• 1. CHINA NETCOM BUSINESS PLAN UPDATEDecember 14, 1999 - Beijing
    • 2. TODAY’S OBJECTIVESReview the overall analysis of regulation, competition, and market development Discuss the specific implications for CNC strategy Review the specific draft business models for CNC How we might attack the business, carrier, and IDD/DLD markets What key trade-offs we need to make What are the key success factors and assumptions? Discuss the initial economics of these business models, and of CNC overall Discuss the specific next steps in two key areas: How to finalize and endorse the overall CNC business model How to move forward with the refined financials, organization design, and the plan for implementation
    • 3. CONTENTKey strategic principles Regulatory overview Market overview Competition overview Business models Next steps
    • 4. KEY CNC STRATEGIC PRINCIPLESThe objective of these principles is to provide further clarity in the development and evaluation of the CNC business model These will be revisited and refined as the project progresses Our development of business models will seek to be aligned with the principles Strategic principles and key assumptions will be considered in the following aspects CNC shareholders Backbone business Local access business International gateway business Regulatory strategy Competitive strategy Marketing strategy
    • 5. CNC SHAREHOLDERSPrinciples: Leverage shareholders’ concerted vision in creating a new generation IP broadband communication infrastructure and a profitable, successful company Leverage shareholders’ existing backbone assets, local access assets, research capabilities and local government relationships to secure time-to-market and create competitive advantage Set a model of a new type of “SOE”, as efficient as the best of the FIE/private companies; beat shareholders’ expectations by delivering high return Assumptions: CAS: strong interest in seeing CNC successful and profitable quickly, stands ready for favorable regulatory influence and research support; wants opportunities for networking R&D MOR and SARFT: extensive backbone and access assets and ROW main focus on using MOR for right of way, helping them participate in telecom will not use SARFT local CATV in the near term, but they are a key user of the backbone Leverage CATV’s city backbone MOR will continue to favor CNC over Unicom, provided CNC meets the expectation as a profit center for MOR Shanghai Municipal Gov't: Eager to make Shanghai as China’s test ground for building high-tech infrastructure (e.g. the integration of telecom, CATV, Internet) in particular, testing HFC to deliver broadband internet one issue: SPT and ATT JV; how to handle?
    • 6. BACKBONE BUSINESSPrinciples: Target advanced backbone among 15 key cities in Eastern China; start with 2 cores from shareholders but quickly build own network Use backbone for a variety of wholesale and retail voice and data services Maintain the leading edge IP/packet network, both for technological superiority and to fulfill shareholder mission Maintain the best cost position, using ROWs, purchasing clout, right technology Build reserve capacity/conduit to deter others Assumptions: The costs of construction and ROW will be the majority of new network costs virtually all of the network will use MOR or SARFT right of way Network will be IP or packet in nature Deployment plan will keep costs low, build out quickly, and create good position lay large number of conduits in one time later fill, light up, and color fiber strands lease out conduit/fiber/bandwidth to maximize return (utilization)
    • 7. LOCAL ACCESS BUSINESSPrinciples: Be very focused in local deployment, targeting priority business areas only Emphasize broadband to the customer, using FTTB and LMDS where logical Seek strong local market share, especially in new data services growth areas Differentiate from China Telecom by superior services, quality, responsiveness Build strong local team to enable fast service response Emphasize “end to end” network ownership and management Assumptions: SH, BJ, GZ, and SZ business districts as targets for the near term These account for the vast majority of business telecom demand Need to set specific estimates for timing and sequence of deployment Assume that CNC will have access to key city ROW, such as subway systems Will need specific assumptions about the ease and cost of hooking up buildings Assume that primary emphasis will be on FTTB, but that LMDS can play an important role, especially in initial deployment and in secondary cities Will need specific assumptions about timing of adding secondary cities
    • 8. INTERNATIONAL GATEWAY BUSINESSPrinciples: CNC will be one of few players with a full international license Vital to enable CNC to provide end-to-end services, global data services, and higher margin IDD service HK gateway link may be strategically important Assumptions: HK as one of the key location for international connection a major traffic destination a major relay location International voice remains highly profitable segment in medium term future Expect high growth together with steep price drop in international services Actively plan ahead for joining international sub-oceanic cable consortium
