Financial innovation gets kind of a bad rap, and one of my favorite parts of this job is when I get to celebrate it just for being itself. Sometimes this means breathtaking magic like the derivative on its derivatives that Credit Suisse sold to itself, or elegant executions of classic ideas like the Coke shares that SunTrust sold for regulatory purposes but not for tax purposes. Other times it’s a more prosaic combination of already-existing building blocks to allow people who were comfortably doing something to keep comfortably doing it in the face of regulations designed to make it more uncomfortable.
Yesterday a reader pointed me to a Bond Buyer article that, while perhaps neither all that scandalous nor all that beautiful, is sort of cozy. It’s about a new issue of callable commercial paper issued by a Florida municipal financing commission, and here’s the joke:
JPMorgan came up with the new product as a solution for variable-rate municipal issuers facing impending Basel III regulatory problems. The proposed regulations would require banks to have a certain higher value of highly liquid assets to be available to turn into cash to meet liquidity commitments that could be drawn within 30 days. Maintaining higher liquidity would be expensive for banks, which may try to pass on costs to its issuers, according to an analyst at Moody’s Investors Service. “What we did, starting over a year ago, is ask what we can do to change the product that will still work for all the players, including issuers, investors, and the rating agencies,” Lansing said. “And the ultimate result was this product.” The new product allows banks to continue to support variable-rate products after the regulations are implemented. The paper has a variable length of maturity, but always at least 30 days. Several days before the paper would have 30 days left to its maturity, the issuer calls the paper.
The joke isn’t that funny, though I giggled at the phrase “a solution for variable-rate municipal issuers facing impending Basel III regulatory problems.” Municipal issuers face no Basel III problems: municipalities are not subject to Basel III. Their banks face those problems – the problem being that, under Basel III, they have to keep some money lying around in case their municipal clients draw on their commercial-paper liquidity facilities. The new Basel III liquidity coverage ratio rules require banks to hold “high quality liquid assets” sufficient to cover certain assumed liquidity outflows, including draws on commercial paper back-up liquidity facilities. Holding HQLAs is, to a first approximation, expensive. Banks being banks, they tend to pass costs on to customers.
So under Basel III, municipalities have to pay more for backup liquidity, or … do this thing. This thing is, instead of issuing 90-day (say) commercial paper, you issue 120-day commercial paper, and you tell everyone you’ll call it for redemption (i.e. roll it over) in 90 days. If you can’t remarket it in 90 days, you have to pay an extra 0.75% on the notes you’ve left outstanding, though you pay that extra 0.75% not to the CP investors but to the liquidity provider, JPMorgan.1
Investors get 90-day CP. The issuer gets 90-day CP. The bank gets to provide its liquidity facility for 120-day CP, which never has less than 30 days of remaining life. That takes advantage of the way the Basel III LCR rules are written: they are aimed at the risk of short-term outflows, and since liquidity facilities are normally only drawn when CP matures and can’t be rolled, the rules exclude facilities backing CP with a longer-than-30-day remaining life.2 So the bank never has to hold any high-quality liquid assets to back this liquidity facility – unless the CP can’t be remarketed on the call date, in which case the bank is getting an extra 0.75% to defray its costs of holding a bigger liquidity reserve.
Is this deal a major work of art? Not really; step-up call structures – “this [bond/preferred/whatever] has a [30-year/perpetual/whatever] maturity, but if we don’t call it within 5 years we have to pay an extra [__]% annual interest” – are a longstanding way for issuers to tell busybodies (capital regulators, ratings agencies, etc.) that they have a long-dated capital structure, while reassuring investors that they’ll get their money back soon. Extending the idea to get around the Basel III LCR is a moderately clever reuse of a very old tool.
Is it terrible and malicious and financial-system-destroying? I mean, meh? It’s municipal commercial paper? That said, there’s a sort of conservation law to financial risk; the fact that bank and issuer avoid the cost of following Basel III rules here means that that cost is somehow shifted elsewhere, in the form of increased risk. So it’s worth tracing where it goes.
Some of it probably goes to investors – “oh it’s basically a 90-day issue, right?,” they think, and so don’t charge enough for the end-of-days optionality.3 But a lot of it probably goes to the bank. The Basel LCR rules are intended to make you hold a liquidity reserve in good times, as sort of a good starting point for dealing with your liquidity problems when times get bad. You don’t have to hold cash equal to all of your potential outflows; instead, Basel makes certain assumptions about how much money will go out the door, including 30% of your municipal commercial paper liquidity commitments.4
If you have a regular-way program of providing liquidity backstops to a lot of municipal commercial paper, some portion of those backstops will always be due within 30 days, so you’ll always hold some cushion of liquid assets. If things get worse in the muni market and issuers actually draw on their liquidity commitments, that cushion will help, though your ultimate risk is that 100% of those commitments end up getting drawn and sending cash out the door.
