Earlier, we reported that the analysts at Facebook's IPO underwriters had cut their estimates for the company in the middle of the IPO roadshow, a highly unusual and negative event.
What we didn't know was why.
Now we know.
The analysts cut their estimates because a Facebook executive who knew the business was weak told them to.
Put differently, the company basically pre-announced that its second quarter would fall short of analysts' estimates. But it only told the underwriter analysts about this.
The information about the estimate cut was then verbally conveyed to sophisticated institutional investors who were considering buying Facebook stock, but not to smaller investors.
The estimate cut appears to have influenced the investment decisions of at least some institutional investors, dampening their appetite for Facebook stock, and crucially, affecting the price at which they were willing to buy Facebook stock.
As I described earlier, at best, this "selective disclosure" of the estimate cut is grossly unfair to investors who bought Facebook stock on the IPO (or at any time since) and didn't know about it.
At worst, it's a violation of securities laws.
This latest chapter in the Facebook IPO story began this morning, when Reuters' Alistair Barr reported that the research analysts at the company's lead underwriters—Morgan Stanley, Goldman Sachs, and JP Morgan—had cut their earnings estimates for Facebook during the company's IPO roadshow. This was highly unusual, if not unprecedented (I've been in and around the tech IPO business for almost 20 years, and I've never heard of it happening.)
Analysts cutting estimates is generally regarded as significant negative news for stocks. This is especially the case when the analysts who cut their estimates are very close to a company—and, therefore, are thought to have particularly good information.
(In the old days, before the implementation of Regulation Fair Disclosure, companies used to manage the market's expectations by telling trusted analysts to change their estimates. Reg FD banned that practice.)
Muppet Beaker
An investor who wasn't told about Facebook's estimate cut.
The fact that some potential Facebook investors were told of the analysts' estimate cuts and others were not would seem to be a major "selective dissemination" issue.
It is inconceivable that a reasonable investor would consider the sudden reduction of the underwriter analysts' estimates to be immaterial to an investment decision.
The SEC and FINRA appear to have acknowledged this, and they may now investigate what happened.
More broadly, everyone is still trying to understand what happened with the pricing of the IPO, which was hyped up to be the offering of the century. We now have some more information on that.
Given the PR and legal disaster that the Facebook IPO is rapidly becoming, most official communications channels have gone silent. Facebook declined to comment. Morgan Stanley did not return a call and email seeking comment.
We have spoken to several sources familiar with aspects of the transaction. We do not have complete details yet, but a general picture of what happened is starting to take shape. For now, please regard most of the information below as scuttlebutt, as it has not yet been confirmed.
The story we are hearing is this...

THE FACEBOOK IPO

WATCH: How Low Will FB Stock Go?
In early May, as Facebook prepared to kick off its IPO roadshow, the research analysts at the company's lead underwriters developed financial forecasts to facilitate the marketing and pricing of the IPO.
Such estimates are usually developed through close collaboration between the underwriters' research analysts and company management. These estimates are viewed by sophisticated investors as having been "blessed" by the company: They are perceived as revenue and earnings targets that the company has reviewed and is confident it will hit. Sophisticated investors use these estimates when they are developing "bids" for the stock, as a tool with which to help determine the price they are willing to pay.
Importantly, these estimates are not published anywhere.
Rather, in conjunction with industry convention (see below*), these estimates are conveyed verbally to institutional investors who are considering investing in the IPO.
(This is an absurd and unfair practice. The estimates themselves are material information--the consensus of smart, well-trained analysts who have worked with the company's management to develop realistic forecasts. Most investors don't even know that these estimates exist, let alone that they're whispered verbally to only a handful of big investors. All potential investors should have easy access to these estimates, as well as to any logic underlying them. The SEC needs to change the rules here.)
The same development and dissemination of estimates also happened with Facebook.
As the Facebook roadshow began, institutional investors who were considering investing in the stock were verbally given the underwriters' initial estimates for the company. And, initially, there was a lot of institutional enthusiasm for the stock.