    • 9. REGULATORY STRATEGYPrinciples: Must actively lobby for favorable regulatory decisions, together with shareholders provide regulators with international benchmarking for best practices align CNC objectives with fair competition, and public interests have effective senior management focus on lobbying issues Be careful about committing investment if regulatory issues too uncertain Ensure CNC strategy addresses national economic development priorities Pay careful attention to managing relations with China Telecom Expected WTO in 2000 will imply greater opening of the market in the future Assumptions: At least a 2-3 year window when CNC can continue to enjoy favorable policy treatment, while also working to straighten out regulatory issues at local level Many key regulations remain in grey areas, where CNC can play a role in shaping the policy Will need to make specific assumptions on a number of regulatory issues, and develop several scenarios Assume that CT and Unicom are the only full service competitors, but several niche players WTO will introduce FDI into China’s telecom market by 2002, but infrastructure play remain tightly controlled (still limited competition) until 2004/5 may be opportunity for some form of partnership with foreign telcos
    • 10. COMPETITIVE STRATEGYPrinciples: Focus on the best service quality, supply what customers’ need, avoid competing on price Build both “highways” and “tollbooths” Prepare for rapidly changing industry structure Be careful in positioning toward China Telecom: complementary and addressing unmet needs growing the whole market The only viable alternative to China Telecom for carriers The only true “end to end” network across China, with clear central management Assumptions: Technology leads to continual change in industry structure The decentralized nature and business oriented behaviors of China Telecom enables CNC to partner at the local level High opportunity for CNC to fill China Telecom product/service backlog Unicom will be a threat to start a “price war” Various niche players emerge in later years, more threat than (carrier) opportunity for CNC
    • 11. MARKETING STRATEGYPrinciples: For local access, will target medium and large businesses in targeted buildings For backbone, provide carrier services For international, support other businesses and also offer IDD and refiling, etc. In all areas, emphasize quality, service, end to end, etc. Get a few key customers early; prove ourselves and then build further Assumptions: Carrier and large and medium sized corporations are the key focus especially those in telecom-intensive industries Assume a growing demand for business telecom services will need both high and low growth scenarios Need to validate the willingness of key customers to switch, and their anticipated areas of future demand growth IP phone still the source of revenue in the near future Dial-up ISP (171) may conflict with ISP carrier interests, but could also serve as strategic inroad for future 3G(1) (1) Not in scope of this project
    • 12. CONTENTKey strategic principles Regulatory overview Market overview Competition overview Business models Next steps
    • 13. REGULATORY HIGHLIGHTSOur key assumptions: MII regards CNC’s IP-based license as Full Service license, but lacks clear regulatory documentation. In case of local interpretation difference, MII is willing to clarify on behalf of CNC e.g. ambiguity on local fixed line (CNC number) CNC will be granted International Gateway license by 1Q 2000 CNC IP network protected by the current fixed/mobile interconnect regulation CNC not required to meet specific coverage targets for the near future (2-3 years) The market will be opened up gradually, with FDI increasing: more value added service providers by 2003 more I-Phone providers by 2001 no new Full Service providers till 2004/2005 Account settlement specific for I-Phone will be regulated after the trial stage ends. It will be lower than the RMB0.14/min rate for basic telecom networks on a per call basis Equal access by prefix/pre-select in 1 year; number portability may take another 1-2 years at least CNC will have LMDS spectrum CNC anticipates a well-intentioned regulator, with varied degrees of control over local incumbent practice None of these assumptions are “guaranteed;” CNC must fight aggressively for them
    • 14. KEY REGULATORY ISSUES AND IMPLICATIONSService licenses Full Service license interpretation Interconnect enforcement Account settlement Pricing and rebalancing ROW and access Frequency spectrum allocation and fees Universal Service Obligation Equal Access Numbering and portabilityIGW license important to competitiveness, size of customer base and # of partners MII notification on each instance adds delay to CNC local interconnection Time to market in each city affected, detract CNC resources Future Interconnect economics Future price competitiveness vs. other players (e.g. CT) economics; price competitiveness vs. CT Case-by-case local coordination and negotiation delay local access build-out; may also be expensive to obtain Time to market for local access build-out by LMDS access solutions Subsidizing incumbent for USO affect cost structure of new entrants Unable to reach CT local line customer Medium term implication: on-net voice not targeted for the near 4-5 years3 full service carrier including CNC More licenses for value-added service