全球宏觀策略英

companiescoveredinMorganStanleyResearch.

212

As761-1009aFebruary

5,

2022

12:42

AM

GMTMORGAN

STANLEY

&

CO.

LLCMatthew

HornbachSTRATEGISTMatthew.Hornbach@Guneet

Dhingra,CFASTRATEGISTGuneet.Dhingra@AndresJaimeSTRATEGISTAndres.Jaime@DavidS.Adams,CFASTRATEGISTDavid.S.Adams@Andrew

MWatrousSTRATEGISTAndrew.Watrous@KelcieGersonSTRATEGISTKelcie.Gerson@+1

212

761-1837+1

212

761-1445+1

212

296-5570+1

212

761-1481+1

212

761-5287+1

212

761-3983MORGAN

STANLEY

&

CO.

INTERNATIONAL

PLC+JamesKLordSTRATEGISTJames.Lord@EricSOynoyanSTRATEGISTEric.Oynoyan@WantingLowSTRATEGISTWanting.Low@GekTengKhooSTRATEGISTGek.Teng.Khoo@JohnKalamarasSTRATEGISTJohn.Kalamaras@AlinaZaytsevaSTRATEGISTAlina.Zaytseva@LorenzoTestaSTRATEGISTLorenzo.Testa@+44

20

7677-3254+44

20

7425-1945+44

20

7425-6841+44

20

7425-3842+44

20

7677-2969+44

20

7677-1120+44

20

7677-0337MORGAN

STANLEY

MUFG

SECURITIES

CO.,

LTD.+KoichiSugisakiSTRATEGISTKoichi.Sugisaki@ShokiOmoriSTRATEGISTShoki.Omori@+81

3

6836-8428+81

3

6836-5466Global

Macro

Strategist

|

GlobalNot

in

Kansas

Anymore

We

think

a

downside

surprise

on

January

US

CPI

inflation

is

all

that

can

save

government

bonds

near

term.We

still

believe

investors

need

more

term

premium

to

protect

against

higher

inflation,hawkish

monetary

policies.Real

rates

to

rise

further.

Toto,I

have

a

feeling

we're

not

in

Kansas

anymore.

GlobalMacroStrategyWe

address

the

topicof

recession

and

why

it's

still

too

early

to

buygovernmentbonds.We

discuss

why

higher

rates

and

a

smaller

balancesheetaren'tall

bad

news,as

well

as

when

we

think

G10

central

bank

QEholdings

will

peak

(August2022).Finally,we

turn

bullish

on

the

EUR.InterestRateStrategyWe

maintain

UST

5s30s

curve

flatteners.We

keep

our

EUR

10y10y

swappayer

and

our

longEUR

5y5y

inflation.We

keep

our

shortEGB

trades:shortOAT

vs

Bund

and

EU,short5y

Bono

ASWand

our

longSep

22FRA/€str

basis.We

enter

a

shortUS

10y

notes

vs

Bund.In

the

UK,we

closeour

GBP1y1y

vs

EUR

4y1y

and

GBP2s10s

steepeners.We

keep

Sep

22MPCreceivers

vs

Sep

23Sonia

future.We

maintain

long20y

JGB

ASWvs

TONAOIS

and

receive

1y1y

vs

pay

5y5y.Currency&ForeignExchangeWe

turn

bullish

the

EUR

amid

ECB

hawkishness,butstay

neutral

on

theUSD

ahead

of

US

CPI.We

recommend

longEUR

versus

GBP,JPY,and

USD,butstay

longUSD/JPY

too.SEKmay

gain

as

hawkish

Riksbank

expectationsrise.Go

shortCHF/SEK.Assetsales

from

the

BoC,RBNZ,and

BoE

may

becomingthis

year.Stay

longAUD/NZD.We

discuss

if

JPY

could

lose

itsfundingcurrency

status

or

not.Inflation-LinkedBondsWe

maintain

short5y

TIPS

vs

longDBRis,and

short5y

TIPS

vs.EDZ3(Dv01

ratio

0.8:1).We

look

atthe

prospects

for

an

upside

surprise

in

January

CPI,

and

explore

two

feedback

loops

for

inflation

in

the

US.We

see

Japan

BEI

continuingto

rise

post-Feb

re-open

auction.We

maintain

longJBI26BEI.

