Key Highlights

• Repo tightened last week, as expected, and we expect similar tightness at the end of September
• Usage of the SRF was an encouraging sign that the Fed’s backstop is effective in calming the water
• Foreign selling of USTs has resumed – at both the front and back ends of the curve

View chart pack

Quarter-end tightness again, but not necessarily structurally tighter funding markets

EXHIBIT #1: MID-MONTH REPO TIGHTNESS

_{Source: BNY Markets, Bloomberg}

Mid-September, as expected, posed some challenges to funding markets, especially the corporate tax payment date last Monday, September 15. As Exhibit #1 shows, the general collateral repo spread versus the rate of interest the Fed pays on reserves turned positive, at 10bp. The Fed’s standing repo facility (SRF) was utilized for $1.5bn in short-term funding that day. Afterwards, as the exhibit reveals, markets experienced a degree of relief, and by the end of the week were well under control, with last Friday’s spread back to –5bp. The tightness at the beginning of the week was not unexpected, and the swift normalization thereafter – during a week which featured the first FOMC rate cut in a year – was encouraging.

With reserves declining – also as expected – to just over $3tn currently, we still look to the September quarter-end as another test of liquidity and funding conditions. Examining Exhibit #1 further, we see that quarter-end this past June also saw materially tighter conditions at the time. Indeed, the SRF saw over $11bn in usage on June 30. It is very possible that we’ll see a repeat this month, but again, that wouldn’t be unexpected. To be sure, spreads at about +10bp would be a far cry from periods of true repo stress, such as the infamous September 2019 episode or during March 2020 in the face of the COVID lockdowns.

We’re particularly encouraged that the SRF has seen usage on days when funding is strained, suggesting that our concerns about any associated stigma discouraging its use were overly pessimistic. In reality, even the $1.5bn of SRF usage last week was a relatively small amount given daily repo volumes, but its presence and the willingness of at least some counterparties to utilize it seemed to have had a stabilizing effect on markets.

The decline in reserves over the past several weeks (after a recent high of $3.4tn in mid-July) has been entirely anticipated as well, given the over $600bn in T-bill issuance and the continued depletion of the reverse repo facility in recent weeks. Money funds (MMFs) have altered their asset allocation away from repo lending to T-bill purchases, a state of affairs that was the norm before the middle of this year when  the debt ceiling prohibited much new issuance from the Treasury, forcing MMFs to lend their substantial cash holdings in funding markets.

EXHIBIT #2: RESERVES STILL FAR AWAY FROM REAL REPO STRAIN

_{Source: BNY Markets, Bureau of Economic Analysis, Federal Reserve Board of Governors}

In September 2019, reserves had fallen to around 7% of nominal GDP (and under 17% of total banking system assets), whereas they currently stand at 10% and 22%, respectively. This suggests they can fall a bit further without leading to a reprise of significant market strain. This doesn’t mean that moving forward, money markets won’t trade tighter, a new normal in funding markets. However, at some point, we will be testing the frontier between an abundant reserve regime presently to a merely “ample” regime. As the Fed has expressed, we won’t know precisely a priori what the quantity of reserves will be at that point, but we will gain guidance from – among other things – the state of the money markets. While it is, as we have stated above, that quarter-end will see a challenge, it remains to be seen if this would only be a calendar effect or the beginning of a test of that frontier.

A resumption in cross-border UST sales, as yields slowly creep higher

EXHIBIT #3: COALESCING DOTS

_{Source: BNY Markets, Federal Reserve Board of Governors}

Since the Fed rate cut last week, which was widely anticipated, the 10y bond yield has risen some 11bp, from 4.03% the day before the announcement to almost 4.14% as of this writing. We attribute some of that rise to Chair Powell’s marginally hawkish press conference, in which he warned that inflation risks have not receded, and the move to lower the funds rate was an “insurance” cut, rather than an overt resumption of a sustained tightening cycle. We think that the Fed will indeed cut rates at each of the remaining two meetings this year, in agreement with the updated Summary of Economic Projections (SEP), also released last Wednesday.

We note that nearly half of participants (nine out of 19) saw rates above 3.75%, suggesting that the Committee is in broad agreement on a restrained pace of policy loosening for the remainder of this year, while one dot was an outlier, advocating for a policy rate below 3% is a clear outlier (see Exhibit #3). There is some merit to the argument that this restrained pace has had something to do with the rise in yields at the back of the curve.

EXHIBIT #4: CROSS-BORDER UST SALES RESUME

_{Source: BNY Markets, iFlow}

However, even before the FOMC, we have noted a resumption in cross-border UST sales, as Exhibit #4 shows. Most of the selling we have observed has taken place at both ends of the curve, at the very front of the curve in the 0–1y maturity bucket, and the long end of the curve in the 7–10y and 10y+ segments. We can’t place our finger on the reason for these foreign sales but suspect that this is an indication of concern that the Fed is moving to accommodate the labor market – which we agree is loosening significantly – and leaving inflation concerns aside for the time being, despite Chair Powell’s admonishment that inflation remains above target and could rise further.

Such overseas selling could also be responsible in part for the gradual and, so far, muted, rise in yields. On the margin, foreign buying or selling can affect these yields and is something we always keep our eyes on. The barbell-like cross-border selling suggests that the preferred part of the curve is in the belly, and we agree. The intermediate maturity segment of the curve is also, in our view, most sensitive to growth concerns, and we have seen buying at these maturities.

Chart pack