An In-Depth Look into Mark Hirschey's Book "Tech Stock Valuation" by Victor Niederhoffer
It is a pleasure to recommend the book Tech Stock Valuation by Mark Hirschey as one of the best books on investment of all time and a worthy successor and follow-up read to the Dimson, Marsh and Staunton book, The Triumph of the Optimists. Hirschey's book contains 10 chapters:
In an interview with Mark Hirschey, he told the Chair the lessons he'd like the reader to come away with after reading the book are:
The interview started with the following exchange:
Chair: "What is the foundation for your work in this field?"
Hirschey: "That knowledge of microeconomics is the key to success in investments. That things like barriers to entry, the structure of an industry, the stages, the cross elasticities, competition, pricing practices, are the key variables. It's amazing that economists like Stigler, Caves, Porter, haven't applied their work more directly to investments. However, the greatest micro-economist of all time is.. . Warren Buffett." (Every year, Hirschey leads a pilgrimage of his students to the Berkshire annual meeting in Nebraska as a practical tribute to the Sage.)
Those who know the Chair's thoughts on this subject will realize that Hirschey's views are not geared to garnering a favorable review, but it's a mark of the book's excellence that despite Hirschey's completely erroneous and harmful views on this subject the Chair can still recommend the book, along with the Aswath Damodaran's Investment Fables: Exposing the Myths of "Can't Miss" Investment Strategies as among the best of the last five years.
The problem with most investment books is that they are either written by practitioners who don't have the scholarly background to support their points with analytical principles and scientific inquiry, or scholars who are so far behind the form or out of line with reasonable applications that the book belongs in an ivory tower. The former are motivated in the main by a desire to promote their current or future activities in the field, and the latter by an effort to cement their positions, or memorialize their previous credentials.
Hirschey gets around this because he is a practitioner. He is the kind of researcher who likes to go through Value Line page by page to find great undervalued stocks that have been hit by a one-off event and are ready to bounce back when the problem is solved. He has obviously lived and lost and made a fortune in the tech bubble, and is wealthy enough and has written enough (his book on Managerial Economics is a classic, already in its 10th edition) and is established enough as a professor at the University of Kansas that he has no need to impress.
The heart of the book's argument that tech is ready to rebound is contained in Chapter 5. The book was published in 2003 and was finished in the middle of 2002 when the Nasdaq 100 was hovering at the 1000 level, down from 5000 in April 2000, amid calls for Nasdaq 300 by the chronic bears, the cults led by the Weekly Financial Columnist and by Hirschey's mentor. The current level of 1500 vindicates Hirschey's main point.
However, the thrust of the argument is a statistical one -- that annual returns in the major averages tend to be negatively correlated. That since the Nasdaq was down 50% by the end of 2001, an imminent reversion to the mean was in the cards. The problem with the author's evidence on this score is that it is highly dependent on the time period and methodology chosen. The author carelessly glosses over such things as the fact that it is guaranteed to happen that after "a sustained market correction" a bull market is likely (in retrospect), and after the 10 worst markets it is likely that no subsequent months will be among the worst. His use of cumulative overlapping 12 month returns (see chart below) is also guaranteed to lead to the appearance of spurious cycles as Slutsky and Yule have pointed out in their academic papers on spurious cycles in moving averages, and all technicians using oscillators and stochastics find out in practice.
He reports as statistically significant a negative 10% correlation in annual returns for Nasdaq from 1971 to 2002 but with only 30 observations the standard error of the correlation coefficient is approximately 2.5 times as great as the retrospective time sensitive correlation he observes. (Indeed an investor who bought Nasdaq on its negative path from 5000 to 1000 based on negative correlation would have found himself a dead duck)
Hirschey's argument for the the rebound in Nasdaq is on much firmer ground with the foundation he lays from microeconomics, with such factors as: stocks in industries with high barriers to entry can maintain higher returns than those with few barriers, that technology and patents make for great barriers, that the major technology companies are becoming like the phone companies of the past as computers become used for communications as well as computing, and that because of the risk involved with constantly changing technology tech investors are graced with higher returns.
The chapters on how investments can be geared to balance sheet analysis are refined and resilient. Particularly interesting is the post-announcement effect on goodwill write-off, where negative moves of 11 % after the announcement continue over the next year. The results on corporate governance and enforcement actions by the SEC are not as clear-cut, as most of the effect occurs during the announcement period, and the effects after the announcement are marred by multiple comparisons.
The Shark Repellent chapter, while very interesting, is also not as useful. The major conclusion is that companies adopting them are good companies with better than average performance measures. Similarly, the chapters on research and development and patent quality while highly suggestive from a barrier to entry economic analysis standpoint but are too granular for any practical investment conclusions to be drawn.
Some of the best chapters in the book involve a description of the woes of the Japanese Economy and AOL, and the excesses of the Tulipmania. The anecdotes here are interesting but the problem is that it's hard to find a bubble prospectively. And despite his poor results from shorting, Hirschey does not draw the conclusion that one should never short stocks as the inevitable upward tidal movement over time, and the irrational heights that bullish sentiment can reach, are too difficult to overcome.
The book raises so many interesting points, the analytical framework is so good, and the presentation of the data is so clear-cut, that despite my reservations about much of the methodology, and many of the conclusions, there are golden nuggets available, lines of inquiries to pursue, and potential investments to make for all readers. This book definitely belongs on every investors' bookshelf.
