版權(quán)說明:本文檔由用戶提供并上傳,收益歸屬內(nèi)容提供方,若內(nèi)容存在侵權(quán),請(qǐng)進(jìn)行舉報(bào)或認(rèn)領(lǐng)
文檔簡(jiǎn)介
2September2024|11:28AMBST
GlobalMarketsAnalyst
LessonsFromG10CuttingCycles
nWithmostoftheG10eitherhavingjustbegunorabouttobegincuttingpolicyrates,welookatthereturnpotentialofavarietyofratesinvestmentstrategiesduringG10cuttingcyclesoverthelastthreedecadestounderstandhowto
positiononcecutsbegin.
nWe?ndthatoutrightlongsgeneratethehighestleveragedreturnsduringcuttingcyclesofalltypes,withthe2ypointconsistentlyonaveragebestforbothtotalandvol-adjustedreturns.2s10ssteepenersofferthebestvolatility-adjusted
returnsamongcurveformats.Mostofthereturnstolongsandsteepenersaregeneratedinthemonthsintoandimmediatelyfollowingtheinitialcut.
nAtcurrentlevelsoffront-endinversion,suchas1y1yvs1y,itwouldbe
unprecedentedtogeneratesigni?cantpositivereturnstolongpositionsintheabsenceofarecession.Andalthoughnon-recessionaryreturnstosteepenersaremixed,theinitial?atnessofthecurvesuggeststherisk/rewardto
steepenerpositionsislikelytobebetter.
GeorgeCole
+44(20)7552-1214|george.cole@
GoldmanSachsInternational
FriedrichSchaper
+44(20)7774-7906|
friedrich.schaper@
GoldmanSachsInternational
_
Investorsshouldconsiderthisreportasonlyasinglefactorinmakingtheirinvestmentdecision.ForRegAC
certi?cationandotherimportantdisclosures,seetheDisclosureAppendix,orgoto/research/hedge.html.
GoldmanSachsGlobalMarketsAnalyst
LessonsFromG10CuttingCycles
ThemajorityofG10centralbankshaveeitherstartedorarepoisedtobegincutting
rates.Theeventualreturnsto?xedincomewilldependontheevolutionofthedataandtheextentofcutscomparedwiththecutsalreadypriced.Tounderstandhowthemarkethaspricedrisksinthepastandtoshedlightonthepotentialpathforreturnsinthe
future,weanalysepreviousG10cuttingcyclesandthereturnstoavarietyofstrategiesacrosstime(Exhibit1).
Exhibit1:G10cutsarebroadening
G10policyrates
10
8
6
4
2
0
-2
%
AUDNOK
CADSEK
EURCHF
JPYUSD
NZDGBP
%
10
8
6
4
2
0
-2
199519971999200120032005200720092011201320152017201920212023
Source:HaverAnalytics,GoldmanSachsGlolalInvestmentResearch
Weidentify74cuttingcyclesinG10(Exhibit2).De?ningacuttingcycledoesrequire
somejudgement,giventhatcyclesareinherentlyuncertainandcanstopandstart
_
dependingonincomingdata.Aseriesofcutspunctuatedbyextendedpausescouldbeconsideredasinglecycle,ormultiplesmallercuttingcycles-hereweseparatecuttingcyclesifthereisapauseofsixmonthsormore.Insomecasesweareconstrainedbydataavailability–ifnotforthepolicyrate,formarketratesacrossthecurve.Inour
analysisweclassifycuttingcyclesas:
1.Initialcuts,forwhichwerequirethatthepriorpolicyratemovewasahike.Thisdistinguishesfromcutsthatmaybearestartingofacuttingcycleafterapause.
2.Cutsthatareassociatedwitharecession.
Thereisoverlapbetweenthecategories,butthisallowsustogenerateresults
dependingonthedifferenttypesofcuttingcycles.Examiningreturnsunderthedifferent?avoursofcuttingcyclethenallowsforaricherdiscussionoftherisk/rewardofeach
strategygiventheuncertainpathahead.
