摩根士丹利-全球-宏观策略-全球利率策略：莱德杯重生-2019.5.3-64页.pdf

Matthew.HGuneet.DTony.SSam.EDavid.G.HShreya.CRobert.J.BKoichi.SShoki.OMORGAN STANLEY&CO.LLCMatthew HornbachSTRATEGIST+1 212 761-1837Guneet Dhingra,CFASTRATEGIST+1 212 761-1445Tony SmallSTRATEGIST+1 212 296-5876Sam ElprinceSTRATEGIST+1 212 761-9491David HarrisQUANTITATIVE ANALYST+1 212 761-0087MORGAN STANLEY&CO.INTERNATIONAL PLC+Shreya ChanderSTRATEGIST+44 20 7425-4740Robert J BrownSTRATEGIST+44 20 7425-4638MORGAN STANLEY MUFG SECURITIES CO.,LTD.+Koichi SugisakiSTRATEGIST+81 3 6836-8428Shoki OmoriSTRATEGIST+81 3 6836-5466Institutional Investor Global Fixed-IncomeResearch Polls are now open.We hope youhave enjoyed our research over the past yearand appreciate your support.Request yourballot.Please click here if you would like to receive our dailyinterest rate market commentary:the TreasuryMarket Commentary.Global Interest Rate StrategistGlobal Interest Rate Strategist|Global GlobalThe Ryder Cup RebornThe macro environment continues to reshape into onesupportive of duration in the US and Canada over Germanyand the UK.Despite recent sideways movement in the RyderCup of Bonds,we expect the cross market yield spread tocompress further in the coming months.Government BondsIn the US,we continue to suggest longs in 2y and 10y notes and long theUS and Canada vs.Germany and the UK in the Ryder Cup of Bonds trade.Inthe euro area,we suggest investors hold shorts in Bunds via the Ryder Cupof Bonds,and point to the possibility of inflation expectations re-ratinghigher,given the recent strength in Brent and core CPI.In the UK,followingthe May MPC,we still favour cross-market shorts in gilts via the Ryder Cup.InflationIn the US,we examine how transitory the decline in core PCE inflation is,alongside the efficacy of trimmed means.We also looked at the decline inunit labor costs.We continue to stick to longs in the 5-year TIPS real yields.In euro area,we continue to like longs in 10y HICPxT breakevens out of theupside data surprise.In Japan,we discuss the upcoming JGBi linker auctionon May 10.Money MarketsIn the US,we recommend September 2019 FRAOIS wideners as we expectthe extreme tightness in Libor-Repo will lead to a switch in foreign bankfunding from Repo to the CP market.We discuss how the fall in 3m Liborfrom 2.60 to 2.56 has been driven by the spread between OBFR and EFFRturning negative rather than by the compression of the term premiumimpeded in the 3m Libor-O/N Libor spread.We expect EFFR to stabilizefrom now until September.DerivativesIn the US,we like belly UST-OIS wideners as we project lower net USTissuance over the coming 2 years.We expect 30y FRAOIS spreads to remainanchored around 25bp given the expectation that the Libor-SOFR spreadwill be fixed around 23bp after the eventual Libor discontinuation.Technical AnalysisUST 10y at 2.525%,support at 2.57%,2.69%,resistance at 2.45%,2.34%.DBR10y at 0.025%,support at 0.05%,0.09%,resistance at-0.03%,-0.10%.UKT10y at 1.22%,support at 1.25%,1.33%,resistance at 1.16%,1.09%.JGB 20y at0.37%,support at 0.38/0.39%,0.41%,resistance at 0.35%,0.32%.Due to the nature of the fixed income market,the issuers orbonds of the issuers recommended or discussed in thisreport may not be continuously followed.Accordingly,investors must regard this report as providing stand-aloneanalysis and should not expect continuing analysis oradditional reports relating to such issuers or bonds of theissuers.Morgan Stanley does and seeks to do business withcompanies covered in Morgan Stanley Research.As aresult,investors should be aware that the firm may have aconflict of interest that could affect the objectivity ofMorgan Stanley Research.Investors should considerMorgan Stanley Research as only a single factor in makingtheir investment decision.For analyst certification and other important disclosures,refer to the Disclosure Section,located at the end of thisreport.+=Analysts employed by non-U.S.affiliates are not registered withFINRA,may not be associated persons of the member and may notbe subject to NASD/NYSE restrictions on communications with asubject company,public appearances and trading securities held bya research analyst account.1May 3,2019 11:40 PM GMT Government BondsUnited StatesWe continue to suggest investors remain long Treasuries,as the market tradesclose to the upper end of the range we think is in place for now.We think themarket will have a very hard time placing less than a 12.5%probability on a 50bpcut(or a 25%probability of a 25bp cut)occurring over the rolling forward-looking12-month period.That should make it difficult for the 2y yield to move above2.45%or the 10y yield to move above 2.62%.We continue to suggest investors express longs in both 2y and 10y notes-in someways suggesting our indifference to curve shape at this point.Steepeners carrynegatively and have done so since late December 2018.If we had conviction thatthe curve was going to trend steeper,say as a result of a series of rate cuts,wewouldnt mind paying the negative carry.But thats