The Impact of Risk Retention Regulation on the Underwriting of Securitized Mortgages

By contrast, Willen (Willin 2014) argues that the financial crisis was, in part, exacerbated by intermediaries having too much exposure to real estate markets on their balance sheet and thus risk retention requirements, to the extent that they would have added more exposure, would be misguided.

Fitch (2017) estimates that only 25% of the roughly 2 billion in residential mortgages that would be subjected to risk retention requirements were securitized in the two years (2014−2015) before risk retention requirements were implemented. The Housing Finance Policy Center (2017) estimates that residential mortgage origination during these two years was approximately 3 trillion. Thus, risk retention rules impacted only 0.016% of the residential market (0.016% = 25% x 2 billion / 3 trillion).

Although outside the scope of this study, it is possible that the implementation of risk retention regulation, by making mortgage underwriting less borrower-friendly, has made previously originated mortgages more likely to default when borrowers attempt to refinance loans that had been originated with looser standards.

Demiroglu and James (2015) provide a useful overview of the issues associated with regulating risk retention.

The debt-service coverage ratio measures the ratio of the income generated by the property (through rents collected, etc.) to the debt service required by the loan. Thus, higher values of DSCR imply, all else equal, a safer loan. This can be thought of as the inverse of the debt service-to-income (DTI) commonly used as an underwriting metric in residential mortgages.

Defeasance requires a borrower seeking to prepay a securitized loan to place Treasury securities into the pool in an amount that would generate the originally promised principal, along with interest payments.

See Department of the Treasury (2014).

The third-party purchaser must specifically negotiate for the purchase of such first-loss position, holds adequate financial resources to back losses, provides due diligence on all individual assets in the pool before the issuance of the asset-backed securities, and meets the same standards for risk retention as the Federal banking agencies and the Commission require of the securitizer. The 5% risk retention requirement can be satisfied if up to two (B-piece) investors purchase the riskiest 5% (by market value) of the securities offered on a pari passu basis and hold these securities for at least five years. See Department of the Treasury, Office of the Comptroller of the Currency, 12 CFR Part 43, Docket No. OCC-2013-0010 page 170.

As part of the pool formation process, B-piece investors could exert pressure on the sponsor in terms of the specific loans being placed into the pools. For example, during pool formation, prospective B-piece investors would be provided details regarding the loans that the sponsor wishes to securitize. The B-piece investor also reviews more detailed information on the ten largest collateral loans, which typically total 50% of the proposed issuance, by balance. This additional information includes the major tenants of the commercial property and the expiration schedules of the property’s significant leases. B-piece investors submit bids to the sponsor, but the bids contain not only a price at which the investor is willing to pay for the riskiest tranches of the deal, but also various stipulations, rights, or flexibility that could affect a sponsor’s profitability on a given transaction. Examples of these non-price terms include the offer to buy a transaction if a certain loan is removed from the pool, or the right to remove a certain number of loans deemed to have excessive risk (called “kick-outs”). These kick-out rights are one way that B-piece buyers could ultimately influence the underlying collateral pool, although during the years immediately prior to the financial crisis, such kick-outs were rare.

Technically, the pooling and servicing agreement of the securitization would typically grant the “controlling class,” which is the security holder in the first-loss position, the right to appoint the special servicer, the institution that controls the workout process.

Regulatory agencies reasoned that, after a five-year period, the quality of the underwriting would be sufficiently evident that the initial third-party purchaser or, if there was no initial third- party purchaser, the sponsor, would suffer the consequences of poor underwriting in the form of a reduced sales price for such interest.

This exemption lasts while they operate under the conservatorship or receivership of the FHFA with capital support from the US Government.

The borrower would have been required to have a DSCR of at least 1.25x for qualifying multi-family property loans, 1.5x for qualifying leased QCRE loans, and 1.7x for all other commercial real estate loans. The loan would have been required to have either a fixed interest rate or a floating rate that was effectively fixed under a related swap agreement. The loan documents also would have had to prohibit any deferral of principal or interest payments and any interest reserve fund, resulting in excluding interest-only loans from qualifying as QCRE loans. QCRE loans further have a maximum amortization period of 25 years for most commercial real estate loans, and 30 years for qualifying multi-family loans, with payments made at least monthly for at least 10 years of the loan’s term. Furthermore, payments made under the loan agreement would be required to be based on a straight-line amortization of principal and interest over the amortization period (up to the maximum allowed amortization period, noted above). The minimum loan term could be no less than 10 years and no deferral of repayment of principal or interest could be permitted. The combined loan-to-value (CLTV) ratio for first and junior loans for QCRE loans are required to be less than or equal to 70% and the LTV ratio for the first-lien loan be less than or equal to 65%; or that the CLTV and LTV ratios be less than or equal to 65 and 60%, respectively, for loans with valuation using a capitalization rate below a certain threshold.

Fabozzi et al. (2015) estimates that 3.58% of all non-agency securitized commercial mortgages between 1997 and 2015 would satisfy the QCRE standards according to regulators’ re-proposal for risk retention guidelines, which were little changed before the final rules.

For the Non-Agency deals, the Supplements also contain information on the major leases within each property.

Standard errors are clustered by both origination month and originator.

In specifications not shown, lagging the interest rate variables to reflect the potential for advanced interest rate “locks” had little effect on the coefficients estimated.

There are 98 unique originators, 13 property types, and collateral located in all 50 states and the District of Columbia.

This suggests that B-pieces sell for a price approximately equal to 50% of face value in order to achieve 5% of deal proceeds, since most other securities sell near par.