2September20242
GoldmanSachsGlobalMarketsAnalyst
Exhibit2:G10cuttingcycles
Assumescuttingcycleendsifratesonpauseforatleast6m
G10cycles
Australia
Canada
EuroareaJapan
NewZealand
Norway
Sweden
Switzerland
US
UK
G10
no.ofcycles
11
8
6
6
5
9
8
4
8
9
74
Averagelength(m)
7
11
10
6
9
10
11
5
8
6
8
Averagedepth(bp)
-124
-220
-125
-43
-190
-197
-191
-100
-180
-137
-155
Exhibit3:Deeprecessionsin?uencefront-endreturnaverages
1y1yrealisedreturnsvsinitial1y1yvs1ylevel
2007/08500Buy-and-holdreturnstolong
1y1yfromfirstcut(bp)
400
higherreturns
300
200
100
Initiallevelof
1y1yvs1y(bp)
0
-150-100-50050100150
-100
steeperfront-endatfirstcut
-200
Source:GoldmanSachsGlobalInvestmentResearchSource:GoldmanSachsGlobalInvestmentResearch
Long1y1y:betterreturnswhencurveissteeper,unlessinrecession
The?rststrategyweconsiderissimplyreceivingandholdingtomaturity1y1yratesacrosscuttingcycles.Thisteststhedegreeofinversionand/orpremiumagainsttherealisedvaluefor1yratesinoneyear’stime.Exhibit3showstheinitialfront-end
inversion,measuredby1y1y–1yswaprates,vs.thegainsmadebyreceiving1y1yratesfor1-year.Inthisinstance,they=xlinewouldrepresentexactlymeetingtheforwardcurve.
We?ndthatreturnsaregenerallypositive,indicatingtheexistenceofpositiverisk
premium.Areasonableprioristhatreturnswouldbepositivewhenthecurveissteep–thisisindeedwhatweseeinthetoprightquadrant.Buttheoverallrelationshipinthefullsampleisnotpositive–andinparticularreceivinginvertedcurveshasgenerated
signi?cantpro?tsinthepast(circledpoints).Thesepointsare,however,G10cutting
_
cyclesthrough2007-2009,wheremeaningfulcutpricingwaseclipsedbytheeventualcutscentralbanksultimatelydelivered.
2September20243
GoldmanSachsGlobalMarketsAnalyst
Exhibit4:Recessionsgeneratereturnsto1y1ylongsdespiteinversion
1y1yrealisedreturnsvsinitial1y1yvs1ylevelinrecessions
Exhibit5:Intheabsenceofrecession,steeperfrontendimplieshigherreturns
1y1yrealisedreturnsvsinitial1y1yvs1ylevelinnon-recessionarycycles
2007/08
500Buy-and-holdreturnstolong1y1yfromfirstcut(bp)
higherreturns
400
300
200
steeperfront-endatfirstcut
100
Initiallevelof
1y1yvs1y(bp)
0
-50
0
50
100
-100
-100
400
higherreturns
Buy-and-holdreturnstolong1y1yfromfirstcut(bp)
300
200
100
Initiallevelof
0
1y1yvs1y(bp)
050100
-100-50
-100
steeperfront-endatfirstcut
-200
Source:GoldmanSachsGlobalInvestmentResearchSource:GoldmanSachsGlobalInvestmentResearch
Toseethiswebreakdownthesampleintorecessionarycuts–thestartofcutting
cyclesassociatedwithrecessionwithin6m–andintocutsthatrepresentthestartofacyclefollowinghikes(asopposedtoapausedcuttingcycle)andthatare
non-recessionary.Thislattercategory?tsourmodalmacroforecastsmostclosely.
Exhibit4andExhibit5showthisbreakdown.Wecanseethatthenegativerelationshipisdrivenbyrecessionarycuts–whenthereisarecession,centralbanksreliablycut
morethanisinitiallypriced.
