Our medium term views

 Our key forecasts

  What we like right now

  The View

 The week that will be

Next week's main event will be the FOMC meeting. At the time of writing, markets are pricing just over 25 bp of cuts for Wednesday - in line with our economists' call - from last Friday's intra-day high of 43 bp.

Our core convictions remain unchanged: even in a soft landing scenario markets will struggle to reduce meaningfully the implied hard landing risk premium. This is likely to leave rates trading somewhat rich, especially in the front-end of the curve. There is room for the curve to steepen, but we continue to favor 5s30s over 2s10s (see Liquid Insight 9 Sep 24). However, with markets still holding onto a deep trough in the cutting cycle, we see tactical value in bear flatteners (see Rates US).

With the FOMC out of the way markets will turn to the BoE. Crucially, we receive inflation data on Wednesday, ahead of the BoE on Thursday. Our economists expect the BoE to be on hold and worry that core inflation may actually pick up a bit in the August print also strengthening the case for patience. We are paid 3s5s7s Sonia and short UK 5y real yields, outright and cross market (see UK Rates Viewpoint 5 Sep 24).

Next, we have the BoJ decision. Our economists do not expect a rate hike before the end of the year, so more relevant for markets could be the interplay between issuance and BoJ purchases (see Rates Japan). We expect the BoJ to reduce purchases of JGBs with maturities up to 10 years and maintain the purchase sizes for superlongs in the Oct-Dec Rinban operations. At the same time the Ministry of Finance shortened issuance at its liquidity enhancement auctions in August, which could be included also in the FY25 JGB issuance plan. We see scope for 10s30s to flatten.

In the Euro Area, the main data of note beyond ZEW could be the details of the August inflation print, as well as the July current account data. The current account data could be a useful reminder that the EA is structurally saving more than pre-pandemic.

The weeks that were

Following last week's payroll print, our economists revised their forecasts for the Fed to consecutive 25 bp cuts (see Federal Reserve Watch 6 Sep 24). The CPI print was not weak enough to sustain hopes of a 50 bp cut (see US Watch 11 Sep 24).

The ECB meanwhile was not dovish enough for markets. 2y German yields are up 7bp on the day and the 2y-30y curve is 4bp flatter. While we believe the ECB has left the door wide open to a cut in October, we see the front-end of the curve as vulnerable and likely to be driven more by priced action in the US and UK, now that the ECB has failed to provide another impetus to the rally. We stay long 2y1y, long cross market and long 30y.

  Rates - US

•      Growth concerns support constructive duration stance; stay dip buying
• Fed likely to cut 25bps next week; dots likely hawkish vs market but not Powell

Growth jitters & debate bait

US rates declined & the UST bull steepened with building downside growth risks. Downside risks stemmed moderating NFP & bank guidance (Ally = worsening consumer, JPM = lower NII). These concerns may have been furthered by market perception Fed cuts won't be fast enough, esp. with sticky core CPI & OER; Waller hinted at 25bps next week & a willingness to speed up if needed, which pushed cuts into the Dec '24 & later (Exhibit 4). Small shifts in election probabilities post-debate may have also pared back further fiscal easing expectations (Exhibit 5). On net, growth concerns pushed credit spreads modestly wider, cross market implied vol slightly higher, & rates lower.

On balance, developments this week did little to change our core views around duration & curve. Duration: we still recommend clients buy the rate dip & add duration exposure on any back up in rates. Our preferred place to add is the belly (5y) given it best represents the Fed cutting trough & is relatively agnostic on the exact Fed cutting path. Curve: we still hold 5s30s nominal steepeners given US growth moderation, Fed cuts, & long end supply / demand concerns. Our recent work highlighted duration & curve trades at the first Fed cut historically outperform forwards (see FAQ: Fed cuts & US rates).

Fed likely to find neutral from below

Fed cuts are very well priced with the trough at 2.85%. The trough is now priced so low because the market anticipates the Fed cutting below neutral to stimulate activity. This contrasts to market pricing at the June FOMC meeting, which expected Fed cuts only to neutral (Exhibit 6). The deeper trough aligns with increased downside growth risks.

