全球基礎設施行業(yè)收費公路中斷

1、GLOBAL FOUNDATIONJune 4, 2019 04:55 AM GMTDisruption Decoded: Global InfrastructureToll Road DisruptionToll roads are a core infrastructure investment class, in our view, but new transport and tolling technologies could disrupt the traditional business m We estimate that shared autonomous vehicles c

2、ould widen toll road valuationranges by 15% vs. a steady state scenario.Morgan Stanley does and seeks to do business with companies covered in Morgan Stanley Research. As a result, investors should be aware that the firm may have aofinterest that could affect the objectivity of Morgan Stanley Resear

3、ch. Investors should consider Morgan Stanley Research as only a single factor in making their investment decision.For analyst certification and other important disclosures, refer to the Disclosure Section, located at the end of this report.+= Analysts employed by non-U.S. affiliates are not register

4、ed with FINRA, may not be associateds of the member and may not be subject to NASD/NYSE restrictions oncommunications with a subject company, public appearances and trading securities held by a research analyst account.獲取報告1、2、3、每周群內(nèi)7+報告；當日華爾街日報、4、行研報告均為公開利歸原作者所有，起點財經(jīng)僅分發(fā)做內(nèi)部學習。掃一掃 關注公號回復：加入“起點財經(jīng)”群。GL

5、OBAL FOUNDATIONContributorsMORGAN STANLEY AUSTRALIA L MITED+Rob KohEquity Analyst+61 3 9256-8932Rob.KohMORGAN STANLEY AUSTRALIA LIMITED+Xindi ShaoResearch Associate+61 2 9770-1115Xindi.ShaoMORGAN STANLEY & CO. NTERNATIONAL PLC+Nicolas J MoraEquity Analyst+44 20 7677-5376Nicolas.MoraMORGAN STANLE

6、Y & CO. LLCAdam Jonas, CFAEquity AnalystAdam.JonasMORGAN STANLEY C.T.V.M. S.A.+Josh Milberg, CFAEquity Analyst+55 11 3048-6133Josh.MilbergMORGAN STANLEY C.T.V.M. S.A.+Ricardo L AlvesEquity Analyst+55 11 3048-6238Ricardo.AlvesMORGAN STANLEY C.T.V.M. S.A.+Lucas T BarbosaResearch Associate+55 11 30

7、48-6636Lucas.BarbosaMORGAN STANLEY & CO. LLCMichael D Zezas, CFAStrategistMORGAN STANLEY & CO. LLCMark T Schmidt, CFAStrategistMORGAN STANLEY & CO. LLCAlexander W VentrigliaStrategistMichael.ZezasMark.Schmidt1Alexander.VentrigliaGLOBAL FOUNDATIONDisruption Decoded: Global InfrastructureT

8、oll Road DisruptionToll roads are a core infrastructure investment class, in our view, but new transport and tolling technologies could disrupt the traditional business m We estimate that shared autonomous vehicles could widen toll road valuation ranges by 15% vs. a steady state scenario.Industry Vi

9、ewAustralia Utilities & Infrastructure Surface TransportationBut capital is flowing into mobility innovation, based on the inter- section of many technologies (app-enabled ride-sharing, autono- mous systems, optical recognition, more efficient batteries, and drones), presenting opportunities and

10、 challenges to traditional tollroads.CautiousIn-LineA globally important asset class: Toll roads are an important insti- tution for policy-makers around the world, to fund and procure trans- port infrastructure. We estimate cumulative nominal investment ofUS$400b in OECD toll roads since 1993, and c

11、urrent listed market capitalisation of US$100b, which are less than 30% of all toll roads. In the United States, 42 out 50 states employ toll roads, withUS$84b of municipal toll road debt outstanding. The McKinsey Global Institute estimates US$18t in total road infrastructure investment to 2035, and

12、 we believe that a higher proportion of roads will be tolled in future, based on the virtuous circle of low operating risk and low cost of capital, helping policy-makers manage finite bud- gets.Traditionally defensive with high leverage: Relative to other asset classes, mature toll road cashflows ar

13、e relatively stable and predict- able (<5% cashflow variability yoy), so toll roads can generally sus- tain high levels of investment-grade debt, and toll road equity investment is typically seen as defensive.We map the landscape of mobility and tolling innovation, which we plan to update on a re