providers (type II) MII supports CNC IP based license as Full Service License But lack of clear documentation may cause confusion at local level Based on cooperation of carriers Arbitration/settlement process exist but time consuming Mild punishment, law suit as last resort For current IP-phone trial, fees not settled Likely future settlement: LD carrier pays local PSTN operator RMB0.14/min Price floor likely specified for incumbent by regulation No imminent initiative for rebalancing Legally feasible for public telecom carrier (e.g. CNC) Practice will have to coordinate with municipalities and infrastructure building Controlled by MII in co-ordination with PTAs Frequency not likely to be auctioned USO for incumbent with contribution from new entrant 2-3 years before transparent and equitable approach Prefix based solution provided by regulation in 1 year Number as national resources controlled centrally Fee will be collected for occupation of number resource 2-3 years before portability regulationsImpact on CNCHighest/ ImmediateHigh/ Medium TermKey issuesAssumptionImplication to CNC
    • 15. KEY REGULATORY BODIES AND RESPONSIBILITIESOther depts, e.g SETCVarious law making bodiesNPCState CouncilDept. of Radio Frequency AdministrationTelecom Administration BureauOther depts.Provincial Telecom Administration BureauChina Telecom National CompanyMunicipal Level Telecom Administration BureauTelecom Operating Entities: e.g. CT fixed lineProvincial governmentMIIAdministrationOperationsNationalProvincialMunicipalityDrafting, passing of telecom laws and statues, e.g. China Telecoms Law International, National and Inter-Provincial scope license granting Highest level of regulatory enforcement/ arbitration International Gateway administration National level telecom resources control (e.g. numbers, frequency spectrums) Provincial level telecom administration Provincial level license granting Provincial level regulatory enforcement focus on coordination In the process of splitting operation and administration A number of Municipal Level TABs are yet to be created Limited enforcement power Operation and administration not separate Settle interconnect and other disputes When not settled by provincial & municipal coordination, propagate up to provincial level and MII…PTAsPTBSource: Pyramid Research; BCG analysis
    • 16. KEY LICENSE ASPECTS AND IMPLICATION TO CNCFull Service and International Value added services ownership Geographic scope Duration license fees Performance targets TechnologyKey AspectsNumber of licenses tightly controlled: at most 3 in 2003, including CNC CNC full service license expected 1Q 2000 International licenses tightly controlled: at most 4 in 2003, including CNC CNC IGW license expected by EOY1999 More than 4 value-added service licenses possible Control-share foreign ownership not allowed in basic telecom and public data transport services Up to 50% of FDI in value added service providers Intra provincial services under control of provincial TABs No limit Regulation will require license fees Regulation will specify clear service buildout targets Regulations will be technology neutral But will give preference to new technologyAssumptionsCurrent regulation only cover interconnection between 7 types of licensed basic telecom carriers (e.g. NLD, local, wireless, international) With IP-phone license, CNC needs to negotiate with CT local branches city by city CNC IP based bandwidth wholesale service not covered in published regulation, adds delay in local negotiation Number of potential customers for carrier’s carrier services Speed of establishment and coverage of targeted customer for carrier’s carrier service Financial strength of new entrant customers Intra-provincial LMDS service require application in each targeted area Competitor’s ability to lockout key service areas Limited degree of impact on economics Adds planning and local coordination pressure to CNC local access solutions CNC IP/DWDM backbone receive positive influenceImplication to CNC
    • 17. NATURE, SCOPE AND NUMBER OF LICENSESSource: BCG analysisVSAT transport serviceInternationalNLDOn-net voiceIP backboneIP-phone (including international)WirelessPagingVAS (e.g.ISP, VPN)Service natureOwning satellite Maintain satellite channel VSAT base stationsOwnership & operation of international gateways Interconnect with foreign carriers Leasing IDD channels and linesBuilding & operating of national switches and trunk lines Maintaining POP is nationwide Provide basic voice & data national transportOwnership & operation of local access networks Ownership of last mile Provide basic local voice/data transport and access to NLD/IDDBuilding and leasing of broadband IP based backbone transport capacityProvide IP based voice service No requirement in infrastructure building Including international IP based LDBuilding & operation of local BSC network Provide wireless voice services Interconnect with national transport carrierBuilding & operation of Local Paging Networks Interconnect with national transport carrierWith or without building & operation of local network Offer web-hosting, VPN, call center services, etc.ScopeNational coverageCross international point of presence Service national transport carriersCross provincial boarders Have access to international PO connects Service local