高盛：新興市場(chǎng)投資周期框架指南

_

10November2022|8:32AMEST

EMinFocus

PuttingourFrameworksTogether:A“CycleGuide”toEMInvesting

Overthepastdecade,ourconversationswithEMinvestorshaveconsistentlymovedCaesarMaasry

+1(212)902-8763|

GoldmanSachs&Co.LLC

inaglobalmacroandtop-downdirection,withafocusonEMinvestmentcyclesandcaesar.maasry@

withspeci?cattentiontothe“EMvs.DM”opportunityset.Accordingly,wehave

JoleneZhong

jolene.l.zhong@

developedanumberofcomplementaryframeworkswhichguideourinvestment+1(917)343-6820|

analysisandrecommendations,oftenindistinctreportsthatareparticularlyrelevantGoldmanSachs&Co.LLC

toacurrentmacrotheme(forexample,interestratesensitivityframeworksduringa

Fedhikingcycle).

Inthisreport,weextendtheseframeworksgiveninvestorquestionsandcomments;

andmoreimportantly,wepiecethemtogetherinattempttoprovidea

comprehensiveoverviewofseveralcriticalinputstoinvestmentanalyses.Thekey

reportsthatwereferencearethefollowing:

n(1)TheEMCycleGauge,whichdistillsmacrodataacrossthemajorEMeconomiesintoan“early,mid,andlate”cyclereadingandcarriesimportantsignalsforsubsequentgrowthoutturns.Thisframeworkoffersabroadinvestmentimplication,butisbestsuitedtoinformgrowthviewswhichinturnshouldbeconsideredinthecontextofvaluationandpolicy.

n(2)EmpiricalstudyofCross-AssetValuationindicators,highlightingwhichsimpleassetratiosandlevelstendtocorrelatethebestwithforwardlong-termperformance.Weargueacross-assetvaluationapproachissuperiorto

“asset-speci?c”metrics.

n(3)EMAssetreturnsaroundthebusinesscycle,aframeworkthatconsiderscrossassetperformanceonavol-adjustedbasisdependentupongrowthphases(withsomenuancearoundinterestratescyclesaswell).OurCycleGaugecanhelpdriveforwardgrowth(and,toalesserextent,interestratepolicy)viewsinputtedintothisframework.

n(4)MacroSliceframework,whichsplitstheEMequitycomplexintodistinctmacrothemesthatcorrelatewitheconomicindicators.Weextendtheanalysistohighlightwhichspeci?cexposureswithEMequitiesarebetterachievedthroughthisstructureratherthanheadlineindexexposure.

Puttingthepiecestogethertohighlightthecurrentstand-outs:WithinEM

equitiesinparticular,wehighlightthatMainlandChineseequitiesscreenasfar

“earliercycle”thantherestofEM,consistentwiththenotionthatpotential

Investorsshouldconsiderthisreportasonlyasinglefactorinmakingtheirinvestmentdecision.ForRegACcerti?cationandotherimportantdisclosures,seetheDisclosureAppendix,orgoto

/research/hedge.html.

_

GoldmanSachs

EMinFocus

re-openingdriveevenheadlineequitiessigni?cantlyhigher.KoreaisoneofthefewAsianEMswhereourCycleGaugeismovingtowardsamid-cyclereadingfromlate;coupledwithasigni?cantlyundervaluedcross-assetsignal,wereiterateKospiasatop“reboundcandidate”.SouthAfricaexportershaveunderperformedsigni?cantly,tendtoofferwidedispersionrelativetoheadlineexposure,andarelikelytobene?tfromrenewedweaknessintheRand.BrazilandIndiahavebothoutperformedthisyear,butthe“guts”oftheralliesaredistinct;wefavorrotatingintoDomesticCyclicalsinBrazilandFinancialsinIndia.

