Regulators require banks to get a certain percentage of their funding from "capital," which loosely speaking means funding that doesn't need to be paid back and that absorbs losses if bad things happen. In practice, "capital" means some combination of:
- common equity and
- other weird stuff.
Common equity is relatively easy to understand, and the definition of what common equity is doesn't change all that much. What other weird stuff counts as capital is permanently provisional, though, as regulators keep fine-tuning the rules to make sure that things that count as capital absorb losses and keep a bank in good health in bad times. And so lots of things that used to work don't work any more.
In Europe, though not in the U.S.,
contingent capital -- or "cocos," bonds that poof into stock (or nothing) if the bank's capital levels get too low -- can count as useful capital. But it's only really useful if the trigger level is low enough: If you issue a bond that converts into capital when your common equity drops below 5 percent of assets, and Europe has, say, a stress test that requires common equity capital of 5.5 percent in stressed circumstances, then that bond is useless for the stress test.
In 2009, Lloyds Banking Group Plc issued a bunch of 2009-vintage capital securities, in the form of 33 different series of cocos called "enhanced capital notes," with 8.4 billion pounds of the ECNs currently outstanding. The ECNs were good 2009-vintage capital, but they are apparently not good 2014-vintage capital, because their 5 percent trigger is too low.
So Lloyds had the sensible idea of replacing them with new cocos with a modern, 2014-vintage 7 percent trigger. Credit Suisse is in a similar boat and has had similar ideas.
It turns out there are two ways to do this. Our ancestors back in 2009 were not as dumb as they looked, and they were aware that the capital rules are subject to change at any moment. So Lloyds's ECNs, like most capital securities, had an explicit regulatory par call, which provided that Lloyds could redeem them at par upon a "Capital Disqualification Event," which would occur if they didn't count as tier 2 capital or if they "cease to be taken into account " by the U.K.'s Prudential Regulation Authority "for the purposes of any 'stress test' applied by the PRA in respect of the Consolidated Core Tier 1 Ratio." And, earlier this year, that happened. (Maybe!)
So Lloyds can just call them at par. Or, not quite; regulators would probably get mad if Lloyds just got rid of even not-so-good old-timey capital, and anyway that wouldn't do anything to improve Lloyds's capital position. Lloyds would have to sell new capital things -- 7-percent trigger cocos, most efficiently, or I guess common stock but hahaha who issues common stock? -- and use the proceeds to pay off the old ECNs at par.
Alternately Lloyds can buy the old ECNs for their current trading levels, which seem not to take into account the fact that they're callable at par, which boggles me but is sort of par for the course. And so Lloyds has in fact announced a series of offers in which it will exchange some of the old ECNs for new, 7-percent-trigger ECNs, and buy back some other ECNs. The purchase/exchange prices Lloyds is offering get as high as 162.5 cents on the dollar.
Which is a good deal! For the ECN holders who are, in fact, only entitled to 100 cents on the dollar. Though some of them are mad anyway:
Hundreds of thousands of Lloyds Bank bondholders have been asked to surrender annual income payments of up to 16 per cent and sell their investments back to the bank. ...
Lloyds says it is willing to pay “market prices” so investors will not lose out, but campaigners argue that private investors are being offered a poor deal. Unlike institutional investors, who can swap their old bonds for new ones, private investors are only being given the option of a cash buyout.
Retail investors loooooooove income, and some of them are calling on the U.K. Financial Conduct Authority to make Lloyds revise the exchange offer to let them keep their income. I don't know. Getting bought out of your yieldy bonds for 106 cents on the dollar is better than getting bought out at 100, which Lloyds could have done.
Arguably it is a less good deal for Lloyds's shareholders (one-third of whom are the U.K. government), since they are in effect overpaying to restructure their capital. But there are other considerations: Banks need to "be weighing up the unidentified cost of angering investors with aggressive regulatory calls that could be reflected in their future cost of funding." Calling these cocos at par and replacing them with new cocos raises the problem that the obvious investors for your new cocos are the current investors in your old cocos, and they may not be amused by getting kicked out at par.
There are some lessons here! One lesson is that there are exciting opportunities for winners and losers and taking advantage of, or being taken advantage of by, the frequent changes as regulators fine-tune post-crisis banking regulation. Lloyds could have -- and Credit Suisse still could -- saved a lot of interest payments with a free par call here due to changes in regulation. Instead it chose to take on the expense and pain of an exchange offer, making it I guess a bit of a loser, though there are other winners. BofA and Goldman are leading Lloyds's exchange offer, and there will be other investment banking mandates to clean up other capital things that need fixing.
Another lesson is that you can't always stand on your rights with providers of capital:
If you'll have a need for capital going forward, you have incentives to do what is perceived as fair by capital providers, even if technically you have the right to stick it to them.
If one were thinking about reforming oh say the U.S. mortgage system in a way that required raising hundreds of billions of dollars of private capital, one might give that lesson some thought.
A last lesson is that retail is dumb. Wait no not that! Sorry, the last lesson is that there are a lot of investors -- "hundreds of thousands" -- who invest in bank securities for income and are very very sad, and sympathetic, when that income goes away. Lloyds's old cocos had a built-in par call if the capital rules changed. It was right there in the document. The capital rules changed, and Lloyds, rather than just buying out its retail investors at par, was nice enough to offer to buy them out at a premium. And they're still complaining to regulators.
