Our medium term views

 Our key forecasts

  What we like right now

  The View

The week that will be: schools out for central banks

Global rates markets receive an August break from central bank meetings, except for the RBA next week. Most major development market central banks will not meet again until September. School is out for most central banks but not quite yet for markets.

Next week has a few risk events, esp US payrolls. We retain a bullish bias across most developed rate markets. August seasonals support this view (Exhibit 4). Detail below.

US: NFP is the key event. Our US economists expect 225k headline vs 175k consensus. If our economists are right, we might guess the market re-prices Sept FOMC meeting by 5bps (current 29bps to 24bps). It will likely take a 275k + strong earnings number to shift Sept to <20bps. If payrolls <125k Sept will likely price 35bps-ish and concerns will rise about the Sahm rule. ISM services & 3/10/30Y auctions are relevant but secondary.

CA: PMIs & employment due. Like US, a strong labor print is needed to shift BoC pricing.

EU / UK / JN: PMIs are relevant but unlikely to meaningfully shift central bank pricing; Aug PMIs are due before each region's next central bank meeting. In Japan, 10 & 30Y JGB auctions on Aug 6 & 8 will be watched for demand post BoJ JGB taper plan.

AU: RBA will discuss a hike but should remain on hold. We expected little forward guidance and a balanced assessment on the risk outlook.

The week that was: slimmer belly for summer

Most global rates declined with belly outperformance as markets pulled forward cut timing and lowered expected cutting cycle troughs. The moves were driven by data softening, a dovish Fed (Exhibit 5), and geopolitical risks (see July FOMC). The exception was Japan where the BoJ hiked and announced details of JGB tapering (see BoJ review).

BoE voted 5-4 to cut the Bank Rate by 25bps (as our economists expected), delivering a hawkish cut. To us, more than one further cut this year seems highly unlikely; Nov seems most likely timing. We continue to expect further dis-inversion in 2s5s part of the curve and revise our QT expectations to £100bn for the new "QT year" (see Rates - UK).

In the US, the Treasury signaled more bill vs coupon supply and funding risks are rising. The biggest risk is that higher repo could drive de-leveraging; this isn't our base case but the tail is growing (see Repo risks rising).

  Rates - US

•      Intermediate tenors likely to lead rate decline as first Fed cut nears
• UST bill shift => coupon delay; 5s30s 50+bps steep likely needs slowdown

No Fed pushback = stop ins

UST rates declined and the belly outperformed. Rates markets were focused on a modest softening of economic data (ECI, ISM, Unit labor cost, ADP) and no Powell pushback on Sept market pricing. On balance the rate moves are in-line with our core views. Last week we wrote: "cutting cycle trough will move lower as Fed cuts near; duration longs are best expressed in intermediate tenors". Price action this week agreed.

Positioning may have also contributed to the large US rate declines. Our weekly flows & positioning report has been flagging declines in fund duration exposure following the peak in June. Some fund managers appear to have faded the lower rate move. We were also seeing CTAs with room to add vs momentum. For detail see: weekly flows report.

Our core rate views: Duration = we continue to favor trading trades with a long bias expressed in the belly since the trough will move lower as the first cut nears. Curve = we hold our long end nominal 5s30s & real 5s10s positions; we expect more steepening in coming weeks. Front end = stay short front end spreads with higher repo risks. Inflation = long 30y B/E hedged with 1y1y inflation short. Spreads = short 30y spread; July refunding is a headwind but not a game changer (discussed below). Vol = lower vol bias with scope for intermediates expiries on the left side to underperform vs the right.

For the remainder of US macro we discuss reaction to Fed & UST refunding + questions about how far the 5s30s curve can steepen.

No pushback Powell = hard to fade the cuts

US rates interpreted July FOMC communications as modestly dovish, on balance. Chair Powell signaled the Committee is moving closer to cuts, mentioned that a cut could be on the table in Sept and focused on risk of further moderation in the labor market.

Powell did little to push back on market pricing of rate cuts; the market likely saw this as an implicit signal the Fed is comfortable with starting rate cuts in Sept. This endorsed our core rate views and supported our bias to trade intermediate rates from the long side.

The July FOMC had no discussion of recent upward pressure in repo rates. We are growing increasing concerned about higher repo rate risks (see: US front end & Repo risks rising). The Fed gave no indication of any concern around UST funding moves.