10November20222

_

EMinFocus

GoldmanSachs

PuttingourFrameworksTogether

EMCycleGauge-overlayingvaluationand“momentum”ofthecycle

Theprimaryaimofour‘CycleGauge’methodologyistocollateaswathofEMmacrodataintoatargetedsignalastothelikelyforwardtrajectoryofEMgrowth.TheinputsincludeGDP(vs.Trend),CurrentAccountBalances(vs.Trend),localinterestratecurves(2yvs.10y),In?ation(Spotvs.theCentralBank’sTarget),andPolicyRates(vs.recentranges).

Exhibit1highlightstheaveragesignalacrossEMeconomiessincethemid-1990s,withthelatestdatapointshowingquiteanextreme“l(fā)atecycle”reading.Asdiscussedinpreviouswork,thecurrentindicatorisafunctionnotonlyofin?ationandhighrates,butalsoduetoasigni?canterodingofcurrentaccountbalances;inshort,thelatecyclesignalisduetoallthecomponentsincludedinourmethodologyandthereforesendsacompellingargumentthatEMgrowthoutcomesacrossmosteconomiesinsubsequentquartersarelikelytobefairlysubdued.

Exhibit1:MostEMeconomiesscreenas“l(fā)atecycle”

+2.0

EM"Earlyvs.Late"CycleGauge

(avg.across19EMs)

+1.5

Current=+1.2

+1.0

"latecycle"

Post-2009,growth&inflationexpectations

used

+0.5

0.0

-0.5

"earlycycle"

-1.0

-1.5

-2.0

-2.5

Source:HaverAnalytics,EikonDatastream,GoldmanSachsGlobalInvestmentResearch

Inrecentreports,wehaveleaneduponthisobservationasarationaleforlowexpectedreturnsacrossEMequitiesinparticular(assumingthatweakgrowthoutcomeslikelyresultinuninspiringreturns).WehavebacktestedthisframeworkwiththeresultsdisplayedinExhibit2,whichsupportthisnotionthatreturnsarelikelytobesubpar.

10November20223

_

EMinFocus

GoldmanSachs

Nevertheless,asthebacktestshows,ourCycleGaugedoesnotpresentsucha“clean”forwardreturnsignal.EquitiestendtoperformstronglyintheEarlyandMidCyclephases,butmuchlesssointheLateCycle(thesereturnsare‘forward’,meaningtheaverage12-moreturndependingonstartingCylceGaugesignal).LocalBondsandCredit(totalreturnsofUSDdenominatedsovereignbonds)alsoperformlessstronglyasthecycleextends,though,ingeneral,thereturndifferentialsbetweenanEarlyandMidCyclereadingarenotparticularlysigni?cantinthisframework.EMFXalsoisnotwelldelineatedbasedonourCycleGauge,anobservationwediscussfurtherbelow.

Exhibit2:TheEMCycleGaugelooselyexplainsreturndifferentialsbasedonstartingcyclephase

EMCross-AssetReturns

2009-Now

basedonindividual"CycleGauge"reading

0.60

0.50

(subsequent6-moreturn,25EMsinsample)

Equity

FX(spot)

LocalBondsSov.Credit

0.40

0.30

0.20

0.10

0.00

(0.10)

(0.20)

(0.30)

(0.40)

(0.50)

Mid-Cycle

Late-Cycle

Early-Cycle

Source:FactSet,GoldmanSachsGlobalInvestmentResearch

Again,ourCycleGaugeismeanttodistillmacrodataintoaclearforwardfundamentalsignal-butthereisnoinclusionofassetvaluationinthatframework(ratheritisbasedprimarilyonslower-movingmacrodata).Expandingthisframeworktoincorporateourrecentworkregardinglong-termEMreturnimplicationsbasedonpracticalvaluationsignals(inwhichourconclusionisthata‘cross-assetvaluationsignal’isfarmorehelpfulthanindividualassetratios),thereisamodestco-movementofassetvaluationcycleswithourCycleGauge(seeExhibit3below).

