Our medium term views

 Our key forecasts

  What we like right now

  The View

 The week that will be

The main focus in the US next week will be the PCE inflation print. Our economists are looking for another good print with a 0.1% mom number in August (see US Watch 12 Sep 24). But with the Fed having demonstrated their focus on the second part of their dual mandate, inflation data may take a back seat. The personal income report will add to GDP tracking, with the Atlanta Fed Nowcast currently at 2.9%. We continue to believe that the combination of a dovish Fed and a resilient economy is a recipe for a steeper curve, including a steeper breakeven curve.

In the Euro Area (EA) we will get sentiment data (PMIs, IFO etc), credit data (M3) and inflation expectations (ECB survey). We also receive the first September inflation prints from France and Spain. We see little in these numbers that would be likely to push the ECB towards cutting in October, further demonstrating the difference in reaction functions. We stay long EUR rates. The Fed is acting to cut the tail risk of a hard landing and an inflation undershoot, the ECB is doing the opposite.

We expect the RBA to stay on hold and the SNB to cut by 25 bp. We are still paid Dec 24 OIS in AUD, and see this week's labor market data as not providing any support for the cuts still reflected in the curve.

Next week sees the LDP leadership election in Japan. The consensus is for a limited market reaction to the outcome. However, the shape of the 2s20s curve could differ depending on the election outcome given divergent views on fiscal and monetary policy (see Rates Japan). Next week's Tokyo CPI data may also bring attention back to fundamentals in Japan. Our economists have revised down their inflation forecasts to reflect the stronger Yen (see Japan Viewpoint 19 Sep 24), but inflation stays around 2% in their new forecasts. This is supportive of further policy normalization in 2025 and ultimately also of further JGB curve flattening.

The week that was

The Fed delivered a 50 bp cut. Our economists now expect another 75 bp of cuts in 2024, and 125 bp in 2025, pretty firmly matching forwards for year-end 2025 (see US Watch 18 Sep 24). We revise our US yield forecasts to reflect the revealed dovish reaction function of the Fed, as well as our economists' new forecasts. We now see US 10y hitting 3.5% before the end of the year, with 2s10s at 25 bp (see Rates US). These forecasts imply lower rates and steeper curves than forwards. We continue to favor steeper curves (5s30s nominal, 5s10s real and in forward breakevens).

The BoE stayed on hold as expected. Our economists have reiterated their view that the BoE will be forced to be more gradual than others (see UK Watch 19 Sep 24). We stick with our bearish real yield views both outright and cross-market. We also remain paid in 3s5s7s Sonia (see Rates UK).

  Rates - US

•      Fed is dovish. Stay bullish duration & in steepeners. We add M5-Z7 steepener.
• We revise rate forecasts lower with faster Fed cuts. QT end pushed to Q1 '25.

Hawkish 50? Not so fast. Fed = dove coop

US rates twist steepened on the week. Sept FOMC saw a remarkable swing in the last week from pricing to a near certain 25bps cut to Fed delivering a larger-than-expected 50bps cut. Rate reaction post Sept FOMC suggested neutral to hawkish reaction; we suggest fading the move & lean into duration & steepeners. Our logic is below.

Actions (& diffusion indices) speak louder than words

The Sept FOMC was unusual given elevated uncertainty on 25 vs 50 & bear steepening following the larger-than-expected rate cut. We suggest fading the rate move post FOMC & "hawkish cut" interpretation. Here is what we learned from the Fed:

Pre meeting guidance: clients have asked: "why did the Fed not better guide the market into Sept FOMC, including via media during blackout?" The most likely answer: they did not object to the market shift. Pre-blackout Waller signaled 25bp cut but clearly didn't object to 50bp. In the 25 vs 50bp debate, the tell on 50bp was lack of Powell sanctioned media pushback. The Fed will correct market if it feels a more hawkish outcome should be priced (Jun '22) but is OK surprising in a more dovish direction (Sep '24). Clients should remember this into 25 v 50bp Nov & Dec FOMC debates.