We are sympathetic to the market's interpretation of Fed reaction function. The Fed does not know where neutral is & will only discover it by relying on feedback from economic data & financial conditions. The market expects the Fed will likely find neutral from below (over cutting) vs find neutral from above (under cutting). A shift to finding neutral from below is also consistent with increased focus on downside growth risks.

A significant number of investors see some overpricing of the Fed near-term. In a back-of-the-envelope calculation, if hard landing scenarios are consistent with 150bp worth of cuts by end-'25, and soft-landing c.50-75bp, the c100-105bp of cuts the market is pricing corresponds to a c.35-40% likelihood of hard landing. We recently recommended US 1y2y OTM payers vs EU 1y2y payers (see US Rates Alpha from 8 Sep '24), and the position is likely to perform well in scenarios where the market fades some of the near-term Fed pricing. Bear flatteners also leverage this type of scenario, and we favor expressing the view thought a long 3m2y payer spread atm/atm+25bp fully financed by selling 3m10y atm+20bp payers. The trade works as a proxy for a bear flattener but avoids the current giveup to forwards in those positions. Carry is positive by c.4bp/m. The main risk on the position is a bear steepening scenario with potentially unlimited downside. The position is also expose to parallel selloffs of more than 25bp.

FOMC: hawkish dots, dovish Powell

The Fed will cut 25bps next week & release a new Summary of Economic Projections. Our economists expect the dots to show total cuts of 75bps in '24, 125bps in '25, 50bps in '26, & none in '27. These dot plot expectations are much more hawkish vs market pricing (Exhibit 7). Rates may initially rise with the dots; we suggest fading the move.

Powell is likely to sound dovish in his press conference. He should highlight downside risks to the labor market & willingness to speed up the pace of rate cuts if needed. Powell will reiterate data dependence but flag downside growth risks. On balance, we expect the market to interpret the Sept FOMC as delivering a neutral to dovish cut.

Regulatory *Barr* lowered

Fed Vice Chair of Supervision Michael Barr gave an outline of the revamped proposal of the Basel 3 (B3) endgame guidelines this week. We do not believe it is a game changer but it may reduce costs and frictions in Treasury markets (lower capital need vs original proposal, GSIB surcharge co-efficient shifts). The reg shift is not enough to change our negative view on spreads. Reduced capital needs should be mildly supportive of swap spreads but the larger underlying theme of heavy Treasury supply should continue to be a main driver of cheaper Treasuries over time, despite the B3 endgame revamp.

Importantly, the B3 endgame shift hints at a new wave of increased private-public partnership in crafting bank and market regulations. As the financial system and regulatory landscape become increasingly complex, the web of interacting regulations, markets and end-users appears to require increasing collaboration between policy makers and stakeholders. We see the B3 rewrite as a more collaborative effort that could set an example for future regulatory changes (see: Basel 3 endgame shaping up).

Bottom line: growth concerns support a constructive duration stance; we hold our existing steepener recommendations. Fed will cut 25bps; dots may be hawkish vs market but Powell likely won't be. For clients who think too many cuts we recommend cross market expressions or bear flatteners in options space.

  Rates - EU

• Given the significant rally in the past two weeks, the lack of dovish tilt from the ECB leaves the front-end vulnerable to moves in other markets in the very short term.
• Investors concerned about an insufficiently dovish Fed should consider our EUR cross market trades, with our received 2y1y €str position temporarily at risk.

Taking stock of investor views and ECB meeting

Markets converging to our baseline…

The last few weeks have seen a clear shift in market narrative. Our investor meetings and the just-released monthly FX and rates sentiment survey, 13-Sep indicate that views around inflation and terminal rate have started to meaningfully converged towards our economists' baseline: sub 2% inflation in the medium term (Exhibit 8) and sub 2% ECB terminal (Exhibit 9). A reduced number of investors challenge our view that the neutral rate in the Euro area may not have moved higher from pre-Covid.