Uponcloserinspection,assetvaluationappearstoleadourCycleGauge,whichisunderstandableasmarketsareforward-lookingandtendtoanticipatemacrodatabeforetheirrelease;butthecurrentreadingisquiteunique:assetvaluationsareundemandingyetourCycleGaugesuggestsEMis“l(fā)atecycleusuallyassociatedwithexpensivevaluations.Thisapparentinconsistencyislikelyexplainedbytheunusualcombinationofslowinggrowthcoupledwithsigni?cantlyelevatedratesofin?ationacrossEMeconomies.

Structurally,asshownontheright-handsideoftheExhibit,thereisastraight-forward

10November20224

_

GoldmanSachs

EMinFocus

relationshipbetweenforwardEMreturnsandtheintersectionofourCycleGaugewithassetvaluation.(Note,theEMCross-Assetperformancereferstoabenchmarkcomprisedof16%MSCIEM,32%EMBIG-Div,32%EMcurrencyspotreturnsagainsttheUSD,and20%GBI-EM,whichiseffectivelyavol-adjustedindexacrossthemajorassetclasses).Ashighlighted,“cheap”valuationscoupledwithanEarlyCyclereadingposethestrongestforwardreturnpotential,whereas“expensive”valuationswithaLateCyclereadingpresentapoorinvestmentopportunityset(the“expensive”andEarlyCyclequadrantalsoappearstoofferanegativereturnimplication,butthereareveryfewhistoricalobservationswiththesecriteria).

Thecurrentbackdropofa“cheap”valuationreadingwithaLateCyclegaugesuggestsmodest,butpositivereturnpotentialacrossEMassetsoverthenext12months.

Exhibit3:CurrentlyourcyclescoreshowsavulnerablemacrobackdropwhileourcombinedvaluationsignalimpliesEMXassetslook“cheap”

1.5

EMCycleGauge

EMX-AssetValuation

(averagez-scoreofEquityP/B,Sov.CreditSpread,

LocalYieldDifferential,FXrealTWI,RHS)

3.0

1.0

2.0

0.5

1.0

0.0

0.0

-0.5

-1.0

(1.0)

-1.5

(2.0)

-2.0

(3.0)

-2.5

"EarlyCycle"&"Cheap"

Source:FactSet,HaverAnalytics,EikonDatastream,GoldmanSachsGlobalInvestmentResearch

AttheindividualEMlevel,wehavehighlightedthatourCycleGaugehelpspickoutperformersagainstunderperformers,albeitwithonlyamoderateexpectedalphawhenlookingsimplyatbuying“early”vs.“l(fā)ate”EMs.Includingavaluefactorcanimprovethisalpha,asdoesameasureof“movement”alongtheCycleGauge.

Exhibit4presentsourCycleGaugeacrossthemajorEMeconomiesandplotsthetrajectoryofthisreadingoverthepast3-monthsaswell.Thereareanumberofnotablemarketsthatstickoutonthisschematic,suchasMainlandChinaappearingasthemostEarlyCycleEM(byawidemargin),KoreascreeningastheonlyAsiaEMthatismovingtoanearliercyclereading(movingleft)alongwithLatinAmericanmarketssuchasBrazilandMexico,andbothPolandandtheCzechRepublicmovingawayfromextremegaugereadingsaswell.ASEAN(Indonesiainparticular)havemovedfromscreeningquiteearlycycle(relativetootherEMs)tomorein-line.

