What’s Jeff Gundlach Thinking?

Our interview took place prior to the announcement in mid-September that Gundlach’s home had been burglarized of 13 pieces of valuable art, in addition to wine, wrist watches and cash. Thieves also took Gundlach’s red 2010 Porsche Carrera 4S. In a dramatic and chaotic press conference in downtown Los Angeles two weeks later, the star fund manager increased the reward offered for the return of his stolen property from an initial $200,000 to $1.7 million. We’ll continue to report new information as it becomes available.

Sure, Jeff Gundlach is given to loud suits, fast cars and quoting Shakespeare. He also has assets under management and a performance record that are the envy of the money management industry. Call him eccentric if you like, but you’d have to follow it with another accompanying superlative (Hint: It rhymes with “zenius”).

The CEO and CIO of DoubleLine Capital is the “it” manager of the mutual fund space, a title he’s enjoyed since being named 2006 Fixed Income Manager of the Year by Morningstar. At the time he was CIO and head of fixed income activities for TCW. Barely three years later he was named as a finalist for Fixed Income Manager of the Decade—a title that eventually went to Bill Gross—when, in a move that shocked the industry, he was summarily fired from TCW (makes you wonder about those stewardship grades). His acrimonious exit led to lawsuits and counter lawsuits, the salacious details of which won’t be repeated here. But Gundlach’s winning ways meant he somehow “lost” the case by being awarded $66.7 million.

If living well is the best revenge then Gundlach is the fund industry’s Juvenal. Since starting DoubleLine in 2010, when the ashes of the TCW debacle were still hot, he’s amassed $43 billion in assets under management ($45 billion by the time this interview is published, he confidently told us). As we said, the envy of the industry.

The star manager status he’s attracted is something we’ve witnessed firsthand. Following his presentation at a conference in early April, Gundlach held forth in the lobby as advisor after advisor—oblivious to or uninterested in the subsequent presenters on stage—asked for his insight for the better part of an hour. It was only after his PR assistant deftly informed him of his next appointment that he was able to extricate himself. As we headed off for our scheduled interview, one lone advisor followed to thank Gundlach for all he does and inform him of the millions the advisor’s firm had just invested with DoubleLine.

Funny thing about Gundlach; he’ll make you his Passepartout for a trip around the investing world in 80 minutes. A person can’t be an expert on everything, but in Gundlach’s case, the truism is wrong, or so it seems. We posited one question and got an earful of master’s level information. In the process, he answered four or five other questions we’d prepared without ever being asked.

We began with his criticism of model-driven investing, our raison d’être for the latest interview, before getting off track. But with a not-insignificant percentage of money managers making a very good living from the strategy, it seemed as good a place as any to start.

“It extrapolates the past into the future,” Gundlach said. “Models by their very nature use statistical regression at their center. The thesis underneath it is that the way things behaved in the past will be predictive of how they will act in the future. What’s wrong with it is that the future doesn’t resemble the past, particularly when there is technological change or innovation.”

The example he cited was the modeling of homeowner behavior. An important question in the world of fixed income and mortgage-related securities is, “When and to what degree will these people refinance?”

“When I started in the business, there was a general rule almost set in stone on Wall Street,” he said. “You would never see higher than a 20% annualized rate of refinancing. Then comes 2003, and that rule no longer works. In that year, 70% of all homeowners in America refinanced their mortgages.

“Obviously, 70% is way higher than 20%,” he wryly added.

Despite surrounding himself with world-class research and the industry’s best and brightest, it was nonetheless his own intuition that alerted him to the danger.

“I got home from work one evening, and there was junk mail from a mortgage originator that described the interest-only loan,” he recalled. “The models that had been developed in those days to predict homeowner behavior, not surprisingly, did not have this type of loan factored into their forward-looking projections for 2004 and 2005. They were operating under the false assumption that the variables and coefficients from the past would stay the same.”

It got so bad, he recounted, that one month’s refinancing results were 250% higher than the highest forecast from the month prior, something he found “very telling.”

“If there’s one thing you would have to believe in if you’re going to believe in model-based investing, it’s that you should be able to predict next month’s prepayment behavior,” he explained. “You know where interest rates are, you know where the refinancing index is, you know what housing starts are, you know what existing home sales have been and you have lots and lots of data, and yet the error was 250%. Model-based investing takes that type of embedded assumption error and multiplies it.”

Easy to criticize, but what does DoubleLine do that’s better?

“We’re selling water in the desert,” Gundlach quipped. “And we have some uniqueness to what we do. We have a product that’s in demand and cash-flow-based. It generates more than competitive returns.”

Color us skeptical, but from the headlines he’s generated and the enthusiasm we’ve witnessed, his “star manager” image has to be a factor. Whether it’s false modesty or he truly believes it’s inconsequential, he wasn’t buying that.

“Our clientele really is independent-minded; they think for themselves and really understand the way in which we think about risk, which is carefully,” he diplomatically answered. “We have a huge following in the very high-net-worth category, and that’s where all the money is. We have the luxury of having our maximum appeal in the demographic that has all the money.”

That’s all well and good, but he still hadn’t told us, specifically, what he does better than model-based managers (or anyone else, for that matter). In a case of “be careful what we wish for,” when we repeated the question, he was more than happy to oblige.