• Cannibalizing Securitization. Some commentators have raised the concern that if legislation for covered bonds were adopted in the U.S. it could cannibalize securitization, particularly mortgage securitization. The comment was raised in a post-crisis environment in which no private sector mortgage securitization was occurring. Even today, more than five years after the crisis, there is little private sector securitization of residential mortgage loans.

Response. Investors in covered bonds generally do not purchase RMBS or other ABS. Covered bond investors tend to purchase sovereign debt and agency debt and find covered bonds to present similar risks. They are attracted by the dual recourse nature of covered bonds, the statutory framework for issuance and issuance by regulated financial institutions. Covered bonds present no convexity risk, no complex class structures and generally are protected by high quality collateral. At a conference in Barcelona before the crisis, HSBC, who was a heavy user of both RMBS and covered bonds, stated that the bank viewed RMBS and covered bonds as alternative means of financing, the choice of which for any particular financing was determined by a variety of market and regulatory factors. There is every reason to expect that RMBS and covered bonds would exist in the U.S. market also as viable alternative means of financing.

• Cost to the Treasury. The Congressional Budget Office has released an assessment of the reduced tax revenue to the U.S. Treasury under H.R.940 as a result of banks using covered bonds rather than RMBS to finance mortgage loans. This loss is estimated to be about $500 million over ten years. The Joint Committee on Taxation sent a revenue estimate to the House Financial Services Committee. David Camp, Chairman of the House Ways and Means Committee, sent a letter transmitting the JCT revenue estimate and enclosing an amendment to H.R.940.

Response. One can quarrel with many of the CBO assumptions in this analysis, but perhaps the most misleading aspect of the conclusions is the failure to acknowledge that, as a result of a change by the FDIC in calculating the assessment for each bank to support the DIF from being based on deposits to being based on all liabilities of a bank, issuing covered bonds instead of securitizing will result in the FDIC collecting higher assessment fees. The amendments proposed by the Ways and Means Committee do not appear to do any harm to H.R.940 as a covered bond statute.

• Encumbrance of Assets. Among the concerns on covered bonds raised by the Federal Deposit Insurance Corporation is a concern that cover pools for covered bonds will tend to consume the better assets of a bank, leaving the FDIC with poorer quality assets to satisfy the claims of depositors and increasing the risk to the deposit insurance fund. The FDIC has also expressed a concern that over-collateralization levels can become larger as a bank’s credit position deteriorates. They point to the nearly 50% over-collaterlization in the Washington Mutual covered bond program shortly before the FDIC took over the bank. The FDIC also points to high over-collateralization in some European covered bond programs.

Response. First, the higher levels of over-collateralization seen in some European covered bond programs are perhaps a misconception. These levels generally are due to structural aspects of the mortgage business in some jurisdictions which result in a bank’s entire portfolio being available to repay covered bonds. Second, many other forms of bank financing encumber assets, including central bank financing, repo financing, securitization, FHLB borrowings and various types of collateralized derivatives. To focus only on covered bonds tends as a policy matter to encourage the other forms of financing over covered bonds, which does not seem to be a wise policy choice. See a more complete analysis at A Battle over Collateral.

In the case of the Washington Mutual program, the high over-collateralization level arose from the peculiar structure used for that program. Under the terms of the program, in the event of the insolvency of the issuer the entire mortgage pool was to be liquidated and the proceeds were to be placed in a guaranteed investment contract. The proposed legislation would not conduct a fire sale of the entire cover pool at insolvency. Instead the assets in the cover pool would be retained in a separate estate and administered to pay the bonds in accordance with their terms.

The former general counsel of the FDIC has reported that prior to his departure the agency had agreed internally that the risk from encumbrance would be ameliorated by an 8% cap on issuance of bonds. This would certainly be an acceptable starting place for U.S. banks issuing covered bonds if it could be agreed to.

• Loss of Repudiation Power. The FDIC has complained that under the proposed legislation it would lose its traditional power to repudiate the obligations of a bank in resolving the receivership of a failed bank in the case of covered bonds. Instead, the cover pool would be separated from the failed institution and adminstered to continue making payments on the bonds.

Response. The FDIC has already lost its repudiation power in connections with several other means of financing bank assets: repurchase agreements, FHLB borrowings, central bank financings, securitizations and collateralized derivatives.

• Only the Largest Banks Benefit. One of the criticisms of covered bonds is that they would only benefit large, money center banks.

Response. The potential benefit to smaller and regional banks seems to be largely overlooked. Covered bonds would permit financing of commercial mortgage loans and loans to municipalities, assets for which smaller and regional banks do not currently have capital markets access.

Moreover, the European jurisdiction with the most similar banking system in terms of number of banks is Germany. Germany has more that 2200 banks. It is the smaller banks in Germany that provide the bulk of the residential mortgage financing and they fund that activity in the covered bond market. Germany has a two-tier covered bond market. The smaller banks finance in the domestic covered bond market, which does not attract international investors but which nevertheless continued to provide funds throughout the financial crisis. The purchasers in the market tend to be other local banks, local insurance companies, smaller retirement funds and other local investors who are able to focus on regional institutions. There is reason to expect a similar market to develop in the U.S.

• Adversely Affect the FHLBs. Several of the Federal Home Loan Banks have raised the concern that covered bonds could damage the FHLB system by reducing usage of FHLB borrowings and that as a result the FHLBs might not be able to provide important support and liquidity during the next financial crisis.

Response. This argument seems to ignore the effect on the FHLBs of securitizations and the support provided by Fannie Mae and Freddie Mac for mortgage loans. Although some institutions relied heavily on FHLB borrowings to finance their mortgage lending prior to the financial crisis, there was extensive use of securitization and the GSEs; nevertheless, the FHLBs moved rapidly to inject massive liquidity into the banking system as the crisis developed. While it is possible that reliance on covered bonds could reduce borrowings from the FHLBs, it is also possible that instead the reduction would, at least in part, be a reduction in RMBS issuance or GSE financings. Moreover, there is nothing to suggest that heavy reliance on the FHLBs prior to a crisis is an esential predicate to the ability of the FHLBs to provide important liquidity as a crisis develops.