On Covered Bonds, Collateral Crunches, And The Circular Logic Of Central Banks
Via Jeff Snider of Alhambra Investment Partners,
Funding In Spain May Get Worse
Since 2009, outside of the megabanks in Europe, the bulk of the rest of the financial system has been completely shut out of the unsecured financing markets - whether that was term debt or overnights (in both euros and dollars). This was and is particularly true for banks domiciled and operating in the periphery. That meant they had to turn to the secured lending markets using some form of “acceptable” collateral. As the crisis turned from US$ mortgage assets to European sovereigns, the ability of these banks to find and pledge sufficient collateral was impeded. Thus, the ensuing scramble for liquidity through secured lending pushed up the prices on what debt was fully acceptable to the wholesale markets (Germany, Switzerland, Holland, France to a lesser extent, etc).
One of the workarounds to this liquidity problem was the reclamation or retention of covered bonds issued by the Eurozone banks themselves. For those not familiar, covered bonds are simply debt securities sold by a bank that rank as most senior - they are legally and severably collateralized by that bank's “eligible” mortgage pool. Instead of a securitized vehicle where mortgage loans are "sold" off balance sheet to an SIV or other entity, the bank has to retain the mortgage loans as collateral for the covered bonds (thus the cover). In light of recent history and contrary to the last decade, retention is now believed to be more desirable. In addition, there are strict rules (which vary by country) about the amount of covered bonds that can be issued for a given eligible mortgage pool. Rules for Spanish Cédulas Hipotecarias, for example, limit the bank to issuing covered bonds no greater than 90% of the eligible mortgage pool (overcollateralization) for residential mortgages and 70% for Cédulas Territoriales (essentially euro-wide public sector loans).
In the event of a bank failure, a special conservator is set up that ensures covered bond investors receive all of the cash flows from the eligible pool before paying any other claims. Legally speaking, the eligible pool is given a separate class from the liquidation/bankruptcy assets. And if there are not enough cash flows from the eligible pool then the entire bank mortgage pool is attached and covered bond investors are given priority. There are additional covenants on the "quality" of the loans that are eligible for use in the covered pool and also limitations to NPLs in the pool (if a certain % of the eligible pool becomes non-performing, the bank is required to replace loans with credits that meet these covenants).
Given these safeguards there is a high element of security through collateral in covered bonds. In the wake of the funding crisis, the ECB has given covered bonds a leg up on other forms of collateral, putting covered bonds just below sovereigns (in good standing) in the pecking order of collateral regimes for its various unconventional funding measures. Because of this, coupled with the high cost of "good" sovereigns, lower tier banks have been using covered bonds to fund their operations in the place of lost traditional funding methods, including, in the case of PIIGS banks, retail deposits.
As the collateral shortage grew acute, Euro banks, Spanish and Italian in particular, began to buy back covered notes from private investors. Further, on new issues, they began to retain a healthy amount. Banco Popular, for example, has a total mortgage pool of about €47bln. The "eligible" mortgage pool runs about €27bln, meaning they are now limited to issuing covered bonds for at most 90% of that eligible pool, or €24.3bln (the actual limitation is less than 90% since only 52% of the total mortgage pool consists of residential loans). Through the end of Q1, Banco Popular had previously issued and "sold" €21.4bln in covered bonds, so as late as March this year the bank was already close to maxing out.
But "sold" isn't the right word for it. Out of that €21.4bln, the bank "retained" €9.2bln, or 43%, of those bonds, ostensibly pledging them to the ECB in probably one or both of the LTRO's. I have not seen any figures from the bank on how much they have bought back out of the 57% that were actually sold to private investors, but I can guess it was not insignificant. The bottom line is that the ECB is likely funding at least 34% of the eligible mortgage pool of the bank directly in various collateral schemes. Perhaps more importantly for future banking and liquidity conditions, Banco Popular has a defined upper limit in the amount of collateral it can issue on its mortgage pool in the form of covered bonds. Once it reaches that defined upper bound, where it is already close to exhausting this route, the bank will be forced to find a further alternate means for funding its existing loan portfolio. That is trouble for a bank in a country that saw 5% of its deposit base walk to some other locale in July and the potential activation and usage of the ELA in August.