Lloyds's old cocos, and its new ones, are also explicitly loss-absorbing securities. That is their purpose: If Lloyds runs into trouble, these bonds stop paying income and flip into (much less valuable) common stock. It's right there in the document. And it's intended to reduce the need for a government bailout: Instead of the government (re-)bailing out Lloyds, coco holders would bail it in. The same coco holders who are now complaining to regulators about getting a better deal than the one provided for in the document. It makes you wonder how much loss absorbing they'd really do.
Updates footnote six and related text with comments from investor activist Mark Taber.
No, this is kind of a lie, sometimes the rules for what counts as common equity change too. But those are questions of accounting, not instruments; common stock is still common stock.
Back in the good old days, for instance, trust preferred stock could count as tier 1 capital in the U.S. and was quite a racket; now it cannot, and is not. (Cannot, that is, except for some grandfathered TruPS of small banks -- see page 9 here. [Update: That link should now work, but just in case, it's Davis Polk's "U.S. Basel III Final Rule: Visual Memorandum" and should be the second result here.])
Because U.S. Basel III rules require that an instrument be classified as equity under U.S. GAAP to qualify as additional tier 1 capital. International rules allow loss-absorbing liabilities to count as AT1 too. This giant PDF from Davis Polk, "U.S. Basel III Final Rule: Visual Memorandum," is the source of much of my bank capital knowledge; the part about GAAP and stuff is on page 73. [Update: That link should now work, but just in case, it's Davis Polk's "U.S. Basel III Final Rule: Visual Memorandum" and should be the second result here.]
Here is the European Banking Authority's announcement of the 2014 stress tests: "In terms of capital thresholds, 8% Common Equity Tier 1 (CET1) will be the capital hurdle rate set for the baseline scenario and 5.5% CET1 for the adverse scenario." The point of cocos is that they should convert into common equity capital in the adverse scenario; if you have an "adverse scenario" and the cocos haven't converted then they provide no capital benefit.
A good place to get a summary of the 2009 ECN structure, and the structure proposed to replace them, is in the comparison of terms in the exchange offer prospectus. Incidentally a fun fact about the ECNs is that they, in turn, were issued in part in exchange for earlier hybrid-ish securities called "permanent interest bearing shares," because the shelf life of weird things that count as capital is like three years.
I gather this is not crystal crystal clear, but it seems clear enough from the structure of the European stress tests. The FT quotes "independent investor activist" Mark Taber saying “Lloyds says it has the right to buy back at par, but that’s contentious. Lloyds should at least get an independent opinion on the deal,” but now you have my opinion anyway.
Update: My opinion might be changing! Taber e-mailed me on March 15 to object, arguing that the "Capital Disqualification Event" in the original ECNs applies if the ECNs no longer count for stress tests of the Core Tier 1 ratio. But what has actually happened is that the stress tests now focus on the Common Equity Tier 1 ratio, not the Core Tier 1 ratio, which makes the ECNs useless for stress tests, but arguably not in the way contemplated by the Capital Disqualification Event language. Taber:
I put it to you that had Lloyds intended the RPC terms to encompass a call being possible if the ECNs fail to qualify for any form of future regulatory stress test, then they would have drafted the terms to say exactly that. Instead, they specifically defined the stress test as being in relation to the 2009 definition of CT1 capital - what was the point of doing so if they had intended something wider?
The point may well have been that our ancestors back in 2009 were in fact as dumb as they looked. Anyway! The controversy here is meatier than I'd first thought, and you can see why Lloyds, at least, would want to pay up to avoid the controversy. (And even that is controversial, since the disclosure in the exchange and repurchase documents, implying that Lloyds might call at par if you don't exchange, might be objectionable depending on your view of the underlying controversy.)
Here is Tracy Alloway at FT Alphaville on Credit Suisse's similarly callable, similarly above-par "Claudius" capital instruments. The link in the text is me making fun of the people buying Claudiuseses at above par, and judging by the reader e-mails I got from that linkwrap I suspect I'm not the only one who's boggled.
That's from a non-U.S. offering memo that I'm not going to link here out of prudence. All of the prices there seem to be above par, with a single exception for which the exchange price is 93. Here is the U.S. offering memo; those series have exchange prices of around 106.
"Exchange price" is sort of a loose term since it assumes that the new securities are worth par but hey why not. Extremely casual Bloomberging (looking up two ECNs, ISINs XS0473106283 and XS0459093521, in HP) suggests that those bonds are currently trading a couple of points above the nominal exchange prices, suggesting that the new ECNs are worth a bit more than par.
In a related vein, magical things have happened to TruPS due to various U.S. rule changes.
Very vaguely related, and in the news today, is the fact that "absolute priority" is not a real rule in bankruptcy. "Oh but I have my rights," the convertible investors might say, but didn't, because getting a deal done sometimes means not sticking to the letter of the law.
A more tenuous, somewhat related lesson is that maximizing immediate shareholder value sometimes takes a back seat to playing nice with other stakeholders, possibly because management correctly believes that overpaying bondholders will be good for long-term shareholder value (by making financing more efficient), but also possibly for agency-cost reasons (managers don't want to get sued, or yelled at, by the bondholders).
Err, maybe not the same ones, since Lloyds is cashing out retail investors rather than exchanging them. But they're demanding a right to exchange, so we'll see.
To contact the author on this story:
Matt Levine at firstname.lastname@example.org
To contact the editor on this story:
Zara Kessler at email@example.com