14、gular basis. Disruption impacts will be different for intra-urban vs. inter-urban roads and for cars vs. trucks, providing relative valuation opportuni- ties, in our view. For example, we see continuing economic drivers to increase freight travel and freight capacity utilisa- tion, whereas commuter

15、travel has higher disruption risk, in our view, owing to lower capacity utilisation (1.5 pax/car) and the potential for unpredictable behavioural change; and1.We introduce our proprietary DCF framework to mthe2.impact of shared autonomous vehicles on a generic toll road. We see potential positives (

16、more travellers, and more road capacity) as well as challenges (less traffic and less predict- able traffic). To highlight the maths, consider that an increase in pax/car to 2.5x would reduce toll road revenue by 40%, but travel demand could grow an extra 1% p.a. over 40 years as cost/km falls and u

17、nserved passengers gain access, and roads might fit an extra 10% of vehicles on a semi-autono- mous basis.GLOBAL FOUNDATIONContents5The Story in Six Charts6Executive Summary9A. Athrough the technology landscape10Mobility technologies14Tolling technologies17B. Ride-hailing, -sharing and -pooling24C.

18、US muni toll road debt: Upside from freight26D. Toll road stock order of preference29Appendix 1: Global listed toll road comparative data30Appendix 2: Generic toll road mGLOBAL FOUNDATIONThe Story in Six ChartsExhibit 1:Toll roads are a globally important infrastructure procurement mecha- nism and a

19、sset investment classExhibit 2:We track a number of new mobility and tolling technologies to help draw out relativitiesSource: McKinsey Global Institute, Aggregate global infrastructure requirement, 2017-2035 (US$t)Source: Morgan Stanley Research. Labels and pathways are random and illustrative only

20、.Exhibit 3:We introduce our proprietary framework to help investors gauge the impact of new mobility technology take-upExhibit 4:We find that valuations are resilient under most scenarios, with disrup- tion risk in the tailSource: Morgan Stanley Research.Source: Morgan Stanley Research.Exhibit 5:We

21、set out our global order of preference for listed toll road stocksExhibit 6:Favor muni toll roads with the highest freight revenue on upside from truck platooningCommercial revenues (% of total)60%57%50%45%44%40%38%30%20%10%0%Ohio Turnpike Commissionllinois State Toll Highway AuthorityPennsylvania S

22、tate Turnpike CommissionOklahoma Turnpike AuthoritySources: Company and media reports, Morgan Stanley Research estimates. Prices as of 31 May 2019.Sources: .Morgan Stanley Research, Moody's MFRAMORGAN STANLEY RESEARCH5GLOBAL FOUNDATIONExecutive SummaryWhy toll roads?OursWe view toll roads as cor

23、e economic infrastructure.We track a number of emerging technologies and highlight some toll road relativities. For freight, we see economics con- tinuing to push the envelope of capacity utilisation. For com- muter traffic, where capacity utilisation is lower, we see higher disruption risk, owing t

24、o the potential for unpredict- able behavioural shifts.1.2.Toll roads often athe arteries into and between cities, sup-porting employment hubs, trade flows, and real property values.A good toll road is a long-life asset (with appropriate maintenance, a concrete road can last 40 years), with a low co

25、st of capital (stable traffic patterns and fixed prices support high financial leverage), so policy-makers globally view toll roads as an attractive option fordevelopment, in the face of limited budgets.Our proprietary ming suggests that shifts in road and3.vehicle capacity utilisation could change

26、equity valuations by15% vs. a steady state scenario. To demonstrate the math, consider that an increase in pax/car from 1.5 currently to2.5x would reduce toll road revenue by 40%, but travel demand could grow an extra 1% p.a. over 40 years as cost/km falls and unserved passengers gain access, and ro

27、ads might fitan extra 10% of vehicles on a semi-autonomous basis.However, the rapid intersection of a number of mobility technologies ( Exhibit 10 ) means that all toll road stakeholders need to explore disruption scenarios and risk allocation. Toll roads are relatively con- trolled environments so