PSTNsCovers local service areas, (e.g. cities municipalities)Cross provincial boarders Access to international points of connectCovers 12-25 trail cities Access to international points of connect Covers local service areas: cities municipalitiesMajority in local scope 4 national coveragePrimarily in local scopeNo. of future licensesRemain 3 for the next 6 years (CT, Unicom, and CNC) 3 In the next 6 years (CT, Unicom, CNC)3 In the next 6 years (CT, Unicom, CNC)4~5 in 1-2 years5-6 possible in the next 1~2 yearsMultiple licenses possible in the next 4-5 yearsNumerous national scope licenses possible in the next 2 yearsNumerous more licenses (with FDI) possible in the next 2 yearsLicense controlled byMIIMIIMIIMIIMIIMIIMIIPrimarily by provincial governmentProvincial governmentNo. of current licenses13223 (CT, Unicom, CNC)4 Trail stage22 national scope; multiple localUnlimitedBasic telecom servicesRemain 1 for the next 6 years
    • 18. INTERCONNECTION ISSUES AND IMPLICATION TO CNCKey IssuesAssumptionsImplication to CNCObligation clearly specified by regulation for interconnection between 7 types of telecom network licensed operators CNC’s IP broadband network receive equivalent protection MII sets technical specification of interconnect solutions Clear provision on cost and ownership of technical asset From time of written request, interconnect complete within: 2 months if only CO data modification is involved 4 months if capacity expansion needed 7 months if new POI (1) needed Time frame not strictly adhered to, varies from 4-6 months Dispute settled primary via coordination Arbitrated by MII and local government Violation fine is very low (30,000 RMB max per offense) Possible to claim for compensation due to non-cooperation of incumbents (the last resort) RMB0.07/min for PSTN access of each end of call path Settlement for IP based will follow scheme when trial finishes Incumbents provide cabinet room and conduit access In case there’s no excess conduit capacity, incumbent has to provide feasible expansion plan Local practice will vary Incumbent charges rental fee for conduit access Incumbent charges for leased line needed for interconnect Infrastructure and network investment related to interconnect are responsibility of both parties, separated from POIIncumbent obligation to interconnect Interconnect timeframe Enforcement Account settlement Access and collocation Charges and feesIf CNC network is not covered by regulation, local negotiation will leave loophole for incumbent and add delay Incumbent local practice varies, add complexity and delay to project Incumbent can take advantage of timeframe loopholes, add delays at local level Central control over local TAB(2) varies due to varied degree of separation, lead to different degree of just and resolution timeframe Upward propagation to MII time consuming, adding delays to resolution Future economics: up to 30% of revenue paid to PSTN based on 0.30/min IP rate Incumbent can take advantage of capacity constraints to delay interconnect at local level Some incumbent local branches require CNC to pay for investment, ownership goes to incumbent, adds cost to CNC serviceNote 1: Point Of Interconnect 2: Telecommunications Administration Bureau
    • 19. CLEAR GUIDANCE FOR INTERCONNECT ALREADY PUBLISHED, LOCAL PTA OBLIGED TO FOLLOWLocal PTA can be resistant, also its organization not suitable for quick execution The policy decision requesting local PTAs to corporate with interconnection trickles from top down: the central Þ provincial PTA Þ city PTA City PTA before implementation will generally ask for lead time for verification with upper levels. E.g. in some cases new entrant has to wait for extended time period for this process Under the Market Department’s coordination, there are up to 18 independent departments within each city PTA to cooperate for the execution and resource planning of interconnect Each of these department can add delay to the process, lead times add up In case of insufficient resource, the requesting and approval process for expansion / upgrade is even longer Local PTA can easily use resource constraint as an excuse for delay in interconnection … But interconnection is feasible CNC has so far been able to reach interconnect contract at provincial levels Most PTAs in CNC’s target cities displayed timely cooperation In order to ensure quick execution, CNC should Understand and incorporate into planning process the status of incumbent network resources Give lead time in planning and communicating intention to interconnect with the incumbent Build strong communication channel with all levels of PTAs Co-ordinate for good timing across all levels; parallel processing to minimize time needed and clear bottleneckBACKUP
    • 20. THE PROCESS OF LOCAL INTERCONNECT IS GENERALY SMOOTH But There’s Great Difficulty in Some CitiesInterconnect in major cities is smooth e.g. BJ, SH, GZ Great difficulty in some cities e.g. WuhanFast agreement reached From negotiation to implementation usually takes 2-3 weeks Takes 4-6 month to complete negotiation Issues: - local CT branch attempted to curtail CNC by forbidding connections of multiple lines at CNC level Resolution: - resort to regulatory provision to reach IP phone interconnect agreement - push off multiple line issue as future negotiation agendaBACKUP