UST refunding recap: bill shift => coupon delay

Treasury kept nominal coupon sizes stable and modestly increased TIPS sizes at the August refunding, as expected. Three aspects of the refunding announcement stood out: (1) future coupon size guidance (2) TBAC bill as % of portfolio guidance shift (3) buyback size increase. We address each below; more detail in Refunding recap.

Future coupon sizes: UST stated it "does not anticipate needing to increase nominal coupon or FRN auction sizes for at least the next several quarters." UST is clearly comfortable with its current financing position and running bill supply higher. Before raising coupons UST likely wants more clarity on the fiscal & QT outlook, new UST staff in place (UST staff turnover typically happens post-election), and debt limit resolution.

TBAC bill guidance: TBAC updated guidance on bills as % of portfolio. TBAC said bills "averaging around 20% over time" provided a good balance & 15% was "lower bound of proper market functioning". We argued this months ago (see May refunding preview).

Shift in bill guidance skews risks later for future coupon size increases. Bills as % of portfolio has long been the primary signal we consider for when UST will boost coupon sizes. Using the prior guidance (15-20% range), we flagged UST may grow coupon sizes in mid-'25 to bend bill % lower over time. We now see lower risk of increases next year. Increases will depend on UST's tolerance around 20% (Exhibit 14). TBAC's guidance makes us think acceptable bill % as 17.5-22.5% or 15-25%.

Buyback size increase: UST increased the max size of its individual liquidity buybacks & signaled a rollout of cash management buyback. UST increased the max quarterly total across liquidity buybacks from $15 to $30bn/q and from $2bn to $4bn per operation. This is consistent with May UST guidance and improvements in operation limitations (i.e. UST raised the 20 CUSIP operation limit). We do not see this as a market signal.

Cash management buybacks will be rolled out in Sept '24. Rollout timing matches Sept 15th corp tax date. Instead of larger bill cuts around the tax date, UST will use buybacks.

Steepening capacity: slowdown needed for 50+bps 5s30s

Clients are asking about curve steepening capacity given the recent rate move, especially around 5s30s. Our view: it will likely take sharp econ slowdown for 50bps+ in 5s30s.

5s30s steepening potential in a bullish dynamic: we generally prefer to look at 2s10s to gauge the potential bull steepening/bear flattening for the curve. The 2s10s bull steepening is contingent on the expectations for the policy trough (c.1.5-2%) and neutral rate expectations (c.2.75-3%). The steepening potential for 2s10s is c.75-100bp at the trough of the easing cycle (see Exhibit 7).

To estimate the expectations for 5s30s, we look at the beta of 5s50s steepeners to 2s10s steepeners in frontend driven bullish dynamics. We see an average beta of 75%, and the 75-100bp of steepening for 2s10s implies c.55-75bp for 5s30s. This is the 5s30s steepening potential at the policy trough and at a 1h horizon we would cap that steepening potential to 30-40bp.

5s30s steepening potential in a bearish dynamic: to estimate the 5s30s bear steepening potential we assume term premium buildup driven by a supply/demand imbalance. The best proxy for this type of dynamic is the steepening of the Gilt curve in Sep/Oct '22. Over that period, we saw the curve steepening c.70bp, and subsequently flatten 40bp once the BoE intervened as a buyer of last resort. The net of these is c.30bp.

On balance, between a bull & bear steepening, we think it will be difficult to see the curve steepening more than 50bps absent a sharp economic slowdown.

Bottom line: July FOMC or UST refunding doesn't change core views: belly rates likely to lead declines as first Fed cut nears and bill shift isn't game changer for 30Y spread short. On the curve, it will likely take sharp econ slowdown for 50bps+ in 5s30s.

  Rates - EU

•  Slightly stronger than expected GDP and inflation print keep us neutral duration short-term, but we keep our medium-term bullish bias outright and cross-market
• Fewer reinvestment flows drive a sharper rise in net issuance in H2. Italy sees support over the summer, France over Q4.

Summer for the brave

Recent data not as hawkish as they may first appear

The euro area's advance 2Q GDP and preliminary July CPI readings were slightly stronger than expected by our economists and the market. Our economists believe the data are not as hawkish as they may first appear and maintain their call for the ECB to deliver two additional cuts this year in September and December (see Europe Economics Weekly, 2 August). Indeed, the market's pricing of ECB policy rates in 2024 were unchanged by these data releases, with almost a full 25bp rate cut still priced in September and two full 25bp cuts still priced in by the end of 2024, i.e. in line with our base case.