SEP says dovish reaction function: market reaction to Powell presser suggested a more hawkish tone due to lack of commitment for future 50bp cuts & the possibility of a higher neutral rate. The Sept FOMC statement also did not show an explicit downside balance of risk bias. We take more signal from the Summary of Economic Projections (SEP). The SEP reflected a dovish reaction shift via FOMC participant assessment of risks to their unemployment (higher) & inflation outlooks (more balanced).

The SEP creates a "diffusion index" from balance of risks to inflation & unemployment (index = {# who say risks to upside - # who say risks to downside} / total). To get a sense of Fed risks to the outlook we look at the difference in diffusion indices for core inflation & unemployment (Exhibit 4). Large upward shifts in the U3 diffusion index tend to line up with large moves in rates. U3 risk higher => rates lower.

Fed dots too hawkish vs rules: Powell called the 50bp cut a "recalibration". The Fed is "recalibrating" to where their policy rules tell them to be (Waller Sept 6 = "Reducing the policy rate now is consistent with many versions of the Taylor rule"). Standard Taylor today = FF at 3.8%. The Fed has 100bps of more "recalibration" before current Taylor.

Standard Taylor can be easily calculated from Fed SEP median economic projections & Fed "star" variables (r*, u*, i*). The Fed's SEP median FF projections are still hawkish vs Taylor, but slowly converging (Exhibit 5). We assume the Fed's dots will continue to converge to Taylor & imply a lower FF rate over time. Lower FF = bullish bonds.

Fed bottom line for rates: dovish. Fed didn't push back on market via WSJ in blackout b/c they are dovish & willing to deliver more cuts than market. Market was disappointed by lack of future Powell 50bp signal but SEP suggests dovish balance of risks. Fed slowly moving to Taylor which means lower FF. Lower FF = bullish bonds. Buy the rate dip.

Dovish Fed & front-loaded cuts = bullish steepeners

Our dovish Fed interpretation & this week's 50bp suggests front loading of rate cuts. Fed front loading suggests reaching the trough faster & curve steepening, esp with still solid US economy. We recommend front end M5-Z7 SOFR curve steepeners. The M5-Z7 spread is currently flat & we target a spread of +50bps with a stop of -30bps. Risks to the trade are a slower pace of Fed cuts & sharper economic slowdown in '26 / '27.

Exhibit 6 shows a comparison of the priced Fed path the day cuts started vs what the Fed delivered across recent cycles. Historically the market has understated the speed and extent of cuts delivered. This cycle, however, the market is already pricing faster cuts in 3/5 recent cycles & near typical cuts observed on average in the first 6 months. This cycle the Fed has already showed its dovish hand. We still see ability for the M5-M7 steepener to perform from bullish side (soft data and Fed delivers greater cuts) and bearish side (Fed pivots too dovish and market begins to price subsequent hikes).

New Fed call & US rate forecast revisions

Our US economics now expect faster & deeper Fed cuts (see Sept FOMC). They expect another 75bps in '24 (Nov = 50, Dec = 25, end Dec '24 FF = 4-4.25%), 125bps in '25 (Jan/Mar 25 each, 3 quarterly 25bp cuts after, end '25 = 2.75-3%). The see a modest economic, labor, & inflation rebound in 2H '25 & '26.

We revise our rate forecasts accordingly (Exhibit 7). Our new 2Y & 10Y: end '24 = 3.1 & 3.5, end '25 = 3.2 & 3.6. Our forecasts are below the forwards & below consensus near term. Our 2Y end '25 forecast includes some expectation for future rate hikes given solid US economy & inflation re-acceleration risk. Our long end forecasts include gradual cheapening & higher rates to reflect tighter spreads & supply / demand issues. We also push out our QT stop from end '24 to end Q1 '25 (for detail see US front end).

Bottom line: Sept FOMC was dovish to us & implies lower FF. Lower FF = bullish bonds. We recommend M5-Z7 SOFR curve steepeners for cut front loading & solid economy. History also says Fed front loads. Forecast changes: rates slightly lower & below forwards; QT end timing is pushed out to end Q1 '25.

  Rates - EU

• Three parts of the EUR rates market are attracting attention: the curve steepening, the drop in implied vols and tightening in periphery spreads. We believe they are all consistent with global price action, and in particular the Fed induced risk-on move.
• We update our regressions to check what this means for 10s30s and swap spreads.