Incontrast,whencutsfollowhikesandthereisnosubsequentrecession,returnsaremoremixed–therearecertainlystillpositivereturns,buttheyareslightlypositivelycorrelatedwithinitialinversion,suchthatgreaterinversionimplieslowerreturnson
_
average.Thisisparticularlyrelevantgiventhatcurrent1y1y-1ylevelsareclosetoorbeyondthemaximumlevelsofinversionwehaveseenatthestartofprevioushikingcycles.
Outrightlongsvscurvetrades&tradetiming
Thenextsetofstrategiesweanalyseareleveragedexpressionsonthespotcurve.Welookatleveragedlongpositionsin2y,5y,10yand30yrates,aswellascurvesteepenerpositionsacrossallcombinations:2s5s,2s10s,2s30s,5s10s,5s30s,and10s30s.We
baseourtotalreturnseriesonswapratesandcapdurationonceyieldsgonegative.Wetakethetotalreturnforeachcurvepointnetofthetotalreturntolong3mrates.Toallowforcomparisonbetweencurvepointsandcurvestrategies,wematchthedv01ofeachexpressionbyadjustingforthedurationoftheposition.Wecalculatetheaveragedaily
returnstoeachstrategyfromthemonthpriortothe?rstcut,andthenforthefollowingsixmonthsortheendofthecuttingcycle,whicheverisshortest.
2September20244
GoldmanSachsGlobalMarketsAnalyst
Exhibit6:Fastanddeepcuttingcyclesgeneratethereturns,regardlessofstartinglevelofrates
Returnstolong5youtrightvsinitial5ylevel,cyclelength,depthandspeed
2.5
2.0
1.5
1.0
0.5
0.0
-0.5
-1.0
-1.5
-2.0
2.5
2.0
1.5
1.0
0.5
0.0
-0.5
-1.0
-1.5
-2.0
avgdailyreturnstolong5y(bp/day)
startinglevelof5yrates(ppt)
0246810
avgdailyreturnstolong5y(bp/day)
cyclelength(m)
0510152025
2.52.5avgdailyreturnsto2.5
long5y(bp/day)
2.02.02.0
1.51.51.5
1.01.01.0
0.50.50.5
0.00.00.0
cycledepth(bp)
-0.5-0.5-0.5
-1.0-1.0-1.0
-1.5-1.5-1.5
-2.0-2.0-2.0
-700-600-500-400-300-200-1000
2.52.5avgdailyreturnsto2.5
long5y(bp/day)
2.02.02.0
1.51.51.5
1.01.01.0
0.50.50.5
0.00.00.0
cyclespeed(bp/month)
-0.5-0.5-0.5
-1.0-1.0-1.0
-1.5-1.5-1.5
-2.0-2.0-2.0
-80-70-60-50-40-30-20-100
Source:GoldmanSachsGlobalInvestmentResearch
_
ThepanelsinExhibit6&Exhibit7showtheunconditionalresults,wherethereturnstoasubsetofstrategies–long5yand2s10ssteepeners–ontheverticalaxisare
scatteredagainstavarietyofcycleparameters,suchasthelength,depth,andspeedofthecuttingcycle,aswellasthestartinglevelofthecurveslope(thishasabetter
relationshipthanthestartinglevelofthepolicyrate).Intuitively,andperhaps
arithmetically,longer,deeperandfastercuttingcyclesareassociatedwithhigherreturnstooutrightlongsandcurvesteepeners.Steeperinitialcurvesareweaklyassociated
withlowerreturns.