28、may well be good testing grounds for thenew technologies.We've written this note not just for investors but also for concession-aires, operators, lenders, policy-makers, and planners.What's next?We plan to continue to monitor and update our map of impactful technologies, and, in view of the

29、potential widening of future out-comes, we think investors, operators, and policy-makers willWhere we are different?With many global OEMs rapidly shifting the transport supply chain to reduced emissions, connected, and autonomous vehicles, we think the advent of 'Auto 2.0' is well appreciate

30、d.Increase focus on risk allocation, e.g., could subscription pricing re-balance pax/car changes?Place higher value on strategic options, e.g., we think toll road concessionaires will be well-placed to work with policy-makers on integrated transport plans.1.But we believe we are the first firm to ap

31、ply the coming changes to2.the traditional toll road business m.Based on the long-range work of our Global Autos & Shared Mobilitycolleagues, we map a range of emerging mobility technologies, aswell as advancement in tolling technologies, and we mplausibletoll road scenario bounds for the take-u

32、p of connected autonomousvehicles (CAV).6GLOBAL FOUNDATIONExhibit 7:Classifying the roads in our coverageExhibit 8:Key toll road characteristicsFrom where to where?Toll roads are typically classed as intra-urban, where the road links a city and its surrounding suburbs, and the primary travel purpose

33、 is commuting to work, or inter-urban, where the road links different cities, and travel purposes may include a higher proportion of trade and leisure.A combination of tolled roads within a catchment area may also be considered a network.Toll roads usually differentiate between vehicle weight and/or

34、 length (e.g., number of axles). Passenger vehicles (cars, and motor bikes) and heavier commercial vehicles (utility vehicles, vans, and trucks) have different drivers (commuting vs. freight) and impose different costs on roads (trucks take more space and are wear out roads faster).Toll roads typica

35、lly charge a fixed toll per trip or km, with a multiplier for vehicle type (e.g., in NSW, trucks pay 3x the car toll), but tolls costs are also a policy tool, so toll structures may vary by time (e.g., peak/ off-peak hours), and be capped within toll road net- works with multiple tolling points, and

36、 toll exemptions may apply (e.g., for low emissions or high occupancy vehicles).In what?Sources: Company reports, Morgan Stanley Research. # heavy vehicle traffic / total traffic. *weighted by proportional EBITDA.For how much?Toll escalation is typically annual, and linked to eco- nomic indicators (

37、e.g., inflation, ideally with a floor at 0%). Toll variability may also be linked to congestion, e.g., some US managed lanes vary the toll on an intraday basis so as to smooth traffic).Toll road concessions are typically a long-term con- tractual agreement or legal instrument (e.g., 20+ years), in o

38、rder to spread the high upfront capital costs over the lifetime users of the road (users pay a small amount so the road gets paid for over the life of the concession).For how long?Source: Morgan Stanley Research.MORGAN STANLEY RESEARCH7VariableCommentaryGLOBAL FOUNDATIONExhibit 9:Minimum DSCR ranges

39、 (x) at project commissioning1-23-45-67-89-1011-121.75NA NA NA NA NA1.75-2.55.0NA1.2-1.11.4-1.22.0-1.42.5-1.755.0-2.5NA<1.11.2-1.11.4-1.21.75-1.42.5-1.53.0x<1.1<1.1<1.2<1.4<1.5<3.0Source: Standard & Poor's, Project Finance Operations Methodology, 16 September 20

40、14 SACP = stand-alone credit profileOPBA = Operations phase business assessment NA = not applicable8Preliminary operations phase SACPOPBAaaabbbbbbToll road debt-sizingInvestors traditionally adopt a somewhat binary approach to greenfield vs. mature toll road developments.Greenfield traffic is less p

41、redictable because of non-linear impacts from toll diversion (users taking alternative routes) and transport network effects (unintended congestion points). Historically, greenfield toll road developments have suffered from inaccurate traffic forecasts, often leading to material capital destruction

42、and secondary buying opportunities.In contrast, free cash flow (FCF) from mature developed market toll road traffic tends to vary by <5% annually, with relatively stable revealed price elasticity (price escalation is generally pre-agreed in the concession), highly predictable workday and seasonal

43、 usage patterns, and relatively simple long-lived physical assets (minimal consumables and few moving parts).In the longer run, macroeconomic factors (population, employment, GDP, and land use changes) generally underpin steady traffic growth, which provides further buffer from short-term volume dow