Short-DurationStrategy

We

continue

to

recommend

payingH3SOFR/FFbasis

and

selling12m

T-

bills

vs

OIS.We

consider

how

differentpaths

of

Treasury

coupon

issuance

impactT-bill

issuance,and

in

turn,how

this

could

affectwhether

reserves

or

RRPfall

firstin

QT.Pleaseclick

hereif

you

wouldliketoreceivethedaily

Global

MacroCommentary.MorganStanleydoesMORGAN

STANLEY

&

CO.

LLCFrancescoGrechi

andseekstodobusinesswithSTRATEGISTFrancesco.Grechi@

+1result,investorsshouldbeawarethatthefirmmayhaveaconflictofinterestthatcouldaffecttheobjectivityofMorganStanleyResearch.InvestorsshouldconsiderMorganStanleyResearchasonlyasinglefactorinmakingtheirinvestmentdecision.Foranalystcertificationandotherimportantdisclosures,refertothe

Disclosure

Section,locatedatthe

endofthisreport.+=Analystsemployedbynon-U.S.affiliatesarenotregisteredwithFINRA,maynotbeassociatedpersonsofthememberandmaynotbesubjecttoFINRArestrictionsoncommunicationswithasubjectcompany,publicappearancesandtradingsecuritiesheldbyaresearchanalystaccount.?1Global

Macro

StrategyMORGAN

STANLEY

&

CO.

LLCMatthewHornbachMatthew.Hornbach@DavidS.Adams,CFADavid.S.Adams@AndrewWatrousAndrew.Watrous@+1

212

761-1837+1

212

761-1481+1

212

761-5287TooEarly

toTalk

Recession...The

pastweek

saw

an

uptick

in

news

stories

coveringinflation

and

also…recession

(seeExhibit1

and

Exhibit2).We

fielded

many

questions

aboutrecession

probabilities

andwhatour

models

suggested

for

them.The

US

nonfarm

payroll

reportfor

January

maypreventthese

concerns

from

growingfurther,atleastfor

now.We

think

its

too

early

to

talk

aboutrecession

when

activity

data

is

strong,butespeciallywhen

many

central

banks

have

yetto

even

begin

tighteningmonetary

policy.Financialconditions

may

be

tighter

than

lastyear,true,butthey

are

only

tighter

on

the

margin.They

are

far

from

tightwhen

placed

into

any

historical

context.Sowhyhaveinvestorsbeenaskingaboutrecession?First,yields

curves

have

beenflatteningas

central

banks

have

turned

more

hawkish

–

both

in

rhetoricand

in

policyaction.As

yield

curves

flatten

and

eventually

invert,models

thatimpute

the

probabilityof

recession

from

yield

curves

startflashingred.Exhibit3shows

the

New

York

Fed's

probitmodel

output,which

relies

on

the

shape

ofthe

Treasury

yield

curve.While

the

Treasury

curve

is

far

from

inverting,other

curves

likereds-greens

and

greens-blues

Eurodollar

futures

curves

are

almostinverted

(Exhibit4).6,5626,0004,0002,000

0Feb-20Aug-20Feb-21Aug-21Feb-221MMAExhibit1:

Dailystorycount

onthetopicof"inflation"fromallmediasources

Dailystorycount

16,000

14,000

12,000

10,000

8,000

Inflationstorycountfromallsources*Source:MorganStanleyResearch,

Bloomberg*NT<GO>

functiononBloomberg4,0003,0002,0001,000

0Feb-20Aug-20Feb-21Aug-21Feb-221MMAExhibit2:

Dailystorycount

includingtheword"recession"fromallmediasources

Dailystorycount

10,000

9,000

8,000

7,000

6,000

5,000

Recessionstorycountfromallsources*Source:MorganStanleyResearch,

Bloomberg*NT<GO>

functiononBloomberg26.010

0Feb-02Feb-06Feb-10Feb-14Feb-18Feb-22RecessionProbabilityofrecessionin12monthsExhibit3:

