电子设备与多产业季度审查：非住宅建设.docx

North America Equity Research 14 January 2020Electrical Equipment & Multi- IndustryC. Stephen Tusa, Jr CFA AC (1-212) 622-6623 Bloomberg JPMA TUSA <GO> J.P. Morgan Securities LLCPatrick M. Baumann, CFA (1-212) 622 0160 J.P. Morgan Securities LLC Abhipsa Sahu (91-22)6157-4230J.P. Morgan India Private LimitedJ.R MorganEE/MIQuarterly Check-Up: Non-Residential ConstructionUS non-res construction PIP data showed improving growth in 4Q19 QTD through Nov, as compared to 3Q19 and 1H19 levels. Top-down indicators including ABI, Dodge, and credit metrics have displayed some sluggishness, overall pointing to a slowdown, similar to forecasts from industry experts, who. however, still call for growth. Bottom-up vertical specific indicators remain mixed on net. with sonic strength in particular verticals, mostly Institutional. 4Q saw a continuation of solid trends in Healthcare and Manufacturing, Office and Education showed accelerated positive growth, while Commercial remains negative, though also showing some gradual recovery. Finally, the overall tone from management teams on the state of the market remains cautiously optimistic with expectations for LSD-MSD type growth in 2020, solid but not necessarily differentiated vs the overall macro backdrop and other end markets, as labor shortage and tariff challenges pose risk for a slowdown. Within our coverage, companies with relevant non-res exposure overall include OWs HON, UTX, PNR, IR. HUBB and HDS, UW-ratcd LII and WSO, and N-ratcd JCI and WCC. Non-res quarterly update. This quarterly report provides a detailed dashboard of macro data related to the US non-residential construction market, which is important to the EE/MI sector, with -15-20% of large-cap sales directly or indirectly related. In addition to the commonly followed top-down data, we look at bottom-up drivers of construction activity with macro data across the major non-res verticals such as office, retail, and education. EE/MI-related US non-res construction improved in 4Q19 above 1Q/2Q/3Q levels. Construction put-in-pl ace in key verticals, as measured by our J PM EE/MI Non-Res Index, was up -3.3% in Nov (-0.9% in Oct, -3.4% in Sept, -1.3% in 3QI9, -0.8% in 2QI9, -+1.1% in 1Q19). Growth in public markets (up 11.4% in Nov, +10.1 % in Oct, +4.7% in Sept) outpaced private markets (+0.6% in Nov, -4.4% in Oct, -6.1% in Sept). Among non-institutional verticals, office continued to be strong and picked up to HSD in Nov vs LSD-MSD prior (vs double digits in 2H18), up 8.2% in Nov, vs up 3.2% in Oct, +3.1% in Sept, up 5.1% in 3Q19, up -7.1% in 2Q19 and up 8% in IQ19. Commercial had a slower negative growth of -10% in 4Q19 QTD through Nov, vs -14% in 3Q19, -15% in 2Q19 and -12% in 1Q19 after decelerating from MSD levels in 2H18. Among institutional verticals, Education showed a positive rise in 4Q19 to +7% in Nov (+5% in Oct, flat in Sept) vs flat in 3Q19, down 3% in 2QI9 and down 1% in 1Q19 while Healthcare remained steady in the LSD-N4SD range in 4Q19 QTD through Nov, up 5% in Nov. vs up 1% in Oct. -+4% in 3QI9,+3% in 2QI9 and +5% in I QI 9. Manufacturing maintained LSD growth mostly, with Nov up 2%. vs up 2% in Oct. -+2% in 3Q19. up -3% in 2QI9, up 6% in IQI9 after remaining negative for more than 2 years (down6% in 1H18 vs down low leens in 2H17). The total non-res index including all verticals like Power. Highway & Street and Transportation, is tracking up 5.1% in Nov vs 2.3% in Oct. 0.4% in Sept,+0.7% in 3Q19,+3.0% in 2Q19, +3.7% in 1QI9 and +3.4% in 2018.See page 31 for