not where our conviction lies,at this point.We compare policy rate paths derived from market prices with the path obtainedfrom surveys of market participants-exactly as Vice Chair Clarida suggested in hisspeech titled Models,Markets,and Monetary Policy.Claridas speech suggested tous that market participant expectations,not just market prices,will play animportant role in the next evolution of rate policy.If market participants begin toexpect either a rate hike or a rate cut at some point this year,FOMC participantsmay feel the need to deliver on those expectations.Finally,we look at the rather disappointing reading on hours worked in the labormarket report and how the disappointing trend has boosted average hourlyearnings since September 2018.We also show the historical relationship betweengrowth in hours worked,personal income growth,and personal consumption.Euro areaEconomic and inflation data surprised to the upside throughout the week acrossEurope,further contributing to the positive tone post the Spanish elections andunchanged S&P ratings review for Italy.Data surprises were most pronounced inItaly,where preliminary Q1 GDP,the unemployment rate and manufacturing allsurprised positively;however,the most significant surprise may have come fromcore CPI,where the majority of our tracking scenarios now point to the potentialfor core and headline to exceed current ECB year-end estimates.We continue to suggest that investors express shorts in German Bunds via theRyder Cup of Bonds trade,but note that a re-rate of inflation expectations couldlead Bund yield higher in the near term.The case for BTP spread tightenerscontinues to strengthen given stronger-than-expected economic data,but weexpect markets to remain hesitant to embrace the better growth data givenheadline risk related to the strength of the government coalition in front of EUParliamentary elections and a likely deficit discussion with the EU this summer.2United KingdomThe MPC delivered a 9-0 hold on Thursday.The tone of the minutes wasbalanced,but the press conference highlighted Governor Carneys underlyinghawkish bias,in our view,particularly given his seeming conviction in a smoothBrexit outcome(upon which the forecast path is conditioned).Over the course of the week,the market had built up some excitement around thepotential for an 8-1 vote,with 1 MPC member voting for a hike.While,in our view,this excitement was somewhat unfounded,the selloff out of the press conferenceeven in spite of a 9-0 vote gives us conviction that the market determined the toneto be hawkish-leaning.As the MPC maintains its position as the only core centralbank exhibiting an underlying hawkish bias,we continue to favour cross-marketshorts in gilts via the Ryder Cup of Bonds.In addition,we continue to favour 2s5sGBP swap steepeners.United StatesMORGAN STANLEY&CO.LLCMatthew HornbachMatthew.H+1 212 761-1837Continue to trade the range with duration,not with curveWe continue to suggest investors remain long Treasuries,as the market trades close tothe upper end of the range we think is in place for now.Recall that our frameworkemphasizes the probabilities the market places on rate cuts over the rolling forward 12month period.As long as market participants continue to expect the Fed to remain on-hold over the coming year,we think the bond market will continue to price in somechance of a rate cut.Exhibit 1 and Exhibit 2 show our simple way of thinking about the relationship betweenthe probabilities placed on a 50bp rate cut and the level of 2y and 10y yields.We thinkthe market will have a very hard time placing less than a 12.5%probability on a 50bp cut(or a 25%probability of a 25bp cut)occurring over the rolling forward-looking 12-monthperiod.That should make it difficult for the 2y yield to move above 2.45%or the 10yyield to move above 2.62%.We continue to suggest investors express longs in both 2y and 10y notes in someways suggesting our indifference to curve shape at this point.Steepeners carrynegatively and have done so since late December 2018.If we had conviction that thecurve was going to trend steeper,say as a result of a series of rate cuts,we wouldntmind paying the negative carry.But thats not where our conviction lies,at this point.Instead,we have more conviction that the market will oscillate between putting a 20%probability and an 80%probability on a 50bp rate cut at some point over the forward-looking 12 months.The probability of such a rate cut could be driven by many differentfactors that could weigh equally on the 10y yield or yields further out the curve.Ofcourse,our suggested long in the 2y note is a negative carry trade as well,but werehappier to pay that negative carry,given where the market trades in our expected range.3For example,Exhibit 3 shows that the relationship between the probability of a rate cutover the forward-looking 12-month period and the shape of the 2s10s curve is non-existent.Exhibit 4 shows a stronger relationship with the 5s30s curve a curve in whicha