What did we do right in 2009?
One year of good return may be just a result of high tide lifting all boats or simply mean-reversion from a terrible year. Nevertheless, my biggest take-away from 2009 was a subtle but important change to my investment philosophy - I have changed my focus from "good and cheap" to "better and cheap". I care more about change in fundamentals - I prefer a bad company that is getting better over a good company with no change in story. This new philosophy has led to solid stock picking, which generally out-performed the market with what I believe to be lower risk ("permanent loss of capital"). Equally important, this new framework gives me better guidelines to size my bets, especially betting heavily in situations where both the story is getting better and stock is cheap.
When I started investing a few years ago, I was firmly in the value investing school - concepts like "intrinsic value" and "Mr. Market", coined by Ben Graham and popularized by Warren Buffett, clicked for me instantly. I spent time studying company fundamentals, coming up with an estimate of the intrinsic value, and trying to buy at a cheap or discounted price. In short, I was trying to buy "good and cheap", and results were satisfactory.
However, I have come to realize the quality of the company and absolute discount to intrinsic value are not everything - one has also to consider the time and factors it takes for the discount to narrow, which typically depend on the business cycle. Thus my new approach comes down to balancing between value and momentum. Value refers to the price paid for the business. Momentum, not to be confused with price momentum in quant and technical analysis, refers to business momentum, i.e. how well the business is doing. Improving momentum can come in the form of higher margin, accelerating top line growth, or improving ROIC. With the exception of select great companies in their growth phase, most companies' stock price and business momentum move in cycles/curves similar to sine waves with peaks and troughs.
These two curves are closely related - when business momentum is good, stock price tends to go up, and vice versa. However, there is often a lag between the two curves, and depending on the part of the cycle, stock price will react to the change in business momentum very differently. I believe this is the crux of investing - how you identify which part of the cycle the company is in, which drivers to watch for and which valuation metrics to use. For example, earning revision is a powerful factor but completely useless at business peaks and troughs. P/E may be a good valuation metric in general, but unadjusted for margins, it is useless or even dangerous at extremes.
For example, assume a retailer's intrinsic value is $20, and buying at $15 may give an expected return of 33%. However, the same $15 price may correspond to two points on the momentum curve - one where the curve is turning up (story getting better) and the other where the curve is trending down. In the former case, you will probably get to $20 in 6-12 months. In the latter case, you may have to wait 18-24 months before the retailer corrects excess inventory and produces positive SSS (curve turning up again) to reach the $20 intrinsic value.
There are two obvious problems with buying at the latter point. First, time adjusted return is obviously inferior. Second, the stock price may first plunge to $6 before recovering. While a pure value investor may think a lower price makes it a better buy (even more margin of safety), reality is that an adverse price movement will slowly but surely inject doubt into my mind. Have I made a mistake? Is this a value trap? Very seldom does stock price move down without some deterioration of business fundamentals and some changes to the initial investment thesis. So unless one has an iron stomach (I don't), it is very tough to keep calm during the price downdraft and continue to average down. There is an even bigger issue - if you are prepared to average down, chances are that you will not buy a full position initially, and inevitably you will end up establishing similar-sized partial positions for all new ideas. Yet some of those ideas will have good business momentum and they are your surer bets, so you lose potential profits in positions that actually have the best risk/time adjusted return.
So doesn't quant investing capture "better and cheap", as preached by the noted quant investor Cliff Asness? Yes and no. I believe there are two problems with quant investing. First, it mistakes cause with effect - price momentum is the result of business momentum, and while the two will resemble each other at certain part of the cycle, they will diverge significantly at critical turning points. Second, the effectiveness of various factors differs significantly from industry to industry as well as at different parts of the business cycle. Quite simply, quant investors lack the domain knowledge of each industry and use the same factors or same weightings across sectors during different points of the cycle.
For example, quant investors will universally use factors such as earning revision, revenue/EPS surprise/breadth to capture business momentum. While this does a satisfactory job overall, it will not capture key drivers for each industry, which often cannot be retrieved from standardized financial statements, such as inventory/store for retailers, or asset inflows for asset managers. Often changes in these key drivers will long precede actual changes in earnings, so generalized quant investing could easily miss the turn. As another example, six months ago, both KIRK and ARO got the highest rating in our internal quant system, yet the two retailers could not be more different in terms of where they were in the business and margin cycle, and the subsequent divergence in stock performance illustrated the flaw in the quant investing approach.
I certainly do not want to leave the impression that other investing approaches are inferior. Indeed, there are many ways to achieve success in investing, and everyone needs to find approaches to fit his or her own traits. I believe I have found mine by balancing between value and momentum. Put simply, I aim to invest in situations where fundamentals are about to turn or have turned while valuation is reasonable. I am certainly not reinventing wheels here, as this is the approach advocated by both Peter Lynch ("catching the turn") and Warren Buffett ("What we really like to see in situations is a condition where the company is making substantial progress in terms of improving earnings, increasing asset values, etc., but where the market price of the stock is doing very little while we continue to acquire it").
Well, this approach may sound good on paper, but how many of these "perfect" situations exist, given how efficient market is with so many hungry and smart investors poring over every corner of the market? I believe these opportunities happen more often than one may think, especially if one can invest in small-cap or micro-cap land. For example, I monitor about 50 names closely in the retail industry (which I shamelessly consider to be my circle of competence). This year alone, I identified 4 separate names that fit the criteria. They respectively returned 50%, 70%, 100% and 900%.
One may counter that retail stocks have done very well in general this year and question whether throwing darts randomly would have generated similar if not better results. I would argue that much of the return (especially the out-sized ones) was hope-based, and rational investors could not have predicted those returns ex-ante with any confidence to place a big bet, as some of those names could easily turn out to be zeros. Yet in all four names I identified, I was reasonably certain of the business momentum and earning surprise, and could accordingly place out-sized bets (10%+), with confidence that even if it did not play out according to plan, I would suffer very small losses due to valuation. While hindsight is 20/20, I could also identify at least two retail names annually over the last few years that fit my "better and cheap" criteria. So they definitely occur, and one just needs to have the patience and courage to bet big when they do come along, usually when market is bad. Those situations can occur in large-cap stocks as well, such as FDX throughout this year. FDX had over $20B market cap, was followed by 25 analysts, yet the stock was at trough EV/sales, even though earnings had bottomed and was poised to recover through cost cuts and market share gains. Earning estimates have moved up 60% in 6 months and stock went up over 150%.
As with anything in investing, there are also drawbacks to my approach. One is depth vs. width - I need to be able to identify and evaluate key drivers for the companies and industries, and this takes significant amount of time. The rarity of these "perfect" situations forces me to turn over a lot of rocks. To date, I am reasonably comfortable with retail industry, and to a much lesser degree with software, asset managers and transport industries. I may soon reach (if not already) a point where I can not physically monitor more names. The other problem is scalability - most of my top ideas are in small to micro-cap land, so it is questionable whether my approach can really handle more than say $50-100M of assets. But that will be a nice problem to have, and I suspect I will just have to make the trade-off between absolute performance and AUM.
What did we do right in 2009?