If your program looks entirely like this thing, though, in the ordinary course you’ll hold no cushion against potential draws, because none of your backstops will ever be due within 30 days.5 Until they are! Once your issuers stop being able to roll their commercial paper, then you’re ultimately on the hook and all of your backstops end up – albeit with a 30-day delay – looking like regular backstops. Just like the regular-way backstopping bank, you may need to ramp up to funding 100% of your commitments, but you ramp up from a much lower starting point, albeit with a bit more time to do it. The time helps, but the bad fact is that you are hoarding even more liquidity at a time when, presumably, so is everyone else.6
Which is not really a big deal: it’s just a little added cyclicality for a firm – JPMorgan – that more or less knows how to deal with its liquidity risk. It follows the letter of the rules cleverly, though you could argue about whether it’s wholly within their spirit. But it’s hard to get worked up about it: as long as markets stay relatively normal, this product really is a bit less risky than a regular liquidity backstop, so why shouldn’t they hold a bit less liquidity against it? As long as markets stay relatively normal, everything is fine. Isn’t that the point?
Morgan Stanley Joins JPM in Offering New Product [Bond Buyer]
Sunshine State Governmental Financing Commission – Commercial Paper Revenue Notes, Series H
1. Page 6 of the CP Memorandum:
The Issuer will incur a surcharge under the Liquidity Facility (the “Surcharge”) to the extent any Notes are not successfully redeemed prior to maturity, for each day during the period commencing thirty (30) days immediately preceding the date such Notes are scheduled to mature …. To avoid such surcharge, the Issuer anticipates entering into a standing order with each Dealer (the “Standing Order”), subject to certain conditions, to facilitate a redemption on the earliest possible Business Day within each Call Exercise Period.
Page 15 notes that the Surcharge is 0.75% on the notes left outstanding past 30 days before maturity. The Call Exercise Period is the 35 days before maturity. So basically you’ll get redeemed 35 days before maturity, 30 at the latest, if things are normal. Btw the 120-noncall-90 in the text is made up; the latest deal is apparently actually 136-day final maturity, but that sounds weird.
2. Paragraph 128 here:
A liquidity facility is defined as any committed, undrawn back-up facility that would be utilised to refinance the debt obligations of a customer in situations where such a customer is unable to rollover that debt in financial markets (eg pursuant to a commercial paper programme, secured financing transactions, obligations to redeem units, etc). For the purpose of this standard, the amount of the commitment to be treated as a liquidity facility is the amount of the currently outstanding debt issued by the customer (or proportionate share, if a syndicated facility) maturing within a 30 day period that is backstopped by the facility. The portion of a liquidity facility that is backing debt that does not mature within the 30-day window is excluded from the scope of the definition of a facility. Any additional capacity of the facility (ie the remaining commitment) would be treated as a committed credit facility with its associated drawdown rate as specified in paragraph 131. …
Paragraph 131(b) & (c) assumes a 30% drawdown of public-sector liquidity facilities but only a 10% drawdown of credit facilities.
3. I guess? I know nothing about the muni CP market, maybe they do charge enough. Also how bad is that optionality? One thing worth noting is that the bank can terminate the liquidity facility for various mishaps, including insolvency or downgrade of the issuer (page 17). One can easily imagine a sequence of (1) failed remarketing, (2) rating agencies notice and downgrade issuer, (3) JPMorgan terminates liquidity facility, (4) CP comes due and can’t be rolled or put to liquidity provider, (5) CP investors eat loss. In other words this is quite wrong-way optionality, for the investors.
4. See note 2. (Is that cross-referencing obnoxious? Probably, right?)
5. I actually don’t know how to read the “additional capacity … will be treated as a committed credit facility” bit of the Basel rules (note 2 again). Does that mean 30% drawdown for 30-day-and-under, and 10% for 30+ day? Probably. In which case this is a 10% buffer, instead of the 30% buffer for a regular-way facility.
6. Of course the 30 days’ delay is worth something. But even there, note this oddity: when an issue can’t be remarketed and so gets within 30 days of maturity, the bank now needs to hold liquid assets equal to 30% of its commitment. But at this point – with short-term muni markets frozen, at least for this particular issuer – the odds that it will have to fund the full amount of its commitment are much higher than in normal times. But it is subject only to the normal, 30% liquidity coverage requirement.