Submitted by Wulf via One Easy Trade blog [4],
In late 1992, just as the Fed was putting the finishing touches on a three-year easing cycle that took the Fed funds rate from 9.75% down to 3%, the 2s/5s Treasury curve peaked at a high of 160 basis points. In the summer of 2003, when the Fed completed another large easing cycle, this one taking the funds rate from 6.5% to 1% over 30 months, the 2s/5s curve peaked at 161 basis points. And in March, 2010, as the Fed made the final purchases of its first quantitative easing program after cutting rates to the zero bound 15 months prior, the 2s/5s curve peaked at 162 basis points.
Today, with the Fed once again in the late stages of an easing cycle (affectionately dubbed QE3), we find the 2s/5s curve steepening yet again, hitting its widest level in 30 months at more than 135 basis points on January 1. Given this curve’s uncanny track record, one can’t help but wonder if it will ring the 160 bell yet again when the Fed completes its final purchase operation of QE3. With tapering now upon us, and the end of QE3 almost in sight, one might even be tempted to sell a few 5s against twos in the hopes of catching that final 25 basis points of steepening.
A closer examination of that trade, however, reveals that one might be better off doing just the opposite.
First let’s consider the timing: At the December 18th FOMC meeting, the Fed announced a $10bn reduction in monthly asset purchases, and if it continues to reduce asset purchases at that pace going forward, it should be done by the end of the year. If, however, improving economic data compel them to reduce asset purchases at an accelerated rate of, say, $15 billion at each subsequent meeting, they could be done as soon as September. This would be the more aggressive timeline and the best-case scenario for a 2s/5s steepener.
Running a simple horizon analysis of a duration neutral 2s/5s expression using current on-the-runs Treasuries, and assuming a 131 basis point spread at entry and a 160 basis point spread at exit on the September 17 FOMC meeting date shows that the spread widening would net you almost exactly one point in capital gains as the 5yr sells off, but would cost you almost as much in negative carry, leaving you no better than if you had simply bought 2s outright and clipped the coupon for nine months.
And if the Fed instead takes a bit longer to wind down its easing campaign, or puts the taper on hold in response to softer inflation data or some unforeseen shock, you’ll want to be long the 5-year point, not short.
Moreover, since the fed funds rate is likely to remain pinned to the floor for a “considerable time after the asset purchase program ends”, the front end of the Treasury curve is likely to stay fairly well anchored for quite some time as well. This may then also place a soft cap on 5-year yields, which one can expect to rise no more than 160 basis points over 2s. So if 2s trade up to, say 60 bps (already higher than they’ve been in years) and 2s/5s reach their 160 basis point peak, you’re looking at a 2.20% 5yr yield, up 45 basis points from current levels. As long as the front end doesn’t come completely unhinged, you now know your likely downside for the next 9 months or so, and your carry is still about as good as it’s been in years.
Investors find great comfort in being able to quantify their downside, and they also like positive carry in a low-yield environment. As such, I would have to think that 5s are getting close to a strategic buy, and I would use any backup towards 2% (or 160 over 2s) to get long.