2September20245
GoldmanSachsGlobalMarketsAnalyst
Exhibit7:Curvesteepenerreturnsimprovewhencurvesareinitially?at
Returnsto2s10ssteepenersvsinitial2s10slevel,cyclelength,depthandspeed
1.4avgdailyreturnsto2s10ssteepener(bp/day)
1.2
1.0
0.8
0.6
0.4
0.2
0.0
-0.2
-0.4
-0.6
-0.8
-1.0-
0.50.00.5
1.4
avgdailyreturnsto2s10ssteepener(bp/day)
1.2
1.0
0.8
0.6
0.4
0.2
0.0
-0.2
-0.4
-0.6
-0.8
0510
startinglevelof2s10s(ppt)
1.01.52.02.5
cyclelength(m)
152025
1.41.4avgdailyreturnsto2s10s1.4
steepener(bp/day)
1.21.21.2
1.01.01.0
0.80.80.8
0.60.60.6
0.40.40.4
0.20.20.2
0.00.00.0
cycledepth(bp)
-0.2-0.2-0.2
-0.4-0.4-0.4
-0.6-0.6-0.6
-0.8-0.8-0.8
-700-600-500-400-300-200-1000
avgdailyreturnsto2s10ssteepener(bp/day)
1.41.41.4
1.21.21.2
1.01.01.0
0.80.80.8
0.60.60.6
0.40.40.4
0.20.20.2
0.00.00.0
cyclespeed
-0.2-0.2(bp/month)-0.2
-0.4-0.4-0.4
-0.6-0.6-0.6
-0.8-0.8-0.8
-80-60-40-200
Source:GoldmanSachsGlobalInvestmentResearch
Onceagain,wecanundertakeasimilarbreakdownbycyclesassociatedwith
_
recessions,andthoserepresentingnon-recessionarycutsfollowinghikes.Whenwedoso,weseethataveragereturnsarehighertooutrightlongsduringrecessions(Exhibit8).Incontrast,non-recessionarycutshavemuchlowerreturnsonaverage.Wealsoseethatforcurvetrades,steepeningexpressionssuchas2s10s,2s30sgenerallyofferthehighestreturns–thisisintuitivegiventhehighestabsolutereturnsaregeneratedby
outrightlongsinthefront-end(Exhibit9).
2September20246
GoldmanSachsGlobalMarketsAnalyst
Exhibit8:Returnstolongdurationmuchhigherinrecession
Leveragedtotalreturnstooutrightlongs1mpriorto6mintocycle
Exhibit9:Returnstosteepenersmuchhigherinrecession
Leveragedtotalreturnstocurvesteepeners1mpriorto6mintocycle
1.2
1.0
0.8
0.6
0.4
0.2
0.0
bp/day
2y
NoRecessionRecession
5y10y
bp/day
30y
1.2
0.45
0.40
1.0
0.35
0.8
0.30
0.25
0.6
0.20
0.4
0.15
0.10
0.2
0.05
0.0
0.00
bp/day
2s5s
NoRecession
2s10s2s30s
Recession
5s10s5s30s
bp/day
10s30s
0.45
0.40
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
Source:GoldmanSachsGlobalInvestmentResearchSource:GoldmanSachsGlobalInvestmentResearch
Littlechangesfortherankorderofoutrightlongswhenweadjustbyvolatility(orby
maximumdrawdown,whichisalmostperfectlyproportionaltothedailyvolofreturns)–long2yremainsthebestexpression(Exhibit10).Butcurvetradesbecomemore
attractive,with2s10sand2s30scompetingwithoutrightlongduration(10yand30y)onarisk-adjustedbasis(Exhibit11).
Exhibit10:2ylongoptimalforoutrightandvol-adjustedreturns
Leveragedtotalreturnsandvol-adjustedreturnstolongs1mpriorto6mintocycle
Exhibit11:2s10soffersbestvol-adjustedreturnsforcurvesteepeners
Leveragedtotalreturnsandvol-adjustedreturnstocurvesteepeners1mpriorto6mintocycle
0.6
0.5
_
0.4
0.3
0.2
0.1
0.0
bp/dayReturnsVol-adjustedreturns
2y5y10y
ppt/day/dailyvol
30y
0.18
10
0.16
9
0.14
8
0.12
7
6
0.10
5
0.08
4
0.06
3
0.04
2
0.02
1
0.00
0
bp/day
2s5s
ReturnsVol-adjustedreturns
2s10s2s30s5s10s5s30s
ppt/day/dailyvol
10s30s
6
5
4
3
2
1
0
Source:GoldmanSachsGlobalInvestmentResearchSource:GoldmanSachsGlobalInvestmentResearch
Nextwelookatwhenreturnsaregeneratedduringthecycle.Todothis,weaveragethereturnsacrossthe12mintoandoutofthe?rstcutofeachcycle(noteourcyclesare
selectedtohavea12mgapinbetween).We?ndthatmostofthereturnstolong
positionsaregeneratedinthemonthsaroundthe?rstcut,withlimitedfollow-through
2September20247
GoldmanSachsGlobalMarketsAnalyst
beyondtwomonthsafterthe?rstcut(Exhibit12).Thispatternistrueforcuttingcyclesofdifferenttypes,althoughwillbestronglyaffectedbycuttingcyclesinresponseto
suddenshocks,forexamplethe2008or2020cuts(Exhibit13).
