Wall Street

Banks That Sold Snap IPO Say It's a Buy

Look at it as a delayed form of congratulations.

All the ghosts are happy ghosts.

Photographer: Patrick T. Fallon/Bloomberg

Here's a Wall Street scandal. An investment bank -- let's call it Bank X -- was hired to underwrite Snap Inc.'s initial public offering. As with any big high-profile IPO, it was a massive team effort. Bank X's capital-markets bankers took cartoon-dog-tongued selfies to impress Snap with their love of the company. Its relationship bankers worked with lawyers to craft a prospectus that tells Snap's story in the best possible light. Its salespeople worked to cajole institutional investors into buying lots of shares in the IPO at high prices. Its retail brokers signed up wealthy clients for IPO allocations. Almost everyone at Bank X went all-out to make sure that the offering was a success.

Almost everyone. Bank X's research analyst didn't participate in this flurry of salesmanship. Instead she sat back quietly, with an enigmatic smile on her lips, barred by regulation from publishing any research until 25 days after the IPO. 1 Snap went public, its stock soared on the first day, and then it bopped around uncertainly for the past few weeks. The analyst said nothing. And today, long after the IPO, she finally released her initial report. And it says, right on top in big red letters: "Sell." Snap, she writes, is a terrible company, terribly overpriced. No one should own it at these levels, and anyone who does own it should dump it immediately.

Aren't you shocked? Dozens of people at Bank X wrote a prospectus and organized a roadshow and held one-on-one meetings and made sales calls, all with the purpose of getting the bank's investor clients to buy Snap's shares. Bank X presented a united front, with all of its messaging carefully controlled to say: "Buy Snap." And then its research analyst -- its resident expert on social-media companies, 2 who, unlike all those bankers and salespeople, is required by law to mean what she says and say what she means -- came out and said, no, actually, sell it. It was garbage all along. Fooled you! In many ways, this behavior is reminiscent of the research-analyst scandals of the last tech boom, when internet analysts had grave doubts about the companies they covered, but kept those doubts to themselves so as not to interfere with their investment bankers' efforts to win underwriting business from those companies. 

It's a big scandal! Or it would be, except that I am making it up. Bank X doesn't exist. No bank would really be that brazen about its conflicts of interest, about the fact that it sells stock to investors without really believing in the stock. 3  Instead, of the 26 banks listed as underwriters on the Snap IPO, 13 initiated research coverage today, nine with Buy recommendations and four with Holds. 4  Price targets range from $24 to $31; Snap closed Friday at $22.74. All of the underwriter banks' research analysts who have published so far have concluded that Snap is at least an acceptable investment. 

Weirdly, people think that that is the scandal. "One explanation for the universal bullishness today is conflict of interest," writes Sam Ro. The conflict of interest, presumably, is that the underwriter banks are giving Snap favorable ratings in order to win underwriting business. "It’s not crazy to think that banks doing business with a company may want to stay in their good graces by issuing bullish calls on the stock," Ro writes.

The slightly odd thing about this theory is that the underwriting banks have already done the underwriting. The checks have cleared! Snap is unlikely to do another IPO any time soon. Sure, it might need other investment banking help -- for an acquisition, say -- but positive research-analyst coverage is less relevant for winning those mandates. And sure, other companies like Snap might want to go public, and might prefer to hire a bank whose analyst is bullish on the sector -- but what other companies are like Snap? 5  

Still it's awkward that most analysts who didn't underwrite Snap have been negative on it, while most who did are bullish. My Bloomberg Gadfly colleague Shira Ovide writes:

The ratings divide isn’t great for the already spotty reputation of investment banks. There’s a perception that Wall Street firms score brownie points by offering generous stock research to land investment banking clients. It’s not always true, but every time the analysts at IPO banks are more kind than their peers at firms that didn’t have a hand in a hot IPO, it fuels the perception of Wall Street grade inflation for favored clients.

So what is going on? Maybe just coincidence! But if you must have a cynical answer, here are two. 

One is a more nuanced version of the conflicts-of-interest argument. A thing to remember about IPOs is that underwriters are traditionally gatekeepers to the public markets, and their historic role included vouching for the quality of the companies that they underwrote. This role has faded a bit in modern markets. Underwriters still do some level of vouching for their issuer clients -- they still do due diligence to make sure that the companies are not outright frauds, anyway -- but the vouching is not so much "this is the next Facebook" as it is "there is some price at which a sophisticated buyer could rationally conclude that the potential rewards of this stock might outweigh the risks."

Still, there is some level of underwriter review. Banking teams go to internal committees to get approval to underwrite IPOs, and part of what those committees want to know is, at some level: Can we be proud of underwriting this deal? Or at least, not horribly ashamed? It helps if the investment bankers can say that they believe in the company. But they are pretty biased: They get paid for doing the deal. It helps much more if the research analyst -- who is less biased, and more of a neutral expert -- can say that she believes in the company. And if she can't, then that takes some explaining. Banks can -- and sometimes do -- underwrite offerings even when their research analysts have a Sell rating on the stock. But it feels weird, and it's somewhat exceptional, for all the reasons I laid out above, in the imaginary Bank X scandal. That looks like a conflict of interest, when your salespeople sell stock that your analysts think is bad. 

