How UBS came undone
A report to shareholders shows in excruciating detail how the Swiss bank bet its reputation on risky mortgage securities - and lost.
(Fortune) -- UBS Chairman Marcel Rohner confirmed today what has been one of the worst kept secrets on Wall Street: An astounding $38 billion in subprime trading losses are forcing UBS to finally abandon its years-long dream of building a global investment bank.
Speaking at the Swiss bank's annual shareholder meeting in Basel, Rohner blandly conceded that those massive losses mean UBS can "no longer aim to offer everything to everyone in investment banking."
He didn't commit to a course of action, but it is difficult to imagine that the 2,500 planned layoffs cited in recent news reports would be adequate if the bank were to truly scale back in everything from debt underwriting and trading to emerging market derivatives.
Regardless of whether UBS (UBS) adopts a proposal from former president and 1% stockholder Luqman Arnold to split the investment bank from the consistently profitable private bank and wealth management units, the damage is done.
No longer will it directly compete with the likes of Goldman Sachs across hundreds of different product lines globally. Instead, UBS will presumably retain a robust franchise in private asset management coupled with standard underwriting and lending operations. In short, UBS is preparing to return to the days before the repeal of Glass-Seagall, when trading and underwriting had a lower profile in its business mix.
The roots of UBS' full-bore retreat from the global investment banking realm are laid out in extraordinary detail in a 50-page report released to shareholders on Monday.
The document is highly illuminating for its detailed analysis of the central role played by its own so-called internal hedge fund, Dillon Read Capital Management, in the debacle.
Dillon Read Capital Management was a $4 billion fund launched in June of 2005 by former UBS investment bank chief executive John Costas, with the help of then-UBS CEO Peter Wuffli.
While its roots are in the internal proprietary trading activities that all investment banks perform, DRCM was a veritable warren of internal contradictions. It managed client money accessed via UBS' massive private client network, but saved millions of dollars annually by using UBS operations and support staff. DRCM also lured away top talent, draining UBS' investment bank of over 100 of its best and most senior personnel, almost all from its fixed-income desks.
Most important, DRCM used the bank's exceptionally low cost of funding to put on trades that were profitable from day one.
Per the report, DRCM never had to plan for the three challenges that all hedge funds must confront: maintaining stable capital at affordable rates, recruiting personnel and operations management.
This created a system of truly perverse incentives: Since all risk was UBS' and DRCM's clients, and since capital was both abundant and inexpensive, the immediate impetus was to buy and position bonds that offered the greatest carry, or differential between the cost of financing and interest paid.
Starting in June of 2005, these securities were most often either subprime mortgage securities or collateralized debt obligations.
It was bad enough for UBS that DRCM had multiple billions of dollars worth of these securities on its balance sheet - which is to say UBS' balance sheet. What was worse was that UBS largely replicated all these trades.
In other words, within UBS' investment bank, its trading desks and internal proprietary trading desks also had billions of dollars worth of these trades on, making losses systemic within the firm.
It was a perfect storm of truly, appallingly bad timing and poor judgment for UBS.
Simultaneous to the launch of DRCM in mid-2005, the report notes, the investment bank started an aggressive attempt to catch up to the fixed-income market share held by rivals like Lehman Brothers (LEH, Fortune 500) and Goldman Sachs (GS, Fortune 500).
For example, UBS committed billions of dollars in capital in late 2005 and early 2006 in a bid to dominate the CDO markets. At a time when Goldman, Lehman and even Bear Stearns (BSC, Fortune 500) were pulling back from CDOs, UBS bankers bought just under $21 billion worth of the stuff in the open market and put it on their books.
What's worse, UBS, in a bid to win lucrative CDO underwriting fees, extended what appears to be nearly limitless warehouse lines of credit, using its balance sheet to buy and hold the underlying CDO assets for their clients.
The amount involved in the CDO warehouse gambit? The report puts it this way: "Throughout 2006 and 2007, there were no aggregate notional limits on the sum of the CDO warehouse pipeline."
Like a cherry on top of a demonic sundae, the UBS report baldly confronts the poor risk management that put the bank in its current fix.
In a section titled "Absence of Risk Management," the report states that at one point in the second quarter of 2007, the CDO desk simply ignored its own internal analysis of the CDO market's health and bought billions of dollars more in mortgage-backed CDOs. At the same time, the CDO desk management was petitioning for capital limit increases.
At another point, the home equity loan trading desk's hedges were rendered ineffective because the hedges were on investment-grade asset-backed securities, while its exposure was to subprime securities - whose prices collapsed far faster than the hedges gained in value.
Since then, UBS has acted dramatically, segregating the troubled assets in a separate portfolio for eventual sale, dismissing legions of managers and publicly acknowledging shortcomings.