The reduced expectations around a potential fiscal shift, towards more supportive policies, are possibly helping the reassessment of the Euro Area neutral (Exhibit 10). The market implied terminal rate has dropped from 2.05% to c.1.75% (pre ECB), finally reacting more forcefully to the front loading of cuts that started to be priced in Jul-Aug. 

… But ECB leaves the front end vulnerable near term.

However, now that the ECB has failed to provide another impetus to the rally, the front-end of the curve appears vulnerable to moves in other markets over the very short term.

1. ECB messaging: no dovish tilt, with low Sep inflation already noted

The ECB delivered the expected 25bp cut but kept the same message of data-dependence and meeting-by-meeting approach. There was no clear dovish tilt. ECB president Lagarde was careful not to provide any signal on the timing or pace of future decisions. She insisted that the cut was only a "step" again, which our economists interpret as as not being clearly on a normalization path yet. If anything, one could

take her comments on looking through the likely weak inflation print in September or her renewed emphasis on the December forecast as very soft signals that an October cut is a difficult proposition (see full ECB review: Que será, será? Será late).

2. Positioning is in steepeners

Our investor conversations over the past two weeks indicate a clear and strong consensus on curve steepeners, both in EUR and US rates. This is also something flagged in the work that our US colleagues conduct on flows (US rates watch, 9-Sep).

3. Lack of confidence on pace of near-term cuts

When it comes to the amount of cuts currently priced in the very front-end (2024), around one third of investors believe that this is exaggerated but are just not fading it at the moment (Exhibit 11). The messaging from other central banks could change that.

Favor cross market trades if concerned

We continue to hold our 2y1y €str received position as a long-term structural trade, recommended on Sep 3^rd (entry: 2.12%, current: 1.9%, target: 1.7%, stop: 2.4%). The main risks are better than expected data or a pricing out of global CB cuts. Still, given the above, we argue that investors concerned about an insufficiently dovish Fed next week or a hawkish BoE should consider the X-market trades we recently recommended:

• Long OTM US 1y2y payers funded by ATM 1y2y payers in EUR. From a mark to market perspective, the risk to these positions is an overly dovish Fed that leads to further outperformance of US rates. From a terminal point of view, the risk would be more persistent EZ inflation, resulting in a larger sell-off in EUR rates.
• Received 3y1y €str vs paid 3y1y CAD OIS (current: 32bp, target: 80bp, stop: 15bp). The divergence in terms of cuts priced between the two central banks is most significant in 3y1y, with the market pricing in that the BoC would extend the cuts to 2.35% (BofA baseline: 3%). The risk to the trade is weak Canadian / US data.
• Receive 5y real rate in EUR vs UK (current: 53bp, target: -40, stop: 90bp). Our inflation strategist entered the trade to position for (1) larger UK short-dated issuance, (2) inflation risks limiting BoE cuts vs ECB, (3) UK's deteriorating IIP/GDP ratio vs the EZ (see note: Liquid Insight 21-Aug). The main risk is a dovish BoE.

We also remain bullish the back end of the EUR curve, with a long 30y Bund position - current: 2.42%, target: 2%, stop: 2.83%. There, positioning is lighter, and supply will be declining. Support can also emerge from structural receiving flows and the macro thematic of a central bank behind the curve. The main risk is reduced PF demand.

 Rates - UK

• BoE QT decision will be important beyond the Gilt supply impact: for debt definition, Gilt liquidity considerations & relative scarcity of low coupon Gilts.

 Quantitative Tinkering

The UK rates market's advance since the start of the month resulted in two 25bp rate cuts being fully priced in by year-end for the first time since early August (Exhibit 12). Sonia swaps now imply 165bp of cuts by year-end 2025 and a 3.17% terminal rate by September 2028 (Exhibit 13). When compared with our forecasts, the market now prices a faster rate cutting cycle (than our expected quarterly pace) and a fractionally lower terminal rate (our base case is 3.25%).