analyst certification and important disclosures, including non-US analyst disclosures.J.P. Morgan does and seeks to do business with companies covered in its research reports. As a result, investors should be aware that the firm may have a conflict of interest that could affect the objectivity of this report. Investors should consider this report as only a single factor in making their investment decision.Figure 13: EE/MI Non-Residential Construction Index vs. ABI C&l andInstitutional (Near-Term View)ABIConstruction PIP, Y/YFigure 13: EE/MI Non-Residential Construction Index vs. ABI C&l andInstitutional (Near-Term View)ABIConstruction PIP, Y/Y25%5%15%-35%Figure 12: EE/MI Non-Residential Construction Index vs. ABI C&l andInstitutional (Historical View)ABIConstojction PIP, Y/Y65.055.045.035.0Total Eb'MI-related Construction ABI C&l25% ABI Institutional-15%Ou «,>0N Z>ON 9>0N g>ON 3>0N m>ON Z>ON >0N 0>ON g>02 80 AON ZOAON '90 AON 2 >02 ，3 >02 3 >02 s>02 'a >02 00 >02 66 AON25.0-3o 二 § 6L宴 '8WN 8工星 匚>0N 二富 9L>0N 9工罡 loL 复 M>0N W A星 RA0N coLAew 'I AON NLAes 二 >0N ，二富 °A0NSource: AIA, U.S. Census.Source: AIA, U.S. Census.Dodge new contractsThe Dodge new non-res contracts are another commonly watched indicator. Because this data series captures large project contracts as opposed to the run rate of activity, it is more volatile and often overshoots to the upside or downside in any given month. The volatility in this monthly series makes it tougher (o draw reasonable conclusions from any short-term trend, but the 12-nionth rolling average has historically been a more reliable indicator. Also note that Dodge data includes 100% of the estimated value of a project in the first month of activity vs. Census data (which relies on Dodge for leads), which assigns spending across the duration of a project. The most recent contracts data from Nov showed new contracts down 5% y/y on a 3mo rolling average, while the 12mo rolling average was down 3% in Nov vs down 4% in Oct, flat in 3Q19, up 5% in 2Q19.Figure 15: Dodge New Non-Res Contracts vs. EE/MI Non-Residential Construction Index (Near-Term View)Dodge New ContractsConstruction PIP, Y/Y30% -10% -10% -Figure 14: Dodge New Non-Res Contracts vs. EE/MI Non-ResidentialConstruction Index (Historical View)Dodge New ContractsConstnjction PIP, Y/Y-Y/Y 3mo Avg Dodge Non-resTotal EE/Ml-related ConstructionJ 66 AONy/y 12mo Avg Dodge Non-res Contracts8 3 S 8 3 8 8 Sse'oN ZLAON 9 二。N g 二。N W AON6 二ON EON 二、ON 0 二。Ne>ON coLAew Z1 ZL富 L>ON 二妻y/y 12mo Avg Dodge Non-res Contracts 寸 biog99zzcoeo6X-5t-TT-XXT-A >A >A >A>A>AC0O<0O<DO<0O<0O<UWNWNWNWNWNW50%-40%-30%20%10%- 0%-10%-20%-30%-40%-50%Source： McGraw Hill Dodge, U.S. Census.Source： McGraw Hill Dodge, U.S. Census.Source： McGraw Hill Dodge. U.S. Census.Tying the above back to the quarterly organic growth for our group, there is little correlation this cycle, where the data sets have decoupled significantly due to large manufacturing projects.10Figure 16: Dodge Contracts vs EE/MI Index Non-Res Organic Growth y/y3Q191Q193981Q183Q171Q173Q161Q163Q151Q153Q141Q143Q131Q133Q121Q123Q11一 Q113QO1QO3Q09一 009 3Q081Q08. Dodge Non-res Contracts (12 month average y/y) . Non-res Company Segments AverageSource: Dodge. U.S. Census.Dodge momentum indexThe Dodge momentum index is a monthly measure of initial reports for non- residential building projects in planning, which generally leads construction spending by a full year. We look at both the overall momentum index as well as the vertical- specific indices such as commercial and institutional