Global versus domestic factors for 10s30s & swap spreads

10s30s under steepening pressure, but 30y flows can create some resilience

We have been surprised by the extent of the 2s10s curve steepening in EUR. Our view of a slow ECB, sticking to quarterly cuts possibly up until March and then pausing at 2% despite sub target inflation, would be consistent with a flatter curve and a slower pace of steepening in the next couple of quarters. Still, in the rally, the market was quick to transpose the Fed's expected reaction function to the ECB, while the post Fed US bear steepening and risk-on move also affected the EUR curve. It provided an additional impetus to the positive carry forward steepeners that had been already in vogue.

Our view on the long-end is still that receiving flows that create some resilience. So far, looking at overall price action in the past week, month, and quarter, we find that the 10s30s steepening matched what the 2s10s steepening and decline in vol suggested (Exhibit 8). The 10s30s curve trades with a stable residual of c.5bp (too flat - Exhibit 9).

We expect receiving flows to intensify, making the residual more negative for 10s30s (we are less optimistic for 30s50s given our view that future Dutch pension fund receiving is likely to shift lower on the curve ahead of the switch to DC). One way we positioned for the resilience of 10s30s in overall curve steepening was via a paid position in the 5s10s30s fly. One could also consider a paid in 2s10s30s (0.21:0.79 weights to match the above betas). However, paying the latter is more negative carry, and is less consistent with our view of a flatter 2s5s curve. A risk to both is a further decline in vol, but we would argue that the selloff in gamma is starting to appear stretched considering the level of rates and what we think is a large mispricing of the neutral rate in EUR. At 65bp/annum, 3m10y implied vol is lowest since Jan-22.

German swap spreads: an updated regression is warranted

The risk-on environment, with lower implied vol and tighter periphery spreads can be seen as supportive of further cheapening in German swap spreads. However, given the extent of the tightening observed already over the past year, many are uncertain about the additional potential. The regressions we used over the last few years (to determine whether swap spreads are trading consistent with other markets) have been showing a persistent negative residual since the end of last year (bonds looking too cheap).

We interpret that as suggesting that the regressions may not capture enough the impact of ECB QT and/or market perception of present and future collateral abondance. To address this, we modify the analysis below, by introducing the German General Collateral repo (GC) spread to €str as additional variable from Mar-23, when ECB QT started. We also split the German Specific Collateral repo (SC) spread to €str in 2 separate variables: SC-GC and GC-€str to easily read the added sensitivity of swap spreads to GC-€str since QT started. Resulting coefficients are in Exhibit 10, using data since May-18. We compare the resulting residuals to those from our previous regressions in Exhibit 12 & Exhibit 13.

Current residuals in Exhibit 11 point to marginal cheapness in 5y bonds vs swaps, while 2y and 10y spreads appear close to fair. We continue to favour longs in 2y spreads vs 5-10y spreads (the above would suggest 10y are a better short than 5y now). 2y would benefit more from a recovery higher in vol and any risks around French politics / periphery, while 10y spreads appear more sensitive to further reduction in collateral scarcity.

For 30y spreads, we use the level of rates instead of vol as variable and consider a shorter period (past 3y) given the different structural changes that have occurred in terms of receiving flows in the back of the curve between 2018 and 2020. As expected from the traditional dynamic hedging flows from Dutch Pension Funds (receiving in swaps when rates break certain levels to the upside), we find that 30y spreads tend to cheapen when rates rise. That said, this directionality can diminish in our view in the rally, as we expect the pension reform to drive receiving near term even in rallies.

   Rates - UK

• BoE QT decision delivered a degree of fiscal respite. Our refreshed BoE balance sheet projections now see reserves falling to around £530bn in Jan'26.

 Respite

The Bank of England (BoE) voted 8-1 to keep the Bank Rate unchanged at 5.0% at its meeting today. The voting was more hawkish than we had expected. The BoE's commitment to a gradual approach to easing resulted in a slight trimming of the market's rate cut expectations, but pricing continues to imply a somewhat faster-than-quarterly pace to mid-2025. Terminal rate pricing rose to around 3.4%, now above our base case of 3.25% which we see being reached in mid-2026 (Exhibit 14, Exhibit 15).