After mixed labour market and growth data, the inflation numbers are the only potentially influential prints remaining before the 19 September rate decision (see Labour market: Healthy employment, 10 September and GDP: Flat, 11 September).  Our economists expect the MPC to remain on hold next week - in line with market pricing - after a finely-balanced decision to cut in August (with the MPC's Chief Economist dissenting) and with inflation persistence concerns to the fore. We expect a slower-than-priced easing cycle with these concerns proving slow to subside.

Our preferred way to position remains via Sonia 3s5s7s fly (pay belly, receive wings). The trade should benefit from pricing out of the rate cutting speed expected by the market. Our conviction is higher for the balance of 2024: we expect cuts in one of the three meetings while the market expects two; two are possible, but one appears much more likely than three (every meeting), in our view. We also expect the fly to benefit from shifting UK banks' flows from structural receiving to mortgage paying (see Back to school: tough tests this term, 5 September). We entered the trade at -12bp with a target of 10bp and a stop at -21bp. Current level is -12bp. Risk to the trade is UK government increasing Stamp Duty tax at the Budget, slowing housing market activity (see BoE preview: Hold with unchanged guidance, 13 September).

 Many ways in which QT decision matters

 The BoE's target for the APF Gilt stock reduction over the year from October, and its Gilt sales schedule for 4Q24, should be announced alongside the September MPC decision. Our assumption is that the vote will be to reduce the stock by £100bn over the year, requiring £13bn of active Gilt sales. However, the Bank's expressed confidence in QT's progress "in the background", as expressed in August's MPR, hints at upside risks.

We assume the BoE will hold one Gilt auction per month to achieve its £13bn active QT target. With the Bank now targeting sales evenly across maturity sectors, but measured in initial proceeds terms, we expect the 4Q24 auction schedule to feature a £0.9bn auction of short Gilts in October, a £0.6bn auction of longs in November and a £0.8bn auction of mediums in December. There are of course many ways the BoE could conduct this £13bn of active QT. For more on Gilt supply from the DMO and BoE combined, see Gilt supply in 4Q24: modestly lighter but upside risks beyond October, 3 September.  

Beyond the Gilt supply to the market, the QT decision will be important in other ways:

•  Debt definition considerations: BoE losses crystalised on QT sales and redemptions are incorporated in the debt measure that the previous Conservative government used in its fiscal rules (PSND ex BoE). The Labour government can change the debt measure used in its fiscal rules in the October budget to reduce (or potentially even remove) the impact of these losses. If it doesn't, the QT decision could have implications for the OBR's assumption for the future QT pace and hence the fiscal space. Currently, the OBR assumes £48bn of active sales per QT year (the amount done in 2023/24) and total QT to fluctuate each year based on the uneven profile of redemptions. Our £100bn/year scenario would reduce QT losses expected in the near term.
•  Gilt liquidity considerations: The QT decision might also matter in the context of deteriorating Gilt liquidity relative to USTs and Bunds (Exhibit 14). Calls for the Bank to sell sub-3y Gilts have resurfaced. This is something we have recommended for a long time - for instance, last December we said: "We do think there are strong reasons for adjusting the buckets, selling fewer long Gilts and introducing sales of 1-3y Gilts" (see Liquid Insight, 6 December 2023). This would improve liquidity in scarcer issues, reflect the reality that the Bank's portfolio has shortened a lot and smooth the transition away from an abundant liquidity regime (by providing banks with more desirable HQLA).
•  Relative scarcity of low coupon Gilts: It was noteworthy that BoE sales of short-dated Gilts in July (two £800mn ops, 15 July and 29 July) saw £1.1bn of the operation size (in market value terms) being purchases of UKT 0.125% 2028, with most of other Gilts acquired in that operation also low coupon issues. In August, the one short-dated Gilt sale saw £0.63bn allocated to UKT 0.125% 2028 (Exhibit 15). If we are right that BoE sticks with £100bn QT - implying £900m short Gilt sales per quarter by our estimates - then that potential source of low coupon Gilts will slow to a trickle.