10November20225

_

EMinFocus

GoldmanSachs

Exhibit4:EMeconomiesvarywidelyintheirCycleGaugereadings

2.0

1.5

1.0

0.5

0.0

-0.5

-1.0

-1.5

-2.0

-2.5

"EarlyCycle"Score

AverageValuation

(Z-scoresacrossEquity,

FX,CreditandLocal

Bondssince2012)

Mnld.China

"Mid-Cycle"Score

Kuwait

Saudi

Arabia

UAE

"LateCycle"Score

India

Taiwan

QatarIndonesia

Malaysia

CzechRep

Thailand

Peru

Mexico

Chile

Turkey

Greece

Brazil

Philippines

Korea

SouthAfrica

Egypt

Hungary

Poland

EMCycleGauge

(calibratedusingGDP

Colombia

&C/ABalanceTrends,

Inflation,andInterestRates)

-3.0

-1.0-0.50.0+0.5+1.0+1.5+2.0+2.5+3.0+3.5

Source:FactSet,EikonDatastream,GoldmanSachsGlobalInvestmentResearch

Inourinitialreport,wehighlightedthatasimplecross-sectionalrankingofourCycleGaugeresultedinoutperformanceofearlyEMs(moresoinequitythanin?xedincome),andExhibit5enhancesthatbacktestbyanchoringoffthechangesinourCycleGaugeratherthantheircurrentreadings.Asshown,along/shortstrategyofEMswhereourCycleGaugeismovingtowardsanEarlyreading(vs.towardsLate)hashistoricallydrivenalphaacrossallthemajorassetclasses,fairlyconsistentlyoverthepast12years.

Asmentionedfurtherbelow,themacrodatathatdrivesourCycleGaugeareprimarilygearedtowardsprovidingaforwardsignalongrowth(corporatepro?tsinparticular)

ratherthanaframeworkthatprovidesadirectinvestmentimplication.Inourview,thesebacktestssuggestthattheforwardgrowthsignalembeddedintheCycleGaugecarriessomeinvestmentsignal-butasdiscussedbelow,thatgrowthsignalneedstobeconsideredinthecontextofmonetarypolicy(andvaluationasmentionedabove)toenhanceitsimplement-ability.

10November20226

_

GoldmanSachs

EMinFocus

Exhibit5:OurbacktestresultshowsEMsthatmovedtoearlycyclesoverpast3monthstendtohavebetter1-yrforwardcross-assetsreturnsthanEMsthatmovedtolatecycles

EarlyvsLateCycleEMPerformance

(annualrebalancing;top9vsbottom9score3-mochanges;all19EMs)

Average1yForwardReturn

Across12Backtests(Startingat

125everymonth-end)

LocalBondTotalReturn

120

FXSpot

115

Equity(in$)

CreditTotalReturn

105

100

95

90

200920102011201220132014201520162017201820192020202120222023

110

Source:FactSet,EikonDatastream,GoldmanSachsGlobalInvestmentResearch

CombiningaGrowthSignalwithaBusinessCycleOverlay

Ourmacroinvestmentframeworksadoptanumberofdifferentapproaches,andoneimportantdistinctionbetweentheCycleGauge,whichutilizesreadilyavailable,reportedmacrodata,andthe‘BusinessCycle’frameworkoutlinedbelowisthatthelatterrequiresanassumption(orforecast)ofmacrodata.BothframeworksoperateundertheassumptionthatgrowthdataarecriticaldriversofEMassetperformance-buttheCycleGaugeaimstopredictgrowthoutcomesbasedonavarietyofrealizeddatameasure,andtheBusinessCycleframeworkspeci?eshowindividualEMassetsarelikelytotradedependentuponsubsequentchangesinthemarket’sviewonthegrowth/policymix.

Forthisframework,wefocusonforwardviews-whichwethinkarewellrepresentedthroughConsensusEconomicsrolling“nextfourquarter”realGDPforecastsand1-yearlocalinterestrates(whicharearepresentationofthemarket’sforecastforlocalpolicyrates).Assetreturnsaremorehighlycorrelatedwiththechangeintheseexpectationsratherthantheirspeci?clevels,andtheswirlExhibitbelowillustratesthecurrentdynamicacrossEMeconomies.