buildings. The overall reading was 155.3 for Nov vs 150.9 fbrOct, 142.7 in Sept,139.8 in 3QI9 and -143.5 in 2Q19. The y/y 3-month rolling average was down 4% in Nov (compared to down 9% in Oct, down 14% in Sept, down -15% in 3Q19 and down 10% in 2Q19). For the sub-indices, the Nov commercial index reading came in al 171.9 (170.6 in Oct. 175.8 in Sept, 162.1 in 3Q19 and 151 in 2Q19), while institutional came in at 134.8 in Nov (126.5 in Oct, 101.8 in Sept, 112.3 in 3Q19 and 134.2 in 2Qi9).28%24%20%16%谶8%4%0%-4%Figure 17: Dodge Momentum Index vs. EE/MI Non-Residential Construction IndexConstruction PIP, Y/YDodge Momentum IndexNov 19Aug 19May 19Feb 19Nov 18Aug 18May 18Feb 18Nov 17Aug 17May 17F8-17Nov 16Aug 16May 16Feb - 6Nov 15Aug 15May 15Feb - 5Nov 14Aug 14May 14Feb1 4Nov 13 ri 3Total EE/MI-related ConstructionDodge Momentum IndexSource: McGraw Hill Dodge, U.S. Census.Employment dataEmployment growth in the construction sector can be viewed as another high-level proxy for activity levels - this historical pattern has varied, with employment sometimes lagging pul-in-place data and other times leading. Non-Rcs Building11Construction employment declined by 1.1% in Nov vs down 1% in Oct, down 1.2% in Sept, down 0.9% in 3Q19, *+0.2% in 2Q19, -+1.5% in IQ19 while growth in non-farm payrolls was mostly steady at 1.5% in Nov vs 1.4% in Oct, -1.5% in 3Q19, 1.6% in 2Q19, -1.8% in IQ19. The decline in employment could be due to rising wage inflation, which was up 3.2% in 3Q19,(4% in 2QI9, 3.5% in IQ, 3.7% in 4Q, 3.4% in 3Q/2Q) vs the 10-year average of 2.4% until 2017.EE/Ml-Related Non-Res Non-Res Building Consti Total Non-Farm Payrolls20.0%-10.0%-0.0% -10.0%-20.0%Figure 18: Non-Res Building Construction Employment & Non-Farm Payroll Employees vs. EE/MI Non-Residential Construction Index (Historical View)Construction PIP, Y/YEmploymentBL AON ZL>ON 9LAON 巴AON 三AON S2A0N ? >ON 二 >ON OLAON SAON §>ON Z0AON 8>ON 90 aon 名>ON 8 4 si a >02 8 5 66>ON 96>ON Z6AON 96 AON IO6AONSource: BLS. U.S Census.Construction PIP. Y/Y20% EE/MI-Related Non-Res>*>>*>>*CT5OCD OCUW N W N WFigure 19: Non-Res Building Construction Employment vs. EE/MI Non-Residential Construction Index (Near-Term View)12Source: BLS. U.S. Census.Figure 20: Office Employment vs EE/MI Non-Residential Construction Index (Historical View)Construction PIP, Y/YEmployment30%EE/WI-Related Non-Res0%20% -10%Office Emplo;-10% -20% -8.0%16.0%4.0%2.0%0.0%-2.0%-4.0%6.0%o% 69 二。N ZLAON 9=0N GZON 二 AON J2A0N CM= ON 二 >0N 0=ON SAON §>0N Z0AON 8>oz 号。N 3>oz 2>oz 8 >02 3>oz 8>oz 66 AON 86 AON Z6>ON 96 AON JIO6AONSource: BLS. U.S. Census.Financial indicators mixedStability in financial markets is an important precondition for construction spending and remains relevant, particularly given the role that credit played in the last peak and subsequent downturn. The quarterly FRB loan officers' survey provides a good directional indication of both Ioan demand and lending standards for non-residential construction loans. Historically, however, the data tends to be better at calling inflection points in (he cycle than it does absolute levels of growth mid-cycle. In the 4QI9 survey, a net 13% of loan officers indicated a tightening of lending standards for non-residential construction loans (7% in 3Q19, 11% in 2Q19, 12% in 1Q19, 4% in 4Q18). In terms of loan demand, a net -5% saw stronger loan demand in the 4Q19 survey (-6% in 3QI9, -17% in 2Q19,-11% in 1Q19,-Il%in 4QI8, -7% in 3Q18).net % (tighteningyioosening standardsFigure 21: Commercial Real Estate Lending Standards and Loan Demand (Fed Loan Officers Survey, Historical View)%97024O2OO-2O4O-6O-8O1OOFigure 22: Commercial Real Estate Lending Standards and Loan Demand (Fed Loan Officers Survey, Near-Term View)%Source： FRBSource FRB13Figure 23: Commercial Real Estate Lending Standards vs. EE/MI Non-Residential Private Construction IndexConstruction Private, Y/YLending Standards (%)Figure 23: Commercial Real Estate Lending Standards vs. EE/MI Non-Residential Private Construction IndexConstruction Private, Y/YLending Standards (%)bP寸bbbbbbbbbbbbbbbbbb寸bb ooacoaooooooooaooooooaoc9Z86OLZSb99Z86O 一 CJCOb99ZCO6 66660000000000LLLLLbLLT- LEE/MI Private Non-Res Y/Y (Ex-Mfg) net % (tightening)/looseningstandardsFigure 24: Loan Demand vs. EE/MI Non-Residential PrivateConstruction IndexConstruction Private, Y/YLoan Demand (%)Source: FRB, U.S. Census.Source: FRB, U.S. Census.Tying the above back to the quarterly organic growth for our group, there is little tangible historical correlation, which doesn't make this data conclusive, but we think it's necessary to watch closely.Figure 25: FRB Survey vs EE/MI Index Non-Res Organic GrowthNon-res company averageFRB Survey (increase in tightening/loosening of standards)Source. FRB. U.S. Census.Looking at CMBS spreads for BBB- bonds vs others, after some moderation in 2017, data has somewhat stabilized in 2018 YTD.Figure 26: CMBS Spreads bpsAPL3Dec. 19Aug, 19 APL9Dec.18 Aug.18 Apr£8Dec£7Aug. 17 Apz.7Dec6Aug, 6 APL6Dec5 Aug，15APL5 06914Aug-14 Apr，14AA CMBS AAA Jr CMBSBBB-CMBSSource: Bloomberg.Commercial Real Estate (CRE) Price IndexThe CoStar Commercial Repeat-Sales Index is an indicator of CRE price changes in the US. The index is constructed using a repeat sales methodology that measures the price change in the sale of a commercial property relative to the previous sale of the same property. This helps measure price movements due to change in market conditions and has been a good leading indicator of overall spending trends. In Nov, the CRE Value Weighted Price Index was up 7.3% y/y compared to 6.8% y/y in Oct, 8.2% in Sept, up -9.9% in 3Q19, up -5.9% in 2Q19,6.2% in 1Q19.Figure 27: CRE Value-Weighted Price Index vs. EE/MI Non-Res SpendingConstruction PIP, Y/YCRE Price Index Value-Weighted-25.0%于子矛吝于子王舌于于25.0%15.0%<s< <S 55.0%-5.0% -15.0% -N N N o o o < < <30.0%20.0%10.0%0.0%-10.0%-20.0%-30.0%-40.0% NNDOO c <<8 8 2O O Q 2 3HTotal EEjWI Non-Res SpendingCRE Price Index Value Weighted-35.0%Source: Costar, U.S. Census.Third-party forecastsSeveral third-party sources issue forecasts for the non-res conslruction markets that are typically (but nol always) based on the U.S. Census pul-in-place data; in other words, they may not be directly comparable. Below, we show the AIA consensus construction forecast, an average of the miijor forecasting agencies including McGraw Hill Dodge. Global Insight, and others (updated twice yearly). Current 2019/2020 consensus calls arc 3.8% / 2.4% growth, respectively.152Q19E2020ENon-Residential Total3.8%2.4%Commercial Total2.5%1.1%Office6.9%2.2%Retail/Other Commercial-0.8%0.3%Hotel3.6%0.5%Industrial Total6.4%3.3%Institutional Total4.9%3.7%Healthcare2.7%3.7%Education6.1%4.4%Religious4.5%2.8%Public Safety8.2%5.9%Amusemenb'Recreation4.4%1.0%Table 5: AIA Consensus ForecastsSource: AIA.2016Y/Y Growth in.20172018201920202H15 forecast8.20%1H16 forecast8.30%6.70%2H16 forecast5.80%5.60%1H17 forecast5.60%4,90%2H17 forecast3.80%3.60%in io lorecasi4.UU%4.yu%2H18 forecast4.70%4.00%1H19 forecast4.40%2.40%2H19 forecast3.80%2.40%Table 6: Non-Res Construction Forecasts Have Consistently Been