The last point is one of the factors that will influence low coupon/high coupon Gilt spreads (another being potential changes in appeal following the Budget). Another is whether the DMO decides to respond to demand with discretionary tenders (as it did this week). We continue to hold our sell UKT 4.5% 2028 Gilt to buy UKT 0.5% 2029 trade, monitoring on a z-spread basis (entered at -8bp with a target of -20bp and a stop-loss at +4bp). Current level is -8bp. Risk to the trade is a change in the tax treatment of Gilts for retail (see Back to school: tough tests this term, 5 September).

 Rates - AU & NZ

Swap spreads in focus

Investors had lightened up on risk ahead of payrolls on our trip to Singapore and Hong Kong last week but the outlook for swap spreads against a backdrop of rising public spending in Australia was a popular topic of discussion. We continue to see public spending as a tailwind for the Australian economy over the next 6-12 months and we still like paying front-end RBA dates. Yet our fair value framework for 10y swap EFP (i.e., swap rate vs 10y ACGB futures yield) suggests more balanced risks and we recommend waiting for wider credit spreads before adding risk.

Fundamentals favour tighter swap spreads…

Investors were looking for the ideal entry point to position for bonds to cheapen vs swaps given the increase in bond supply from rising fiscal deficits. To be sure, the fundamental drivers of swap spreads suggest investors should position for tighter swap EFP. We currently see four compelling, fundamental reasons to receive swap spreads (i.e. position for cheaper bonds vs swap EFP):

1. Combined state and federal fiscal deficits are set to increase by almost 80% between 24/25 and 25/26 financial years (Exhibit 19).
2. Kangaroo bond issuers, who tend to receive swaps at the tenor they issue as part of their routine hedging programs, have extended out the curve (Exhibit 20). The attractiveness of issuing in AUD vs USD and rising superannuation fund demand for fixed-income paper means this trend is likely to continue for the foreseeable future (Exhibit 21).
3. We see tighter basis markets as super funds' demand for cash and fixed-income products outpaces supply over the next few years. A likely reduction in major banks' HQLA portfolios should also reduce demand for hedged purchases of semis (i.e. less asset-swap paying from bank treasuries) (see Bullish on basis 29 August 2024).
4. Markets are pricing in a synchronised global easing cycle and our global rates team recommends buying duration (Trades for cutting cycle: a historical comparison 05 September 2024). Although AUD rates are likely to lag any rally given the divergence between the RBA and Federal Reserve, the global rates backdrop means it's unlikely we will see corporates switching from floating to fixed-rate debt in the near term.

… but tight credit spreads are a risk

10y swap spreads are currently a little more than 1 standard deviation tighter than our fair value framework implies (Exhibit 16, Exhibit 17). The main reasons for the divergence from fair value are credit spreads (tighter than anticipated) and issuance patterns (less paying flow than anticipated), offsetting the impact from semi G-spreads (wider than anticipated) (Exhibit 18). While we see fundamental reasons for semi-G spreads to remain wide and issuance-related paying flow to remain depressed, the signal on credit risk suggests the risk/reward for receiving swap EFP is not particularly attractive.

Cyclical credit risks = wait for better levels to receive

A sudden widening of credit spreads is a tail-risk scenario. As we approach the end of the business cycle, there is a risk of deteriorating credit quality. Insofar as Australian company insolvencies are a leading indicator, the recent surge sends a negative signal. Yet BofA's Australia credit loss indicator, which was showing more acute signs of credit stress in 2023, has gradually eased over the past 6-9 months, suggesting the worst may be behind us (BofA Australian Bank Credit Loss Indicator 12 August 2024). Against this backdrop we would trade swap EFP more tactically and wait for the spread to widen to around 15-20bps before adding risk.