ProbabilityofaUSrecessionin12months

%

100

90

80

70

60

50

40

30

20Source:MorganStanleyResearch,

FRB

NewYork,

Bloomberg0.130.02-Feb-02Feb-06Feb-10Feb-14Feb-18Feb-22Reds-GreensGreens-BluesExhibit4:

Reds-GreensandGreens-BluesEurodollarfuturescurves

%

1.6

1.4

1.2

1.0

0.8

0.6

0.4

Source:MorganStanleyResearch,

BloombergWe

don'tthink

investors

should

worry

aboutyield

curve

flattening–

atleastnotyet.The

yield

curve

simply

tells

whether

Fed

policy

is

in

restrictive

territory

(an

invertedcurve)or

in

accommodative

territory

(a

positively

sloped

curve).And

justbecause

policy

is

restrictive

doesn'tmean

the

economy

is

close

to

a

recession.The

intentof

restrictive

policy

is

to

weigh

on

demand

and

economicactivity.Whetherthatputs

an

economy

into

recession

depends

on

how

quickly

the

economy

is

growing.Restrictive

policy

applied

to

an

economy

growingat4.6%Y/Y

in

real

terms

-

whatoureconomists

envision

for

the

US

economy

in

2022-

is

differentthan

restrictive

policyapplied

to

an

economy

growingat2%Y/Y

-

the

economy

thatexisted

pre-pandemic.Second,consumer

confidence

appears

under

stress.Consumers

neither

have

enjoyedhigher

prices

for

goods

nor

the

constantthreats

posed

by

COVID-19.Our

model

for

theUS

consumer

suggests

they

should

be

more

confidentthan

the

indexes

indicate.132.2113.8

30

10-105070130110

90-30

Jan-97Jan-02Jan-07Jan-12Jan-17Jan-22ConferenceBoardconsumerconfidence2-FactorModelExhibit5:

USconsumerconfidenceand2-factormodelestimate

Index

150Source:MorganStanleyResearch,

ConferenceBoardNote:Linearregressionmodeluses

U-3unemploymentrate,

2yreturnoftheS&P

500index-18.4-20-4004020-60

Jan-97Jan-02Jan-07Jan-12Residual+/-1StdevExhibit6:

USconsumerconfidenceresidualfrom2-factormodelestimate

Residual,consumer

confidence

points

60>0=confidencehigherthanmodelestimate<0=confidencelowerthanmodelestimate

Jan-17

Jan-22Source:MorganStanleyResearch,

ConferenceBoardNote:Linearregressionmodeluses

U-3unemploymentrate,

2yreturnoftheS&P

500index3-6.6-1.64

0

-5-101163146617691

106

121

136

151

166

181

196

211

226

241

256Business

daysfrom

startofcalendar

year200120082016200220092017200320102018200420112019200520122020200620132021200720152022

While

consumers

may

have

to

deal

with

higher

prices

for

a

while

longer,they

may

not

have

to

deal

with

another

COVID-19variantof

concern.Any

reprieve

from

the

pandemic,

especially

headinginto

the

summertime,should

boostconsumer

confidence,atleast

relative

to

our

model

estimate,we

think.

…AndTooEarly

toBuy

GovernmentBonds

The

upcomingUS

CPI

reportmightbe

the

lasthope

to

save

global

governmentbonds

from

their

worststartto

the

year

since

2009(see

Exhibit7).An

upside

surprise

next

Thursday

would

mean

further

talk

of

the

Fed

raisingrates

50bp

in

March.Ata

minimum,

calls

for

the

Fed

to

hike

atevery

meetingthis

year

will

look

much

less

off-base.Exhibit7:

BloombergGlobalTreasuriesAggregateIndexcalendaryeartotalreturns

Totalreturn,

%

20

15

10

5Source:MorganStanleyResearch,

Bloomberg

Our

economists

continue

to

call

for

4x25bp

rate

hikes

from

the

Fed,as

they

still

don't

see

anythingin

the

data

to

justify

more

than

4hikes

and

the

startof

balance

sheet

normalization.Butthe

consensus

is

at5hikes,and

some

projections

look

for

7

hikes.

Meanwhile,the

marketprices

between

5-6hikes

this

year

(134bp

to

be

exact,or

9bp

more

than

5x25bp

hikes).

