of the overall eligible group. But even if they can't, the ranking is essential to allow me to reduce the number of stocks under consideration from the thousands, to a number an investor could plausibly own (I assume 20 positions).

The approach I developed is based on three components: Value, Growth and Statistical Stability.

The Value and Growth components are the exact same ones I developed for the bear-market strategies. The only thing being done differently now is that I'm applying them to a play-offense list of eligible sectors, rather than to a defensive group.

The Statistical Stability component looks at the same factors as are used in the bear market model. The difference, besides the eligibility list, is in the way the factors are sorted. When playing defense, I seek stability. Now, I'm looking for volatility. For example, in the bear-market model, lower betas were preferred. Here, higher beta stocks are favored.

This modeling is still a work in progress. So for now, weightings among factors within a component remain equal. Based on that, here's what I have at present:

  1. VALUE
    • Earnings Yield
    • Enterprise Value/EBITDA
    • Enterprise Value/Estimated EPS
    • PEG Ratio
    • Price/Sales
    • Price/Book
  1. GROWTH
    • EPS and Sales Growth, latest quarter
    • EPS and Sales Growth, trailing 12 months
    • EPS and Sales Growth, last 5 years
    • EPS and Sales Growth, last 10 years
  1. STATISTICAL STABILITY
    • Beta
    • Beta Relative to Industry
    • Standard Deviation of Share Price
    • Change in Standard Deviation
    • Change Short Interest
    • Standard Deviation of EPS trend
    • Range in current-year EPS estimate
    • Range in long-term growth-rate estimate
    • Stability in current-year EPS estimate

Table 2 summarizes the results of the more promising Portfolio123.com backtests.

Table 2

 

Average "excess" (vs. S&P 500)
share price change when . . .

Market is Up

Market is Down

All Stocks

2.23%

-0.20%

All Eligible "Play Offense" Sectors

3.92%

-1.12%

 

Value

2.30%

-0.06%

Value (50%) + Growth (50%)

2.40%

-0.15%

Value (50%) + Stability (50%)

3.38%

-1.32%

Computations are for stocks with market capitalizations of at least $250 million

At first glance, it looks like the Value and Value+Growth models don't deliver. On reflection, though, they have much to recommend them. The up-market-down-market profiles match the full stock universe. But the number or choices is now manageable (20). And as noted, given my reservations about areas like finance and housing, I suspect that going forward, the full universe might not perform as well as the backtest might infer.

The Value+Stability model doesn't quite match the up-market-down-market profile of its "benchmark." Bear in mind though that it preserves the general flavor and uses only 20 stocks, thus making it investable, something that cannot be said of the all-eligible-stocks group.

The Value Model

Value is usually seen as a conservative style that's best used in bear markets. Actually, recent testing has raised questions about this approach.

Nowadays, when so much information is so widely available, it's hard to imagine there are many "cheap" stocks out there that don't in some way deserve to be cheap. The most recent examples, of course, were the low price metrics seen on the unraveling housing and financial stocks. So bear market investors who favor this approach need to adopt more discriminating approaches (as will be discussed inn a subsequent article).

Another interesting observation is the role value can play on the upside. As noted, stocks today often get pounded because investors find bona fide cause for worry. But when recovery starts, those concerns are often abandoned quickly prompting stocks to move, even well ahead of earnings data reflecting better times. And recall that this particular model conditions the universe to eliminate companies that seem most vulnerable to misplaced hope.

The factors are the usual suspects. Note, though, that the standard P/E was flipped upside down and expressed as an earnings yield. I do this to avoid the tendency of databases to assign NM ("Not Meaningful") entries to companies with losses. When investing for early-stage recovery, I do not want to abandon these and am happy to accept negative earnings yields.

Table 3 shows the 20 stocks that presently appear in the value recovery model.

Table 3

The Value Recovery Model

Companies sorted by market capitalization

AU Optronics Corp. (ADR) (AUO)

Electronic Instr. & Controls

Seagate Technology (STX)

Computer Storage Devices

Magna International Inc. (USA) (MGA)

Auto & Truck Parts

The Goodyear Tire & Rubber Co (GT)

Tires

Avnet, Inc. (AVT)

Electronic Instr. & Controls

Manpower Inc. (MAN)

Business Services

Arrow Electronics, Inc. (ARW)

Electronic Instr. & Controls

Ingram Micro Inc. (IM)

Computer Hardware

Autoliv Inc. (ALV)

Auto & Truck Parts

AutoNation, Inc. (AN)

Retail (Specialty)

Tech Data Corporation (TECD)

Computer Hardware

Celestica Inc. (USA) (CLS)

Electronic Instr. & Controls

Office Depot, Inc. (ODP)

Retail (Specialty)

Amkor Technology, Inc. (AMKR)

Semiconductors

Benchmark Electronics, Inc. (BHE)

Electronic Instr. & Controls

Oshkosh Corporation (OSK)

Auto & Truck Manufacturers

Himax Technologies, Inc. (ADR) (HIMX)

Semiconductors

Navios Maritime Holdings Inc. (NM)

Water Transportation

Lear Corporation (LEA)

Auto & Truck Parts

Sierra Wireless, Inc. (USA) (SWIR)

Communications Equipment

 

The Value-Growth Model

Here, the aforementioned value approach is supplemented by consideration of growth. The latter is widely accepted as a sound way to play a recovery. But the devil is in the details: How should growth be measured?

