At a time when making stock research truly independent is a top priority in the brokerage industry, Thomson First Call -- Wall Street's data clearinghouse -- is pushing analysts to stay with the flock.

In the wake of Wall Street's $1.4 billion settlement with regulators, analysts are under renewed pressure to offer original, honest insight into companies they cover. That means, among other things, saying when true earnings are being obscured -- for instance, by the repeated use of "one-time" charges and gains. Under the exacting standards of a post-Enron world, failing to flag such abuses can bring on widespread scorn.

So it's understandable that some Wall Streeters are frustrated with First Call. Several sell-side analysts say the firm has pressured them to change their earnings estimates, simply because their numbers didn't match their peers'. These analysts say when they stood firm, First Call threatened to pull their estimates from its all-important consensus survey -- and shove their research off the front pages of its Web site.

Hit hardest by the rules are Wall Street's sell-siders, who have been feeling a bit touchy to begin with. Criticized from nearly all corners in recent years for everything from publicity-seeking to apparent conflicts of interest, analysts are more sensitive than ever to issues of independence -- even as they remain tethered to the likes of First Call to reach a wider audience.

First Call's research chief, Chuck Hill, defends the practice of pushing for agreement, saying the firm's overriding goal is consistency. But critics say such heavy-handed tactics serve only to manufacture the appearance of consensus -- while squelching dissent that investors might be better off seeing.

"I don't like it at all," Legg Mason analyst Timm Bechter says. "It's like being told to retie your shoes incorrectly."

William Tell

Take a recent example offered by Lehman Brothers' Steve Levy. Last month, telecom gearmaker Nortel ( NT) released earnings boasting a penny-per-share profit. Levy took exception, pointing out that the company used 4 cents of net income from discontinued businesses to offset the 3 cents of losses from continuing operations.

The earnings-quality push that has swept Wall Street since the accounting scandals of 2001 and 2002 would seem to dictate that the 3-cent loss carry the day, since that's the result of Nortel's ongoing business. Yet, since the majority of Nortel analysts called the quarter a penny profit, First Call was soon on the horn with Levy telling him to change his number or he'd get bumped.

While defending his practice of providing the Street with a single number, First Call's Hill says "there is no right answer on this stuff. What one person considers a nonrecurring item another will call recurring," referring to the typical gains and charges that may or may not be considered one-time events.

Levy says that in order to keep his quarterly and yearly projections available on First Call, he eventually gave in and joined the profit consensus.

Like Lehman's Levy, Legg Mason's Bechter says he once caved to First Call pressures. Last year, Bechter found himself in the minority on a Lucent ( LU) earnings call. Despite arguing unsuccessfully in his report that his peers were overlooking an element in their earnings calculation, he changed his estimate to stay in First Call's good graces.

"I understand their need for apples-to-apples comparisons, but the problem comes when the majority isn't looking at it the right way," says Bechter.

Regarding First Call's practice of not recognizing nonconforming estimates, Hill says analysts are free to publish and distribute their unique insights in their own reports. But analysts say it's a case of First Call's way or no way, if they want their differing estimates to reach First Call's broad audience.

Standard Bearer

What First Call is doing isn't without precedent. Rival Multex, now called Reuters Research, also aims for a strict consensus average. But unlike First Call, Multex will keep out-of-line estimates in with the rest of the individual projections. First Call, on the other hand, omits not only the divergent number but also the analyst's quarterly and yearly figures.

Even so, most people interviewed said the notion that First Call has a one-size-fits-all mandate only underscores Wall Street's continuing obsession with the consensus number. Even investors who rely on their own calculations say they still use the tallies from First Call and its rivals Multex, Zachs and IBES as a sort of reference point for where the majority happens to be.

"I believe in democracy -- the more numbers the better," says TIAA-CREF money manager Bill Newbury. "But there is also a need for a standard." Of course, "that's not saying First Call should be the one to apply the standard."

That said, the critics of the First Call practice see wider ramifications for a market still recovering from past excess. Levy says the thought process behind the push for consensus is similar to the one that led to First Call recognizing tech companies' use of pro forma earnings calculations. Posting those numbers allowed loss-making companies to sell their stories to investors by presenting financial results that excluded a host of expenses in their net income numbers.

It isn't that analysts weren't going along with the pro forma scheme, Levy says. But First Call "blessed the number. They condoned it," says Levy. He says that accelerated Wall Street's acceptance of the pro forma standard, which in turn helped fuel many acquisitions that otherwise would have been seen as clearly uneconomic.

Levy points to Nortel's $7 billion acquisition of Bay Networks in 1998; had Nortel included in its earnings the actual costs related to the acquisition, investors may have balked, he says. Nortel went on to rack up billions of dollars in additional acquisitions -- deals that nearly brought the company down when the telecom market inevitably cooled.

"It's problematic if they drive the consensus and it leads to a false game of beating that number," says a money manager who asked not to be identified. "Then they've become too powerful and end up driving perception one way or another."

Hill says he's heard the arguments and throws the charges right back at the analysts, saying they have the job of "gatekeepers." The process works best, says Hill, when they're "doing their job, thinking about what should or shouldn't be excluded and alerted more."

Sure, dare to be different -- just not so different that it risks your place on First Call.