Later this month embattled UBS will hold its big annual investor day in New York at which interim chief executive, Sergio Ermotti, is expected to spell out his plans for the troublesome investment banking arm.
A slimming down seems inevitable. But if Ermotti is having second thoughts about turning UBS into Julius Baer on steriods, he could do worse than cast an eye over one of the latest report from JP Morgan’s banks analyst Kian Abouhossein.
He’s reckons there’s another way. A Swiss way.
UBS should merge its investment bank operations with those of Credit Suisse.
We believe the pro-forma 50:50 JV would be in both shareholders’ and Swiss gov’t interest resulting in less IB capital at risk and assets for Switzerland, higher ROE through material cost cutting mainly outside Switzerland, and material capital release for shareholders. We argue the JV over time would have to be funded separately and see a rationale for moving the legal entity to the UK. We analyse the 1997 SBC-UBS merger as a roadmap.
The Bank of England will love that; another G-Sifi domiciled in the City of London.
Still from a Swiss perspective the idea has merit.
We believe a JV in IB divisions would lead to material reduction in RWAs of 47%, implying reduced risk and lower capital requirements. Post a JV, UBS and CS would generate 29 per cent and 45 per cent of ‘13E group profit respectively from IB on our estimates. We also estimate the IB to consume 29 per cent of group capital for UBS, 37 per cent for CS post a JV, down from 56 per cent and 62 per cent pre-JV. A JV should also lead to smaller trading and derivative position as was seen in the 1997 merger of UBS and SBC.
There would be a human cost, of course, in the form of 13,000 redundancies, or about 38 per cent of the combined headcount, says Abouhossein. But just think about those cost savings.
On our estimates, an IB JV would lead to combined staff reduction of c.13,000 or 38% of the combined IB division headcount pre JV, reducing total cost by SF7.5bn (40% of IB cost base). We estimate net SF3.7bn restructuring charge and pro-forma revenue loss of 33% due to overlap in businesses. We estimate a JV to operate with C/I ratio of 68% in ‘13E, down from avg.74% pre-JV. IB ROE in ‘13E improves from avg. 13% to 20% on B3CET1 capital. A JV should be complementary in FICC, and geography.
Now, UBS would have to offer some cash to get the deal done.
We estimate an IB exit multiple for CS at a 20 per cent premium to that of UBS would lead to SF3.1bn one-time payment from UBS to CS at the start of the JV. With UBS SF51.3bn in tax losses not recognized as DTAs in addition to SF8.7bn of on balance sheet DTAs, a JV should help better monetize these DTAs through improved profitability outlook in the IB.
However, Abouhossein says it would be worth it.
A self-funded IB JV would in our view be forced to focus on improving RoAs vs. global peers leading to greater efficiency. We see return on B3CET1 [core tier one capital under Basel III] declining to 16 per cent from 20 per cent in ‘13E, still above the avg. IB ROE pre-JV in a scenario of 156bps increase in post tax cost of funding with SF50bn senior debt issuance.
And it’s got us thinking about another national investment bank solutions. Could Barclays and RBS pool their operations to create BARBS? What about BNP Paribas and Societe Generale. Their JV could be called Banque Arrogant. JP Morgan and Morgan Stanley could be reunited as Morgan^2.
SocGen scraps dividend to meet capital shortfall – FT