44、nturns.Accordingly, well-positioned mature toll roads can currently achieve investment-grade credit ratings with very low cashflow coverage relative to industrial issuers, e.g., Debt Service Coverage Ratios (or DSCR, equal to Cashflow Available for Debt Service divided by interest and amortisation)

45、of 1.2-1.4x ( Exhibit 9 ), and amortisation within 2-5 years of concession expiry (e.g., 35 years for a 40-year concession).GLOBAL FOUNDATIONA. A landscapethrough the technology we see dozens of technologies affecting toll roadsExhibit 10:Related research we see differential impacts on road and traf

46、fic types chal- lenging the traditional relative investor preference for intra-urban roads over inter-urban roadsUrban Air Mobility: Wing: A New May 2019)tone for Urban Air Mobility (15The Morgan Stanley Global Electric Vehicle Market Monitor (12 Apr 2019)Investment Implications of Autonomous Urban

47、Air Mobility (8 Feb 2019)Autonomous Driving: Who Is Best Positioned to Monetize the Content we view economic and behavioural changes as the key cata-lysts ahead of regulatory evolutionOpportu? (3 Oct 2018)We map out a dozen impactful technologies in the following pages. which we split into two group

48、s mobility technologies (how wetravel) and tolling technologies (how a toll road operator works).Auto 2.0 Global Chart Book (20 Sep 2018)US Infrastructure: Laying an Investment Foundation (9 Apr 2018) Electric Vehicles: On the Charge (31 Aug 2017)Exploring the Bear Case: Distracted Driving + ADAS =

49、$7 Trillion of Used Values at Risk (1 Feb 2017)Cities & Shared Mobility - A Policymaker & Investor Guide (7 Jun 2016) Global Investment Implications of Auto 2.0 (6 May 2016)Our technology list builds on the foundational work of our Autos & Shared Mobility colleagues ( Exhibit 10 ). We ta

50、ke their work and think through the opportunities and challenges for traditional tollroads.Source: Morgan Stanley Research.Exhibit 11:Consensus view of toll road typesOur list is neither exhaustive nor mutually exclusive, and we intendto revisit this map in future research as technologies and busine

51、ssms evolve, and data sets become available.In general, we see high, and perhaps underappreciated, disruptionrisk for toll road business ms on a 20-year view, highlighted bytwo standard disruption markers:l the inevitable 'collision' of technological change and slower moving regulatory frame

52、works; andl pockets of low-capacity utilisation held up by long-standingbehaviours, providing new opportunities.When we collate our impact analysis ( Exhibit 12 ), we see a positive risk skew for heavy vehicles and inter-urban roads (driven more by economics), and a wider risk skew for passenger veh

53、icle traffic on intra-urban roads. Put simply, we think there is a strong economic driver for freight efficiency, but changing habits could influence pas-senger mobility.Source: Morgan Stanley Research.Exhibit 12:Summary of impact analysisSource: Morgan Stanley Research.MORGAN STANLEY RESEARCH9Traff

54、icIntra-urbanInter-urbanPassenger vehicle Heavy vehicle+-+-+-+-Intra-urbanInter-urbanPassenger trafficFreight trafficLeast elastic(commuting to work), most predictable (linked to employ-ment)Most elastic(business and hol-iday); moderately predict-able (linked to economic activity and weather)Quite i

55、nelastic(flow of essentialgoods); moderately predict-able (linked to industrialproduction)Quite elastic(trade in goods); moderately predict-able (linked to economic activity, e.g., creditgrowth)GLOBAL FOUNDATIONMobility technologiesExamplesUber, Lyft, DidiTake-UpOpportuLower cost per km stimulates t

56、rip counts (more trips, and more vehicles on roads, i.e., 'empties')ChallengePotential for ride-sharing to decrease trip counts (see ride-pooling analysis below)ExamplesElectric vehicles; hydrogen fuel cell vehiclesTake-upOpportuEmissions abatement incentives may cata- lyse take-up of other technologies through fleet turnover;ChallengeOther policies may have shorter-term traffic impacts, e.g., London's Ultra Low Emission Zone adds a congestion