MostEMsarecurrentlyinthe“GDPdown,RatesUp”quadrant,asforwardexpectationsofrealGDPgrowtharedecliningdespitein?ationremainingpersistentlyhigh(andthemarkettakingupitsviewofwherepolicyratesarelikelytogo).Therearesomeexceptions,suchasBrazilandMainlandChina—andthestrongoutperformanceofBrazilianassetsinrecentmonthsisnosurprisewhenlookingatthisdiagram.

10November20227

_

EMinFocus

GoldmanSachs

Exhibit6:MostEMsarein“GDPCuts,Ratesupphase”,exceptforafewearlyhikerslikeBrazil

EMGDPvsRatesScreen

"GDPUpRatesUp"

SouthAfrica

Peru

Thailand

Indonesia

Mexico

Korea

Taiwan

Colombia

-1.0pp

CzechRepublic

-1.5pp

"GDPCuts

RatesDown"

Hungary

2.0pp3.0pp4.0pp5.0pp6.0pp

Source:ConsensusEconomics,Datastream,GoldmanSachsGlobalInvestmentResearch

UnliketheCycleGauge,wheretheinvestmentimplicationisabroadoneof“marketAislikelytooutperformmarketBaspeci?cfocusondistinctgrowthandratesviewshelpsoutlinetheassetclasswithinanemergingmarketthatislikelytoprovidealpha.Exhibit7outlinestheaveragevolatility-adjustedreturnsofeachassetclass(theaverageacross

3-mochangesof1yforwardGDP

"GDPUpRatesDown"

Brazil

3-mochangesof1yRate

"GDPCutsRatesUp"

-2.0pp-2.0pp

Mnld.China

Philippines

Malaysia

Poland

-0.5pp

-1.0pp

0.0pp

0.0pp

0.5pp

1.0pp

Chile

India

individualcountry-assetperformance)basedonthefourquadrantsoftheschematicabove.

Thekeytake-awaysfromthisframingarethat?xedincomeprovidethebest

vol-adjustedreturnsintheearlycycle(growthupandratesdown),whereasequitiesleadinthesubsequentgrowthup/ratesupquadrant.Asthecyclematuresandgrowthbeginstosurprisetothedownside,FXtypicallyoffersthe“bestshort”—thismaybecounter-intuitivegiventheincreaseinfront-endrates(andthereforeFXcarry);butoursenseisthattheseepisodeshavehistoricallycenteredaroundrecessionsduringwhichEMcentralbanksneededtohikeratestodefendagainstprecipitousFXweakness.

Sovereigncredithashistoricallyprovedtobethemostdefensiveduringthelastpartofthebusinesscycle,wheregrowthratesarestilldroppingbutcentralbankshavechangedtackandthemarketispricingcuts.Thisreadingisprobablyupwardlybiasedbasedonthepastdecadewhichfeaturedanumberofepisodesofsofteningand‘muddlethrough’EMgrowthwithaccommodativepolicyenabledbyDMquantitativeeasing(whichhasdisproportionatelybene?tedEMcreditduetoitsdirectUSdurationexposure).

10November20228

_

GoldmanSachs

EMinFocus

Exhibit7:Fixedincomeisthe“bestlong”earlyinthebusinesscycle,FXthe“bestshort”asthecyclematures

ReturnsbasedonLocalEconomicCycle(Avg.across25EMs)

1.00

EquityFXLoc.BondsCredit

0.80

0.60

0.40

0.20

0.00

(0.20)

(0.40)

(0.60)

GrowthExpectationsUPPolicyRatesCUTTING

(1)Early

Recovery

GrowthExpectationsUPPolicyRatesHIKING

(2)MidCycle,

HikingStarts

GrowthExpectationsDOWNPolicyRatesHIKING

(3)Mid-LateCyclePotentialOverheating

GrowthExpectationsDOWNPolicyRatesCUTTING

(4)LateCycleCuttingStarts

Source:FactSet,Datastream,GoldmanSachsGlobalInvestmentResearch

Intheaboveanalysis,wehaveupdatedourcross-assetschematictofocusonassetreturnsatthemarketlevel(asopposedtotheheadlineEMbenchmarkindicesinpriorwork),thoughthebroadconclusionsremainunchanged.Below,welinktheCycleGaugetothisanalysisastheGaugeprovidesaforwardviewonsubsequentConsensusGDPrevisionsand1-yearinterestratechanges.