Yesterday's MPC did not alter our UK rates views materially:

We expect UK real yields to underperform vs. EUR: UK's fiscal slippage is evident & presents upside risks to 5y RYs, while falling European inflation should allow the ECB to cut real & nominal policy rates swiftly. UK's deteriorating IIP/GDP ratio vs the EZ is another factor supporting this view. We keep pay 5y UK real rates to receive 5y Euro real rates entered at 43bp pickup with a target of -40bp and stop at 90bp (Liquid Insight, 21 August). Current level: 46bp. Risk to the trade is a recessionary threat seeing the Bank of England prioritize underpinning growth over bringing down inflation (as it did in 2016 after the Brexit referendum).

Further underperformance of 5y vs. to 3y and 7y in Sonia: beyond terminal rate pricing, we expect potentially more mortgage hedging (paying fixed) ahead to support underperformance of the 5y area of the curve. We keep Sonia 3s5s7s fly (pay belly, receive wings) entered at -12bp with a target of 10bp and a stop of -21bp (UK Rates Viewpoint, 5 September). Current level: -10.7bp. Risk to the trade is UK government increasing Stamp Duty tax at the Budget, slowing housing market activity.

Low coupon/high coupon trade to benefit from less low coupon Gilt supply from the BoE now that the Bank will be selling only £13bn in the new QT year. We keep selling UKT 4.5% 2028 Gilt to buy UKT 0.5% 2029 entered at -8bp with a target of -20bp and stop at 4bp (UK Rates Viewpoint, 5 September). Trade risk is a change in the tax treatment of Gilts for retail.

Long Gilts to find some support from BoE QT and DMO Remit outcomes, with long-end swap spreads faring slightly better than 10y UKTs and 30y USTs (with the Fed's Powell giving no indication yesterday that the Fed is contemplating a further slowdown/stop to QT yet). We also note the BoE said that Q325 "is likely to consist of only short and medium maturity sector auctions, depending on the movement in Gilt prices and the realised distribution of sales throughout the year". We keep long UKT 4 3/8 07/31/2054 vs. T 4 5/8 05/15/54 on ASW entered at 1bp (z-spd) with a target of -15bp and stop at 10bp (UK Rates Alpha, 12 July). Current: 2.8bp. Risk to the trade is a meaningful increase in long-end DMO Gilt supply from the "unallocated".

QT decision delivers a degree of fiscal respite

Although £100bn was the market's median prediction for the QT pace over the year beginning October, comprising £87bn passive and £13bn active, the fatter tail of expectations was to the upside, we would say, with a bigger minority expecting a larger active component than the one expecting no active sales at all.

So, to an extent, the news provided the market and the government with a degree of respite - in terms of the effective total Gilt supply to be absorbed and in terms of the prospective losses to be crystalised (with implications for the debt path and funding need). The 2bp richening in 30y Gilt ASW spreads on the news, lifting them off an eight-month low, was therefore understandable.

Although there were several calls (including from ourselves) for either the addition of a 1-3y Gilt sales bucket to the QT programme or a skew shorter to sales, this recommendation was always something commentators felt the Bank should do rather than something anyone felt it would do.

BoE balance sheet update: incorporating QT pace for the new year

We refresh our BoE balance sheet projections in light of today's BoE QT pace announcement. We now expect reserves to fall from £800bn at end-2023 to around £530bn in January 2026 (Exhibit 16).

We continue to assume rising TFSME repayments from 3Q24, as banks attempt to avoid any increased cost of replacement funding associated with what would otherwise be a £56bn TFSME maturity "cliff-edge" in 4Q25. TFSME repayments have picked up so far this quarter, already exceeding the prior quarterly peak (Exhibit 17). We do not know how banks will choose to manage down the cliff-edge issue, but sequentially raising quarterly repayments from late 2024 would make sense to us. We assume a £13bn repayment in 4Q24, rising by £2bn/quarter thereafter.

Increasing TFSME repayments, together with quarterly APF Gilt redemptions totalling around £29bn per quarter between 1Q25 and 3Q25, imply a rapid BoE reserves shrinkage over that period.

