[quickthrowman]

You’re still exposing yourself to duration risk, right? What’s the average duration of your short-term MBS portfolio?

MBS bonds pay a risk premium for a reason, you’re virtually free of credit risk, but you’re assuming interest rate/duration risk (not particularly relevant if duration is low, I’m not familiar with the duration of short term floating rate MBSes)

Also, what happens in a Silicon Valley bank type scenario, let’s say you have lots of withdrawals and you have to liquidate at under face value. Who eats the loss?

[JackFr]

They said “short duration” not “short term”. The real risk is from spread duration rather than simple interest rate duration, and assuming they don’t lever up, that should be minimal.

The beauty of MBS floaters is that you’re relatively insensitive to prepayments because to a first approximation they’re always priced at par.

From an investor standpoint, as they say, you’re making maybe SOFR + 1.5%. That’s not a very sext return. But let’s say your banks repo desk is willing to finance the purchase at 5% down. Then you can lever up your investment 20x and now you’re a big shot making SOFR+30%, which is very sexy. But what’s that, when your lever like that, a tiny decline in price wipes out your entire stake (Welcome to 2008).

[sam_palus]

Very well put. And yes, to your point, we don't lever up.

And yes, SOFR + 1.5% isn't very sexy, but we're competing against existing treasury product that use money market funds and pay SOFR (or less, after fees). So that 1.5% is meaningful.

[quickthrowman]

Thanks for the informative reply, that makes sense.