In January of this year, UniCredit, Italy’s largest bank by assets, issued €25bln in covered bonds to fund its existing mortgage pool in the wake of the Italian banking distress. The bank also mentioned that it reserved the right to retain a portion of the issuance to pledge as collateral with the ECB (given that LTRO2 was only a month away and UniCredit was heavily involved in the unconventional monetarist scheme it is very likely that a large portion was retained). There is something very unnatural about these “retained” arrangements since the issuing bank is essentially retaining its own liability to pledge as collateral to the central bank to fund its assets. Can you really “own” your own liabilities? Since circular logic pervades the current realm of central banking, this is wholly unquestioned. In reality, retained covered bonds are just the accounting gloss on direct monetization of past and existing mortgage loans.
Given these atypical collateral dynamics that have taken root in Europe, and the temporary calm covered bond collateral provided, UniCredit’s covered bond sale last week may have sounded a new collateral alarm. The company was successful in placing €750mm in covered bonds (no word on how much was retained and how much was actually sold to actual outside investors), but it had planned on issuing as much as €1bln. The shortfall in actual placement was due to the lack of additional available mortgage credits in the “eligible” mortgage pool. It seems as if UniCredit has run into its upper limit.
UniCredit is not alone. In France, Société Générale (one of the banks rumored to be close to Lehman status on December 8, 2011) surprised the markets last week by tapping an unsecured funding arrangement instead of the expected covered bond sale - again, same story, not enough loans in the eligible pool.
On Tuesday of last week, Santander did the exact same thing for the exact same reasons.
For Santander and SG, the unsecured markets remain an option, though we can be fairly certain that the unsecured funding arrangements were far more costly on both a marginal and nominal basis. For UniCredit, Banco Popular and their periphery compatriots (Santander, despite the Spanish name is largely a British bank with Spanish exposure) there is no unsecured funding option, even at the senior levels. Collateral shortage is a very real issue and in many ways parallels the sovereign debt problems, but it also represents an element all its own - something that is not widely known or covered (pardon the pun). If banks are running out of eligible mortgages, they will be forced to turn to other means of raising funds, including asset sales. That offers a very good explanation as to why sovereign yield curves in Spain and Italy flattened and plateaued so dramatically in July – with covered bond issuance hitting against the upper bound Mario Draghi was forced to issue his blanket promise to try to calm unsettled funding markets.
This is all occurring at the same time some of the Basel III requirements are beginning to hit and capital reserves are going to be at a premium. Banco Popular, for example, is not yet at the 10% capital level to be in “compliance” with the January 1, 2013, Basel "observation" period. As it is, the size of the bank’s eligible mortgage pool in comparison to its overall mortgage pool is quite revealing about potential capital shortfalls. Taking into account other security and collateral arrangements, about €18bln of the bank’s mortgages do not qualify as eligible primary collateral for covered bonds. From that we can infer a lot about the “quality” of those loans, and thus the primary reason these banks are stuck in the covered bond collateral arrangements without other options.
The liquidity noose gets tighter and tighter as banks continue to exhaust funding avenues for simply maintaining their existing bank books. Since they can barely fund what they have now, expanding lending not just in the Eurozone (euro banks are/were heavy marginal lenders in dollar credit assets through the eurodollar market) is completely impossible - the "broken transmission mechanism" of monetary policy. Speaking in terms of liquidity, it is possible that as we reach this upper limit in covered bonds that the liquidity crisis will renew right where it left off before the LTRO’s (which renders a failing grade on the LTRO’s in terms of what they were supposed to accomplish, just ask CIF).
