Facebook muppet of the day: UBS

June 8, 2012
Maria Bartiromo today, under the headline "UBS May Have Facebook Trading Loss of $350 Million":

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google,mail" data-share-count="false">

CNBC has a breathless report from Maria Bartiromo today, under the headline “UBS May Have Facebook Trading Loss of $350 Million”:

UBS is sitting on losses that could be as high as $350 million stemming from its investment in the Facebook initial public offering, and is preparing legal action against Nasdaq as a result, people familiar with the matter told CNBC…

These people said UBS wanted 1 million shares, but when it did not receive confirmations, it repeated the order multiple times and was left with much more than it intended…

Apparently, UBS tried to unload the stock at $35 a share, but could not catch a bid and sold some of the positions under $30 a share.

The math here just doesn’t add up. Let’s say that UBS put in an order for 1 million shares, didn’t get a confirmation, so canceled the first order and put in a second order. And then didn’t get a confirmation for that, either. Let’s say it did this, oh, half a dozen times in all. And that each time, the order went through but the cancelation didn’t. Which would mean that UBS ended up with 6 million shares at the opening price of $42. And then sold them all at an average of $33. That’s a loss of $9 per share times 6 million shares — which works out at $54 million.

So, where does the other $300 million come from?

The fact is that if UBS ended up losing anywhere close to $350 million on Facebook stock, it has no business being in the equity capital markets at all. On the day of the IPO, it makes perfect sense that UBS put in orders to buy stock, since it was surely receiving orders from its clients. And I can also believe that due to Nasdaq glitches, UBS might have ended up buying more stock than it wanted.

But if UBS only had orders for 1 million shares from its clients, and it found itself with many more shares than that, it was incumbent on UBS to sell the excess stock — immediately. UBS has no business holding millions of shares of Facebook on its balance sheet for no good reason other than that Nasdaq made a mistake.

What’s more, it was abundantly clear on that first day of trading that Morgan Stanley would buy back as the market wanted to sell, at the IPO price of $38 per share. If UBS had more shares than it wanted, there was a willing buyer at that price.

So what happened? Kid Dynamite has got some screenshots from Twitter, which show that UBS was buying aggressively in the market, at 11:40am, at $40 per share. To the tune of some 86 million shares 860,000 shares. Anything from 11:40am onwards can’t be attributed to Nasdaq glitches, it’s a simple bet that Facebook was going up rather than down.* By Monday morning, UBS advertised more shares traded even than Morgan Stanley — over 100 million in total.

The most likely thing, here, is that UBS simply bollixed up its Facebook trading strategy in the worst possible way, going massively long at exactly the same point in time as everybody else was going massively short. It was a bold and enormous bet, and like many bold and enormous bets, it wound up going spectacularly wrong. But if that’s what happened, you can’t really blame the Nasdaq. After all, if the bet had worked out, and Facebook stock had soared, you can be sure that UBS — and its traders — would be taking full credit for their genius and prescience.

*Update: Apologies to UBS, the screenshot shows actual shares for sale, not lots of 100 shares as Kid Dynamite first assumed. And there were some Nasdaq glitches even after 11:30am. But the fault here still looks as though it lies at least as much with UBS as it does with the Nasdaq.


Comments are closed.