With

marketpricingjust9bp

above

consensus

expectations,wecontinuetothink

marketsaren'tpricingenoughrisk

premiumforthechanceofamoreaggressivepolicy

path.Yes,it's

true

thatmarkets

didn'tprice

more

than

the

consensus

expected

in

previous

cycles,as

we

discussed

in

Pricingthe

Pace

of

Rate

Hikes.Butconsensus

expectations

didn'tchange

as

quickly

as

they

are

today.

An

upside

surprise

on

January

US

CPI

nextweek

followed

by

further

strength

in

nonfarm

payrolls

and

even

more

inflation

in

February

could

getexpectations

of

a

50bp

hike

in

March

to

solidify.Thatoutcome

would

certainly

joltexpectations

for

Fed

policy

this

year

-

and

investors

aren'tadequately

protected

againstthat,given

currentmarket

prices,we

think.

The

same

applies

to

the

pricingof

Fed

policy

in

2023and

2024,where

markets

price

Fed

policy

to

remain

below

2.00%.Given:

4Covid-19

1.

the

median

FOMC

participantthinks

2.5%represents

neutral

policy,

2.

Chair

Powell's

claim

earlier

this

year

thatpolicy

belongs

back

atneutral,and

3.

the

median

marketparticipantthinks

the

terminal

rate

in

this

cycle

is

between

2.00-

2.25%;marketpricingin

2023and

2024incorporates

a

negative

term

premium

-

perhapsreflectingfears

of

a

recession

we

discussed

above.We

think

investors

should

demand

apositive

term

premium

instead.Wheremightanegativetermpremiumbejustified?Intheverylong-endoftheyieldcurve.Amore

aggressive

Fed

policy

path

should

weigh

on

longer-dated

inflation-riskpremiums,as

hikingcycles

have

done

in

the

past.Atpresent,we

don'tsee

much

termpremium

-

either

positive

or

negative

-

in

the

longend

(see

Exhibit8and

Exhibit9).Still,historical

cycles

may

notbe

a

helpful

guide

today.The

Fed

changed

its

guidance

onthe

policy

path

(rates

and

balance

sheet)much

faster

than

in

the

pasttwo

cycles.Andeconomicdata

-

firston

inflation,and

now

on

the

labor

market-

has

been

much

morevolatile

relative

to

expectations.Realtermpremiums(thebedfellowofinflationriskpremiums)shouldbehigherthanbefore,asaresult.Exhibit8:

UST30y1mimpliedforwardyieldvs.thelongerrunneutralratepertheSMP2.022.002.52.01.54.03.51.0

Feb-16

Feb-17

Feb-18

Feb-19

Feb-20

Feb-21

Feb-22UST30y1mLongerrunneutralratefromSMP%Source:MorganStanleyResearch,

FRB

NewYork

SurveyofMarketParticipants

(SMP),BloombergExhibit9:

UST30y1mimpliedforwardyieldlessthelongerrunneutralratepertheSMP

0.0-0.5

0.53.01.0DifferencebetweenmarketrateandFRB

NYSMP%LasthikingcycleTrumpelection

0.02LDIDelta

Omicron-1.0

Feb-16

Feb-17

Feb-18

Feb-19

Feb-20

Feb-21

Feb-22V-shapedrecovery,ARPA

stimulusSource:MorganStanleyResearch,

FRB

NewYork

SurveyofMarketParticipants

(SMP),BloombergUS

Treasuries

offer

even

less

value

to

FX-hedged

buyers

nowIn

mid-January,we

discussed

the

value

proposition

of

US

Treasuries

for

euro

and

yen-based

investors.We

suggested

investors

should

assess

value

by

using12-month

FXforwards,instead

of

the

typical

3-month

versions,given

how

hawkish

markets

pricedcentral

bank

policies.Since

then,the

value

proposition

of

Treasuries

has

worsened.Exhibit10

and

Exhibit11

show

the

annualized

currency-hedged

yield

thata

EUR-basedinvestor

and

JPY-based

investor

mightgetfor

buyinga

10y

US

Treasury

note.The

timeseries

shows

the

annualized

currency-hedged

yield

using3-month

FXforwards,as

wellas

the

annual

yield

the

investor

would

getusinga

12-month

FXforward.