Several days later, however, on May 9th, Facebook filed an amended IPO prospectus with the SEC.
This prospectus contained new disclosure language that had not previously appeared in Facebook's SEC filings. The language was on page 57 of the prospectus, in a section discussing the company's recent financial and user trends:
Based upon our experience in the second quarter of 2012 to date, the trend we saw in the first quarter of DAUs increasing more rapidly than the increase in number of ads delivered has continued. We believe this trend is driven in part by increased usage of Facebook on mobile devices where we have only recently begun showing an immaterial number of sponsored stories in News Feed, and in part due to certain pages having fewer ads per page as a result of product decisions.
The appearance of this language unnerved some sophisticated investors and analysts, who took it as a  sign that Facebook's business might have deteriorated. The language was vague, however, and it did not make clear that Facebook's second quarter was weaker than expected. (To infer that message from the language, you had to know that Facebook's first quarter had been weak--and that the cause had been the divergence between user growth and revenue growth.)
Soon after Facebook amended its prospectus, all three analysts at the company's lead underwriters—Morgan Stanley, JP Morgan, and Goldman Sachs—cut their estimates for Facebook's Q2 and the full year.
These estimate cuts were conveyed verbally to sophisticated institutional investors.
And, not surprisingly, these investors viewed the estimate cuts as a startling and negative development.
One important question, of course, was why all three underwriter analysts cut their estimates.
Had they all read the new sentence in the prospectus above and realized that the second quarter was weak? Or had they been tipped off?

It seemed inconceivable that all three analysts could have read the language above and concluded independently that Facebook's Q2 was weak and therefore decided to take the highly unusual step of cutting estimates in the middle of a company's IPO roadshow.
More likely, it seemed, someone had directed the analysts to cut their estimates—most likely someone with inside knowledge of how Facebook's Q2 was progressing.
And we have now heard from one source that that is what happened.
One of the underwriter's analysts has said he was told by a Facebook financial executive to cut his estimates.
According to another source with insight into the Facebook IPO process, until the underwriters' analysts cut their estimates, demand for Facebook's stock among sophisticated institutional investors was high. Once these investors heard about the estimate cut, however, they became more cautious about the IPO.
(Again, an estimate cut like this during a roadshow would be hard to interpret as anything but negative. One institutional investor I spoke to said he has looked at more than 1,200 IPOs over the course of his career, and he has never heard of this before. This is especially true because the underwriter estimates aren't really "estimates"--they're more like company guidance.)
The estimate cut, moreover, was followed by three additional pieces of information that were interpreted negatively by some institutional investors:
1) The price range for the deal was increased, which made the deal even less attractive in light of the estimate cut,
2) The size of the deal was increased, which meant that more stock would be sold, and
3) Many smart institutional Facebook shareholders like Goldman Sachs decided to sell more stock on the deal—the "smart money," in other words, was cashing out.
Meanwhile, during private roadshow meetings, Facebook executives were reportedly "signalling" to some sophisticated investors that Facebook's advertising revenue would not grow as rapidly as some potential investors had hoped. Facebook's advertising business is driven primarily by company-to-company sales efforts, not by the self-serve ads that drive Google's business. Facebook executives reportedly made clear to sophisticated investors that this would limit the rate at which Facebook's ad business could grow.
By the second week of the roadshow, after the estimate cut and price increase, some institutional investors became more cautious about the IPO. According to one investor who looked at the deal, institutions "got the willies" and started to talk about paring back their stock orders.
Meanwhile, out in the real world, demand for Facebook stock was hitting a fever pitch. One senior stockbroker at a major brokerage firm reported that he "had never seen such demand" for an IPO.
These individual investors, needless to say, were not likely aware that the research analysts at the company's lead underwriters had cut their estimates for the company. They were also, presumably, unaware that Facebook's Q2 was weaker than expected.
At the end of last week, the time came to decide on the IPO price for Facebook's stock.