That might force the ECB to loosen collateral restrictions further (restrictions were adjusted not that long ago on June 22), but in the covered bond space such a move might further upset the marketplace since covered bond investors will not take kindly to losing collateral protections and covenants. As it was, the potential "bail in" solution for Spain floated earlier in the year coupled with the various bonds that have ended up at the ECB threaten to diminish the seniority of the covered bonds since the ECB refuses to take losses (the precedent set in the Greek PSI earlier this year). The more bonds pledged to the ECB the greater the potential for ECB erosion of overcollateralization and then cramming down losses, pushing covered bond investors down the capital structure to par with unsecured senior debtholders. The ECB and its dependents are painting themselves into a smaller and smaller corner.
Marginally, covered bonds as collateral to the ECB is an extremely important bridge holding the shaky liquidity system together as it is now. Despite all the monetary measures and collateral bypasses, the funding reality remains largely unchanged over the past year – private arrangements are little to non-existent. The two variables that have changed most dramatically are the willingness of retail depositors to remain in place given the dynamics of each bank’s undesirable loan book and the ECB’s willingness to take up the slack created by that deposit flight (through both unconventional collateralized arrangements and TARGET imbalances). Draghi and his rate caps have no impact here. If, however, the covered bond bridge is pushing up against its very real statutory and quality limitations, that might mean the ECB is now fighting a two-front funding war – retail deposit flight and collateral diminishment at the same time. In any event, the diminished potential of covered bonds for liquidity means that much more funding flexibility, one more funding source or marketplace, is no longer available to the second tier and below banks. That gets problematic given the exposure of top tier banks to these markets and these banks themselves (French banks, in particular, seem to have a lot of periphery-based subsidiaries that run into periodic funding and capital issues).
It is vitally important to remember how these plumbing issues came about in the first place. One of the primary lessons for central bankers coming out of decades of studying the Great Depression was the unfortunate liquidity conditions that saw “good” banks ruined with “bad” banks. That meant, for central bankers, the public could not be trusted with discerning between “good” and “bad” banks, meaning central planning in the banking system itself was actually more desirable than free markets. From that it was inferred that contagion could be avoided through centralized liquidity schemes. Therefore central banks could perform the public service of sorting good banks from bad through collateral arrangements – if a bank has enough “good” collateral on its books it must be fairly called a “good” bank worth saving.
In light of that original framework for central bank intervention, the twisting and distortion of collateral schemes belies that original intent. If banks possess enough “good” collateral to be salvageable, the opposite must also hold true. Banks that do not possess enough “good” collateral have self-selected themselves for extinction and resource re-allocation. That is the only interpretation that would preserve a central bank philosophy consistent with central bank goals of actual public/free market service. Anything outside those bounds is venturing away from true economics and monetary economics into politics.
Central banks and mainstream economists continue to fall back on the idea that the banking system and the global economy are one and the same, inseparable. Therefore the banks must be saved at all costs or we risk total economic chaos. Yet, that would only be true of the “good” banks. A market system needs to feature a self-correction that destroys “bad” banks lest they interfere with the vital task of intermediation in the real economy – “bad” banks give out bad loans so a healthy economy must contain only a minimal proportion of bad banks. Therefore, central banks have it exactly backward. The “bad” financial economy must actually be separated from the real economy lest we remain stuck in economic reversal and malfunction. Since these “bad” banks have shown themselves to be as such to central bankers by appealing to unconventional collateral schemes, the next step should actually be the easiest. Instead of perpetual liquidity problems, central banks could simply wind down these institutions and remain consistent with the public/free market philosophy they espouse.
Reality is, of course, far less straightforward. Central banks are not paragons of economic virtue despite their PR campaigns promising otherwise, and that is no more apparent than at the ECB. At least we can call a spade a spade in light of unending collateral problems. There is no economic argument for maintaining self-selected bad banks. Free markets demand their extinction. Anything short of that will result in escalating and perpetual liquidity and solvency crises until the real economy is freed from the yolk of bad banks and their dis-intermediation.
There is no real wonder as to why we have exactly that right now – the intrusion of politics done in the name of economics. The cost of this political intervention is malaise and re-recession. Markets, however, are far less conflicted (outside of interventions). The potential upper bound of the covered bond regime demonstrates once again that the intrusion of politics also has an upper bound.