May 10, 1999
Interview with Legg Mason Focus Trust Manager
Robert G. Hagstrom
Recently, Dale Wettlaufer (DW) and Yi-Hsin Chang (YC) traveled to Wayne, Pennsylvania, to talk with Legg Mason Focus Trust portfolio manager Robert G. Hagstrom. Hagstrom, the author of the best-selling The Warren Buffett Way, was ahead of the curve in finding great value in NASCAR and in writing The NASCAR Way, and has recently written The Warren Buffett Portfolio.
In his latest book, Hagstrom looks at investing the Warren Buffett way from a multidisciplinary viewpoint, challenges long-held academic definitions of risk in portfolio management, and even takes on the mutual fund industry's approach to the way the world works. We will present Yi-Hsin and Dale's talk with Robert over the next few days, beginning today with what Hagstrom tried to accomplish in his new book.
DW: Robert, what did you want to achieve with The Warren Buffett Portfolio?
Hagstrom: If you look at The Warren Buffett Way in 1994, it was a very different time in the market. First of all, we came off that 1991 recession, markets were sloppy, nobody was making double-digit returns, much less triple-digit returns. It was a very difficult time. It was the first book that came out that kind of went into the case study analysis, and so when it hit, it really sparked quickly.
Since then, we've had about five books on Buffett. Roger Lowenstein did a great job [with Buffett: The Making of an American Capitalist], Andy Kilpatrick I still think does a great job with keeping up with the history [with biannual editions of Of Permanent Value: The Story of Warren Buffett], and then you had the Buffettology book and Buffett Speaks [Warren Buffett Speaks: The Wit and Wisdom from the World's Greatest Investor], and then Cunningham came out with the Cordozo Law Review compilation. So there's been a lot of stuff in the last three years, so we're reaching saturation on Buffett.
The second thing is, the market has been so powerfully strong that I think people are not looking for solutions because it's been rather easy. Bull markets typically make it easy. If you can just catch the trends wherever they are, you can generally make money. Also, technology is such a big part of the investment return, and Buffett is very well known as not having a high degree of participation there. So the idea [with investors in general] may be, "there's nothing here, I don't need help, I'm making money, technology's where it's at, and I already know that Warren Buffett's not there," so it's a different time today than it was in 1994.
I wrote the book for two reasons. One was very self-serving, because of Legg Mason Focus Trust, and the other was, I think that really the literature had not discussed what I thought was a really important part of the Warren Buffett process, which is now that we've got the stock selection process down, how do we make big-bet portfolios? The Buffett literature that was out never got into portfolio management too much. It's [presented] as kind of buy-and-hold, but it wasn't why you put 20-25% into something and what can happen when you do that. I thought this was an area that was completely underserved, but I was convinced it was also an area that explained how he got excess rates of return. You have to pick the right stocks, no doubt about it, but if he had bought 50 or 100 things, I'm not sure he would have gotten the returns that he got. Clearly, making big bets on high probability events was driving the high excess returns in the portfolio.
We figured that out when we started doing this research for AIMR [Association for Investment Research and Management] with Joan Lamm-Tennant, who was over at Villanova University and now works for General Re. The problem that I ran into when I first mentioned Legg Mason Focus Trust in 1995 was that there was a great deal of skepticism about, "did we know what we were doing and is this the right way to do it?" We didn't have that many data points to look at. You have Buffett, Charlie Munger, [GEICO chief investment officer] Lou Simpson, and [Sequoia Fund's] Bill Ruane. You only had four or five data points to look at.
Most of the people we interacted with said, "Well, that could all be intellectual fruit from the same tree," and they were really concerned that there weren't that many focus funds out there. If it was such a great idea, why hadn't Wall Street imitated it? So the research we were doing for AIMR and that we did with Lamm-Tennant was meant to try to give us some academic bearing on why concentrated portfolios have the potential to outperform broadly diversified portfolios.
The research paradigm was random selection. We had a database of 1,200 companies and the computer randomly selected 3,000 observations. According to the statistical people, once you get 3,000 observations, you've washed out enough noise that you can make some kind of conclusions. You can at least draw some conclusions. So when we did that first research and looked at the distributions of different portfolio sizes over a 10-year period, it was striking. It just jumps out at you immediately that, as you reduce the number of stocks in your portfolio, you begin to measurably increase the probability of generating high excess returns. The problem is you also increase the probability of generating very negative returns.
That was the first insight. The second insight was, if you had a 100-250 stock portfolio, the probabilities of your generating high excess returns relative to the market� the probabilities just weren't there for you. You were in the 1 in 3,000 or 5 of 3,000 that could get it done. We didn't adjust, and we're going to do some of this in a [yet-to-be-published] monograph, for market impact cost, expense ratios, and that sort of thing. According to Fidelity, at an 80% to 100% turnover ratio, you can get market impact costs of two percentage points in the portfolio, just the buying and selling action. So you're two points behind there. You have another point or so in expense ratio and you're now three points behind.
DW: And that's not even counting taxes.
Hagstrom: Absolutely. Already, you're three points behind before taxes and you've got a 100-200 stock portfolio, which statistically isn't going to be one of the portfolios that can get you ahead of the market. You kind of recycle back and look at the fact that since 1990, 97% of the money managers haven't been able to outperform the market. I'm beginning to think the reason why is that we're structured all wrong. Did you read Peter Bernstein's article on "Where Are the .400 hitters?"
It was in the last chapter of The Warren Buffett Portfolio. Peter came about it in a very interesting way, which is that perhaps the competition has gotten so intense that we can't get high excess returns. When we went through Peter's paper, it seemed to me that he was absolutely right structurally, but he left the back door open. The back door was that you can get this done if you're willing to assume high standard deviations. He said that if you're willing to accept that, you probably can be a .400 hitter. Well, when we went back and looked at Buffett, Munger, and Ruane, all of them were high standard deviations relative to the market.
Very bouncy roads, along the lines of Warren Buffett's [preference for] a bumpy 15% compound annual return over the long term rather than a smooth 12% long-term return. I just basically started to put this stuff together and said perhaps the reason why we're not performing is that we have too many stocks, we're too concerned about the smooth ride, and we're just not going to get there from here. I mean, you just cannot generate those high excess returns. So if we're paid to beat the market and clients are expecting that... it seems to me that this is the smart way in which to do it.
The book was written to explain that. One, we under-emphasized that in The Warren Buffett Way. And two, we wanted to really drive home the point that, you first have to pick the stocks and secondly, you've got to put them together in a portfolio. The thing that really kind of took me aback after I wrote The Warren Buffett Way was that it seemed like everyone had "Warren Buffett conversation" down: "We only buy the right businesses, we want good managers, we're looking for high returns on capital, good profit margins, cash flow and earnings, and we always buy them for less than they're worth." That's the Warren Buffett Way. Then you find out about the portfolio they manage; it's got 140 stocks, turnover ratio of 90%. They got the stock selection part down, but not the portfolio management aspect. That was part of the driving force in writing the book.
When we started Focus Trust, there was a lot of skepticism about this. "Are you sure 10-12 stock portfolios are the way to go?" we were asked. "Sure, Warren's doing it." And they would respond "Well, that's right for Warren, but what do you know about it?" So the book was written to flesh out the academic defense for doing this. And that was the first half of the book.
DW: Do you feel good about what you accomplished?
Hagstrom: I think so. Mike Mauboussin at CS First Boston was helpful with the book, went through it and said "yeah, this looks good." [Legg Mason Value Trust portfolio manager] Bill Miller went through the book with me and read every chapter and went through it all and said "yeah, makes sense to me." Joan Lamm-Tennant is a Ph.D. and is statistically competent, writes papers for the CFA group, and they said it's fine. Keith Brown down at AIMR said "yeah, looks good to me." So we've got all these people who are saying the research design looks fine, so we think we went about it correctly. Now the big question begs itself: "If it's correct, why aren't we [the mutual fund industry] doing it?" I've got guesses why, but I really don't know why we don't do more of this. Why aren't there another 200 or 500 focus funds out there?