Uncategorized
Step-ups drive non-QM debt costs down
admin | May 31, 2019
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.
Issuers of non-QM securitizations have started offering step-up coupons to address the negative convexity created when the issuer has an option to call the deal on specific dates. The call largely stops the debt from trading much above par, and the step-up similarly limits the debt from trading much below par. The step-up has improved ‘AAA’ execution for at least one issuer by as much as 20 bp and has significantly improved overall execution across the capital structure. In fact, pairing a step-up coupon with the issuer date call may have trimmed the weighted average pricing spread by as much as 25%.
What’s a step-up worth?
Deals issued by Invictus on the VERSUS shelf offer the cleanest measure of the impact of the step-up. Invictus first issued VERSUS 2019-1 with no date call and no step-up coupon and later issued VERSUS 2019-2 with both a 3-year date call and a 100 bp step-up on the ‘AAA’ through ‘A’ classes. Fortunately, the ARRW shelf issued deals that help filter out the impact of the step-up. ARRW 2019-1 came with no date call and no step-up, and ARRW 2019-2 came with only a 3-year date call.
Focusing for the moment on the simple case of the ‘AAA’, the VERSUS 2019-1 class came 5 bp tight to the ARRW 2019-1 class. The VERSUS 2019-2 class, however, came 25 bp tight to the ARROW 2019-2. Of the 25 bp difference, 5 bp arguably is the baseline spread difference between the two shelves’ ‘AAA’. The remaining 20 bp in better execution looks due to the step-up coupon.
Broadening the analysis to the full deal involves calculating a weighted spread for placing a consistent 93.75% of debt (Exhibit 1). The weighted average spread on the VERUS 2019-1 deal was 130 bp while the same measure on ARRW 2019-1 was 123 bp. Ostensibly the ARRW deal achieved better execution because that collar of risk is only comprised of ‘AAA’ through ‘A’ rated bonds whereas the same collar of risk in the VERUS deal is comprised of both investment and non-investment grade bonds.
Comparing the two latter transactions, the VERUS cost of issuance tightened by 32 bp, with the cost of issuance tightening to just 98 bp to sell the same corridor of risk as the earlier transaction, representing a 25% decrease in weighted spread. By comparison, the cost of issuance on the latter ARRW deal improved by just 10 bp to 113 bp or an 8% reduction in liability spreads. The VERUS execution on the second set of transactions was 15 bp better than the ARRW deal in spite of the fact that the corridor of risk sold included non-investment grade bonds while the ARRW deal did not. (Exhibit 1)
Exhibit 1: Calculating the cost of issuance across non-QM shelves
Source: Bloomberg LP, APS Note: Spreads and credit enhancement levels as of deal issuance. Cost of issuance is calculated as a spread over applicable index rate multiplied by weighted average life to pricing call assumption weighted by each tranche as a percentage of total risk sold.
The tighter spread on the second VERUS transaction looks to be driven more so by the step-up increasing the likelihood that the short-dated call gets exercised. Spreads tightened materially at the top of the capital structure on bonds affected by the step-up. Additionally, the deal was priced to the first date call as opposed to the earlier deal which priced to the deal’s 30% collateral clean up. This had the effect of curtailing the average lives on longer sequential mezzanine bonds by just over a year. Curtailing those cash flows contributed to the reduction in the overall deal level cost of issuance and likely tightened spreads on those bonds given the additional incentive to call the deal provided by the step up.
The step-up feature appears to have materially improved execution on the latter VERUS transaction. This feature should begin to pop up with increasing frequency in non-QM deals as a potentially low-cost option to issuers that can materially reduce their funding spreads across not only bonds impacted by the step-up but lower rated mezzanine classes as well.
Thoughts on relative value
A 20 bp tightening on ‘AAA’ execution based driven by the introduction of a short-dated call and coupon step-up may seem like a significant nominal tightening. However, the introduction of the step-up creates additional incentives for the issuer to call the deal up to a 100 bp increase in bond yields. The step-up coupled with the overall increase in cost of funds to the issuer resulting from the deal de-leveraging over time as higher cost liabilities becoming a greater proportion of the outstanding deal should create significant incentives for the issuer to call the deal. Given this, ‘AAA’ through ‘A’ rated bonds look and feel like three-year amortizing bullets at spreads of 85 or more over interpolated swaps. This makes for a compelling relative value against comparable duration ABS, CMBS and investment grade corporates.