And so it happens that most of the time, when a bank underwrites an IPO, its analyst is at least lukewarm on the company. That could happen by omission: If the analyst has strong and vocal doubts about the company, the company might just not hire her bank. (Or maybe her doubts are so strong and compelling that the bank turns down an underwriting role. It could happen!) Or it could be a happy coincidence, where the analyst liked the company to begin with. Or it could be the result of long discussion, where the company (and the banking team) spend time with the analyst, educating her on the company, addressing her doubts, and making changes to the offering to respond to her concerns. Sometimes those discussions might get contentious, and the analyst might feel pressured, rather than persuaded, to get to a Buy or Hold rating. 

But there's another, probably more important pressure on the analysts. Imagine if Bank X -- or Morgan Stanley or Goldman Sachs Group Inc. or any of the actual underwriting banks -- had come out with a Sell rating on Snap today. Would Snap be annoyed at them? Sure, whatever, probably. But all it could do would be to withhold future investment banking business 6 ; it's not obvious that there's all that much of that, and in any case it doesn't influence the analysts' paychecks. So the analysts probably don't worry that much about annoying Snap.

But you know who'd be really annoyed at an analyst who came out with a Sell rating? The investors who had bought Snap shares in the IPO. And those investors do influence the analysts' paychecks, by routing commissions to their banks and by filling out the sort of accuracy-and-helpfulness scorecards that help determine analysts' bonuses. If those investors had bought Snap shares in the IPO, at the recommendation of Morgan Stanley or Goldman Sachs salespeople, and then a Morgan Stanley or Goldman Sachs analyst told them that they were fools to do so, they'd be understandably annoyed. Also, if the analysts' recommendation caused the stock price to drop, those investors would be directly harmed. That's not great client service.

Last week, when I predicted that (1) the underwriter banks would soon come out with positive ratings on Snap and (2) that would inspire "a round of cynical groaning about analyst conflicts of interest," I wrote that analysts "have conflicts, sure, but their key conflicts aren't conflicts of interest" 7 :

Giving a mediocre company a buy rating can be good for your investing clients, if they own the stock. If no one owns Snap -- if, say, it's a private company coming to market for the first time -- and your clients come to you to ask if they should buy it, you should tell them what you think. But Snap has already had an enormously hyped IPO. If you are publishing a research report weeks later, you know that a bunch of your investor clients bought Snap in the IPO. You know others were pitched the stock in the IPO, and didn't buy, and presumably won't now. Putting a buy rating on the stock will help the former group (the stock will go up!), without much hurting the latter group (what do they care?).

In some abstract general sense, one role of an equity research analyst is to tell investors which stocks to buy. That is not the only role, or the most important one; after all, the job of a professional investor is to decide which stocks to buy, so most professional investors have to rely on analysts for something else. (Supporting analysis, or general ideas, or management access.) But in any case, it isn't that helpful for an analyst to tell investors whether or not to buy an IPO 25 days after that IPO. At that point, it's much more helpful -- and more welcome -- for her to congratulate them on having bought it already.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

  1. Weirdly, the official regulatory ban is actually 10 days but everyone waits 25 days due to other, more complicated regulatory reasons. Call it 25 days.

    Incidentally, in the real world, analysts can and do participate in the IPO selling process -- talking to investors and so forth -- before they publish their first notes. Bank X's analyst is staying quiet here mostly for dramatic effect.

  2. Camera companies, whatever.

  3. Oh I'm exaggerating. It does happen. (It didn't happen with Snap.) It's just rare, because it looks so awkward!

  4. Here is a roundup of research coverage. I include "overweight," "buy," "outperform" and variants thereon in "Buy," and "hold," "neutral" and "market perform" in "Hold." Among non-underwriter analysts, from Bloomberg, I see four Buys, seven Holds and six Sells.

  5. Also, no one believes this, but it's generally really hard for an investment banker to influence a research analyst's recommendation, even if he wanted to. The analysts don't get paid for bringing in banking business, and are legally required to be independent. They're generally not even allowed to talk to the bankers unchaperoned, though in an IPO process there is usually more contact between bankers and analysts than for an already-public company.

  6. Well, it could also withhold management access -- another interesting research conflict that people think leads to undeserved high ratings.

  7. This is also how I think about the role of credit-rating companies in the financial crisis. Did the raters conspire with the banks to slap AAA ratings on risky securities and sell them to unsuspecting investors? Sure, probably, sometimes. But the more compelling explanation, much of the time, seems to be that the rating companies and the banks and the buyers all wanted risky things to have AAA ratings, and so the rating companies gave the buyers exactly what they wanted. 

To contact the author of this story:
Matt Levine at mlevine51@bloomberg.net

To contact the editor responsible for this story:
James Greiff at jgreiff@bloomberg.net

Before it's here, it's on the Bloomberg Terminal.