Momentum and Growth

I make no pretense to being anything more than a value manager.  It’s what I’ve done for the past fifteen years, with pretty good results.  Granted, my new methods over the past seven years attempt to incorporate industry rotation in two ways:

  • Industries where pricing power is near there nadir, such that the only direction is up, given enough time.  Strong companies in weak industries survive weak pricing cycles, and do well when the cycle turns.
  • Industries where pricing power is underdiscounted, and it will pay just to wait for future earnings to validate a higher P/E.

But no, I don’t explicitly focus on earnings growth, though I do look at forecast earnings for next year, which embeds a future ROE forecast.  I ignore growth forecasts for several reasons:

  • Growth forecasts tend to mean-revert.  Low growth companies tend to surprise on the upside, and high growth on the downside.  With a little help from pricing power, I tend to get more good surprises.
  • ROEs also tend to mean-revert.  Competition enters spaces with high ROEs and exits spaces with low ROEs.
  • It’s rare for a high growth, high P/E company to grow into its multiple.
  • Low growth, low P/E companies can be treated like high-yield bonds.  A P/E of 10 implies an earnings yield of 10%; I may not capture all of that 10% in dividends and buybacks, but a modestly good management team will find ways to deploy excess cash into other organic growth opportunities which will grow earnings in the future.  With a little good management, I can see my company with a P/E of 10 grow its intrinsic value by more than 10% in a year.

As for momentum, my rule of thumb is that momentum persists in the short run, and mean-reverts in the intermediate term.  I have to size my trading to the rest of my strategies.  Value emerges over the intermediate term, not rapidly.  The same tends to be true of industry rotation; it works with a lag, but it works.  I have to become like Marty Whitman at that point and say that often the fundamentals and price action are lousy when I buy, and for me that’s fine, because:

  • I focus on balance sheet quality,
  • Accounting integrity, and
  • Cheapness.
  • I have my rebalancing discipline standing behind me, which often has me buy more before the turn occurs.
  • I also stay reasonably well-diversified.

The turns usually do occur.  I never make a ton of money on any trade, but typically 80% of my trades make money. And, my losses are typically small, so this method works well for me.

Anyway, that’s why I embrace negative momentum and don’t explicitly embrace growth.  It can place me in the “caricature” camp for value managers, because my valuation metrics are usually lower than most.  Given my longer holding period, I’m fine with that, because low valuations tend to produce their own catalysts for change, if one has done reasonable research on the shareholder-friendliness of the corporation, and the strength of its financials.  Besides, as intrinsic value grows with companies having low valuations, there is a strong tendency for the stock to rally.  Think of PartnerRe, which has never had a high valuation; as it puts up good earnings year after year, the price of the stock keeps running.  Just another example of an underdiscounted trend in the markets.

Full disclosure: long PRE

1 Comment

  • Josh Stern says:

    Thanks for the quick reply! You write “Industries where pricing power is near there nadir, such that the only direction is up, given enough time.” Okay, that is a kind of industry based growth forecast for earnings. But I’m unclear on what aspect of your methodology suggests that it selects candidates where the industry is near its nadir, as opposed to early innings of a long decline or else headed towards obsolescence? Do you make those determinations subjectively, or is the idea that being contrarian on industries by using relative strength over some given trailing time period (which one?) has been statistically predictive of superior future performance in the past, and that relationship is likely to continue going forward?

    You noted that “I don’t explicitly focus on earnings growth, though I do look at forecast earnings for next year, which embeds a future ROE forecast,” but my understanding was that you ranked earnings yield for next year (or similar) and not any measure of change in earnings between current year and next. Is that right? If so, then that is not capturing change in ROE either.

    Your comment that “I never make a ton of money on any trade, but typically 80% of my trades make money. And, my losses are typically small, so this method works well for me.” Keeping losses small is definitely important. Successful technical traders frequently emphasize cutting loses and letting winners run, and often report high gains even with sub-.500 winning percentage. I’m not sure what my winning percentage is, but I tend to do something similar wrt valuation instead of performance, and often find that my annual gains are dominated by a relatively small number of big winners.

    I’d like to add that I have a lot of respect for the significance and difficulty of trying to do objective modeling – for stock performance prediction and other domains.