5Aeuro-based

investor

would

only

pick

up

26bp

and

a

yen-based

investor

would

giveup11bp

to

buy

a

10y

US

Treasury

note.As

a

result,wethink

theonlymajorsourceofoverseasdemandforTreasurieswillcomefrominvestorswillingtotakecurrencyrisk.And,given

the

ECB

has

become

more

hawkish

-

threateningto

strengthen

the

euro,wethink

thatdemand

for

Treasuries

withoutan

FXhedge

would

mostlikely

come

frominvestors

in

Japan

-

keepingusbullishonUSD/JPY,andturningusbullishonEUR/JPY(see

Turningbullish

EUR).0.760.50.0Exhibit10:

Yieldpick-upofferedby10yUSTover10yDBRforaEUR-basedinvestor

%

1.5

1.0

0.27Using12minstead

-0.5

-1.0

'13

'14

'15

'16

'17

'18

'19

'20

'21

'22

UST10ypick-upforEURinvestor

rolling3mfwdSource:MorganStanleyResearch,

Bloomberg0.57

0.0-0.5-1.00.5Exhibit11:

Yieldpick-upofferedby10yUSTover30yJGBsforaJPY-basedinvestor

%

1.5

1.0

-0.11Using12minstead

-1.5

-2.0

'13

'14

'15

'16

'17

'18

'19

'20

'21

'22

UST10ypick-upforJPYinvestor

rolling3mfwdsSource:MorganStanleyResearch,

BloombergSeasonality

unkind

to

global

government

bonds

until

later

in

FebruaryNo

letup

is

in

sightfor

the

governmentbond

market,in

partbecause

seasonality

isn'tsetto

help

until

the

2nd

half

of

February.That's

when

supply

dynamics

in

the

UKand

alarge

indexextension

in

US

Treasuries

has

helped

boostreturns

in

the

past.Even

then,the

supportis

modestatbest,and

notwith

the

type

of

confidence

interval

for

whichyou

would

hope

(see

Exhibit12and

Exhibit13).Given

January

US

CPI

reports

on

February

10,a

strongnumber

could

push

governmentbond

prices

lower

once

more.Forinvestorslookingtoputmoneytowork

ingovernmentbonds,wethink

aJanuaryUSCPIupsidesurprise-drivenriseinyieldswouldpresentanopportunity.We

doubtlevered

investors

would

have

more

duration

to

sell

after

a

strongCPI

report(they

would

likely

be

maximum

short).Likewise,we

doubtinvestors

who

have

beensittinglongduration

would

be

able

to

hold

on.As

such,astrongJanuaryCPIreportmightseebondmarketpositioningreachpeak

bearishness,near-term.

6Apr1HJan

1HJan

2HApr2HMay1HMay2HJun1HMar1HJan

1HJan

2HMay1HMay2HMar1HMar2HJun1HJun2HFeb1HFeb2HMar2HApr

1HApr

2HJun2HFeb1HFeb2HHigherRates,

SmallerBalanceSheetNotAllBadNewsFor

decades

now,savers

in

the

developed

world

have

complained

aboutlow

interestrates,understandably.Borrowers,meanwhile,have

been

silenton

the

issue

-

alsounderstandably.Whatis

good

for

one,can

be

bad

for

the

other.Savers

have

placed

theblame

squarely

on

the

Fed,and

their

complaints

have

some

merit(see

Exhibit14andExhibit15).0.610.00-0.540.32-0.040.400.030.77-0.510.39-0.070.41-0.2-0.4-0.61.00.20.0Exhibit12:

Globalgovernment

bonds(unhedged)totalreturnsbyhalf-monthperiodthroughJune

%

-0.8

Median

half-monthreturn(last20years)Source:MorganStanleyResearch,

Bloomberg7353828567905096887862876050100

90

80

70Exhibit13:

Confidenceintervalforhalf-monthlyreturnsinglobalgovernment

bondsthroughJune

%

40

ConfidenceintervalSource:MorganStanleyResearch,

Bloomberg7.668Jan-82Jan-92Jan-02Jan-12Jan-22Netinterestincomeas%oftotalincomeTrendExhibit14:

USpersonalnet

interest

incomeasa%oftotalpersonalincome

%

20

18

16

14

12

100.2520Jan-82Jan-92Jan-02Jan-12Jan-22FedfundstargetrateTrendExhibit15:

Target

fedfundsrate

%

16

14

12

10

8

6

4Source:MorganStanleyResearch,

FederalReserveSource:MorganStanleyResearch,

BEA

Atthe

same

time,in

nominal

terms,personal

netinterestincome

has

increased

(see

Exhibit16).The

dramatica

rise

in

aggregate

US-dollar

denominated

debtoutstanding

goes

a

longway

to

explainingwhy,despite

the

decline

in

yields,personal

netinterest

income

has

risen

(see

Exhibit17).

7Naturally,US

Treasury

securities

have

contributed

significantly

to

the

increase

in

overallUS

debtoutstanding,even

after

removingFed

purchases

(see

Exhibit18).As

a

result,when

the

Fed

raises

or

lowers

its

policy

rate

relative

to

trend,and

debtmarkets

respondwith

higher

yields,netinterestincome

relative

to

its

trend

follows

(see

Exhibit19).1.11.01.4Jan-82Jan-92Jan-02Jan-12Jan-22TrendExhibit16:

USpersonalnet

interest

income

US$trillions

NetinterestincomeSource:MorganStanleyResearch,

BEA25.525201510

5

030Jan-82Jan-92Jan-02Jan-12Jan-22Exhibit17:

BloombergUSAggindexdebt

outstanding

US$trillions

USAggregateindexdebtoutstandingSource:MorganStanleyResearch,

BloombergExhibit18:

TotalmarketableUSTreasurydebtoutstanding22.61510

5

0

Jan-82

Jan-92

Jan-02

TotalUSTmarket

outstanding…

14.4Jan-12

Jan-22

...ex-FedholdingsUS$trillions

25

20Source:MorganStanleyResearch,

BEA,

FederalReserveExhibit19:

Deviationofthetarget

fedfundsrateandnetinterest

incomefromtrend500-50-100-150-200

2

1

0-1-2-3-4

Jan-82

Jan-92

Jan-02

Jan-12

Jan-22FedfundsdeviationfromtrendNetinterestincomedeviationfromtrend(rhs)%6543US$billions

200

150

100Source:MorganStanleyResearch,

BEA,

FederalReserveHigher

Fed

policy

rates

should

then

boostpersonal

netinterestincome

this

year.Inaddition,the

publicwill

earn

additional

intereston

the

securities

no

longer

purchased

bythe

Fed.In

thatsense,rate

hikes

and

balance

sheetnormalization

aren'tuniformly

badfor

the

economy.Still,netinterestincome

only

makes

up

7.6%of

personal

income

in

the

US.So

highershort-term

rates

-

which

will

putupward

pressure

on

longer

term

borrowingrates

forhouseholds

and

corporations

-

and

more

Treasury

issuance

-

which

will

drain

broadliquidity

-

should

have

a

net-negative

impacton

the

economy.

8Speakingof

SmallerBalanceSheetsAs

central

banks

wrap

up

their

assetpurchases

and

embark

on

their

post-COVIDtighteningcycles,we

expecta

key

area

of

investor

debate

will

be

the

future

of

theirassetholdings.Of

the

eightG10

central

banks

with

QE-related

holdings,we

expectsixcentral

banks(Fed,BoE,Riksbank,RBNZ,BoC,and

RBA)to

end

2025with

fewer

holdings.Some

of

these

central

banks

may

choose

to

allow

their

QE

holdings

to

roll

of

theirbalance

sheetas

they

mature;others

(like

the

Fed)are

likely

to

limitthe

pace

of

runoffthrough

the

use

of

"caps."Still

other

central

banks

may

accelerate

the

pace

of

roll-off

by

actively

sellingtheassets

they

acquired

through

QE

programs.For

example,the

Bank

of

Canada

has

notedthat"sellingcertain

assets"is

an

action

itmay

take

to

"normalize"its

balance

sheet.Likecaps,active

sales

may

also

be

used

to

smooth

the

pace

of

balance

sheetshrinkage

byoffsettinglumpy

maturities.Our

economists

currently

projectthatthe

ECB

will

conclude

assetpurchases

in

thecomingquarters,butatthis

time

do

notanticipate

the

GoverningCouncil

thatwill

allowPEPPand

APPassets

to

roll

off

the

balance

sheet.Similarly,our

Japan

economistsexpectthe

BoJto

continue

purchasingassets

atleastuntil

the

end

of

2023.Exhibit20:

Weexpect

QE-relatedholdingstoshrinkatvariedpacesuntil2025...-20%-30%-40%-50%-60%202220232021

Fed

BoCBoERBA

2024RiksbankBoJ2025

RBNZ

ECB

%ofCurrentQEHoldings

10%

0%-10%Source:CentralBanks,

Bloomberg,

MorganStanleyResearchExhibit21:

...but

inaggregate,QEholdingswillpeakduringsummer2022191817

20212022202320242025USDtn2120G10CentralBankQEHoldings(Aggregate)

August2022Source:CentralBanks,

Bloomberg,

MorganStanleyResearchExhibit22shows

the

pace

atwhich

we

expecteach

central

bank's

QE

holdings

to

declinein

the

comingyears.We

do

notinclude

Norges

Bank

or

the

SNB,which

did

notconducttraditional

QE-like

purchase

programs

duringthe

pasttwo

years.Similarly,since

weinclude

only

traditional

QE

assets,we

do

notinclude

the

substantial

TLTRO

repaymentsthatwe

expectto

shrink

the

ECB's

overall

balance

sheetover

the

nexttwo

years.

9SomeconclusionsstandoutfromthisexerciseAggregate

G10

central

bank

QE

holdings

are

setto

continue

expandingthrough

thissummer

and

peak

around

August.Unless

central

banks

announce

plans

to

actively

selltheir

QE

assets

in

the

comingquarters

(or

the

ECB

concludes

its

assetpurchasessubstantially

earlier

than

our

economists

currently

expect),aggregate

G10

central

bankQE

holdings

will

end

2022atroughly

the

same

levels

atwhich

they

currently

stand.Also,the

maturity

structure

of

the

Bank

of

Canada's

QE

holdings

stands

outamongG10central

banks.BoC

QE

holdings

will

mature

(as

a

proportion

of

the

currentlevel)fasterthan

other

central

banks'balance

sheets.The

BoC

may

choose

to

smooth

the

course

of

this

runoff

by

usingcaps

(if

itis

inclined

tobringthe

pace

of

balance

sheettighteningcloser

in

line

with

the

Fed)or

by

usingactivesales

(if

itis

inclined

to

return

fairly

quickly

to

conductingits

monetary

policy

through

acorridor

system).The

BoC's

relevantframework

notes

thatin

the

case

of

assetsales,such

sales

"would

be

carried

outan

appropriately

measured

pace."Turningbullish

EURThe

ECB,in

our

view,sets

up

an

attractive

opportunity

to

turn

bullish

the

EUR

and

werecommended

buyingEUR/USD

in

the

wake

of

the

meeting.We

also

maintain

longEUR/GBPand

look

to

add

longEUR/JPY

(in

addition

to

longUSD/JPY)and

shortCHF/SEK,as

we

think

SEKmay

be

the

"nextshoe

to

drop"as

investors

look

for

a

hawkishpivotfrom

the

remainingdovish

central

banks.For

more

see:EUR

|Bullish

EUR.EUR/USD

has

become

increasingly

correlated

to

front-end

rate

differentials

(Exhibit22).EUR/USD

weakness

in

recentmonths

was

explained

by

risingfront-end

US

rates.So

toocould

we

see

EUR/USD

strength

driven

by

a

catch-up

from

the

front-end

in

Europe.Positioning,seasonality,and

the

equity

rotation

narrative

outof

growth

and

into

valueall

supportEUR/USD

too.Exhibit22:

EUR/USDhasbeenchieflydrivenby2yratedifferentials,reflectingmonetarypolicydifferentialsSource:Macrobond,

MorganStanleyResearchExhibit23:

Market

expectationsforECBhawkishnesshavebeenbuildingrelativetothat

oftheFed

FSource:Macrobond,

MorganStanleyResearch10The

hawkish

response

of

the

ECB

fits

nicely

into

our