The latest impediment to non-agency securitization stemming from the Dodd-Frank Act (DFA) is the proposal to create "premium capture cash reserve accounts" (PCCRAs). This provision, contained in the recent interagency document that proposed how the "risk retention" provisions of the DFA would be defined and implemented, would require the creation of a cash account to prevent structurers from attempting to monetize what they define as "excess spread." In my view, this proposal could be highly damaging to consumer mortgage lending and is premised on a flawed understanding of securitization practices. It also represents another in a series of impediments to the revival of the market for non-agency securitizations.

After outlining a narrow definition of "qualified residential mortgages" (QRMs) that are exempt from the 5% risk retention requirements, the release then describes a potential loophole:

By selling premium or interest-only tranches, sponsors could thereby monetize at the inception of a securitization transaction the "excess spread" that was expected to be generated by the securitized assets over time. By monetizing excess spread..., sponsors were able to reduce the impact of any economic interest they may have retained in the outcome of the transaction and in the credit quality of the assets they securitized.

The release goes on to state that:

(T)he Agencies propose to adjust the required amount of risk retention to account for any excess spread that is monetized at the closing of a securitization transaction. Otherwise, a sponsor could effectively negate or reduce the economic exposure it is required to retain under the proposed rules.

The solution proposed by the agencies would require sponsors seeking to structure MBS that don't meet QRM standards (and thus require the holding of a 5% risk retention piece) to fund a cash account. The account's dollar value would represent the difference between the proceeds of a securitization and 95% of its par value (accounting for the 5% risk retention already required by DFA). The reserve account would be junior in priority to the retained interest and would have to be held for as long as bonds in the structure remain outstanding, making its present value extremely low.

The most damaging aspect of the PCCRA provision would be the impact it would have on the pricing and issuance of non-QRM loans, which comprise a large share of prime loans that are not eligible for agency execution. Keep in mind that lenders' offerings are quoted as "points" given various note rate strata, typically displayed as a matrix with a variety of rate/point combinations. Relatively low rates are accompanied by positive points to be paid at closing, while higher rates are quoted with negative points (which are essentially rebates and are often used to fund both closing costs and lender add-ons and price adjustments). The offerings are calculated based on the net proceeds received for all of a loan's components when sold to investors, typically through a securitization vehicle. Since excess interest is effectively rendered worthless under the proposal, lenders' ability to offer reasonable rates with negative points would be limited, if not eliminated. This would in turn hurt cash-strapped borrowers that must allocate their available savings toward their down payments.

The proposal could, however, have an unintended impact on lenders' ability to offer lower-rate loans requiring positive points. The proposal's definition of "premium" is the excess of a transaction's proceeds less 95% of its face value. As I read it, this means that the sponsor would need to hold cash in the capture account against loans with net securitized values below par but above 95%. This would make these loans uneconomical to securitize. However, deeper discount loans could also be a problem under the DFA. A clause that revises the Truth in Lending Act states that loans can have a maximum of 3% in points and fees in order to be classified as "qualifying loans"; otherwise, they are treated as high-cost and potentially abusive financial products. Ultimately, the interaction between the PCCRA proposal and the qualifying loan provision may have the unfortunate result of limiting originators' ability to securitize any non-QRM loans.

The logic behind the proposal is itself highly questionable and betrays a naive understanding of securitization mechanics. The stated rationale is that structurers could figure out ways to eliminate or transfer their exposure to retained risk by monetizing "excess spread" through the sale of IOs and premium securities. However, the proposal's utilization of the term "excess spread" is flawed. While it's technically defined as the difference between the collateral's interest inflows and the tranches' coupon payments, the term typically is applicable to securitizations of "alty" and subprime loans that have note rates high enough to allow interest payments to be utilized as part of a transaction's credit support.In writing that the monetizing of excess spread "...created incentives to maximize securitization scale and complexity, and encouraged aggressive underwriting," the document is confusing prime and non-prime securitizations. As a result, it would impose a set of complex and burdensome requirements on sponsors in order to effectively outlaw structures that are not economical for most current production. This is especially important in light of the narrow definition proposed for QRM loans; the PCCRA provision would impact a large portion of issuance that would not qualify as QRMs but nonetheless have relatively strong credit attributes.

The PCCRA proposal also suggests that structurers could, by creating some form of IO security, render a horizontal risk-retention piece as economically worthless. In that case, a superior solution would simply be to mandate vertical risk retention, which would force sponsors to hold a pro-rata share of the collateral rather than a first-loss piece.

In fairness, the interagency proposal includes a provision that may allow sponsors to avoid the PCCRA requirement. The document states that "...the proposal would not require a sponsor to establish and fund a premium capture cash reserve account if the sponsor does not structure the securitization to immediately monetize excess spread..." However, it does not elaborate on how sponsors could avoid triggering the requirement when securitizing premium loans. Originators may be able to avoid the need for premium capture by holding excess servicing rather than securitizing it. (In senior-sub structures used to securitize prime loans, excess servicing is typically structured into WAC IOs. These are pro-rata interest cash flows stripped from the premium loans backing the transaction, and thus are neither senior nor subordinate in priority.) However, this raises a series of additional issues and concerns. Servicing is a volatile asset that is difficult to value, hedge and fund; the safety and soundness of the banking system is arguably not enhanced by forcing lenders to hold more servicing. It will also conflict with Basel III, which will create onerous capital requirements for institutions that hold servicing that accounts for more than 10% of Tier I capital.

This proposal typifies the difficulties that regulators are having, and will have, in effectively implementing the DFA's risk-retention concept without crippling the non-agency securitization market. These issues have important implications for both housing and the financial system. If the PCCRA concept is adopted, I expect that the origination of non-QRM loans will be limited to lenders able to hold them in portfolio. Given the limited appetite of depositories for residential mortgages, the proposal risks starving the housing market of funds at a time when home prices remain under pressure. Moreover, this will serve to concentrate the risk of housing in the banking system, rather than dispersing it into the much larger capital markets.

However, the most disturbing aspect of the proposal is that it represents another element in the evolving regulatory framework that is hostile to the securitization of residential mortgages outside of the agencies' auspices. There is nothing in this proposal that is consistent with the goal, as stated in the Obama administration's white paper on housing finance, of phasing out the GSEs and enticing more "private capital" into the system. Meanwhile, the housing market remains in a funk that is partially attributable to tight lending standards and originators'inability to securitize non-agency mortgage production; this situation will only grow more problematic when the ceiling on conforming balances is reduced this spring. It's time for regulators, legislators and the administration to stop paying lip service to the recovery of the non-agency MBS market and begin to outline and implement policies that support the widespread availability of mortgage funds to qualified borrowers.


Bill Berliner is a mortgage and capital markets consultant based in Southern California. His email is and his Web site is

Subscribe Now

Access to a full range of industry content, analysis and expert commentary.

30-Day Free Trial

No credit card required. Access coverage of the securitization marketplace, including breaking news updated throughout the day.