Inflation outlook drives our bullish stance…

The main difference between our and the market's expectations on ECB policy remains beyond 2024. In 2025, the market is pricing in c. 90bp of cuts by the ECB, which is less than the 125bp expected by our economists with the divergence becoming evident in 2H 2025 (Exhibit 8). Our expectations of more cuts partly reflect the inflation outlook: in 2025 we expect headline CPI in the euro area to be 1.5%, vs 1.8% currently priced in by inflation swaps, and 2.2% forecast by the ECB.

… but markets may need more data to be convinced

For the market to reprice in more cuts, we will probably need data prints to signal disinflation is closer to our expectations. The stronger-than-expected July CPI print may temporarily delay such repricing and make it difficult for the market to price in a lower terminal rate in the euro area in at least the next 2-3 months. This keeps us neutral on 10y Bunds near-term but we maintain a bullish forecast of 2.10% by YE 2024 vs forwards of 2.20%, and 1.85% by YE 2025 vs forwards of 2.23%.

We maintain a bullish bias on euro rates vs US rates…

Our US economists still forecast the first rate cut by the Fed to be in December after the July FOMC meeting, although risks of a September cut have increased (see US Watch, 31 July 2024). In any case, the market is pricing in c. 200bp of cumulative cuts by the end of 2025 in the US, which is more than our economists' forecast of 125bp. When compared with our forecasts, more monetary easing priced in for the US would stand at odds with less easing priced in the Euro area, given the stronger economic outlook and fundamentals in the US.

… while being mindful of uncertainty around US elections

The outcome of the US elections is a source of uncertainty for cross market performance between the US and euro rates. In our view, the associated impact will depend on which policies take centre stage post-elections (see Global Economic Viewpoint, 31 July 2024). For US rates, the impact could be lower if trade policy and immigration dominate the agenda, and higher rates on the back of more expansionary fiscal policy. Meanwhile for euro rates, we believe the risks are skewed towards a slightly faster rate cuts by the ECB.

 Marginal increase in net supply after August…

The summer may, yet again, not allow much time for investors to delve into the details of supply/demand for Eurozone Government Bonds heading into September and beyond. Indeed, the equity sell-off in European Autos and Banks along with rekindled geo-political risks may prove the dominant drivers of EUR rates and spreads, along with central bank communication.

Nevertheless, it is still useful to highlight the timings and differentials relative to what markets are typically used to, flow-wise. September is expected to see an increase in net EGB supply from the second week onwards, more so than in 2023 (Exhibit 9).

… driven by QT, with Italy seeing support over the summer and France over Q4

The driver of this (marginal) increase is not so much on the issuance side, which despite the initial upside surprises to deficits is now largely in line with run rates, but because of ECB QT (Exhibit 10). In terms of country composition, Italy likely sees support in August-September, France in Q4 (Exhibit 11).

   Rates - UK

• More than one further cut this year still seems highly unlikely to us. We revise our QT expectation to £100bn for the new "QT year" starting in Oct.

A hesitant cut

The Bank of England (BoE) voted 5-4 to cut the Bank Rate by 25bps from 5.25% to 5.0%, as we had expected. Mann, Haskel, Pill and Greene voted to keep rates unchanged. While we are concerned about inflation persistence in the UK, we also thought that the BoE would be willing to tolerate and explain away some upside data strength to justify a cut. This is exactly what we got with the BoE highlighting "some progress in moderating risks of inflation persistence" and pointing to declining less volatile parts of services inflation. The BoE expects inflation persistence to fade but highlights upside risks.

In our view the BoE delivered a hawkish cut. The decision was characterized as a "slight reduction" in the degree of policy restrictiveness. The BoE gave no clear guidance on the future rate path and emphasized a meeting-by-meeting approach based on evidence on inflation persistence. Going forward, the BoE expects inflation persistence to fade, but highlights upside risks. Moreover, the decision to cut today was finely balanced and the minutes noted that the Bank "will not cut rates too much or too quickly". All of this points to the fact that the BoE is not rushing to cut rates again very soon.

No change in our Bank rate call

Our view remains that the stickiness in domestic inflation implies that the BoE's rate cutting cycle is likely to be slow and shallow. We expect one more cut in 2024 (Nov), four quarterly cuts in 2025 and two cuts in 2026, to reach a terminal rate of 3.25% by mid-2026. But risks are rising that the BoE can deliver less cuts compared to our base case of one more cut this year and four cuts next year.

