By the Numbers
DUS spreads should widen as rate hikes flatten the curve
Mary Beth Fisher, PhD | March 18, 2022
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.
The swap and Treasury yield curves are mostly flat from three out to 30 years, while shorter rates are expected to steadily march higher as the Fed raises rates throughout 2022. Fannie Mae DUS spreads are likely to widen as the curve flattens, with shorter spreads widening more than longer spreads. DUS spreads will not be alone but should widen in tandem with other products. The good news is that spreads on longer maturity DUS should widen less than spreads on shorter maturities, with lower price impact on longer term DUS. This creates an opportunity for investors to migrate out the curve, increase carry and improve total return.
Back to 2019
The most recent time the 2s10s curve was this flat to slightly inverted was in 2019. The Fed began an easing cycle in July of 2019 and gradually lowered the Fed funds target rate from 2.50% to 1.75% over three consecutive meetings. During the six months prior to the easing cycle, the 2s10s Treasury curve was relatively flat, range-trading from 15 bp to 25 bp, and it proceeded to invert modestly in anticipation of subsequent rate cuts in the latter half of 2019. The level of rates was also quite similar to where rates are now, with the 10-year Treasury yield traveling through a range from 2.75% to 1.50% from early 2019 to just ahead of the pandemic.
Exhibit 1 shows a graph of DUS spreads (5/4.5, 10/9.5 and 15/14.5) versus the slope of the 2s10s Treasury curve since 2016. The current 2s10s slope is 24 bp and the DUS spreads for each tenor are highlighted in red.
Exhibit 1: DUS spreads versus the slope of the 2s10s Treasury curve
Note: Daily spreads since 2016.
Source: Bloomberg, Amherst Pierpont Securities
The very wide spreads on the right-hand side of the graph, when the 2s10s curve was 100 bp to 150 bp steep, date to the first quarter of 2016. A trifecta of idiosyncratic events occurred at the beginning of the year that caused credit spreads broadly to widen across multiple markets and products; but spreads recovered by the end of the first quarter of 2016 and began a steady tightening trend to end 2017 at historically tight levels.
DUS spreads revisited and moved through those levels to new tights during the pandemic when the curve was steep. The recovery from the pandemic and the beginning of Fed hikes has flattened the curve and widened spreads – a trend that is expected to persist throughout 2022.
Impact of a flattening scenario
The overnight index swap (OIS) curve is currently pricing in 150 bp to 175 bp of hikes through the December 2022 FOMC meeting. Assuming that trajectory is realized, the Treasury yield curve could be essentially flat by year-end at 2.25%. Based on that scenario, the spread projections are as follows:
- DUS 5/4.5 spreads should widen by 28 bp, from 30 bp to just under 60 bp;
- DUS 10/9.5 spreads should widen by 13 bp, from 59 bp to 72 bp;
- DUS 15/14.5 spreads should widen by 8 bp, from 87 to 95 bp.
Exhibit 2: Projected price changes due to spread widening
Note: Assumes par priced DUS.
Source: Bloomberg, Amherst Pierpont Securities
The projected impact of the spread widening on DUS is shown in Exhibit 2. The longer duration DUS has the lowest price impact because their spreads are likely to widen the least. Longer maturity DUS spreads are already at their wides for the current slope of the curve, in part because as rates have sold off investors have moved into shorter duration securities. The effect has been to tighten spreads in short duration DUS and widen spreads in longer duration DUS. The larger spread duration of the longer maturity securities is offset by the significantly smaller amount of expected spread widening. Investors who are do not expect a further sell-off in the level of rates should begin moving out the curve now into longer duration DUS such as 15/14.5, because it is projected to have a smaller price decline than 5/4.5 as short rates rise and the curve continues to flatten. The higher coupons of the 15/14.5 also benefit total return accounts and leveraged accounts that finance their positions as the carry is better.
Possible spoilers
Arguably the Ukraine situation could create another credit event, which would almost certainly initially push spreads wider as opposed to tighter. Persistently high inflation could recalibrate long term inflation expectations, causing the long end to sell-off. That would likely trigger a pricing in of faster hikes by the Fed, tempering any trend towards re-steepening. The net result would probably be a parallel shift up of the curve amidst an ongoing flattening. A renewed, serious outbreak of Covid could stall the economy and put the Fed on pause. Otherwise, it is difficult to see a scenario for tighter spreads as the Fed continues to hike rates and the curve flattens.
Treasuries versus swaps
The same DUS spread versus yield curve slope analysis can be run using the swap curve. It doesn’t change the trend or the projected spreads at all. The swap curve is already flatter so it implies an inversion a bit sooner, but the spreads are projected to continue widening as the curve flattens if historical levels are used as a guide. LIBOR swaps are becoming passe, and new issues are being priced off either SOFR swaps or Treasuries, so using these curves as a base for projections is reasonable even though the DUS spreads shown are to the swaps curve.
DUS is not alone
DUS is not alone in its response to a flattening yield curve. In agency MBS, for example, a flattening yield curve in theory should also widen spreads. As the curve flattens, implied forward rates steadily move lower. If the curve actually inverts, forward rates drop below spot rates. And as forward rates drop, the probability of some round of refinancing in the future increases. Agency MBS needs to trade at wider spreads to compensate for that rising risk. Of course, the slope of the yield curve is not the only thing that determines spreads in practice. But it is an important influence.
A flatter yield curve also tends to coincide with wider swap spreads. And to the extent the market interprets a flattening curve as raising the probability of slower economic growth or outright recession, spreads in credit should widen, too.
Since no assets can trade in their own bubble, a flattening curve, all else equal, should soften spreads across the market.