(MONEY Magazine) – We admit we had some doubts whether Eliot Spitzer's campaign to clean up Wall Street research would produce better returns for investors. After all, just because research is unconflicted doesn't mean it's any good. "It could be wrong for different reasons," Spitzer, New York State's attorney general, readily admits.

Well, the early returns are in, and it appears our doubts were misplaced. A study conducted for MONEY shows that analysts' stock-picking prowess has improved markedly since Spitzer began his crusade to end Wall Street's practice of using inflated stock recommendations to woo lucrative investment banking clients.

We asked analyst tracker Starmine to construct model portfolios for 3,000 Wall Street analysts--buying their buys and shorting their sells--and compare the performance of those portfolios with indexes for the industries the analysts cover. Result: The average Wall Street analyst outperformed his or her industry benchmark by 3.2 percentage points during the 12 months that ended Feb. 29, and by 2.2 percentage points during 2003. By comparison, analysts trailed their benchmarks by an average of 0.1 percentage points in 2002.

Two or three points may not sound like a lot, but it's one grand canyon by Wall Street standards. "A believer would argue that the Spitzer reforms have really paid off," says Starmine's David Lichtblau. "A cynic would say that 2003 was more of a momentum market, which helped analysts because they tend to be trend followers."

It's no surprise to hear that Spitzer prefers the optimist's take. "What I hear anecdotally," the AG tells us, "is that there is an entirely different ethic in the research departments and a greater focus on independence and hard analytical work." Let's hope it continues. --J.B.