  Rates - JP

• We expect BoJ to reduce purchases of sub-10yr JGBs in Oct-Dec, while maintaining purchase sizes for superlong JGBs
• MoF shortened its maturity for JGB issuance at August liquidity enhancement auctions

BoJ reduction of JGB purchases likely to focus on sub-10yr maturities

During Oct-Dec Rinban operations, we expect the Bank of Japan (BoJ) to reduce purchases of JGBs with maturities up to 10 years and maintain the purchase sizes for superlong JGBs. In addition, considering that the Ministry of Finance (MoF) shortened the JGB issuance maturity at its liquidity enhancement auctions in August, the FY25 JGB issuance plan could also include reductions in the issuance of superlong JGBs and increases for JGBs with maturities up to 10 years.

Expect BoJ to maintain superlong purchases in 4Q24

The BoJ is scheduled to release its "Schedule of Outright Purchases of Japanese Government Bonds" for 4Q24 on 30 September at 5pm Japan time. The planned reduction of JGB purchases announced at the BoJ's July Monetary Policy Meeting (MPM) came as no surprise and focused on reducing purchase amounts for 1-3yr, 3-5yr, and 5-10yr issues (see Japan Watch: BoJ Review: Hawkish shift 31 July 2024). Considering that (1) BoJ's purchases vs the MoF's monthly issuance are still high in maturities up to 10yr (Exhibit 22); and (2) JGB yields have been stable since BoJ launched de-facto quantitative tightening (QT), we expect reductions of outright purchase (Rinban) operations in Oct-Dec to focus on JGBs with sub-10yr maturities, while recent purchase amounts for superlong maturities are maintained. We expect the BoJ to reduce planned monthly purchases of 1-3yr and 5-10yr JGBs by ¥100bn each and 3-5yr JGBs by ¥200bn from current amounts.

If the cutbacks are as we anticipate, BoJ monthly purchases as a percentage of issuance will be 50% for 1-3yr, 57% for 3-5yr, 58% for 5-10yr, 45% for 10-25yr, and 12% for 25yr+ maturities. Based on those figures, BoJ purchases of sub-10yr maturities as a percentage of monthly issuance will remain high, and we believe reductions of its purchases of superlong JGBs is unlikely until its purchases as a percentage of monthly issuance for sub-10yr maturities falls to around the same level as 10-25yr issues. We therefore do not foresee the BoJ scaling back purchases in the superlong zone until the start of 2025, at the earliest.

Potential flattening of 10-30yr JGB yield curve

Considering (1) an increase in net JGB supply would be a factor which raises yields of sub-10yr maturities; and (2) a likely focus on shorter maturities for new JGB issues, we see potential for a flattening of the 10-30yr yield curve. The current 10yr-30yr spread is c.120bp, but we believe this could narrow to c.110bp.

• According to our estimates, net JGB supply (gross issuance minus redemptions and BoJ's net purchases) will surge from ¥2.3tn in 2023 to ¥40.4tn in 2024. As noted above, the BoJ will likely reduce its purchases of JGBs with maturities of up to 10 years, while maintaining the recent sizes for purchases in the superlong zone. We therefore believe the yields with maturities up to 10yr are more likely to rise than yields on superlong bonds. We accordingly expect the upcoming JGB auctions to generate tails that will lead to a gradual rise in JGB yields.
• MoF reduced the issuance amount in its bimonthly liquidity enhancement auction of existing issues with remaining maturities of more than 15.5yrs and less than 39yrs to ¥400bn at the August 2024 auction, down from ¥500bn at previous auctions. On the other hand, at its monthly liquidity enhancement auction of existing issues with remaining maturities of more than 5yrs and less than 15.5yrs, the MoF increased the issuance amount to ¥650bn, up from ¥600bn at previous auctions. We believe these changes are aimed at (1) alleviating tight supply-demand for sub-10yr JGBs; and (2) reducing the duration in JGB issuance. Thus, the JGB issuance plan for FY25, which will be announced in late December, may include an increase in issuance of JGBs with maturities up to 10 years and a reduction in issuance of superlong JGBs.