Rates: a hesitant cut

Bank rate cut expectations increased following BoE's rate cut decision, now implying more than two additional 25bp cuts by year-end (Exhibit 12). Pricing of terminal rate was brought lower - now trading in line with our base case of 3.25% - but market continues to expect a slower cutting cycle than we think will be the case (Exhibit 13).

With the trough in the market still only priced for early 2029, we think that the current curve inversion between 2026 and 2029 should be trading closer to nil. We keep our short SFIM9 vs. SFIM6 futures entered at -19.5bp with a target of 10bp and stop at -35bp (-0.5bp currently). See Rates-UK section in Motion sickness published on 14 Jun.

Far from a ringing endorsement for a rate cutting cycle

Although we had been expecting a cut, we were nevertheless surprised by the extent of the rally in Nov and Dec Monetary Policy Committee (MPC)-dated contracts following the announcement. This was a very nervous, or hesitant, cut. A 5-4 vote is itself a close call. And given that of the five members that voted for a reduction "for some of these members, the decision was finely balanced", this was about as far from a ringing endorsement as would seem possible.

To us, this makes it highly unlikely that we see more than one further cut this year, with that cut (if it comes) presumably arriving in Nov. That was reinforced, in our view, by the Governor's response to a question about public sector pay awards, where he said: "The next step in this process is the Budget on 30th October…. We will wait for that news and then we can fully process that, as we always do, as announced government policy".

So the 30 Oct Budget will have a big influence on the November MPC decision, and a cut in advance of that information's arrival would need material weakness in inflation or activity beforehand, we would say.

In UK Rates Alpha published on 19 Jul, we recommended an Aug-Dec MPC-dated Sonia steepener at -38bp. Although we continue to believe that too much further easing is priced by year-end, with the Aug contract fixing yesterday we closed the trade at a spread of -40bp for reassessment.

We now think QT will continue running at £100bn in the new QT year

On Quantitative Tightening (QT), the Committee stated that it will vote on the target reduction of Asset Purchase Facility (APF) Gilt stock over the period from Oct'24 to Sep'25 at the 19 Sep meeting. Box A in the the Monetary Policy Report (MPR) contained an update on QT progress, with the Bank judging that QT had only a small impact on Gilt yields and market functioning. According to the Bank's estimates, of the 275bp rise in 10y Gilt yields from Feb'22 to Jun'24, 75bp represented an increase in Term Premium (TP). Of that 75bp TP increase, the QT impact was believed to be only 10-20bp.

We read BoE's confidence in QT performance as a signal that it might keep QT pace at £100bn in the new "QT year". Bank's note on the degree of uncertainty about the QT interaction with Bank Rate during rate reduction period implied some caution, suggesting that an increase in total QT pace was perhaps less likely. We continue to think that stopping "active" QT in the new year would make a lot of sense given BoE reserves fast approaching the ceiling of the Preferred Minimum Range of Reserves (PMRR). But given BoE's confidence outlined in the MPR we now expect QT to continue running at £100bn. Small size of active sales required in the new year means that the BoE could choose to either conduct active sales in 4Q24 only (with no passive QT in that period) or space auctions out over the four quarters.

 Rates - AU

• We expect the RBA to hold rates unchanged during its Aug policy meeting. 2Q trimmed-mean CPI was weaker than expected.
• The RBA will likely view inflation as moving in the right direction but the slow progress means cuts are not imminent either.
• We want to fade cuts priced into the 2024 RBA curve and like AUDNZD longs but wait for better entry levels.

Expect the RBA to stay on hold

The RBA is set to hold its next policy meeting on August 6 and we expect policy rates to stay unchanged at 4.35%. This meeting follows the 2Q inflation data released earlier this week, which showed that underlying inflation momentum has moderated, albeit at a slow pace.

Specifically, trimmed-mean CPI increased by 0.8% qoq, weaker than market expectations of a 1.0% increase. On a yoy basis, trimmed-mean CPI reached 3.9%, a touch higher than the RBA's own forecast at 3.8% (Exhibit 1).

We therefore expect the RBA to acknowledge that the current policy rates are working to reduce price pressures in the right direction. However, the slow progress in returning to target likely means that rates will be on hold for a period of time. The Board is also likely to highlight the relative weakness in the consumer sector as inflation-adjusted retail sales contracted by 0.3% qoq in 2Q, its second consecutive quarter of decline.

Fading RBA cuts looks like the right trade

We closed our flattener recommendation last Friday because positioning looked stretched, and the risks were skewed to the downside. A softer-than-expected underlying inflation pushed the 3s10s curve 9bps steeper and dragged down the AUD on crosses. As we noted yesterday, the sell-off was more pronounced in AUDNZD, likely because longs were especially crowded (see Australia Watch: 2Q CPI review: trimmed price pressure 31 July 2024).