AsshowninExhibit8,EarlyCyclereadingsareassociatedwithfairlysigni?cantGDPrevisionsoverthefollowing3-months(+30bp)andmodestlylower1-yearinterestrates(-3bp);andLateCyclereadingsofferthereversesignal(-17bptoGDPestimatesand+12bpto1-yearinterestrates).MidCyclereadingsofferamutedGDPsignal(-10bp)andasimilar1-yearratesignalasanEarlyCyclereading(underscoringthenotionthatourCycleGaugeisabetterindicatorofforwardgrowthoutcomesthanpolicy).

Separately,wehavedevelopedanumberofframeworksforidentifyingpotentialshiftsinmarketpricingregardingEMcentralbankaction(someofwhicharefeaturedinourlocalratesmonitor),andincorporatingsuchratesviewswithgrowthimplicationsfrom

frameworkslikeourCycleGaugeareessentialineffectivelypickingtherightspotsacrossspeci?ccountryormarketlevelassetpairsinEM.

10November20229

_

EMinFocus

GoldmanSachs

Exhibit8:“LateCycle”indicatorssuggestsubsequentcutstoGDPestimatesandhigherfront-endrates

2004-Now

EMGDPRevisionsandInterestRateChanges

0.4

basedonstarting"CycleGauge"

GDPRevisionChangein1-yrRate

0.3

0.2

0.1

0.0

(0.1)

(0.2)

Mid-Cycle

Late-Cycle

Early-Cycle

Source:ConsensusEconomics,Datastream,GoldmanSachsGlobalInvestmentResearch

WheredoesMacroEquityFitIn?UtilizingourMacroSliceequityframework

“Cross-over”investinghasbecomeincreasinglypopular-bothintermsofequityinvestorsutilizingmacroassetsforhedging(inFXandratesspace)aswellasfundsadoptingmulti-assetbenchmarksandconsideringEMequityvs.?xedincomeallocations.Nevertheless,we?ndthereremainsadistinctionbetween“macroassetswhichtypicallyincludecurrencies,sovereign?xedincome,andcommodities,andthemicroassetclassessuchasequities.

Wehavepushedbackonthisconcept,arguingthat,whenstructuredproperly,EMequitiesofferstraightforwardmacroexposuresthatreliablytradewithsimilardataoftenusedtobaseinvestmentdecisionsinthe‘macro’assetclasses.Furthermore,asshownbelowinExhibit9,thecorrelationofequitieswithFX,rates,andcreditisjustashighas

withinthe‘macro’assetspectrum(hereweuselocalinterestrates,nottotalreturnoflocalbondsasabove).Whatmakesequitiesunique,however,istheabilitytodivewithintheindexandstructuresub-componentsthatareuncorrelatedwithheadlineindices.

10November202210

_

TheMacroSliceframeworkisexplainedindetailinitsinceptionreport,butinitssimplestformDomesticCyclicalsaregearedtotradethedomesticgrowthcycle(belowshownrelativetolocalManufacturingPMIdata);Financials(almostwhollyBanks)tradeinterestrate-premia,heremeasuredassimplelong-endratedifferentialsagainsttheUS;non-commodityExporterstradewith‘DM’exposures,whichcanberepresentedthroughgrowthdifferentials;andCommoditycyclicalsanchoroffphysicalcommodityprices.

