For those interested in the plumbing of the financial system and thinks like RRP (Reverse Repos), I always find ZP of CSFB — some like him and some like to hate him — to be insightful.
Not that I AGREE with everything he says but rather I know what I don’t know about the nuts and bolts — sorry for mixing my ‘trade’ references here — and this part of the market.
The STIR community got so caught up with the fierce re-pricing of OIS curves and the imminent threat of QT (the Fed soon letting its balance sheet shrink) that hardly anyone asks about the adjustments to the Fed’s administered rates that will accompany the first interest rate hike. We discuss two scenarios below.
First, the Fed adjusts the target range for the fed funds rate to 25 to 50 bps by adjusting the rate on the o/n RRP facility 20 bps higher to 25 bps, and the rate on the o/n RP facility 25 bps higher to 50 bps. IOR would be at 40 bps – or 10 bps below the top of the target range like it is currently. Effectively, these changes would widen the repo-corridor from 20 bps currently to 25 bps.
Importantly, in this scenario, the Fed aligns the RRP rate with the bottom of the target range as opposed to setting it 5 bps above it as currently. We think such an adjustment is justified with liftoff, as once we move away from the ZLB, money funds don’t need “social” support from the Fed to cover variable costs (adjusting the RRP rate from zero to 5 bps was social support plain and simple).
Second, the Fed could also adjust the administered rates so that the IOR rate forms the top of the target range (like it did from 2015 to around mid-2018) and everything else is set relative to IOR. In this scenario, IOR would be hiked 35 bps to 50 bps, RRP would be adjusted 20 bps higher from 5 bps to 25 bps (like above), and the rate on the o/n RP facility would be adjusted to 75 bps, such that we have a repo-corridor that’s 50 bps wide with IOR in the middle.
Why is the second scenario not unlikely?
First, IOR was the top of the target range in the past. The Fed cut IOR relative to the top of the target range to make banks lend more in the repo market when GC repo was trading above the fed funds rate in 2018 and 2019, and then during the pandemic we got stuck with an IOR rate “structurally” below the top of the target range, and with the funds rate trading closer to the bottom of the range than to the top (south of the mid-point of the range, not north). But maybe the normal state for IOR is to serve as the top of the target range, and now that it’s time to tighten fast (rates and QT), it’s reasonable to expect that the Fed will want to set IOR and see the funds rate higher within the range.
Second, it’s healthy to have a 25 bps spread between the IOR and RP rates (as opposed to only 10 bps currently) so bank portfolios can police repo rates before the Fed starts to lend – the Fed is a lender of last resort, not first resort.
Whether the first hike widens the IOR-RRP spread to 15 or 25 bps from 10 bps, the fed funds-SOFR basis should be wider than the 3 bps implied by futures.