    • 21. TERMINATION SETTLEMENT MAY REDUCE CNC MARGIN (I)CurrentLikely scenarios for Y2000End userCustomer pays local PTA for PSTN usage e.g. RMB 0.1 every 3 min to PTA No settlement paid by CNC to PTA Customer pays local PTA for PSTN usage e.g. RMB 0.1 every 3 min to PTA No settlement paid by CNC to PTA CNC pays RMB0.14/min to PTA for both origination and terminationCNC charges customer IP-phone carrier rate e.g. RMB 0.3/min paid to CNC (debit from prepaid IP phone card) CNC charges customer IP-phone carrier rate e.g. RMB 0.3/min Per minute IP rate ¥0.44/min paid to CNC otherwise CNC see margin reducingNo settlement No settlement No settlementCNCCNCEnd userPTASwitchSHPSTNSwitchCNC backbone(1)SwitchSwitchBJPSTNPTATransmission (paid by CNC)Fees:12Transmission (paid by CNC)Note 1: Currently leased from CT. Own backbone by July 2000Example Inter- connectionBackup
    • 22. TERMINATION SETTLEMENT MAY REDUCE CNC MARGIN (II) Example: National Long DistanceBackupCNC pays PTACNC pays PTAKeep IP-phone price the sameCustomer pays PTA(1)IP-phone priceCustomer pays PTAIP-phone priceNew IP-phone priceCurrent IP-phone Trial Stage scenario: no settlementCNC Alternative ICNC Alternative IIPossible scenario for 2000: no settlement1RMB/minNote 1: PTA charges customer on 3 minute units, RMB0.18 per 3 min. Calculation for this example uses average per minute rate of RMB0.06, assuming customer calls long enough duration. Source: BJ PTAPossible scenario for 2000: RMB0.14/min settlement rate for both origination and termination ends2Increase IP price to cover settlement cost Customer doesn’t pay PTA but perceive IP-phone price increase CNC maintain current IP-phone revenueKeep IP price same Customer doesn’t pay PTA nor perceive IP-phone price increase CNC pays PTA out of IP-phone income CNC revenue = 54% of current IP-phone revenue0.30.440.360.36CNC retains 100% of IP-phone price as revenueCNC retains 100% of IP-phone price as revenueThe RMB0.14/min settlement rate specified by current regulation is based on traditional fixed line economics. CNC would lobby for new settlement rate specific to IP-phone economics
    • 23. KEY PRICING REGULATION ISSUES AND IMPLICATION TO CNCKey IssuesAssumptionsImplication to CNCRegulation follows cost-based pricing guideline Carriers obliged to publish accounting data assisting pricing regulation Price floor specified for incumbent New entrant has flexible price range around incumbent rate Local, NLD, IDD, wireless, satellite National leased lines Price Bureau within each provincial government sets prices according to local standard of living and inflation rates Local PTAs will have autonomy Since local networks have been built by local PTA investments, local PTA has strong influence on local price levelsGeneral pricing rules Differential Treatment Central control Local control / Geographic practice Usually high initial CNC cost position due to leasing access and settlement Short term uncertainty in price based competitiveness Some flexibility in pricing of CNC product offering Advantageous long term competition capability over incumbent Clarity in pricing is favorable to CNC But limits CNC pricing flexibility Disadvantageous local price based competitive position in some geographies Varied cost structure across geographies complicates product management process, difficult for CNC to have consistent pricing, and hence, image.
    • 24. LEASED LINE COST HIGHEST AMONG INTERCONNECT COSTS Approximately RMB 134K/month to Link up One CityItemCost per month (in RMB1,000)Leased line to local PTA Capacity: 2 M bps Number of channels: 30 Domestic LD transport(1) Total for connecting up one city Leased line to international gateway E1 capacity Total for IDD from one city9 100-150 109 - 159 Þ 134 average 320 454 average(1) Before CNC backbone construction completes, bandwidth for this segment is leased from China TelecomBackup
    • 25. ROW AND ISSUES RELATED TO BULIDING LOCAL LOOPKey IssuesAssumptionsImplication to CNCTheoretically, CNC can assume privileged access to public resources, e.g. bridges and roads There is no specific definition of ROW in most localities In practice, there’s no free access On time fee charged for digging as high as RMB 20K per km reimbursement to local municipalities Labor account for 5-10% of total cost Conduits in bridges and commercial buildings owned by municipal governments Conduits in residential area largely owned by CT Conduit access auctioned by local government when building completes Once auctioned, additional conduits can not be easily added for 4-5 years Utility privilege Access fees and cost Ownership & control SchedulingTheoretical utility privilege does not guarantee local cooperation Local access to conduit in most cities primarily rely on renting from resource holders e.g. CT, utilities etc. Negotiation of rental largely on a case-by-case basis Case-by-case negotiation will delay local loop buildout Conduit access rights have to be purchased back from exiting holders Planning and coordination needed to fit schedule
