Setting up for a bond pickers market next year
admin | December 18, 2020
This document is intended for institutional investors and is not subject to all of the independence and disclosure standards applicable to debt research reports prepared for retail investors.
Shorter investment grade debt took the lead this year in delivering the most efficient returns across mortgage credit, but the picture for next year looks a little different. Prepayment and issuer call risk look likely to limit returns in non-QM credit while some RPL and CRT picks have better prospects. Return of forbearance and excess spread could lift returns in legacy RMBS. It should be a bond pickers market next year.
Shorter investment grade debt led in 2020
Stacking up returns across mortgage credit this year shows that certain investment grade exposures delivered top results. Debt with shorter spread duration fared best as pro-rata ‘AA’ and ‘A’ tranches of non-QM deals and M1 classes of GSE Credit Risk Transfer deals printed the best Sharpe ratios followed by both RPL and non-QM ‘AAA’ rated tranches (Exhibit 1).
Exhibit 1: Short duration investment grade mortgage credit fared best in 2020
Conversely, legacy and more levered exposures to mortgage credit significantly underperformed on a risk-adjusted basis and in some cases even underperformed a high yield benchmark. Over the course of 2020 the Citi High Yield Index posted an average monthly return of 43 bp with a standard deviation of 0.046 and a Sharpe ratio of 0.31. Certain non-investment grade exposures to mortgage credit outperformed the index including B1 classes of CRT as well as a market weighted exposure to legacy subprime, Alt-A and prime credits outperformed a high yield benchmark.
Exhibit 2: Some non-investment grade exposures outperformed. high yield
Negative convexity limits prospects for non-QM
Within the various sectors of mortgage credit certain cohorts that underperformed in 2020 may continue to do so. While investment grade sectors of the non-QM market performed well, ‘BBB’ and non-investment grade rated tranches underperformed. And extracting attractive absolute or risk adjusted returns from lower rated tranches of non-QM trusts may be challenging into next year. Negative convexity in the sector looks to be on the rise both from elevated prepayments and declining delinquency rates, and investors in non-QM trusts are short both prepayment and short-dated call options. Those options look increasingly likely to be exercised by issuers if borrower performance continues to improve. These short-dated issuer calls effectively cap potential price appreciation as investors become wary of premium risk to a call scenario. Because tranches originally rated below ‘A’ in non-QM trusts are locked out, this insulates them to some extent from elevated prepayments but not the short-dated call option. If credit performance in the sector continues to improve, investors in lower rated classes will likely get called at par with limited potential to roll down the spread curve. On the other hand, if the trajectory of borrower performance stalls and the incidence of permanent loan modifications increases, credit support from both excess spread excess spread and hard subordination may erode, potentially driving prices of those locked out bonds lower.
Better prospects in RPL and CRT
Conversely, sequential structures that offer investors the ability to deleverage and roll down the spread curve should fare better into next year despite already tight nominal spreads across most mortgage credit exposures. Mezzanine tranches of re-performing loan securitizations should offer investors a reasonable source of total return into next year but that will be highly dependent on those deals continuing to deleverage via elevated prepayments and as such returns will likely not be uniform across the sector. Cleaner RPL deals backed by loans sourced from commercial bank or GSE portfolios with smaller amounts of securitized principal forbearance should exhibit faster prepayments giving investors more upside to deleveraging and potential ratings upgrades.
In CRT, spreads have compressed materially with new issue CRT deals trading back to spreads not dissimilar to those prior to the pandemic. Investors will continue to benefit from rolling down the spread curve in sequential structures but they can likely extract additional total return by identifying dislocations in spreads between mezzanine and subordinate tranches and rotating in or out of those credits as the basis between tranches widen or tighten.
One potential issue with regards to CRT performance into next year is the supply technical. While Freddie Mac has provided clear guidance with regards to 2021 issuance, questions around when and if Fannie Mae will come back to the market persist. If Fannie Mae remains on pause into next year this is likely a resoundingly favorable supply technical for CRT as Freddie Mac will likely only contend with modest amounts of additional supply from mortgage insurers and private issuers. However if Fannie Mae resumes normal issuance or potentially increases issuance on risk retained in the guarantee book while on pause it could create a supply technical that would cause spread widening across the capital structure.
A bond picker’s market in legacy RMBS
Legacy RMBS will likely continue to be a bond picker’s market into next year although some broad fundamental drivers should buoy performance. Elevated recoveries of forborne principal have persisted throughout the pandemic and appear poised to do so into next year, While the curve may steepen and rates may rise modestly next year, the short rates that determine floating-rate legacy liabilities should remain low. And as bond coupons remain depressed, the amount of excess spread available to legacy trusts should remain elevated. Mezzanine, subordinate and residual classes of legacy trusts with leverage to both forbearance recoveries and excess spread should outperform into next year.