Tom Gardner wrote today's Rule Breaker column: http://www.fool.com/portfolios/rulemaker/2000/rulemaker000602.htmPast performance may not predict anything, but it does correlate with future success. On average, in a free-market system, isn't it true that the longer the run of triumph, the more cash a great company will store up for investments in superior future results?I don't think you can say that past performance correlates with future success. While it's true that a long run of triumph sometimes allows accumulation of a substantial amount of cash that can be used for investments, there is no evidence whatsoever that those investments perform superiorly. Past performance does not correlate with future performance, except for horizons shorter than a year. In that spirit, think about Rule-Making filmmaker Steven Spielberg. Past success doesn't guarantee that his next five movies will be blockbusters. But reputation, his level of experience, the loyalty of his creative teams, and his access to capital make it an odds on favorite.I think this example is somewhat flawed. Clearly Spielberg's reputation makes him a favorite in terms of future being a blockbuster, i.e. selling many tickets for big revenue. Odds are for that. However, his abilities and reputation do not guarantee that his future movies will be business successes. While his future movies likely will rank high in terms of revenue, they will also be among the most expensive movies produced. Certainly, because Spielberg has been so successful in the past his honorarium is humongous. Therefore, by hiring Spielberg to produce your next movie you are paying a high price for high expected revenue, but there is no telling whether the odds of your movie becoming a business success are best with or without Spielberg.And that's what we're looking to own in this portfolio. We're buying Secretariat as he storms down the stretch at the Derby. We're investing in the final three seasons of Seinfeld. Again, you are paying very high prices. Furthermore, you only mentioned your successes here. You also end up paying high prices for Mike Tyson's before his last match and the final three seasons of Melrose Place. If you could split your portfolio ex post into the best performing and the worst performing stocks then clearly the best performing stocks will have performed well. That doesn't validate your strategy though.And in corporate America, that odds-on-favoriteness, borne out of stellar past performance and strong present standing, leads naturally to the discounting forward of future values. The price goes up because the probabilities of success have gone way up.Yes, at the point in time you buy them the companies are already very expensive. Therefore, even if their odds of future business success are good, we can't say that our odds of good returns have increased.As investors in Rule Makers, we embrace that circumstance. We'd rather pay more for greatness -- watching it compound for years and years -- than get a fair price for mere competence, or score mediocrity on the cheap, neither of which is a great vehicle for compounding.You seem to base your arguments on the assumption that you can pick great companies that outperform in the future. Assuming that I can pick future winners that's certainly what I would do rather than picking mediocre companies that underperform.There are a wide variety of risks tied to stock-market investors, but virtually no existing taxonomic work to differentiate one from the other. For example, business schools and mutual fund companies have been suggesting for years that beta -- the volatility of a stock price -- is a measure of risk. It is, but only for traders. It isn't for investors. While America Online (NYSE: AOL) rose 150 times in value over the last ten years, did it matter how ziggy the price zigs were or zaggy the zags were along the way? But to an investor, the price movements were not risky. Risk was tied to the company's level of performance and its valuation, not its temporary price movements.These statements seem to rely on a misunderstanding about the concept of risk. Surely price movements were risky to an investor as they reflect new information being embedded into prices about how successful the company would become and the likelihood of bankruptcy. Risk is really an ex ante proposition, i.e. given everything we know today what risks will the company face in the future. That AOL ends up exceeding all expectations ex post has nothing to do with ex ante risk. Many companies that were as risky as AOL ten years ago no longer exist, i.e. they have realized the worst possible outcomes ex post.Now, the appealing feature about risk is that it's somewhat stable over time. For example, a historically high risk company like AOL by nature of its business also faces high risk in the future. The broadband transition is one example. Therefore, a risk measure like beta (if you like that one) estimated on historical data will likely be useful in evaluating risk for AOL going forward. Interestingly, volatility is way more persistent than return. Therefore, buying the best past performers will likely not lead to future outperformance while buying historically risky company will likely lead to future risk.the case of Rule Maker investing, I'd like to distinguish between valuation risk and performance risk. The former is the risk of paying too much for a company. The latter is the risk of not getting what you paid for. This again relies on the supposed link between past performance and future. As argued above, while a company that has done well in the past likely will do well in the future, it's already expensive enough to render expected future stock performance average. In other words, any performance risk that I can think of should already be build into the valuation step. There's no separation.I herby encourage the valuation superstar out there who can use all the available data (there's tons of it) to show me the valuation methodology that: in 1980, proves Wal-Mart (NYSE: WMT) will be worth $260 billion today in 1990, proves Microsoft will be worth $350 billion today in 1994, proves Cisco will be worth $450 billion today I'm not your valuation superstar but I do know that valuation is an ex ante concept. That is, given everything we think will happen in the future, and the expected risks, what's the stock worth today. Ex post, of course, we can hope that on average the past valuation was correct. However, actual ex post realized market caps will display way more variation than ex ante predicted. In that distribution, a few companies will turn out to do extremely poorly and no longer exist, while a few will have performed extremely well. You are asking us to go back and ex ante value the three companies with perhaps the best realized ex post performance. This really has nothing to do with valuation. The only reasonable question you can ask is whether we can go back in time and correctly value all stocks on average relative to what we have realized today. Even then, you have to be very careful in accounting for any unanticipated new information that have reached the market in the meantime.If our Rule Maker companies are required to justify their valuations based on Wall Street earnings estimates, well, we're in trouble. But if our companies -- more than slips of tradable paper -- are to be evaluated as businesses, we should use cash flows and the Flow Ratio. When we do, I think right away it becomes clear that we'll need considerably less earnings growth than Bill does to justify today's valuations.I don't believe in historical earnings, cash flows, or the flow ratio as valuation tools. I also don't recall any serious empirical evidence that cash flows predict future stock returns better than earnings. Moreover, I have never seen any evidence on the effectiveness of the flow ratio. Do you any credible evidence that for example the flow ratio has predictive power for future returns? Anticipating that you don't have the evidence, how can you justify the strong statements you make in today's rule maker column?Datasnooper.
Best Of |
Favorites & Replies |
Start a New Board |
My Fool |
BATS data provided in real-time. NYSE, NASDAQ and NYSEMKT data delayed 15 minutes.
Real-Time prices provided by BATS. Market data provided by Interactive Data.
Company fundamental data provided by Morningstar. Earnings Estimates, Analyst Ra