Subsidies masking inflation: fade RBA cuts

The market has subsequently stabilized around these levels. and we generally like to fade these moves because subsidies are masking underlying inflationary dynamics in the Australian economy. For rates investors, fading RBA cuts priced by year-end 2024 also looks attractive, especially given how rich the US SOFR curve is. Yet August (il)liquidity is a concern, and we want to see better levels (around 15-20bps priced for November) before recommending a paid Nov or Dec '24 RBA position.

Outright paid positions > selling RBA vs 2y/3y

Some investors have expressed an interest in hedging this risk by receiving further out the curve (e.g., vs 3y bonds/ bond futures or 1y1y swaps). The market is certainly pricing a shallow easing cycle over the next 2-3 years. We generally prefer selling Nov RBA vs 3y bonds/ futures because the very flat 2s3s curve means the spread to the US is much less extreme at the 3y vs 2y or 1y1y points. However, this week's price action has pushed this spread to new lows and outright paid positions look preferable to curve trades at this point (Exhibit 2).

Even shorter-dated hedges like buying Dec '24 or Mar '25 bank bill futures are too expensive, in our view. The spread is about 0bps for December and 20bps for March, although in both cases bills-OIS basis should tighten so we prefer buying futures to receiving OIS as a hedge. Ultimately, the main reason we prefer outright positions, though, is rich pricing in the FOMC curve. The pace of cuts priced for the Federal Reserve is far beyond our team's forecasts and gives us comfort that an offshore-led rally could further weigh on pricing in the front meeting dates. For now, though, we recommend patience before adding risk.

The composition of inflation matters

The composition of inflation matters and the 2Q data paints a picture where goods inflation is picking up while services inflation is moderating slightly. This is good news, in our view, for two reasons. First, the pickup in goods prices, such as that of tobacco and clothing, is unlikely to be sticky as goods are in a disinflationary environment globally. Second, and perhaps more importantly, sticky services prices, such as health prices, have shown signs of moderation. That said, rents still remain stubbornly elevated, reflecting low vacancy rates and a tight rental market. Overall, this suggests to us that inflation could see further easing in 3Q, which should leave the RBA comfortable that re-acceleration risks are fairly contained.

Slight revisions for the RBA forecasts

We do not expect major changes to the Bank's economic forecasts but minor adjustments following the 2Q data release are likely. We expect the RBA to slightly lift their year-end forecast for trimmed-mean CPI %yoy but it should still return to target in 2025. Growth assumptions could also be slightly altered given than the consumer sector continues to show relative weakness.

  Rates - JP

• The BoJ hiked to 0.25% and confirmed plans to halve JGB purchases by Jan-Mar '26. Hawkish guidance points to further hikes.
• Rates: Cuts to JGB-buying as expected; 2024 net supply set to jump.

BoJ Review: Hawkish shift

   At the July monetary policy meeting (MPM), the BoJ policy board voted 7-2 to raise the target for the uncollateralized rate to 0.25% from the current 0-0.1%.

In addition, the BoJ announced long-awaited details of its plan to reduce its JGB purchases over the next 1-2 years, as promised in its previous MPM (for further details, see the discussion in the rates strategy section below).

Although the JGB purchase reduction plan was largely in line with market expectations, a 15bp rate hike was not the base case for the majority of analysts, and only about 40% priced in as of yesterday morning (note: BofA's forecast was for a hike; see: BoJ preview: Another step towards normalization. 26 July 2024).

Rising risk of Oct-Dec hike as BoJ suggests more to come

With the move to 0.25% out of the way, the markets focus has turned squarely to the pace and timing on future hikes. On this point, the BoJ's communications-both in the policy statement and Governor Ueda's press conference-were more hawkish than we and the markets had likely expected.

The latest guidance in the policy statement, for example, stresses that, while the future policy outlook hinges on the data, considering that "real rates are at significantly low levels," the BoJ will "continue to raise the policy rate" should economic activity and prices continue to develop in line with the central bank's baseline scenario outlined in its latest Outlook Report. While this idea was first presented in the April Outlook Report and thus not entirely new, we think its notable that the BoJ chose to highlight it in the policy statement.