Exhibit12:Returnstolongsgeneratedattimeofcut
Averagemonthlyreturnstolong5y.Zeroistimingof?rstcut
Exhibit13:6mpriortocutsgeneratesmuchlargerreturnstolongs
6mpre-vs6mpost-initialcutreturnstolong5y
2.0bp/daybp/day
1.5
1.0
0.5
0.0
-0.5
-12-10-8-6-4-2024681012
2.0
2.0
1.8
1.6
1.5
1.4
1.2
1.0
1.0
0.8
0.5
0.6
0.4
0.0
0.2
0.0
-0.5
bp/dayreturn6mintocutreturn6moutofcutbp/day
unconditionalrecessioninitial+recessioninitial/norecession
2.0
1.8
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
Source:GoldmanSachsGlobalInvestmentResearchSource:GoldmanSachsGlobalInvestmentResearch
Flatcurvespointtobetterrisk/rewardforsteepeners
Currentlevelsofinversionsuggestahighhurdleforfront-endoutperformance.Exhibit
14showsthecurrentlevelof1y1yvs1yarealmosteverywherelowerthanthe
maximuminversionatthetimeofpreviousinitialcuts.Intheabsenceofrecession,thissuggeststhatthebartorealisedreturnsonlong1y1yishigh.For2s10s,theUS,EUand
Canadaaremoreinvertedthanaverageatthetimeofinitialcuts(Exhibit15).Thissuggeststhatsteepenerriskispotentiallymoreattractivegiventhe?atnessofthecurve.
_
2September20248
GoldmanSachsGlobalMarketsAnalyst
Exhibit14:Currentfront-endinversionraisesthebartoreturns
Maxinversionof1y1yvs1ytimeof?rstcutvscurrent1y1yvs1y
Exhibit15:Betterrisk/rewardtosteepenersinUSD,CAD,EUR
Flattest2s10sattimeof?rstcutvscurrent
40
20
0
-20
-40
-60
-80
-100
-120
-140
bpMax1y1yvs1yinversionattimeoffirstcutbp
Currentinversion
JPYCHFCADAUDUSDEURSEKNZDGBPNOK
40
60
20
40
0
20
-20
0
-40
-20
-60
-40
-80
-60
-100
-120
-80
-140
-100
bp
bp
Max2s10sinversionattimeoffirstcutCurrentinversion
CHFJPYUSDCADEURAUDSEKNZDGBPNOK
60
40
20
0
-20
-40
-60
-80
-100
Source:GoldmanSachsGlobalInvestmentResearchSource:GoldmanSachsGlobalInvestmentResearch
However,therecentrallyinratesandcurvesteepeninginthelastthreemonthshasgeneratedreturnssimilartoorabovetheaverageforthemonthssurroundingcuttingcycles(Exhibit16andExhibit17).Togetherwiththeinitiallevelofinversionalready
priced,thissuggestsfurtheractivitydataweakeningwouldbenecessarytogeneratesigni?cantreturnstolongs.