  Front end - US

•  Debt limit client questions rising, we update our projections. X-date should be in in summer '25, July now most likely

Debt limit interest rising… it's almost that time again

Client questions on debt limit (DL) & funding impact have picked up. We update our note from May (see US funding & debt limit) following new TGA guidance, deficit profile tweaks, & recent price action. Bottom line: DL timing hasn't changed much. DL suspension period ends on Jan 1 '25. X-date likely in June or July '25 (July most likely). DL will lower TGA, increase cash in system, & temp ease funding into X-date. Post X-date TGA will rise & tighten funding. Trade = SOFR/FF curve flattener.

Debt limit background: suspension period & X-date

The debt limit, which is currently suspended, will reset to the level of UST debt outstanding on Jan 2 '25. Once at the DL Treasury will need to rely on the remaining cash in the TGA and employ an accounting tool known as "extraordinary measures" (EM). EM provides some headroom under the DL for Treasury to issue a limited amount of debt. Congress will need to raise / suspend the DL before UST cash & EM are exhausted; if no increase the US gov't will face technical default risk (i.e. "X-date").

X-date tweaks: July now most likely

To project the X-date we rely on 3 key factors: (1) TGA level (2) EM size (3) deficit size.

TGA: we update our forecast for YE TGA balance from $650b to $700b following the most recent quarterly refunding announcement. Treasury forecasts a TGA of $700b at YE '24, down from $850b at end Q3 '24. It must decline because the law prohibits TGA at DL suspension period end from being "above normal operating balances". We interpret "normal operating balances" to be in-line with 5D of UST outflows, which have recently averaged around $700b, in line with the Treasury's forecast.

EM: we leave our EM forecast little changed at $350b-500b. There is likely a 1-time EM measure available on Jun 30 of ~$119b; if UST can make it to June 30, they can likely make it through early July. The 1-time measure explains our large EM range.

Deficit: the X-date will ultimately depend on the level of deficits in 2025. We revise our monthly financing need forecast to smooth out our prior deficit forecast and align more with historical trends. If we assume a lower deficit in 1H '25, it may provide Treasury with more time before the debt limit becomes binding.

X-date: we continue to project the X-date will bind in June or July '25, with risk now skewed to July (we acknowledge high uncertainty & low confidence). We assume US election outcomes do not meaningfully impact debt limit timing since DL increases are always politically difficult votes even with unified gov't.

DL impacts: bill supply, excess liquidity, & QT timing

The DL is likely to impact bill supply, excess liquidity, & QT timing.

Bill supply: the DL will limit bill supply in 1H '25. Why? When the DL binds, Treasury cannot grow USTs outstanding. UST debt managers prefer to keep coupon auction sizes stable since they have largest duration risk; debt managers then must cut bill supply to offset coupon growth. We project bill cuts of $476b in 1H '25.

Excess liquidity: lower bills and higher liquidity from TGA drawdown should temporarily drive ON RRP take-up higher. Why? Lower bill supply will richen bills vs OIS & displace MMF investors into ON RRP. While higher ON RRP may suggest more liquidity in the system, the cash should quickly leave once the DL is resolved via a bill financed TGA rebuild. Signs of a liquidity build will be temporary.

QT timing: we expect the Fed will judge it prudent to end QT in late '24 or Q1'25 at the latest to avoid the risk of over draining. Lower TGA will temporarily create a loss of money market signal or Fed "blind spot" in their assessment of "abundant" to "ample" reserves. The Fed likely won't want to keep draining & overdo it when losing the money market signal. These DL dynamics drive our earlier vs consensus timing for QT stop at end '24; however, if Fed officials don't acknowledge these DL dynamics soon we will likely push out our QT stop timing to Q1 '25. The later the QT end in DL "blind spot", the greater the risk of over draining.

Market impact: 1H '25 stable funding, 2H '25 tighter

DL will support very different funding dynamics in 1H vs 2H '25. In 1H '25, upward funding pressure should moderate or reverse some of the recent rise in SOFR. This will be driven by lower bills outstanding & TGA drop. In 2H '25, funding markets will tighten with higher bill supply, a TGA rebuild, & Fed ON RRP drain.