This process was handled by Facebook's lead underwriter, Morgan Stanley, and Facebook executives.
According to one source (unconfirmed--this really is just scuttlebutt), based on the book of orders submitted by both institutional and retail investors, Morgan Stanley found that there were two distinct price levels at which investors were interested in buying stock.
Institutional investors, having digested the news of the underwriter estimate cut, were comfortable buying Facebook stock at $32 a share.
Retail investors, meanwhile, who were presumably unaware of the estimate cut, were comfortable buying Facebook at $40 a share.
Knowing that a big percentage of the IPO stock could be sold to retail investors instead of institutional investors, Facebook and Morgan Stanley decided to price the IPO at $38.
Although the precise allocations could not be learned, a source says that Morgan Stanley allocated a far larger percentage of the Facebook deal to individual investors than is normally the case in an IPO like this.
On Friday, May 18th, Facebook's stock opened at $42. It spent most of the day above $40, giving short-term traders a chance to flip for a quick gain, and then sank quickly. With heavy support from Morgan Stanley, the stock closed on Friday at just above the IPO price.
Given the amount of stock that had been sold, Morgan Stanley could not support Facebook's stock price indefinitely without exposing itself to huge losses. In two trading days this week, as the IPO hype wore off and news of the analyst-estimate cut spread, Facebook's stock plummeted.
The stock closed today at just over $31 a share, about the price that institutional investors were reportedly comfortable paying for it.
The SEC and FINRA have already said they may look into the Facebook IPO process. The Massachusetts Secretary of the Commonwealth has also just announced that has subpoenaed Morgan Stanley over the issue.
So, at some point soon, we will likely get the full story.
In the meantime, it's hard to conclude anything other than this:
In one of the biggest IPOs in history, in which a huge amount of stock was sold to small investors, privileged Wall Street insiders once again got top-notch information...and individuals got the shaft.
UPDATE: Reuters' reporters Poornima Gupta and Alexei Oreskovic have uncovered the before-and-after Facebook estimates that were verbally conveyed to big investors (below). As you can see, both sets of numbers are carefully coordinated across the firms. There is no way that the estimates would be this close without explicit direction from Facebook. There is also no way that an investor could get this information from Facebook's IPO prospectus. These disclosure rules need to be changed...
Here are the detailed figures from the four banks, according to one of the investors who received the new numbers.
Lowered full year revenue estimate for 2012
Morgan Stanley -- $4.854 bln (new)from $5.036 bln (old)
Bank of America -- $4.815 bln (new) from $5.040 bln (old)
JPMorgan -- $4.839 bln (new) from $5.044 bln (old)
Goldman Sachs -- $4.852 bln (new) from $5.169 bln (old)
Lowered estimates for second-quarter 2012
Morgan Stanley -- $1.111 bln (new) from $1.175 bln (old)
Bank of America -- $1.100 bln (new) from $1.166 bln (old)
JPMorgan -- $1.096 bln (new) from $1.182 bln (old)
Goldman Sachs -- $1.125 bln (new) from $ 1.207 bln (old)
Lowered 2013 Earnings per share estimate
Morgan Stanley -- 83 cents (new) from 88 cents
Bank of America -- 64 cents (new) from 66 cents
JPMorgan -- 66 cents (new) from 70 cents
Goldman Sachs -- 63 cents (new) from 68 cents
* Several readers have asked why underwriters are allowed to verbally disseminate analyst estimates to big clients. This is an excellent question. The short answer is that they shouldn't be allowed to: It's very unfair to anyone who doesn't happen to be a big client.
The reason this rule developed, ironically, was to protect small investors from sleazy brokerage firms that might hire analysts to publish puffed-up estimates only to sell a deal. Big investors, the thinking went, were grown-up enough to be able to evaluate the estimates for themselves, whereas smaller investors weren't.
The Law of Unintended Consequences then led to the current situation, which is worse: Institutions get highly material estimates shaped by the company. Small investors don't.