    • 26. FREQUENCY FOR WIRELESS BB NEED TO BE ALLOCATED Currently Only Narrow Band Wireless Spectrum AssignedKey IssuesAssumptionsImplication to CNCNo clear regulation available and fierce competition expected in frequency bands for LMDS (25-38GHz) 1890-1900MHz and 1960-1980MHz reserved for FDD Mode WLL 1900-1920 reserved for cordless standards e.g. DECT Wireless datacom in 821-825 and 866-870MHz range, NO voice traffic is permitted in this range Frequency management policies implemented by local DTAs, in conjunction with PTAs and PTBs Local frequency allocation is issued after PTA test and MII examination, a process of 6 - 12 months Allocation remains fixed for 3-5 years on average Frequency policies handled by local DTAs, problems resolved locally and appealed to MII 1975-1980MHz band occupied by DGT’s wireless access network in SH, BJ, GZ, SZ, FZ and Xiamen 1890-1895 and 1970-1975MHz occupied by Unicom in CD, TJ and CQ Local wireless access licenses need to be obtained from local PTAs Primarily to be controlled instead of frequency auctions No overall regulation backing onetime solution Roof top access and building access depend on case-by-case negotiations with property ownersFrequency band allocation Administration of frequencies License of service Frequency fees Roof top accessReserved frequency and available LMDS modulation affects viability and quality of CNC access solutions Long circle of frequency allocation process will impact time-to-market for CNC solutions Occupied frequency spectrums in key target cities expect to take long cycle of negotiation and migration Cost of local wireless access solutions local license application cycle delay and variation Will introduce delays in key cities e.g. BJ, SH and GZ Time to market to key target customers
    • 27. OTHER REGULATIONS AND OBLIGATIONSKey IssuesLikely scenarioImplication to CNCLess transparent practices likely to remain for 4 -5 years before equitable policies are made During the interim, incumbent is subsidized for USO by other carriers, reflected in account settlement prices: RMB0.07/min instead of RMB0.044/min New entrant not expected to be obligated … Instead indirectly participate in paying incumbent subsidy via higher prices Expect to be thoroughly addressed over 1-2 years time Regulation provides general end user right to choose LD carrier, access provider to assign default LD carrier Enforcement is responsibility of local TAB within provincial governments Technical solution requirement for incumbent not specified Numbers, as national telecom resources, regulated by MII and local TABs, granted via allocation or auction Regulation will specify coverage, service targets and timeframe Regulation will specify usage fee for assigned numbers Specific regulations and technical solutions not likely in 2-3 years Currently no carrier has portability practicesUniversal Service Obligation Equal Access Numbering plan Number portabilityIndirect USO subsidy affects price-based competitive capability and economics Non-transparent USO contribution practices creates loopholes for incumbent to predate pricing Customers’ willingness to switch carrier Unprotected numbering resources affect viability of future CNC offerings Timeframe requirement adds planning pressure and difficulty to solution Customers’ willingness to switch carrier
    • 28. KEY CNC BUSINESS OBJECTIVES AND CORRESPONDING REGULATORY IMPERATIVESThe Carrier’s Carrier, Enterprise Solutions and Offnet Voice business models require CNC to achieve the following business objectives, and hence actively lobby for favorable regulations3(3)333Carrier’s CarrierEnterprise SolutionsOff-net VoiceFiber local access solutionOff-net voice interconnect solutionLMDS broadband wireless access solutionBusiness ObjectiveImportant to realizing business model
    • 29. OBTAIN FREQUENCY SPECTRUM IN 6 MONTHS TIME FRAME To Ensure LMDS Solution Is Available Before June 2000Frequency Assignment Feasibility Trial Roof Top Access Service License Frequency FeeAdvocate MII to reserve the suitable 25-38GHz frequency range for LMDS applications. Clarity in provision should be obtained in the near future. Advocate MII to sub-allocate suitable LMDS frequency spectrum to carriers in the near future - finish national level planning for number of carriers - provide a short list of alternatives to determine allocation scheme sub-bands within the feasible 25-38 range to planned carriers e.g. 25 - 25.5GHz range reserved for CNC or specific frequency spectrum auction plan - frequency allocation should have nationwide coverage, not geography specific Advocate MII to set specific migration procedure for occupied frequency spectrum with timeframe specified Feasibility for LMDS solution should be technology specific instead of carrier specific(1), MII should publish acceptable solutions based on previous trials to avoid duplication of effort Obtain approval from MII for feasibility trial in one key city (e.g. BJ or SH) for two considerations - control of trail time frame - early customer lock-in Obtain carrier utility right and compensation procedure that has nationwide application from MII, avoid case-by-case cooperation Obtain nationwide applicable service provider license based on spectrum allocation Advocate MII to set frequency usage fee based on number of operational base-stations, multiplied by a per base-station feeKey Issues For The LMDS Business ObjectiveCNC Imperatives deploying LMDS In 15 major citiesLMDSCNC PriorityHighest/ ImmediateHigh/ Medium TermNote 1: Unicom will start trial in ChengDu before EOY1999, expected to finish trial around June 2000.
