IN terms of intellect, Ben S. Bernanke may be to the Federal Reserve what John G. Roberts Jr. is to the Supreme Court. And like Chief Justice Roberts, Mr. Bernanke, the nominee to replace Alan Greenspan at the Fed, has left a paper trail worth studying. What can it tell us about the sort of Fed chairman he would be?

In general, Mr. Bernanke's work has been solidly in the mainstream - a mainstream he has helped define since he began publishing papers in major economic journals since 1981. He has written repeatedly about ways of using mathematical models of a dauntingly complex economy to set monetary policy. When he has strayed from that subject, his conclusions have sometimes raised eyebrows.

For example, in 2001 he joined Refet S. Gurkaynak of Bilkent University in Ankara, Turkey, in urging colleagues to adopt the sort of savings-driven models of the economy that the White House used to justify its tax cuts. David H. Romer, a professor of economics at the University of California, Berkeley, whose earlier work was discussed in the paper, said the statistical support for their argument was unconvincing.

In 1999, Mr. Bernanke and Mark Gertler, chairman of the economics department at New York University, wrote that financially fragile countries should not adopt fixed exchange rates, because they had been associated with crises in the past - a logic that other economists have disputed.

"Most of my American colleagues seem to go in this direction, but it's not a widely shared view around the world, and especially not in developing countries," said Charles Wyplosz, a professor of economics at the University of Geneva's Graduate Institute of International Studies. "The experience with floating interest rates is that they tend to float too much, so some sort of harnessing of the exchange-rate movements is useful for small, open economies."

These topics, however, are not at the core of what Mr. Bernanke would be concerned with at the Fed. There, his opinions about domestic monetary policy would be more important. One tenet of Mr. Bernanke's philosophy could not be clearer: that the central bank should use a model, not just hunches, to decide about interest rates and the money supply.

This is how he put it in 1997 in a paper with Michael Woodford, now a professor of political economy at Columbia: "We conclude that, although private-sector forecasts may contain information useful to the central bank, ultimately the monetary authorities must rely on an explicit structural model of the economy to guide their policy decisions."

Mr. Bernanke has examined exactly what data a central bank should use to calibrate its models. In 2003, he and Jean Boivin, an associate professor at the Columbia Business School, wrote that there was little advantage in waiting for "final" figures for government statistics, and that preliminary figures worked just as well for economic forecasts. The types of data used by the Fed could also change, said Anil K. Kashyap, a professor of economics and finance at the University of Chicago's business school.

"Greenspan was famous for looking at all these strange things like boxcar shipments," Professor Kashyap said, adding that Mr. Bernanke might prefer "indicators motivated by theoretical considerations," like banks' loan commitments.

Yet even in the "data-rich environment" that Mr. Bernanke and Professor Boivin described, models don't always encompass every possible outcome. Though models can explain various facets of the economy's behavior, even several at a time, no one has come up with a single formula that explains virtually everything.

To wit: a recent theme of Mr. Bernanke's work, one that he expounded as a Fed governor, is that the central bank should not try to prick bubbles in asset prices - mostly because it's too hard to tell when a bubble is really present. In that 1999 paper, he and Professor Gertler supplied a model to show how monetary policy should react to changes in asset prices.

Rudiger Dornbusch, the late professor of economics at the Massachusetts Institute of Technology, said the model lacked one important feature. "When it comes to monetary policy and asset price volatility," he wrote, "the interesting issue is not the gentle part of the trip but rather when it crashes." At that point, credit and liquidity can dry up.

"Neither of these considerations has a place in the Bernanke-Gertler model, which is just price-based and lacks rationing and liquidity," Professor Dornbusch concluded. The absence of liquidity and the onset of credit rationing, he argued, can be extremely important factors in forming policy at times of crisis.

Interestingly, Mr. Bernanke and Professor Gertler wrote a few years earlier that monetary policy might be most powerful during periods of tight credit, because of its effects on banks' balance sheets. But the periods of tightest credit - usually just before, during and after a crash in the markets - are the ones that are most difficult to model.

That is not to say that Mr. Bernanke would shy away from radical action in a crash. In an early version of a paper with Professor Gertler that was eventually published in 1990, they wrote that "under some circumstance, government 'bailouts' of insolvent debtors may be a reasonable alternative in periods of extreme financial fragility."

PROFESSOR ROMER at Berkeley said he believed that Mr. Bernanke would know when to discard his models. "He of course understands that even in normal times, the best model is just a guide," Professor Romer said. "If something extraordinary happens, like either Russia goes under or the stock market goes down by 20 percent, anyone with a modicum of common sense knows that the model's not going to be a reliable guide."

Despite their differences, Professor Romer gave Mr. Bernanke high marks as a potential Fed chairman. The professor's view carries some weight; in 2004, he and his wife, Christina D. Romer, also a professor of economics at Berkeley, published a paper entitled "Choosing the Federal Reserve Chair: Lessons From History."

"It's not your political abilities or who you know," David Romer said. "What does matter is your understanding of the economy and the effects of monetary policy. By that standard, Ben is the best person you could choose, basically."