Our view coming into this meeting was that a July hike would be followed by additional 25bp hikes in January 2025 (with risk that the move comes earlier in December '24) and in Jul-Sep '25, to take the end-CY25 policy rate to 0.75%. We keep this call for now. However, the BoJ's latest communications raise the risk that the next move could come earlier than our baseline, i.e. in Oct-Dec '24, and that the terminal rate in this cycle could be slightly higher than our current assumption of 0.75%.

Rates: Expect sharp increase in 2024 net JGB supply

At its 30-31 July monetary policy meeting (MPM), the BoJ raised the policy rate to 0.25% and gave details on planned cuts to its JGB purchases. JGB futures rose and the 10yr yield fell in response to the meeting statement. Yields mainly for medium-term issues had already risen from the market open on an early-morning media report that leaked the BoJ's deliberations on a hike to 25bp, but the cuts to JGB purchases were not a surprise. We would note that long and superlong JGB yields also rose toward the close. The details of the cuts to the BoJ's purchases are as follows.

Pace of cuts, terminal purchase amount: The BoJ plans to reduce monthly purchases by ¥400bn per quarter, to roughly ¥3tn in Jan-Mar 2026. In short, it will gradually scale back its monthly buying from ¥5.7tn in July to ¥5.3tn in Aug-Sep and ¥4.9tn in Oct-Dec, eventually reaching around ¥2.9tn in Jan-Mar 2026. It will also release an interim assessment of its cuts to outright purchases and details of its purchase policy for April 2026 onward at the June 2025 MPM. BoJ Governor Kazuo Ueda mentioned at the post-meeting press conference that the expected 7-8% decline in the BoJ's JGB holdings over the next two years would still leave them overly high, suggesting that it could continue to scale back its buying in April 2026 onward.

Which maturities and when: The BoJ will switch to announcing specific offer amounts per auction rather than ranges, which will determine its offers for the relevant quarter. It will reduce its offers for 1-3yr and 5-10yr issues by ¥25bn versus recent levels, to respectively ¥350bn and ¥400bn; for 3-5yr issues, where it has thus far bought the highest percentage of issuance, it will cut purchases by ¥50bn, from ¥425bn to ¥375bn. However, it will maintain its recent offer amounts for 10-25yr and 25yr+ issues. It will continue to announce purchase sizes per auction and the schedule for rinban operations at the end of each month.

Rates market implications

Our simulation suggests that net JGB supply (gross issuance minus redemptions and the BoJ's net purchases) will jump to ¥39.4tn in 2024 from ¥2.3tn in 2023, suggesting the risk of rise in the JGB yields. As noted, long-end yields indeed fell immediately after the MPM statement was released but rose toward the close.

The BoJ's recent monthly JGB purchases as a percentage of issuance are 58% for 1-3yr, 74% for 3-5yr, and 65% for 5-10yr maturities, versus 45% for 10-25yr and 12% for 25yr+ issues. We therefore see it as unlikely to reduce superlong purchases at least until its buying as a percentage of issuance for sub-10yr maturities falls to around the same level as 10-25yr issues. We therefore do not expect it to begin scaling back superlong purchases until at least the start of FY25, and think the near-term uptrend in yields will likely be confined mainly to sub-10yr maturities. We estimate end-2024 JGB yields of 0.5% for the 2yr, 1.25% for the 10yr, and 2.3% for the 30yr issues.

  Front end - US I

• UST repo rates are rising with elevated UST supply, slow cash drain, and constrained dealer balance sheets

Repo risks rising: watch out above

We see increasing risks of a more rapid shift higher in UST repo rates in 2H '24. The drivers: (1) large UST net settlements (2) constrained dealer balance sheets (3) slow cash drain. Our concerns increased after the persistent widening of TGCR and ON RRP rates, which reflect repo market frictions (see plumbing frictions). Our concerns make it hard to be long SOFR/FF or SOFR swaps at any tenor, especially the front end; we suggest trading spreads from the short side. There is too much UST supply vs available cash.

If repo rises rapidly the next 2 events to watch are: (1) willingness / ability of commercial banks to shift cash out of IORB and into repo (2) potential UST RV HF de-risking, which could accelerate repo rise. If banks prove unable to contain repo & UST RV HF de-risk, Fed will be needed. Fed SRF likely will be tested and is likely to prove porous; next step = QT end, Fed OMOs, and admin rate change. If UST market functioning deteriorates more extreme steps will include UST bill and/or coupon purchases + regulatory easing. We see rapid repo rise as a growing risk, but not our base case. Detail below.