GoldmanSachs

EMinFocus

Exhibit9:Majorassetclassesarecorrelatedacrossmarkets(individualcountryindicesusedhere);ourMacroSlicesofferuncorrelatedexposures

AllMajorEMAsset

ClassesareFairlyHighly

Correlated

0.50

Equity"MacroSlices"

(whicharelong/shortagainstrespecitveheadlineequityindices)arefarlesscorrelatedwiththemajorEMbenchmarks

0.20

0.10

0.00

(0.10)

(0.20)

Exporters(non-commod)

vs.vs.FXRates

Equity

Source:FactSet,Datastream,GoldmanSachsGlobalInvestmentResearch

Apartfromofferinganuncorrelatedexposurewithheadlinemacroassets,MacroSlicesaregearedtowardsspeci?ceconomicvectors.Theslicesarenotexactlysectors,whicharede?nedbyindustrycategorization,butrathertheydependuponthespeci?crevenueexposures(forexample,weconsiderautomanufacturersinKoreaasexporters,butautomakersinIndiaarepartofthe‘DomesticCyclicals’macroslicedespitehavingthesamesectorclassi?cation).Thisspeci?cationdrivesourmacroequityframeworkinwhichtooperate.

Cross-AssetReturnCorrelation

(monthlyreturnssince2014)

vs.

Headline

Equity

vs.

Headline

Equity

DomesticCyclicals

Equity

vs.

Credit

Ratesvs.

Credit

Equity

vs.

Rates

vs.

Headline

FX

vs.Credit

Equityvs.FX

FX

vs.Rates

vs.Credit

vs.Credit

vs.Credit

vs.Rates

vs.Rates

Financials

0.70

0.60

0.40

0.30

vs.FX

vs.FX

Thereisanaspectof“applestooranges”whencomparingmacrosliceswithFX,rates,andcredit,inthatweoftendisplayaslice’srelativereturnagainstDefensivesortheheadlineequityindex(along/short);whichhedgesoutbroadbetaexposure.ButtheintuitionhereisthatDefensives(orheadlineequities)offeraconsistenthedge—forFXtrades,incontrast,along/shortpairing(beitagainsttheUSDoranotherEMcross)naturallytakesonadistinctmacroexposureoutsidetheEMonthelongside.

10November202211

_

GoldmanSachs

EMinFocus

Exhibit10:DomesticCyclicalstradewithPMIdata

Exhibit11:FinancialsarecorrelatedwithEMInterestRateDifferentials

120

135

130

EMex-China2

DomesticCyclicals(vs.Defensives)

EMex-China56ManufacturingPMI55

LocalYieldDifferentialvs.US

115USDPriceIndex

110

105

100

95

90

85

80

75

70

Financialsvs.HeadlineEquities

USDPriceIndex

54

125

3

53

120

52

51

50

49

115

4

110

105

48

100

5

47

95

46

45

44

90

6

85

2020

202120222023

20152016201720182019

2020202120222023

20152016201720182019

Source:FactSet,GoldmanSachsGlobalInvestmentResearchSource:FactSet,GoldmanSachsGlobalInvestmentResearch

Non-commodityexportersaretheleastcorrelatedwithaspeci?cmacrotrendcomparedwithmacroslicepeers;theytendtoperformfairlysimilarlytoDMvs.EMequitiesandtheDMvs.EMgrowthdifferential(thoughtherearetimessuchasearly2019and2020whenthisrelationshiphasbrokenapart).Partofthesigni?cantoutperformanceofthismacrosliceisduetotheweightingofTechnologystocks,whichhaveoutperformedmacrorelationshipsduetobothahighergearingtowardslowinterestratescomparedwithequitypeersaswellasstructuralimprovementinpro?tmargins.

Exhibit13:EMExporterstendtooutperformwiththeDMvs.EMgrowthdifferential

Exhibit12:EMexportershaveoutperformedsimilarlytothe‘DMvs.

EM’equitytrade

150

175

DMvs.EMEPS(USD,RHS)

DMvs.EM(MSCIWorldvs.MSCIEM)

Exporters(vs.HeadlineIndex)

150

165USDPriceIndex

155

Exporters(vs.HeadlineIndex)

USDPriceIndex

140

USDPriceIndex

140

145

1