Muni Yield Curve: Flatter but Not Imitation

Boston - The municipal and the Treasury yield curves are different. This is never more evident than when the Treasury curve inverts.

There have been multiple historical incidences of Treasury 2's to 10's inversion in the last quarter century (when 2-year Treasury notes yield more than 10-year Treasury notes). However, since the municipal 2-year AAA benchmark measured by Thomson Reuters Municipal Market Data (MMD) began on Jan, 3, 1995, the municipal curve has never seen a 2's to 10's inversion.

In fact, the municipal curve has been very resistant to inversions of any sort. Going back to 1995, there have been only 40 trading sessions when the 2-year benchmark yield has exceeded the 5-year benchmark (0.45% of the time), and the maximum inversion has been 3 basis points. In comparison, the current inversion between Treasury 2's and 5's has persisted for 89 of this year's 92 sessions, and the maximum Treasury 2's and 5's inversion was 154 basis points in 2010.

We would argue that various factors explain the persistence of a positively-sloped municipal curve. They include investors' preference for maturities inside of 10 years, issuance weighted toward the longer maturities, and the dominance of retail investors who tend toward buy-and-hold rather than continuous repositioning.

Flatter is a challenge

Based on historical precedent, we think it is unlikely that there will be a muni curve inversion, but there has already been a curve flattening. The current MMD benchmark spread, between 2's and 10's, is 38 basis points. This is 1.35 standard deviations flatter than the average spread of 127 basis points since 1995 (standard deviation measures the dispersion of a data set relative to the mean, or average). The current curve presents different challenges than the average historical curve in regards to positioning.

Given that the curve has flattened, we wanted to explore different hypothetical portfolio structures to assess how they might respond, assuming various market scenarios/changes.

Specifically, we constructed three portfolio structures: a 1-10 year equally weighted ladder, a barbell composed of 1-2 and 9-10 year buckets, and a bullet composed of 4, 5, 6 and 7 year buckets (see "Definitions" at bottom of this post).

We created these three hypothetical portfolio structures (ladder, barbell and bullet) using a pool of bonds that met our credit standard (only A- rated and above) and have traded within the last calendar quarter. We chose 1-10 year maturities because this is the range of the Treasury inversion that investors are focused on.