     Rates - JP

•  LDP leadership election will effectively select new PM; market consensus is for limited rates market impact
• Bond market reaction could be outsized depending on the winning candidate

Updates on LDP leadership election

The LDP leadership election is scheduled for 27 September. The consensus from a survey of bond market participants is that the outcome will have little impact on the rates market. However, we think the 2yr20yr curve could either steepen or flatten depending on the winning candidate.

Election overview

The election was triggered by the end of current party leader Fumio Kishida's term on 30 September. It will be held on 27 September, with nine candidates running. Given the LDP's position as dominant party in the ruling coalition and the fact that Mr. Kishida is not in the race, the change in LDP leadership will also result in a new prime minister. The result will be decided by the votes of 367 LDP members of parliament plus a further 367 votes allocated to party members, with the new party president elected for a three-year term. If none of the candidates wins a majority of the 734 (367 x 2) total votes, the top two will proceed to a runoff election in which the candidate with the largest share of a combined 414 votes (367 MPs¹ + 47 votes for prefectural LDP branches) is elected as party leader.

Recent developments

Media reports suggest that at this point no candidate is likely to secure a majority of votes, increasing the likelihood of a runoff election².

 Bond market participants' views

52% of respondents to a Quick survey thought the results of the LDP leadership election would have virtually no impact on the bond market, 40% expected a moderate impact, and 6% foresaw a major impact. This likely reflects the record number of candidates in the race (nine), the fact that many are not proactively seeking fiscal consolidation³, and the lack of specific measures even from those who favor fiscal consolidation⁴. As such, in our view the survey result stems from a lack of clarity on how the candidates differ on monetary and fiscal policy.

  Front end - US

Fed not focused on QT, so end date must shift

The minimal discussion of the Fed's balance sheet policy at the Sept FOMC indicates risk of a later end to QT. We had thought QT would be slowed or stopped with a 50bps cut; our logic = the Fed would not want to send mixed messages with the policy rate & balance sheet policy, esp when making such a large FF move. The Sept FOMC shows the Fed sees moves in the policy rate & balance sheet as independent.

We have long expected the Fed to be prudent in their QT decision making to avoid a repeat of the repo market volatility of 2019. The debt limit (DL) (back in early '25) creates a funding "blind spot" we thought the Fed would avoid.

The lack of Fed QT focus now pushes our modal base case for QT stop from end '24 to end Mar '25. We continue to believe the Fed *should be* cautious about continuing QT during DL. The lack of QT focus suggests later end date & more potential funding vol.

DL blindside: fool me once, shame on you, fool me twice…

The DL will temporarily cloud the Fed's ability to accurately monitor liquidity and funding in 1H of '25 (see debt limit). During the DL, we typically see the Treasury's cash balance decline which will offset the impact of QT on liquidity. The TGA drop temporarily halts or reverse upward pressure in funding markets. We saw this occur in the last episode of QT, where TGA drained but reserves and SOFR remained relatively stable (Exhibit 20).

Once the debt limit is resolved, there will likely be significant upward pressure in funding as the TGA rebuild swiftly drains liquidity. The debt limit resolution in Aug '19 was soon followed by significant repo market volatility in Sept '19. Notably, the TGA drain and corresponding refill will be roughly 2x larger than in '19. This will likely amplify DL funding distortions. We are concerned the Fed may not be strongly considering debt limit dynamics in their QT end expectations, which risks repeat of prior mistakes.

To analyze further, we project the Fed balance sheet assuming no debt limit impact (Exhibit 21). In doing so, we assume (1) a steady TGA from the Treasury's $850b forecast for Q3 '24, (2) 100% of the QT drain from ON RRP until ON RRP $0, (3) BTFP drain comes out 50/50 from both ON RRP and reserves until ON RRP is $0 (4) no change in redemption caps. In this scenario, we find that the ON RRP would fall to $0 by Jan '25 and reserves would quickly fall from $3.3tn to <$3tn (or 10% of GDP) by May '25.

Conversely, if we assume a DL debate that brings TGA near $0 as we have seen in recent episodes, TGA drain could drive up to $850b into financial markets (Exhibit 22). As a result, ON RRP will likely remain elevated, reserves flat, and funding pressures minimal. The longer the Treasury is at the DL the more cash will enter the system and the more significant the backside liquidity drain. We assume DL down to wire regardless of election outcome given it is an unpopular vote.

