Part 24 (2/2)

By midafternoon I had interviewed five men and a diminutive woman who could've been Donna Shalala's twin sister. All agreed that Underwood had not seemed depressed prior to his death. All denied knowledge of any kinky side to his personality. Three mentioned New Paradigm Systems. I hailed another cab.

New Paradigm Systems was located in a nicely landscaped office park in the Boston suburbs. The building was five stories of greenish marble and smoky gla.s.s. A young security guard at an octagonal black kiosk near the entrance asked me to sign in. I told him I had business with NPS and he directed me to the fifth floor. Large metallic letters above the entrance to the suite spelled out ”New Paradigm Systems.” Smaller letters beneath that announced the company's line of work-”Economic Consulting.”

The suite had been decorated by a professional. Gla.s.s doors led to a reception area with scarlet linen wallpaper, rich walnut paneling, and elegant French provincial furniture. A highly polished receptionist's desk was centered in front of the rear wall, but there was no receptionist. A heavy wooden door to the receptionist's left guarded the rest of the suite. I noticed security cameras in two corners. I started toward the door, but it opened before I reached it.

”I hope you haven't been waiting long,” the man said. ”I just noticed you on the monitor.” He was my age, a few inches taller, and thin. Gray slacks, blue oxford-cloth s.h.i.+rt, no tie. He had something resembling a Roman haircut, and it was short enough that I could see the tops of his ears, which tapered to rounded points. ”I know,” he said, ”I look like Spock.” I smiled to show appreciation for his self-deprecating humor.

”I just walked in the door,” I a.s.sured him.

”Good,” he said. ”What can I do for you?” He used his left foot to prevent the door from closing.

”My name is Pepper Keane,” I said. ”I'm a private investigator.” I handed him a card. ”I'd like to speak with someone about Donald Underwood.” He noticed the J.D. after my name.

”I used to practice law,” he said. ”Hated it.”

”It's an illness,” I said. ”Like gambling or alcoholism. I'm thinking of founding a twelve-step program for lawyers who want to get out of it.”

He grinned. ”That's good,” he said. ”I'll have to remember that.”

”Do you have a few minutes?”

”We really liked Don,” he said, ”so I'll be happy to talk with you, but I'm not sure I'll be much help.”

”I'm not either,” I said. ”I'm just trying to cover all the bases.”

”Russ Seifert,” he said as he extended his right hand. ”C'mon back.” He held the door for me, then led me down a hall. Considerably less had been spent decorating the suite's interior; it was modern and functional. There were workstations and offices, but most were unmanned. Much of the s.p.a.ce was occupied by a gla.s.s-enclosed, climate-controlled computer room.

Except for the stock quotes scrolling across the digital display on the wall opposite his desk, Seifert's office was much like that of any moderately successful insurance salesman. He had a nice view of the parking lot and much of the office park. One wall was dominated by framed etchings of old clipper s.h.i.+ps. His desk held some nautical trappings, including a bra.s.s barometer, and I guessed he might have a thing for sailing. He motioned for me to sit, then fell into an oxblood leather executive chair behind his desk. ”I thought the Underwood thing had been put to rest,” he said.

”The local cops and the feds have put it to rest, but-”

”The feds?”

”Let me explain,” I said. I repeated the story for the umpteenth time: Two other specialists in fractal geometry had been murdered and I was looking into the possibility that the deaths might be related.

”I had no idea,” he said.

”I'm surprised the FBI didn't interview you,” I said. ”I spoke with some of his colleagues at Harvard, and it's no secret he did work for your company.”

”He designed software for us.”

”You paid a Harvard professor to write code?”

”It's sophisticated software,” he said.

”What exactly does your firm do?” I asked. ”'Economic consulting' is a broad term.”

”In essence,” he said, ”we a.n.a.lyze data and try to predict what a market or security is going to do in the future.”

”Using principles borrowed from fractal geometry?”

”Fractal geometry, chaos theory, nonlinear statistics. It's all intertwined.”

”You're the president?” I asked. I took a card from his gold-plated holder: ”Russell J. Seifert, M.B.A., J.D., LL.M.”

”I started the company five years ago,” he said. ”I practiced law on Wall Street for eight years and decided there had to be an easier way to make money.”

”Looks like you've found one.”

”We've done well,” he said. ”When I started, it was just me. Now we have ten employees.” One of them, a bookish woman in jeans, stopped at the door immediately opposite Seifert's office, entered a code in the keypad, and stepped into what looked like a small library. ”That's Dr. Long,” he said. ”She's an economist.”

”Why all the security measures?” I asked. All the doors had keypad locks.

”It's a very secretive business,” he said.

”Why is that?”

”We're not the only ones doing this,” he explained. ”There are other firms offering this type of service. Some of the larger banks and brokerage houses have in-house teams doing exactly what we do.” I nodded.

”All of these firms,” he continued, ”have access to the same information. What matters is what you do with the information.”

”I'm not sure I follow you,” I said.

”We rely on certain theoretical ideas we have about market behavior-part of Underwood's job was to design software to implement those ideas-and those are all that distinguish us from the compet.i.tion.”

”You're afraid someone might steal your ideas?”

”It happens,” he said. ”You can't put a value on a good theoretical model.”

”What makes one model better than another?”

”Consistency,” he said. ”n.o.body can predict the future with certainty. We don't measure success by how much money we make; we measure it by how often we're right. When you provide investment advice to people managing hundreds of millions of dollars, you'd better be right more often than you are wrong.” I ma.s.saged my temples and thought about what he'd told me.

”You said you can't put a value on a good model, but I'm trying to get an idea of what one of these models would be worth. Suppose I came to you and tried to sell you on a model I'd developed. How would we arrive at a value?”

”It doesn't work that way,” he said. ”In order for me to evaluate the model, you'd have to reveal it to me and show me data to prove it works. And once you'd done that, I wouldn't need you, though I might offer you a job if I felt you'd be an a.s.set to the firm.”

”So,” I continued, ”if you're in the business of developing these theoretical models, you can't really sell them door-to-door?”

”Correct,” he said. ”You can't copyright an idea.”

”I could copyright the software,” I said.

”Yes, but the model is what has value. If I liked the model, I could buy the software from you or pay someone else to design it.”

”So if I develop the mother of all theoretical models, how do I make money on it?”

<script>