To position for these dynamics our ideal expression would be a '25 SOFR/FF spread curve flattener. We expect that Mar '25 SOFR/FF will widen with bill paydowns & TGA decline while Sept '25 SOFR/FF will lag due to expectations for future TGA rebuild. The spread between these tenors is currently flat & we would target a spread of 3.5bps. However, liquidity in the Sept '25 SOFR/FF contract is thin. An alternative expression is long Mar '25 SOFR/FF vs short 1y1y SOFR/FF. This spread is currently 1bps & we recommend clients target 4bps with a stop of -1bps. Risk is an early debt limit resolution or very near term SOFR spike.

We currently prefer expressing DL dynamics as a SOFR/FF spread trade vs outright long Mar '25 SOFR/FF due to year end funding concerns. If USD funding tightens into year-end it will result in tightening pressure on SOFR/FF. We expect to have a better sense for year-end funding dynamics by mid/late October (street balance sheet tends to tighten into Oct 31 Canadian fiscal year end). If year-end funding pressure appears benign, we would have more confidence to recommend outright long Mar '25 SOFR/FF.

Bottom line: we update our DL expectations with most recent data. We still expect X-date in summer & now see it most likely in July (with elevated uncertainty). DL will lower TGA, increase cash in system, & temp ease funding into X-date. Post X-date TGA will rise & tighten funding. Trade = SOFR/FF curve flattener.

 Front end - EU

• We expect the Eurosystem balance sheet to fall to €6.4trn by end-2024, incorporating our latest MRO/LTRO expectations as the refi-depo corridor narrows.
• We stay paid EUR FX-Sofr basis and close our 2y 3s6s widener.

ECB balance sheet update

We forecast the Eurosystem's balance sheet to fall from €6.9trn at end-2023 to €6.4trn by end-2024 and €6.1trn by end-2025. The European Central Bank (ECB) will narrow the spread between the main refinancing (refi) and the deposit facility (depo) rate from 50bp to 15bp, effective 18 September 2024. This will be the tightest refi-depo spread since the ECB was established. Recent cross-border flows suggest that the initial main refinancing operation (MRO)/ longer-term refinancing operation (LTRO) take-up by banks in September will be between €50bn and €65bn.

The ECB's current set of open market operations will not help banks satiate demand for long-term funding, in our view. This suggests term premium will continue to build across the curve over time. But recent euro interbank offered rate (Euribor) fixings have prompted our 2y EUR 3s6s widener recommendation to breach its stop and therefore we close that recommendation.

As the ECB continues, and is set to accelerate, quantitative tightening (QT), we expect banks to rely more on central bank funding to meet their reserve demand over time. The ECB's shift to a demand driven floor system is different from the US Federal Reserve's ample reserve system: scarcity of euros vs US dollars will grow over time. We stay paid EUR FX-secured overnight financing rate (Sofr) basis.

      Technicals

•   US 2Y yield tests 3.55% (its 2023 low) and US 30Y yield test 3.94% (it's 52wk low). These levels, if broken, could extend the decline in US yields in H2. We're sympathetic to some bounce signals at key levels, but no bottom patterns yet.

US 2y & 30y yield test key levels. Big tops or big ranges?

Our base case view called for peak yield by US Memorial Day and lower yields thereafter. We look at the charts of US yields below and see potential for big tops, rather than ranges, as the pendulum swings from up to sideways and then to down. This is part of the reason why we advocated for being long USTs by Memorial Day and buying dips in H2. Now we see US 2Y and 30Y yield teetering on a cliff edge of 3.55% and 3.94%, respectively, as we approach the Fed meeting next week. US 5Y is getting close to its level of 3.20%. The 10Y yield has already made new 52wk lows, but its 2023 lows are still a bit lower at 3.25%. We're sympathetic to some short-term signals that suggest yields have come a long way and may see a bounce higher, such as oscillator divergences and a TD Sequential 13. Usually yields form bottom patterns and/or longs capitulate to provide substance to refute tops/favor ranges. For now, our medium-term bias remains.

 Rates Alpha trade recommendations