Fannie and Freddie’s Credit Risk Transfers

The Urban Institute’s Housing Finance Policy Center has released its February 2017 Housing Finance at a Glance Chartbook, always a great resource for housing geeks. Each Chartbook highlights one topic. This one focuses on GSE credit risk transfers, an important but technical subject:

The GSE’s credit risk transfer (CRT) program is growing and tapping into a more diverse investor base, reducing the costs of CRTs and improving liquidity in this market. At the same time, the continued reliance on back-end transactions is cause for concern
.
Freddie Mac‘s first two capital markets CRT transactions of 2017 have been different from previous Structured Agency Credit Risk (STACR) transactions in one important way. Unlike the pre-2017 deals, in which the first loss piece (Tranche B) was 100 basis points thick, the first loss piece (Tranche B2) in the latest transactions is only 50 basis points thick while second loss piece (B1) is also 50 basis points thick. Splitting the old B tranche more granularly in this manner is a noteworthy development for a few reasons.
Although this is hardly the first improvement the GSEs have made to their back-end CRT execution, it is an important one. Splitting the offering into more granular risk buckets will force investors to price the tranches more accurately, thus facilitating more precise price discovery of credit risk. More granular tranching will also help increase the demand for STACR securities. Investors who were previously willing, but unable to invest in the B tranche because investment guidelines prohibited them from taking first loss credit risk will now instead be able to invest in the second loss B1 tranche, which offers a higher expected returns than the previous second loss tranche (M2). Growing and diversifying the investor base is important because it makes the bidding process more efficient and minimizes the cost of risk transfer for Freddie Mac and the taxpayer. A larger, more diverse investor base also bodes well for the liquidity of the CRT market, which is still in its infancy.
Clearly, these innovations are important steps towards improving the efficiency of back-end CRT. But at the same time, they must be viewed in the context of the broader objectives of credit risk transfer and housing finance reform which have near unanimous support: reducing taxpayer risk, passing the benefits of CRT on to borrowers, facilitating broad availability of credit through the economic cycle, ensuring adequate access for lenders of all sizes, and promoting a variety of CRT executions, including at the front end to facilitate an understanding of which programs are most favorable under which circumstances.
Although the GSEs have experimented with front end mechanisms like lender recourse and deeper MI, these transactions have been few and far between, and with very little transparency about pricing and other terms. But more importantly, the GSEs’ continued and significant reliance on back-end capital markets transactions doesn’t put us on a path towards achieving some of the program objectives outlined above. This matters because it signals that the GSEs’ current strategy for credit risk transfer, which revolves largely around the success of back-end transactions, may ultimately keep the program from realizing its full potential. (5)
 So, all in all Fannie and Freddie are taking a step in the right direction, but it is just a small step on the road to housing finance reform.

Transferring Risk from Fannie & Freddie

The Federal Housing Finance Agency has posted its FHFA Progress Report on the Implementation of FHFA’s Strategic Plan for the Conservatorships of Fannie Mae and Freddie Mac. As its name suggests, it provides a progress report on a range of topics, but I was particularly interested in its section on credit risk transfers for single-family credit guarantees:

The 2014 Conservatorship Strategic Plan’s goal of reducing taxpayer risk builds on the Enterprises’ previous risk transfer efforts. Under the 2013 Conservatorship Scorecard, FHFA expressed the expectation that each Enterprise would conduct risk transfer transactions involving single-family loans with an unpaid principal balance (UPB) of at least $30 billion. The 2014 Conservatorship Scorecard tripled the required risk transfer amount, with the expectation that each Enterprise would transfer a substantial portion of the credit risk on $90 billion in UPB of new mortgage-backed securitizations. FHFA also expected each Enterprise to execute a minimum of two different types of credit risk transfer transactions. FHFA required the Enterprises to conduct all activities undertaken in fulfillment of these objectives in a manner consistent with safety and soundness. During 2014, the two Enterprises executed credit risk transfers on single-family mortgages with a UPB of over $340 billion, which is well above the required amounts. (14)

Risk transfer is an important tool to reduce the risks that taxpayers will be on the hook for future bailouts. The mechanism for these risk transfer deals are not well understood because they are pretty new. The Progress Report describes how they work in relatively clear terms:

The primary way that the Enterprises have executed single-family credit risk transfers to date has been through debt-issuance programs. Freddie Mac transactions are called Structured Agency Credit Risk (STACR) notes, and Fannie Mae transactions are called Connecticut Avenue Securities (CAS). Following the release of historical credit performance data in 2012, each Enterprise has issued either STACR or CAS notes that transfer a portion of the credit risk from large reference pools of single-family mortgages to private investors. These reference pools are comprised of loans that the Enterprises had previously securitized to sell the interest rate risk of the loans to private investors. The STACR and CAS transactions take the next step of transferring a portion of the credit risk for these loans to investors as well. Each subsequent credit risk transfer transaction is intended to provide credit protection to the issuing Enterprise on the mortgages in the relevant reference pool. (14)

The Progress Report provides more detail for those who are interested. For the rest of us, we may just want to think through the policy implications. How much credit risk can Fannie and Freddie offload? Is it sufficient to make a real dent in the overall risk that the two companies pose to taxpayers? It would be helpful if the FHFA answered those questions in future reports.