Exhibit16:Recentreturnshavebeenelevatedvsaveragearoundinitialcuts
G10averagedailyreturnsin6maroundcuttingcyclesvslast3m
Exhibit17:1y1yreturnshavebeensubstantialinrecentmonths
1y1y3mchangevs1y1ycuttingcycleavgrealisedreturns
1.4
1.2
1.0
_
0.8
0.6
0.4
0.2
0.0
bp/dayavgdailyreturnsincuttingcycleslast3m
Long2y2s10ssteepener
bp/day1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
160
bp
140
120
100
80
60
40
20
0
NZD
3mchange1y1yrealisedreturnsfromfirstcut
USDSEKNOKAUDCADEURGBP
bp
CHFJPY
160
140
120
100
80
60
40
20
0
So
溫馨提示
- 1. 本站所有資源如無(wú)特殊說明,都需要本地電腦安裝OFFICE2007和PDF閱讀器。圖紙軟件為CAD,CAXA,PROE,UG,SolidWorks等.壓縮文件請(qǐng)下載最新的WinRAR軟件解壓。
- 2. 本站的文檔不包含任何第三方提供的附件圖紙等,如果需要附件,請(qǐng)聯(lián)系上傳者。文件的所有權(quán)益歸上傳用戶所有。
- 3. 本站RAR壓縮包中若帶圖紙,網(wǎng)頁(yè)內(nèi)容里面會(huì)有圖紙預(yù)覽,若沒有圖紙預(yù)覽就沒有圖紙。
- 4. 未經(jīng)權(quán)益所有人同意不得將文件中的內(nèi)容挪作商業(yè)或盈利用途。
- 5. 人人文庫(kù)網(wǎng)僅提供信息存儲(chǔ)空間,僅對(duì)用戶上傳內(nèi)容的表現(xiàn)方式做保護(hù)處理,對(duì)用戶上傳分享的文檔內(nèi)容本身不做任何修改或編輯,并不能對(duì)任何下載內(nèi)容負(fù)責(zé)。
- 6. 下載文件中如有侵權(quán)或不適當(dāng)內(nèi)容,請(qǐng)與我們聯(lián)系,我們立即糾正。
- 7. 本站不保證下載資源的準(zhǔn)確性、安全性和完整性, 同時(shí)也不承擔(dān)用戶因使用這些下載資源對(duì)自己和他人造成任何形式的傷害或損失。
最新文檔
- 【正版授權(quán)】 ISO 2475:2025 EN Chloroprene rubber (CR) - General-purpose types - Evaluation procedure
- 2025裝飾裝修工程分包合同
- 2025技術(shù)傳授合同樣式
- 二零二五年度高端裝備價(jià)格保密合同3篇
- 2025年度綠色建筑示范項(xiàng)目建房協(xié)議書3篇
- 二零二五年度企業(yè)內(nèi)部停車場(chǎng)車輛使用免責(zé)協(xié)議3篇
- 二零二五年度智能家居系統(tǒng)發(fā)起人投資合同3篇
- 二零二五年度歷史文化街區(qū)物業(yè)用房移交及文化保護(hù)協(xié)議3篇
- 二零二五年度社區(qū)食堂兼職煮飯人員協(xié)議3篇
- 二零二五年度內(nèi)部員工保密協(xié)議模板:企業(yè)核心競(jìng)爭(zhēng)力保護(hù)3篇
- 2025年上半年河南省西峽縣部分事業(yè)單位招考易考易錯(cuò)模擬試題(共500題)試卷后附參考答案-1
- 深交所創(chuàng)業(yè)板注冊(cè)制發(fā)行上市審核動(dòng)態(tài)(2020-2022)
- 手術(shù)室護(hù)理組長(zhǎng)競(jìng)聘
- 電力系統(tǒng)繼電保護(hù)試題以及答案(二)
- 小學(xué)生防打架斗毆安全教育
- 2024-2025學(xué)年九年級(jí)英語(yǔ)上學(xué)期期末真題復(fù)習(xí) 專題09 單詞拼寫(安徽專用)
- 網(wǎng)絡(luò)運(yùn)營(yíng)代銷合同范例
- 2024年新人教版七年級(jí)上冊(cè)歷史 第14課 絲綢之路的開通與經(jīng)營(yíng)西域
- 植保無(wú)人機(jī)安全飛行
- 醫(yī)療糾紛事件匯報(bào)
- 2024年村干部個(gè)人工作總結(jié)例文(3篇)
評(píng)論
0/150
提交評(píng)論