Drivers of higher repo: supply, cash drain, dealer limits

Repo has shifted higher earlier than we expected. SOFR, SOFR 75^th pctl (bi-lateral repo proxy), and tri-party repo rates have all shifted higher. The higher repo drivers include (1) UST supply (2) cash drain (3) dealer constraints. We elaborate on each.

UST supply: The Treasury has issued $1.2tn in net supply over 2024 YTD, including $942b in net coupons. There is a notable increase in SOFR on these large coupon settlement dates. Looking ahead, we expect to see this trend continue, especially given expected large coupon settlements through the remainder of 2024. Supply will not be lightening up and repo likely will keep rising on settlement dates.

Cash drain: YTD Fed QT and BTFP winddown has drained $507b from the system, including $158b from reserves and $420b from ON RRP. Additionally, YTD $54bn in growth in the TGA has also removed cash from the system. Looking ahead, we forecast roughly $275b in additional drain from QT and BTFP through year end. This should be partially offset by a $115b TGA drop. Cash keep draining & funding markets will show it.

Dealer constraints: dealer sheets have grown roughly $95b YTD, including $62b in USTs alone. UST supply growth has pressured dealer holdings and financing needs higher. Inability for dealers to intermediate MMF cash out of ON RRP suggest capacity limits. Sheet pressures extend beyond UST repo and can be seen in equity financing. These limits are likely to get worse in 2H '24 and into year-end.

Each of these factors is expected to extend in 2H '24. UST settlements will rise $150-200b/m, the Fed is not seriously discussing QT cessation, and dealer constraints will likely worsen. These factors collectively point to sustained higher funding rates.

If repo sustains its upward rise the next 2 events to watch are: (1) willingness / ability of commercial banks to shift cash out of IORB and into repo (2) potential UST RV HF de-risking, which could accelerate the move. We offer thoughts on each.

         Front-end - US II

• Fed funds to move higher as FHLB discount notes cheapen
• FHLBs tend to require a 3-5bps spread of FF to O/N discos; as discos cheaper, FF higher

Fed funds will rise with FHLB discount note rates

Clients have increasingly asked: when will the effective federal funds (FF) rate rise? Our answer: when FHLB discount note (DN) rates increase. DNs will rise as bill supply builds and cheapens money market paper. We expect signs of upward FF moves in August with the arrival of +$240b of bill supply.

Fed funds background: FHLBs lend, foreign banks borrow

FF is highly idiosyncratic. Short FF refresher is below; detail in Fed policy plumbing.

Lending: dominated by FHLBs who comprise 90-95% of activity. FHLBs hold cash liquidity to fund member advances. If no advances, FHLBs lend excess cash in FF because they cannot earn IORB. FHLBs value flexibility to invest FF late day and have cash returned early. FHLBs fund FF partially via DNs and target a 3-5bps spread of FF-DN.

Borrowing: currently concentrated with foreign banks (FBOs). FBOs borrow FF to arbitrage IORB. FBOs do FF-IORB arb because they have fewer balance sheet constraints vs domestic banks. FBOs will use FF for other money market arb, including lending in GC repo.

FHLBs will demand higher FF when disco rates rise

FF will rise with FHLB DN rates. FHLBs will demand greater return on FF with higher DN funding costs. Historically FF-O/N DN spread and FF volumes are correlated as discos richen vs FF, FHLBs increase FF lending (and vice versa). If DNs too cheap vs FF, FHLB volumes decline until FF rises. We expect to see this dynamic in coming months.

In theory, DN rates correlate with DNs outstanding, bill rates, and bill supply. In practice, the results are mixed. We overweight theory vs practice. DN rates should cheapen with DN issuance and bill supply + cheaper bills. DN issuance is driven by (muted) FHLB advances; bill supply will rise in Aug and likely cheapen bills & DNs.

FF has not yet moved because O/N DN rates are not high enough. We suspect higher bill supply will cheapen bills & DNs, eventually leading to upward pressure on FF.

SOFR / FF disconnect: plumbing frictions

Historically there is a strong relationship between higher repo and FF. Higher repo should theoretically be correlated with higher bill & DN rates. At present, money market frictions are resulting in a surprising wedge between repo and bill / DN rates (see plumbing frictions). This gap could widen and take SOFR higher vs FF.

Bottom line: FF likely will rise with higher short-dated bills & DNs. Once short-dated bills & DNs start cheapening, FF likely will increase. Higher FF pull should result in a modest knee-jerk widening of SOFR/FF. FF isn't stationary, it likely will move higher.