Mark Hirschey adds: I greatly appreciate your kind comments on my Tech Stock Valuation book. Still, your comments make me think I could be clearer in making my point.
In Ch. 5 of my book, I try to distinguish between the familiar statistical "regression to the mean" concept, and what I call "return reversal." They are distinctly different. The return reversal concept is based upon economic theory concerning the mean reversion in business profits over time, and the overreaction hypothesis, which is based on investor psychology.
1. For both companies and industries, expansion and contraction occurs based upon the relationship between the internal rate of return on investment and the marginal cost of capital. Capital expenditures rise when the internal rate of return on investment exceeds the marginal cost of capital. Capital expenditures fall when the internal rate of return on investment is less than the marginal cost of capital. At any point in time, firm and industry profit rates vary widely. Over time, however, expansion and contraction cause these profit rates to converge toward the overall average annual rate of return on invested capital. During the twentieth century, the overall average annual rate of return on invested capital has averaged roughly 10 per cent per year. During the Post World War II period, the overall average annual rate of return on invested capital has averaged roughly 12-14 per cent per year.
2. Experienced and rational investors know that competitor entry and growth in highly profitable industries with minimal barriers to entry causes above-normal profits to regress toward the mean. Conversely, bankruptcy and exit allow the below-normal profits of depressed industries to rise toward the mean. However, inexperienced and emotional investors appear to extrapolate recently good or recently bad industry performance into the future. This causes positive overreaction and stock prices that are "too high" in periods following robust firm revenue and earnings performance, and negative overreaction and stock prices that are "too low" in periods following poor revenue and earnings performance
3. For some investors, especially those with a strong background in statistics, the idea of return reversal in market returns might be misinterpreted as a simple "regression to the mean." However, the statistical regression to the mean concept fails to explain return reversals in the S&P 500, where returns following vicious bear markets substantially exceed long-term averages rather than regress toward long-term market norms. The regression to the mean concept also fails to explain return reversals in Nasdaq following vicious bear markets and boisterous bull markets.
In short, the statistical regression to the mean concept predicts normal stock-market returns in periods following either abnormally good or abnormally bad performance. The return reversal concept, based on economic theory and investor psychology, predicts abnormally good long-term (12-month) returns following long (12-month) periods of abnormally bad stock-market performance. The return reversal concept also predicts abnormally bad long-term (12-month) returns after long (12-month) periods of abnormally good performance.
Nobody should be surprised about the lack of negative serial correlation in daily, weekly or monthly returns. That's just noise. For the effects of investor psychology to come into play, longer time frames must come into play. Bull markets have height and duration. Bear markets have depth and duration. In a bear market, like 2000-02, investors eventually "puke out" their positions when the bad news never seems to stop. You could say that I looked at 12-month returns to get a simple and convenient handle on return reversal over an arbitrary "long" period. However, many investment managers are paid on an annual basis, and my results suggest that 12-months may be as much indigestion as the pros can stand before "puking" out their losing positions.
I was predicting abnormally good tech stock performance in the period following 2002, not merely normal market returns.
I hope this clears up my point.
Vic asks Professor Hirschey:
I wonder if your worship of your mentor extends to his analysis of balance of payments accounting and his reasons for being short the dollar, as well as his avoidance of individual stocks?
Professor Hirschey responds:
I also enjoyed your comments about Buffett. My college-age daughter Jessica jokingly refers to Buffett as my hero. I've followed him since 1969, hold lots of BRK.A (but less than Jimmy Buffett does), and carefully consider everything he has to say. He's a really smart guy, and I do find him interesting, but the "hero" idea seems a stretch.
He's had an odd personal life, living in Omaha for years with a woman that his wife (who lived in San Francisco) introduced him to. Buffett has also had a much less than ideal relationship with his kids, which put his current comments about the importance of family in an interesting light. He's older and perhaps wiser now, and trying to make amends, I guess.
Buffett has also made lots of arbitrage investments that do not square with his public persona to buy and hold quality companies. Buying into the airlines (US Air) and the investment bankers (Solomon) with convertible bonds were mistakes, despite the advantages of fixed income instruments for an insurance company. Buying shoe companies was also a (now admitted) mistake; I think he may also come to regret taking on a big position in furniture retailing, a business that is now morphing into very risky electronics retailing. Power utility stocks look like a sensible way to 10% returns going forward, but they are no path to big returns, that's for sure. Buffett's currency bet is a big departure for Berkshire shareholders. I also think it's a bad bet. I could be wrong, of course.
IMHO, Buffett is very, very good. He's also been very lucky at key points (Washington Post, Cap Cities, Buffalo News). He's been very good and very lucky. His $43.2 billion payoff is at the way upper tail of a reasonable distribution of what might be expected. I've attached a spreadsheet that might give some perspective on Buffett's performance. As $43.2 billion, Buffett's results are such an incredible outlier that they just shouldn't be able to occur (you really couldn't flip a coin and get heads that many times). The only way I could get any kind of odds on it was to set up the simulation assuming that you have the best of the best manage a portfolio with an average return of 15% with a standard deviation of 20% and let them take over in 1966 with $25 million. If you would have done that you would have had a 3.74% chance of ending with $43 billion. The catch is, of course, that Buffett earned a 29.5% annual return for 13 years to get to $25 million in 1966.
Essentially, most mortals have little chance of becoming the next Warren Buffett. Not me, for sure. I'm working real hard, and I'm not worth 1/10,000th as much.