In Exhibit 23, we compare our "no debt limit" scenario to our "debt limit" scenario forecasts for June '25. The no debt limit scenario would show a lower level of liquidity in the system with ON RRP at $0 and reserves at $2.9tn. This would be a level of liquidity around the 10% of GDP threshold the Fed has previously communicated would be an appropriate time to end QT. If the Fed was continuing QT during this scenario, we believe they would have likely ended QT earlier. However, if they continue QT into the debt limit period, liquidity will falsely appear more abundant because of the TGA drain. If they wait until June or July to end QT, once the debt limit is resolved, liquidity will likely quickly become difficult to source, driving funding pressure and market volatility.

We continue to believe the Fed should be careful in considering when to end QT. It will be difficult for the Fed to communicate why they are ending QT during a period where ON RRP take-up is positive and reserves still appear abundant. We believe it would be necessary to end QT by March '25 while reserves are still ample to avoid a significant spike in funding rates post debt limit resolution. In doing so, we project Fed liquidity will remain above 10% of GDP through the remainder of 2025.

Bottom line: Sept FOMC indicated minimal focus on QT end, esp with unusually large 50bp Fed cut. We push our QT end date from end '24 to end Mar '25. DL creates a funding blind spot for Fed that should otherwise see QT stop by Q1 '25. Risks are building the Fed might repeat mistake of past & push QT too far in DL period.

         Front-end - EU

• Low €2bn take-up from first MRO by banks with new corridor; we see downside risks to our OMO take-up near-term that could keep front-end basis tight for now
• We maintain a structural euro bases widening view and we stay paid EUR FX-Sofr basis

Low MRO take-up under tighter corridor

The ECB conducted the first MRO under its new 15bp refi-depo corridor, vs 50bp previously. 29 banks borrowed €1.935bn. In the previous week, 24 banks borrowed €2.053bn. MROs have a maturity of one-week.

Our base case for MRO plus LTRO take-up in September 2024 is between €50bn and €65bn (see European Rates Watch, 3 July 2024). The MRO take-up this week suggests MRO and LTRO take-up this month could be low vs our expectations. Possible reasons for the low take-up in this MRO are 1) cross-border flows continued with reserves moving from cash rich banking systems to cash poor banking systems, 2) stigma from banks' perspective of tapping too much from the first MRO, 3) bank preferring to borrow from LTRO rather than MRO, and 4) banks willing to hold less reserves for HQLA purposes. LTROs have a maturity of three-months.

For our base case to be realised, we will need to see an increase in the take-up from next week's MRO and LTRO, both will be allotted on 24 September and settle on 25 September. In our view, this increase is quite meaningful and therefore there are downside risks to our expectations near-term. We are also cognisant of potential desire by banks to shift LTRO demand from the September operation to the October operation where possible, as the latter will cover the year-end turn.

The low take-up from the first MRO under the new refi-depo corridor could keep front-end euro specific bases, such as FRA-€str and 3s6s tight near term, as the market interprets the low take up as a signal of very limited euro funding pressure. We believe sentiment on euro funding could turn when usage of the ECB's OMO rises meaningfully.

Our structural view that banks will be increasingly reliant on the central bank for funding has not changed. The ECB has chosen a demand-driven floor system as it continues QT, which means banks are expected to turn to it for funding as reserves continue to decline. The low MRO take-up this week may delay but, in our view, is unlikely to prevent widening pressures on euro FRA-€str and 3s6s bases over time.

On a cross market basis, we see euro funding pressures rising vs US dollar funding on divergent QT outlooks between the ECB and the Fed. We stay paid 9Mx12M EUR FX-Sofr initiated in May 2024 (current: -6.4bp, target: -2.0bp, stop: -10.2bp). Risks are the Fed continues QT for longer than expected, the ECB does not accelerate QT, and strong take-up in the ECB's structural credit operations when available.

  Spreads - EU

• Curve dislocations across Euro government bonds are not normalising post summer break
• Flows related to French political turmoil as well as longer lingering auction discount may play the main role for this

Euro govie bond dislocations not adjusting post August

A number of metrics show a sustained increase in Eurozone Government Bond yield dislocations around each sovereign curve since June in particular.

