I realize it doesn't actually work this way but I always imagine that sell-side analysts at big banks who cover other big banks enjoy sabotaging each other a little. "Take that, you Deutsche Bank jerks!," Jernej Omahen might have thought as he hit send on this one:
Deutsche Bank AG fell the most in more than five months after Goldman Sachs Group Inc. cut the company to sell from hold, saying it may have to transfer $13 billion to its U.S. unit under new capital rules.
Deutsche Bank slid as much as 6.2 percent, the biggest intraday drop since Sept. 26, and traded at 33.07 euros at 1:40 p.m. in Frankfurt [closing at 33.66 / down 4.6%]. The stricter requirements may hurt profit at Europe’s biggest bank by assets and require it to ask shareholders for more money, Goldman Sachs analysts including Jernej Omahen wrote in an e-mailed report from London today.
Goldman's note addresses two impacts of recent Fed moves to make international banks' US operations safer: the capital impact, and the funding impact. The capital stuff is wholly imaginary, though I guess the economic consequences might be real enough. Start with this chart, and note that "Taunus" is basically shorthand for "Deutsche Bank's US operations":
If GS is right - I have no idea, I'll just assume they are, but there are some assumptions and guesses here - the problem with Deutsche's U.S. operations isn't that they're undercapitalized; it's that they have negative capital. That is, Deutsche's U.S. business is "worth," using Basel capital accounting, a negative amount of money: has liabilities that exceed its (risk-weighted) assets. Why not close it down? The answer is some combination of "that's not true, because capital accounting rules don't reflect economic reality," and "that's not true, because isolating the U.S. operations doesn't reflect actual economic claims," but the first one is probably more important. We occasionally talk about how surreal the bank capital conversation can be - often using Deutsche Bank as an example - and this is a good reminder.
The surrealism extends to the solution, too. Goldman thinks that to fix this negative capitalization and leave Deutsche US well enough capitalized to appease the Fed, Deutsche will need to add $13bn of equity capital. It can do this by just saying it did it - just poof "intragroup debt" into "intragroup equity" - but that would reduce the equity capital of Deutsche non-US (though not, of course, Deutsche-as-a-whole). This is the sort of thing that sounds like a problem to regulators, but you might pause to wonder if it is a problem to anyone else. If you are a creditor of Deutsche Bank in Frankfurt, and they shuffle some capital to New York, does that make them less likely to repay you? Goldman says "our interpretation of the FED proposal is that this portion of capital transferred is only available to absorb US losses," and I suppose that is true in the jump-to-default case. I wonder about the likelihood of Deutsche Bank Europe going under while Deutsche US sits around fat and happy and unable to help.
Anyway if you believe it this is what happens:1
This would still leave Deutsche's non-U.S. business better capitalized than Commerzbank, but that is apparently not saying much, as Goldman thinks that European regulators would respond by demanding that Deutsche raise more external capital.2 More capital = issuing contingent capital notes, more CoCos = 7-7.5% interest, more 7-7.5% interest = less profit. The Germans don't love this:
Elke Koenig, chief of German regulator Bafin, criticized the U.S. plan yesterday. It was a “unilateral decision” and a “step backward” for international coordination on regulation, she said on the sidelines of a banking conference in Frankfurt.
Sure, true, though for non-coordination it could be worse. As a first cut, recapitalizing the US business weakens the European business, in a straightforwardly zero-sum way. As a second cut, Bafin can always just ask Deutsche to raise more capital. The race-to-the-top effect here might annoy Deutsche but it's not clear why regulators would get that upset.3
Anyway that's all imaginary, though I guess it can cost real money if both sets of capital regulators really do stick to their guns of higher capital as a percentage of imaginary numbers. The real, or real-ish, thing is funding. Here is Goldman:
In our view, the FED aims to remove a key funding vulnerability uncovered during the crisis – the reliance of foreign banks on short-term US$ funding. This would be achieved through a liquidity stress test on par with that of US BHCs, in our view. Our analysis indicates that DBK could be affected in two ways:
- First, we estimate the US$ maturity mis-match to be US$73 bn (end 2011, last reported). With the FED encouraging banks to reduce the maturity gaps, this could result in incremental funding costs.
- Second, the European banks (and in our view, DBK) have been using the FX swap markets to transform unsecured long-term € liquidity into US$ liquidity. The FED proposal does not explicitly address this issue and it could well be that this area remains “as is”. However, were this to be curtailed, the earnings impact could be substantial, potentially altering the prospects of the business model.
They don't do much to quantify the second, but on the first, they assume that Deutsche US's maturity gap (excess of short-term liabilities over short-term assets) will have to be cut in half, or by some $36bn, at a cost of 150bps to increase funding duration.4 True? I dunno, the argument here seems sort of arbitrary.
If it turns out to be true, though - if Deutsche is required to fund longer-term and more expensively due to its bigness and internationalness - what does that say about the too-big-to-fail subsidy? Bloomberg had another editorial yesterday about how to reduce that let's say nebulously quantified subsidy, considering usual solutions like taxation, increased capital, and bail-in debt. I think of "more long-term unsecured funding" as very much on a continuum with bail-in debt and equity capital: it's a more stable (for the bank) and riskier (for the investor) form of funding for bank assets than short-term secured debt. So it makes banks safer. Also it costs more. Which may soon be an issue for Deutsche Bank.
- Notice how the U.S. banks are all on the left, with relatively low leverage ratios, while the European banks are largely on the right, with relatively high ratios of total assets to equity? That's largely an artifact of different accounting rules.
- Are you surprised that there does seem to be some correlation between total leverage ratios (total assets to equity) and Basel III capital ratios (risk-weighted-assets to equity)? Discuss.
2. And possibly retaliate by demanding that U.S. banks better capitalize their European operations, which possibility Goldman, the Fed, and I all find sort of meh.
3. I guess the answer is because it hurts German bank competitiveness: US banks only need to be well-capitalized overall, while European banks need to be well-capitalized separately both in the US and Europe. As a simple arithmetic matter that shouldn't be that hard? If you need X% capital, then you can easily have X% capital in Europe and X% in the US and have it average out globally to X%, right? I dunno.
4. Here is a table of Goldman's math on short-term assets and liabilities, and how the gap has come down over time: