Agency MBS: Different tools?

Research evidence excerpt

Agency MBS: Different tools?

r than in an-all out effort to get

mortgage rates modestly lower ahead of the midterms. Perhaps of note, it looks like the

GSEs added far more Treasuries in the quarter than MBS, likely in an effort to stabilize their

net income (though it also means that their duration gaps continued to expand).

Figure 3ThesecondquartersawsignificantUSTpurchasesatbothGSEs,thoughtheretainedmortgageportfolioactivitywascomparativelymuted is a quick summary of the various disclosures across the investment and deriva-

tives sections of the quarterly reports, with the differences in categories between Fannie and

Freddie reflecting the granularity available from each GSE. Fannie became net short anoth-

er 4.5bn TBA in the quarter, but it’s harder to make heads or tails of Freddie’s derivative

reporting (which adds notional amounts for longs and shorts). Of note is the pickup in Trea-

sury holdings, which seem likely to have been a driver of the increasing duration gaps (Fig-

ure 5…thoughitsemsthatTreasurypurchasesmayhavepropeledtheGSEdurationgapshigher). At the risk of parsing the 10-Qs with too much precision, we did notice that Fannie

added the phrase “such as U.S. Treasury securities” to this paragraph (which was otherwise

largely unchanged from their first quarter 10-Q):

Historically, our interest-rate risk management strategy focused primarily on maintain-

ing an asset duration closely matched to our liability duration, net of derivatives, to

minimize exposure to interest-rate movements. Beginning in the fourth quarter of 2025,

we adjusted our interest-rate risk management strategy to balance this objective with

the goal of managing the volatility of our earnings associated with short-term interest

rate changes.