Our colleague Jennifer Hughes cites the worries of fund managers over the underlying collateral that backs covered bonds in Europe, which reminds us that we’ve been meaning to write something about another recent proposal meant to jumpstart a covered bond market here in the US.
We say “another” because politicians pushing covered bonds in the US isn’t a new trend. FT Alphaville has been writing about the possibility for some time now, and there have been several attempts since the crisis ended to get this market moving.
A proposal to this effect nearly passed as part of Dodd-Frank, but was taken out at the last minute by Democratic Senators because of protests from the FDIC, which we address below. The Fed, as if to pave the way, has since given covered bonds their own collateral category.
We think the benefits these bonds provide — sorely needed private participation in housing finance and an added reason for banks to mind the quality of their balance sheets — would outweigh the costs. Early signals are that US investors would be interested in buying them.
That said, we’re also sensitive to some of the worries about them, not all of which can be easily dismissed. So let’s take a closer look at a few.
Covered bonds and the Federal Home Loan Banks
In a recent report, Moody’s finds that a covered bond market would be damaging to the FHLBanks, whose “advance lending programs and repurchase facilities are the primary secured funding facilities utilized by US banks.”
It’s true that the FHLBanks would have to compete with covered bonds, but as we explain below, that’s actually a feature, not a bug
Moody’s argues that FHLBanks performed better during the financial crisis than did European covered bonds: the banks’ debt spreads widened less and the banks themselves kept making advance loans while covered bond issuance shrank.
Fair enough, and it’s true that the FHLBs remain an important part of the housing finance market, but they weren’t exactly without their own crisis-related problems.
There’s a reason the Treasury department’s new proposal to overhaul the housing finance system included a reform of the FHLB’s, restricting the size of their lending advances and investment portfolios:
Prior to the crisis, the FHLBs suffered from inadequate regulatory oversight, and were allowed to build large investment portfolios that subjected them to excess risk, while providing concentrated funding to banks engaging in unsound business practices. Today, eight of the twelve banks are under regulatory orders with respect to their capital or have voluntarily suspended dividends or the repurchase of excess stock.
Because each of the twelve FHLBs is also liable for the losses of other FHLBs, additional losses could adversely affect the entire FHLB system, damaging the mortgage finance market and potentially constraining access to capital for financial institutions. Reforms to the FHLB system are necessary to restore its important primary role of providing a stable source of mortgage credit for financial institutions of all sizes.
Nor are we really sure that the performance of the covered bond market in Europe says much about how a hypothetical (and much, much smaller) US covered bond market would behave in a crisis.
Regardless, covered bonds aren’t meant to replace the FHLBs — nobody has suggested that — but to provide a complementary source of financing during a time in which the roles of Fannie and Freddie will be wound down and the activities of the FHLBs themselves restricted.
The FDIC’s limited recovery
First, recall that there’s nothing right now legally preventing US banks from issuing covered bonds — in fact, it’s been done before. (Only twice, before the crisis, but still.) The reason they don’t issue more is that covered bonds here lack the secured creditor protections that they have in other markets.
In the event that the issuing bank goes under and is placed into FDIC hands, the FDIC right now has a number of options — including taking over the pool and paying investors the lesser of par or market value. The idea that investors in these bonds could lose access to the underlying collateral sort of negates the whole point of covered bonds.
Via Moody’s, here’s how the new proposal addresses this:
The legislation avoids exposing the entire cover pool to market-value risk at once following an issuer default. For bank issuers, it requires the FDIC to either sell the entire covered bond program to another issuer within 180 days, during which time it must pay the covered bonds and comply with all the issuer’s other obligations under the transaction documents, or transfer the cover pool to a separate legal estate to wind down gradually. Under the wind-down approach, the estate’s administrator only need sell assets when the cash generated by their natural amortization is insufficient to make scheduled payments to investors.
This wouldn’t leave the FDIC empty-handed, but clearly the corporation would rather have more options to maximise its recovery rather than fewer. Perhaps, as Indiviglio notes, there are further steps the FDIC could also take, such as charging banks that issue covered bonds higher assessments.
But to be blunt, there should be a test of relative importance here, and we agree with Felix’s take that solving the mortgage financing problem is simply more urgent than protecting the FDIC’s access to the cover pool.
Meanwhile, we suspect the FDIC also isn’t too keen on this part of the law:
The legislation promotes consistency among different types of issuers since, unlike the prior bill, it charges a single regulator to write the regulations. The legislation directs the Secretary of the Treasury to establish a regulatory oversight program covering objective standards for approving new programs, eligibility standards for assets, and rules for safe and sound asset-liability management. The Secretary of the Treasury also sets the minimum overcollateralization levels for covered bonds backed by each of the eligible asset classes.
What about the collateral?
In the US, HR 940 actually would allow multiple asset types to be eligible as covered bond collateral, but each cover pool is restricted to a single asset type. So mortgage loans can’t be in the same cover pool as small business loans, which can’t be mixed with credit card receivables, etc…
The main issue cited in the FT article about covered bonds in Europe is transparency:
But fund managers at M&G warned that investors might be allowing the perceived safety to blind them to a lack of transparency about the underlying loans.
“While many of these assets are exceptionally robust, too few let you see inside,” said Richard Ryan, senior credit fund manager at M&G. …
Many of the bonds are issued under strong legal frameworks, but even this might not be enough, M&G warned.
“Even with the best structure in the world, if the underlying collateral is poor, there’s not much benefit to being able to get your hands on it,” said Tamara Burnell, head of financial institutions’ analysis at the fund manager.
True, and such skepticism is always healthy — though remember that the way covered bonds work is that bondholders get paid even if the collateral goes to pot (unless, of course, the bank itself also goes broke).
But here we’ll re-emphasize the hugeness of the covered bond market in Europe, where it is of the order of some $3 trillion. It is now approaching the point where some analysts worry it could even start crowding out senior bank debt issuance (with good reason).
These are good reasons to make sure the US banking market never becomes over-reliant on covered bonds, but such fundamental issues just aren’t relevant yet in a country where the covered bond market hardly even exists.
Issues related to collateral quality are important, but they should be addressed by making sure that regulators (Treasury, mostly) write smart asset eligibility standards rather than used as an argument against trying covered bonds at all.