steepener is less negative carry than a 2s10s steepener.Therefore,if we had to choosea curve trade,we would probably enter a 5s30s steepener.But we arent thrilled withthose steepeners either.So why dont we like the 5s30s steepener?Exhibit 1:Probability of 50bp cumulative rate cuts in thenext 12 months vs.CMT 2y yieldSource:Morgan Stanley Research,Federal Reserve,BloombergExhibit 2:Probability of 50bp cumulative rate cuts in thenext 12 months vs.CMT 10y yieldSource:Morgan Stanley Research,Federal Reserve,BloombergExhibit 3:Probability of 50bp cumulative rate cuts in thenext 12 months vs.CMT 2s10s curveSource:Morgan Stanley Research,Federal Reserve,BloombergExhibit 4:Probability of 50bp cumulative rate cuts in thenext 12 months vs.CMT 5s30s curveSource:Morgan Stanley Research,Federal Reserve,Bloomberg4In part,we worry about the crowded nature of positioning.Forward starting curvesteepeners have been a popular trade over the past year.The structures benefit whenthe belly of the curve(4-7y sectors)outperform the wings(2y,10y,30y sectors).Thesepositions could be at risk if financial conditions were to tighten dramatically led by riskyassets.In part,we dont have sufficient reason enough to dislike the long end of thecurve on its own,given we see a healthy amount of term premium already there(seeExhibit 5 and Exhibit 6).Trade idea:Maintain long UST 2y notes at 2.33%Trade idea:Maintain long UST 10y notes at 2.54%Trade idea:Maintain long US(TY)and Canada(CN)futures vs.short Germany(RX)and gilt(G)futuresSpeaking the same language as the Vice ChairIn his Friday speech titled,Models,Markets,and Monetary Policy,Vice Chair Claridaspoke about extracting signal from noise in financial markets dealing with monetarypolicy.Clarida spent time on the market price for monetary policy and how that pricemay include more than just a pure rate expectation.Clarida explains:For this reason,it is useful to compare policy rate paths derived from market priceswith the path obtained from surveys of market participants,which,while subject tomeasurement error,should not be contaminated with a term premium.Market-andsurvey-based estimates of the policy rate path are often highly correlated.Exhibit 5:UST 30y1m implied forward yield vs.the FOMCmedian longer run projected policy rateSource:Morgan Stanley Research,Federal ReserveExhibit 6:UST 30y1m implied forward yield less the FOMCmedian longer run projected policy rate-1.5-1.0-0.50.00.51.0May-14May-15May-16May-17May-18May-19Difference=Term premium%0.46Source:Morgan Stanley Research,Federal Reserve5We perform exactly this analysis and have for many years.Exhibit 7 compares the ratepath expected by market participants and that which is priced into the fed funds futuresmarket.We put the“rate path”in terms of the number of 25bp hikes that markets priceor market participants have expected in the 12 months starting around each FOMCmeeting in the past.Exhibit 8 then differences the two series to estimate how much term premium isimbedded into market prices.Of course,as Clarida suggests,this estimate is subject tomeasurement error.Clarida continued:But when there is a divergence between the path or destination for the policy rateimplied by the surveys and a straight read of interest rate derivatives prices,I place atleast as much weight on the survey evidence(for example,derived from the surveysof primary dealers and market participants conducted by the Federal Reserve Bank ofNew York)as I do on the estimates obtained from market prices.To us,this suggests that market participant expectations,not just market prices,willplay an important role in the next evolution of rate policy.If market participants beginto expect either a rate hike or a rate cut at some point this year,FOMC participants mayfeel the need to deliver on those expectations.Disappointment on wages masks more disappointment on hoursDespite robust headline payroll gains,average hourly earnings disappointed consensusexpectations at 3.2%Y/Y(see Exhibit 9).The softer wage growth helped contain thebearish momentum that built in the Treasury market since Chairman Powells pressconference.Our economists found the wage trends more upbeat than the headlinesuggested.They noted that their“estimate of the median rate of wage growth jumped0.3pp in April to 3.5%,which marked a high since December 2008”.Exhibit 7:#of 25bp rate hikes expected by marketparticipants in the next 12 months vs.the#implied bymarket pricing at each FOMC meetingSource:Morgan Stanley Research,Federal ReserveExhibit 8:#of 25bp rate hikes expected by marketparticipants in the next 12 months less the#implied bymarket pricing at each FOMC meetingSource:Morgan Stanley Research,Federal Reserve6At the same time,all of these factors masked a rather disappointing reading on hoursworked.Remember that average hourly earnings(AHE)is just a ratio of average weeklyearnings(AWE)to average weekly hours(AWH):AHE=AWE/AWH.This means that AHEcan increase in one of two ways:either AWE increases or AWH decreases.In th