September 27, 2022 | This week, The Institutional Risk Analyst is visiting Washington, DC, to speak with some folks on Capitol Hill about the evolving credit situation in the government loan market. We wrote about the impending implosion of the market for FHA and VA loans in our last comment (“The Bear Case for Ginnie Mae Issuers”), but we were fascinated by a comment on CNBC to the effect that the market was having "trouble" getting the VIX above 30.
The 30-year peak for the VIX Index was just shy of 60 in October 2008. The next major peak for the VIX was, strangely enough, March of 2020 over 53. In 2008, the assets of the Federal Reserve System was just shy of $1 trillion. Today the Fed’s assets are about $8.8 trillion, down from a peak just shy of $9 trillion. Yet the VIX has been strangely subdued even as actual volatility in the fixed income market has exploded.
We felt fully gratified by Fed Chairman Jerome Powell’s comment that the Fed would not be selling Treasury collateral or MBS from the SOMA. We’d reported the FRBNY research papers that said as much several months ago, but the frantic Buy Side mob continues to fret about whether Chairman Powell is going to wake up one morning and hit the bid. But we don't think Powell and FOMC members are looking to add to market volatility.
When the Fed of New York says “no realized losses,” we take that as biblical writ. Pay attention folks. The answer is there for the taking. We wrote back in July ("Questions for Chairman Powell"):
"Importantly, the paper assumes that the Fed will not realize any market losses on the SOMA. This suggests that there will be no outright sales of either mortgage-backed securities or Treasury paper in the SOMA. Is this now Fed policy? Somebody should ask Chairman Powell."
Truth is, the low coupon paper held by the Fed is pretty much unsalable. The Fed owns most of these GNMA 1.5% and 2.5% coupons, MBS that are so volatile and thus costly to hedge that even were the FRBNY to solicit bids, the market price would be weak and for very small size. Nobody wants these securities, either for the MBS or the servicing strip behind the security. Owning GNMA 1.5s is an excuse to lose money.
Take an example. If a loan in a GNMA 1.5% MBS goes into delinquency, the servicer must advance interest, principal, taxes and insurance until the distressed borrower gets back on track or the house goes to foreclosure. The advance on the Ginnie Mae security will cost at least SOFR +150bp, but the FHA will reimburse the servicer at the 1.5% debenture rate of the MBS. Great deal, huh?
For the same reason, the Federal Reserve is losing billions on the interest rate mismatch between what it earns on its portfolio and what is pays on reserves. As interest rates rise back to normal levels, the several trillion dollars in low-coupon MBS created by the FOMC constitute a ghetto of abnormal securities with low cash flows and commensurately elevated volatility. These securities may not be in the money in terms of refinancing the underlying loans for years, meaning that they will be a low-return, high risk portion of the MBS complex.
Remember, the par GNMA MBS today is over 6% for October delivery. If the on-the-run Fannie Mae MBS for delivery in October is yielding 6.25%, how high does the loan coupon need to be for the lender to make money selling the loan? Try a 7.5% coupon rate for the borrower.
One of the major negative factors in quantitative easing or QE is that lower yields mean higher price volatility for bonds, money market instruments and even whole loans. As interest rates rise, the rate of volatility in the markets will hopefully decline, but don’t be fooled by whether the VIX has risen above 30 or not.
The visible rate of volatility in stocks and bonds is higher today than when the VIX reached almost twice current levels. Until the FOMC raises interest rates to more “normal” levels, we don’t expect actual volatility to decline. Indeed, as we proceed with quantitative tightening or QT, we believe that the large body of abnormally low coupons will increase actual volatility and cause substantial market dislocation in the process.