    • 30. OBTAIN REGULATORY SUPPORT FOR UTILITY PREVILEGE Avoid Case-by-case Local Negotiation to Ensure Quick DeploymentFiber Local AccessCNC PriorityHighest/ ImmediateHigh/ Medium TermUtility privilege & ROW Compensation fees for digging Building Access Scheduling for future access and capacityObtain specific regulation defining CNC’s utility right and obligation Obtain specific regulation from MII for cooperation procedure with municipalities for access - advocate MII to coordinate with Municipal Governments in 5 key cities for trial of cooperation procedure - fix regulation based on trial for application in other cities Obtain regulation governing cooperation timeframe Advocate MII to coordinate with Municipal Governments in key cities for overall compensation produce and fee determination based on indicative indexes (e.g. Real Estate Index held by statistical institutions) Obtain specific regulation based on result of MII coordination, for application in other cities For Municipal Government controlled bridges and commercial buildings - advocate regulation to include specific provisions for obligation of building operator for conduit and equipment room access conduit leasing fee scheme guideline conduit provision timeframe For China Telecom controlled buildings and residential areas, advocate regulation for access similar to that of interconnection Advocate regulation for civil planning authorities to publish info and procedures in - conduit capacity and auctioning and fees - conduit ownership and transferring and fees - regulation should specify guideline for fee determinationKey Issues For The Fiber Local Access Business ObjectiveCNC Imperatives In Building Fiber Access in 5 Key Cities
    • 31. OBTAIN STRONG ENFORCEMENT AND FAIR SETTLEMENT RATE To Ensure Optimal Economics, Also Obtain Equal Access ProvisionsOffnet VoiceCNC PriorityHighest/ ImmediateHigh/ Medium TermInterconnect Enforcement Account Settlement Equal Access Advocate serious punishment clauses that are adhered to by all players and geographies, overseen by independent party Advocate for specific regulation for access pricing based on incremental cost of network, eliminate incumbent predatory charge for interconnection Advocate speed up of restructuring of incumbent’s organization to facilitate competition and cooperation Advocate for new settlement scheme different from the current settlement for fixed/mobile telecom network interconnection - clarify with regulation making body that economics difference between IP and traditional calls - advocate settlement rate lower than the current RMB0.14/min rate specified for traditional network interconnection - use international benchmark and LRIC (Long run incremental rate) Advocate completion of new settlement regulation before current trial stage ends Regulation for Prefixed Based equal access (e.g. same length prefix for all players) should be implemented within short time frame of less than 1 year Advocate regulation for technology investment requirement for all carriers to provide solution to identify caller - enables user to avoid dialing long prefixes - enable enterprise users that demand differential service to have different priority treatmentKey Issues For The Off-net Voice Business ObjectiveCNC Imperatives In Offering Off-net Voice
    • 32. SUMMARY: KEY REGULATORY RELATED CNC IMPERATIVESService licenses Interconnect Enforcement ROW and access Frequency spectrum allocation and fees Universal Service Obligation Termination settlement Equal Access Numbering and portability Granting of International gateway license and clarification of CNC full service license Clarification of equal interconnect privilege for IP broadband network (e.g. CNC’s) Serious punishment clauses that are adhered to by all players and geographies, oversaw by independent party Access pricing based on incremental cost of network, eliminate incumbent predatory charge for interconnection Restructuring of incumbents organization to facilitate competition and cooperation Specific regulation defining new telecom entrants’ utility right and obligation, and cooperation procedure with municipalities for access Overall compensation settlement produce and amount determination based on indicative indexes (e.g. Real Estate Index held by statistical institutions) Specific allocation of 25-38GHz spectrum for LMDS applications Approval of six-month CNC LMDS trial in target city e.g. BJ or SH Specific procedures for occupied frequency spectrum reallocation/migration Guidelines and procedures for local frequency resource allocation and fees Transparent USO contribution regulation, eliminate indirect subsidy by higher price Accounting separation of incumbent along business line and geography Reporting and accounting procedure for Cost Based Settlement Specific regulation governing responsibility and obligation for investment in equal access solutions Public number resource allocation/auction/ transfer proceduresKey issuesCNC imperatives, key areas to advocateCNC PriorityHighest/ ImmediateHigh/ Medium Term