The timing of the move aligns with the turmoil from French politics but also with the wider repricing of central bank rate policies and the resulting rates decline.

We slice and dice the data available to us to better understand the dynamics, and the sustainability of this trend. For this, we are mainly relying on the micro relative value framework we developed recently (read European Rates Watch: EGBs, bond by bond 08 March 2024). We like using this framework for the exercise because the curves we use as the reference of the calculation for rich/cheap differentials are fitted to discount factors on a cash-flow level basis: this therefore adjusts for pricing differentials originating from specific features such as the coupon size.

Bond dislocations amongst the top four markets have increased significantly both outright as well as relative the smaller markets (Exhibit 24). In fact, these curve dislocations have now surpassed those of the smaller ones also in absolute terms.

French elections and supply reducing RV mean-reversion

Curve dislocations, including those in the high/low coupon bond spectrum, can appear for a variety of reasons ranging from differences in liquidity, tax treatment, balance sheet impact, convexity and default risk variations. While not unambiguous, the oversized increase in dislocations across very liquid markets vs the less liquid ones may suggest that the driver of this trend is not down to liquidity considerations. Also, these dislocations have increased relative to August, when liquidity seasonally drops.

When further dicing these dislocation measures we note that curve dislocations have also typically increased across the low-high coupon spectrum, with high coupon bonds cheapening on the curve more than usual relative to the low coupon ones. This is especially true 1) in the German and French front-end (Exhibit 25- since June's political turmoil in France) 2) across top markets (ex Spain) in the belly of the curve (Exhibit 26) 3) in the Spanish and French long-end. This may suggest that a phenomenon similar to the UK's reduction of low coupon supply is also now affecting Euro Govie pricing.

Flow deviations from the French bond market implicit downgrade seems to have a role in this given the country split across the three main parts of the curve but we also note that supply events may have started to have a longer lasting impact on a bond by bond basis (Exhibit 28): bonds with a recent supply event tend to trade cheaper on the curve relative to those not being tapped than usual.

  Technicals

•   Ahead of the FOMC we flagged short-term signals suggesting yields may bounce. A post Fed bounce is underway, especially for US 30Y yield. It initially made a new low and, as of writing, is above the high of last week (an engulfing candle).
• Such a candle implies the bounce could have more to go in the coming weeks. We show the top of its channel and 50d SMA at about 4.20-4.15%. Above this is an old breakdown level at 4.30-4.33%.
• For now we view this as countertrend because we do not see bottom patterns in the yield charts yet, the weekly charts still look like cycle tops and wave (c) is typically deeper than we've seen, such as 3.61% on the US 30Y.

US 30Y yield: Technical bounce underway

  Special Topic: Vol dynamic post FOMC

• Vol lower after the Sep FOMC with the left side leading the way. We stay biased lower on dovish Fed response function and as election event risk passes.

Vol dynamic after the FOMC meeting

In Fed, elections & volatility, from 17 Sep 24, we previewed the potential reaction of the US volatility grid to different scenarios for the Sep FOMC meeting. We noted that our baseline has been to see vol drift lower as the Fed starts to ease policy (1y10y moving from c.100-115bp into an c.85-100bp) and for the underperformance of the left side vs the right side (into flat levels) as easing gains momentum. We argued that we were more likely to frontload this dynamic on a 50bp vs a 25bp move in Sep.

Post Sep FOMC we saw vol broadly lower across the grid (Exhibit 30), with 1y10y pushing under 100bp vol (Exhibit 31), and the underperformance of the left of the grid vs the right side (Exhibit 30 & Exhibit 32). We see scope for vol to continue to push lower, particularly after the passage of the next calendar driven event risk which are the US elections in early Nov (see Election pricing in rates & FX vol, from 16 Sep 24).

In Q&A, the 50bp move was framed as a policy recalibration in the context of a material overshoot of policy rates vs neutral in the current cycle (Exhibit 33). We see indications of a more dovish Fed response function, which supports our lower vol bias. Significantly, we see a reflection of the latter in lower slowdown (-5%) and hard-landing (-2%) likelihoods expressed in the dynamic of 10y BEs (Exhibit 34) & 2s10s curve (Exhibit 35).

  Rates Alpha trade recommendations