Notice above that forecasted growth is not part of the picture. I'm usually a big advocate of using estimates-based data. But frankly, scenarios such as the one we now face, where things have been deteriorating contrary to expectations held by many as recently as six months ago and uncertainty as to whether things will get better or worse in the near future, are difficult for analysts. Trend-oriented forecasting techniques used by many don't work at times like these. And the corporate executives who develop the guidance so widely used nowadays likewise find it difficult to wrap their arms around what's taking place.

I therefore prefer to stick with historical growth rates. Use of recent tallies is standard. I'm also willing to stretch farther back in time toward companies that have, relative to others, actually shown themselves better able to grow. The next cycle might witness changes in leadership. But that's for later. I'm assuming that in the earliest stages in recovery, the investors will be viewing the world through familiar leadership lenses.

Growth alone tested poorly, but the approach fared much better when teamed with value, presumably causing it to be more likely to favor stocks for which the so-called shoes have already dropped.

Table 4 shows the 20 stocks that presently appear in the value-growth recovery model.

Table 4

The Value-Growth Recovery Model

Companies sorted by market capitalization

AU Optronics Corp. (ADR) (AUO)

Electronic Instr. & Controls

LG Display Co Ltd. (ADR) (LPL)

Electronic Instr. & Controls

Western Digital Corp. (WDC)

Computer Storage Devices

Lenovo Group Limited (ADR) (LNVGY)

Computer Peripherals

Jabil Circuit, Inc. (JBL)

Electronic Instr. & Controls

Fossil, Inc. (FOSL)

Jewelry & Silverware

General Cable Corporation (BGC)

Communications Equipment

Take-Two Interactive Software (TTWO)

Software & Programming

Federal-Mogul Corporation (FDML)

Auto & Truck Parts

EnerSys (ENS)

Electronic Instr. & Controls

Chartered Semiconductor Mfg. (CHRT)

Semiconductors

Navios Maritime Holdings Inc. (NM)

Water Transportation

Republic Airways Holdings Inc (RJET)

Airline

inVentiv Health Inc. (VTIV)

Business Services

Exide Technologies (XIDE)

Electronic Instr. & Controls

Sigma Designs, Inc. (SIGM)

Computer Peripherals

TBS International Limited (TBSI)

Water Transportation

Multi-Fineline Electronix, Inc (MFLX)

Electronic Instr. & Controls

Hawaiian Holdings, Inc. (HA)

Airline

Sierra Wireless, Inc. (USA) (SWIR)

Communications Equipment

 

The Value-Stability Model

This approach seeks value stocks that had, in the recent past, been causing the most consternation to investors. It's tempting to measure Wall Street angst by looking at declining stock prices. But lately, pretty much everything has been dropping without regard to much except the need of holders to liquidate. I'm not sure I can count on a mirror-image bounce because the most aggressive liquidators may not be in much position to participate in the next up-cycle (in other words, recent sellers may want to buy those same shares back but be unable to do so).

So I choose to measure angst in terms of stability rather than price trend. Beta is an old standard. Standard deviation is a close cousin (volatility for a stock is calibrated on its own, rather than the stock relative to the market). When it comes to factors like short interest or earnings estimates, I don't care so much about direction (as noted, it's hard to give much credence to such expectations nowadays). What I'm most interested is uncertainty in its own right; the level of indecisiveness. I'm looking for situations where investors are likely to turn away from the unknowns and get on board for an unfolding up-cycle, trusting that one way or another, the things that recently wondered about will eventually fall into line.

The back-test results point to this as model as having the highest risk-reward profile. Considering how we're actively courting angst, that seems logical.

Table 5 shows the 20 stocks that presently appear in the value-stability recovery model.

Table 5

The Value-Stability Recovery Model

Companies sorted by market capitalization

J.C. Penney Company, Inc. (JCP)

Retail (Department & Discount)

Macy's, Inc. (M)

Retail (Department & Discount)

Seagate Technology (STX)

Computer Storage Devices

Magna International Inc. (USA (MGA)

Auto & Truck Parts

Western Digital Corp. (WDC)

Computer Storage Devices

Manpower Inc. (MAN)

Business Services

Jabil Circuit, Inc. (JBL)

Electronic Instr. & Controls

General Cable Corporation (BGC)

Communications Equipment

DryShips Inc. (DRYS)

Water Transportation

Office Depot, Inc. (ODP)

Retail (Specialty)

TRW Automotive Holdings Corp. (TRW)

Auto & Truck Parts

Fairchild Semiconductor Inter (FCS)

Semiconductors

Sotheby's (BID)

Retail (Specialty)

ArvinMeritor, Inc. (ARM)

Auto & Truck Parts

Oshkosh Corporation (OSK)

Auto & Truck Manufacturers

Navios Maritime Holdings Inc. (NM)

Water Transportation

Brown Shoe Company, Inc. (BWS)

Footwear

Sigma Designs, Inc. (SIGM)

Computer Peripherals

TBS International Limited (TBSI)

Water Transportation

Lear Corporation (LEA)

Auto & Truck Parts

 

The other side of the coin

I tried to control risk here by limiting the sector eligibility list. Realistically, though, investors who really do anticipate further turbulence ahead can find better ways to invest, even within the equity asset class. I'll therefore follow-up with more detail on bear market approaches.

Source: How to Position Your Portfolio for Recovery