  Supply - US

UST coupon sizes stable, bills to rise in future

 Treasury kept nominal coupon sizes stable and modestly increased TIPS sizes at the August refunding, as expected (see UST refunding preview). Three aspects of the refunding announcement stood out: (1) future coupon size guidance (2) TBAC bill as % of portfolio guidance shift (3) buyback size increase. We address each below.

Future coupon sizes:  UST stated it "does not anticipate needing to increase nominal coupon or FRN auction sizes for at least the next several quarters." UST is clearly comfortable with its current financing position and running bill supply higher. Before raising coupons UST likely wants more clarity on the fiscal & QT outlook, new UST staff in place (UST staff turnover typically happens post-election), and debt limit resolution.

TBAC bill guidance:  TBAC updated its guidance on bills as % of portfolio. TBAC said bills "averaging around 20% over time" provided a good balance and 15% was "lower bound of proper market functioning". We have long felt this prudent ("we think that UST can justify a potentially higher level given strong bill demand", see May refunding preview).

 The shift in bill guidance skews risks later for future coupon size increases. Bills as % of portfolio has long been the primary signal we consider for when UST will boost coupon sizes. Using the prior guidance (15-20% range), we had flagged UST may grow coupon sizes in mid-'25 to bend bill % lower over time. We now see lower risk of increases next year and believe current UST coupon sizes may be appropriate through FY '26 using our baseline deficits. Timing of increases will depend on UST's tolerance around 20%. TBAC's guidance makes us think they will accept bill % as 17.5-22.5% or even 15-25%.

Buyback size increase:  UST increased the max size of its individual liquidity buybacks & signaled a rollout of cash management buyback. UST increased the max quarterly total across liquidity buybacks from $15 to $30bn/q and from $2bn to $4bn per operation. This is consistent with May UST guidance and improvements in operation limitations (i.e. UST raised the 20 CUSIP operation limit). We do not see this as a market signal.

Deficit scenarios suggest limited coupon growth

In Exhibit 14, we show bills as a % of marketable debt under different deficit outcomes. The timing around when UST may consider growing coupons will depend on their tolerance around the 20% average. Our baseline deficit scenario shows bills more than 22.5% by end of FY '26. If tolerance band is closer to 5 PPTs, it would require a $500 -$750bn deficit shock for both fiscal years to see bills exceed 25% by the end of FY '26.

While we see potential for UST to grow auction sizes in mid/ late '25, this will be a function of UST's tolerance above 20% and scale of financing needs. As a baseline we hold nominal coupon auction sizes unchanged through our forecast horizon.

TIPS supply increases likely to continue

As anticipated, UST continued to modestly grow TIPS auction sizes and delivered a $1bn increase in the 5y new issue. Consistent with TBAC financing recommendations, we pencil in an additional $1bn increase to the 10y TIPS new issue next quarter. After these increases, we see TIPS as a share of supply ex-bills stabilizing.

Higher bill supply = lower LT ON RRP

The revised TBAC guidance on bills as a share of marketable debt does not impact our near-term bill supply forecasts or expectations for Fed ON RRP. In the longer term, higher bill supply as a share of marketable borrowing means higher money market rates & lower ON RRP levels over time. We hold our view for QT to conclude by the end of the year driven by funding pressures and debt limit fluctuations in TGA next year.

Bottom line:  Auction sizes were in-line with expectations and UST guided to stable auction sizes for "several quarters." TBAC shifted bill target higher which will see more bill supply over time (negative for front end spreads, less cheapening pressure on long-end spreads). UST buyback max size increase is purely due to operational improvements.

 Technicals

•    US 10Y yield: On July 25^th 2024 at 4.24%, the 50d SMA crossed below the declining 200d SMA. Out of twenty past signals, US 10Y yield was lower 70-80% of the time 25-65 trading days later. This supports our H2 bullish UST view.
• US election year seasonals: The average trend for 10y yield from now to US Election Day is sideways. However, when the sitting President was Democrat, the average trend was up into election day, a risk to our view. (Full report)

 US 10Y Yield

Decline in yields in line with our core view, short-term getting stretched

The 50d SMA crossed below a declining 200d SMA on July 25^th with a closing level of 4.24%. History says yield tends to be below this level especially 25-65 trading days later. Reiterate bullish bias for H2 and buying dips. With 14-day RSI falling below 30 and yield approaching the 61.8% level of 3.92%, the risk of a tactical bounce is increasing. We buy the dip. US election year seasonals a risk to our view.

 Rates Alpha trade recommendations