    • 33. CONTENTKey strategic principles Regulatory overview Market overview Competition overview Business models Next steps
    • 34. KEY FINDINGS IN MARKET ASSESSMENT (I)1998 telecom market size estimated at RMB 239 BN, grow at 19% CAGR to RMB 676 BN in 2004 DLD/ILD voice at RMB 67 BN, with 6% CAGR growth to RMB 93 BN in 2004 Total datacom at RMB 24 BN, with 36% CAGR growth to RMB 154 BN in 2004 Offnet voice (IP-phone) represents the major opportunity near-term, market size passing 30% volume by year 2004; but window for CNC closing as competition intensifies and price declines Datacom growth solid mid-long term, fueled by emerging services (conservatively estimated to be over 20% of total datacom by 2004) Geographically, demand highly concentrated in top cities Voice market 199 Bn in year 1998, 64% in top 60 cities Data market 24 Bn in year 1998, 46% in top 4 cities
    • 35. KEY FINDINGS IN MARKET ASSESSMENT (II)Business customer consumes 54% of voice market, 97% of data market Large, medium enterprises are bulk users in the business sector (55% of Toa voice, >95% of total data) Financial services, high tech, professional services, int'l trade, and top general manufacturing firms identified as major telecom Telecom service quality and service levels are the key purchasing criteria among key (large and medium)customers Carrier market at RMB 15 BN in 1998, grow at 10% CAGR to RMB 27 BN in 2004 ISP and mobile the main carrier segments: over 90% of total carrier market
    • 36. CONTENTKey strategic principles Regulatory overview Market overview Demand forecast Geographic segmentation Customer segmentation Competition overview Business models Next steps
    • 37. DEMAND MODELING METHODOLOGYProduct ACURRENT MARKET SIZE By productProduct BProduct C. . .Driver 1Driver 2Driver 3Growth driversForecast using Historical trend Foreign benchmark China-specific adjustmentProduct AProduct BProduct C. . .Driver 1Driver 2Driver 3FUTURE MARKET SIZE By productIterationsIterations/ verify modelShare assumptions by product by year Competitive environment Foreign benchmark Geographic coverageCNC REVENUE by productFill in gapsBreakdown to growth drivers, test formulae/ relationship
    • 38. 2004 TELECOM MARKET PROJECTED TO BE RMB 676 BN Data And Internet Services Are The Growth Engine Going ForwardData(2)Internet(3)Other (4)(RMB Bn)99’-04’ CAGR (%)Voice (1) Data (2) Internet (3) Other (4)55% 10% 0% 35%239TotalVoice (1)(1) Voice services include local, domestic long-distance (on-net & off-net), and international long distance (on-net & off-net, including Hong Kong, Macau and Taiwan) (2) Data services include leased lines (mobile and paging operators, ISPs & business) and access ports (3) Internet services include internet access (broadband & narrowband), web hosting / collocation, and other emerging services (4) Other services include the sale and lease of dark fiber / conduits, mobile, and paging Source: BCG analysis31236344150658767610%25%102%23%19%48% 9% 1% 42%46% 9% 2% 43%41% 10% 3% 45%39% 11% 5% 45%36% 12% 7% 44%34% 13% 10% 43%Telecom Market Size
    • 39. 1998 TELECOM MARKET IS RMB 239 BNMarket Size (RMB Bn)Local (63)DLD (48)ILD (18)Leased lines(0) (23)Internet access (1)Mobile (69)Paging (15)97’ - 99’ CAGR(0) Estimated (1) Voice services include local, domestic long-distance (on-net & off-net), and international long distance (on-net & off-net, including Hong Kong, Macau and Taiwan) (2) Data services include leased lines (mobile and paging operators, ISPs & business) and access ports; leased lines market estimated with CNC’s team inputs (3) Internet services include internet access (broadband & narrowband), web hosting / collocation, and other emerging services (4) Other services include the sale and lease of dark fiber / conduits, mobile, and paging Source: China Telecom annual reports; CNC team inputs; BCG analysis(1)(2) (3)(4)38%23%20%Overall - 27%Backup
    • 40. MARKET FORECASTBackup
    • 41. BACKUP: DRIVERS AND ASSUMPTIONSBackupTo be updated
    • 42. DATA AND INTERNET SERVICES SHOW SUBSTANTIAL POTENTIAL FOR GROWTH: 37% CAGR 1999-2004Leased Lines - ISPLeased Lines - BusinessMarket Size (RMB Bn)24TotalLeased Lines - Mobile99’-04’ CAGRAccess PortsInternet AccessWeb Hosting / CollocationOther Emerging ServicesLeased Lines - Paging324160831141544%41%1%39%152%69%136%(1)(1) 01’-04’ CAGR (2) 00’-04’ CAGR Source: CNC team inputs; foreign benchmarks; BCG analysis84%(2)37%
    • 43. EMERGING DATA SERVICES WILL ACCOUNT FOR MORE THAN 20% OF THE TOTAL DATA SERVICES MARKET BY 2004Existing data services marketEmerging data services marketTotal data services marketRMB BnYearRMB BnYearRMB BnYear99’-04’ CAGREmergingExisting99’-04’ CAGR99’-04’ CAGR31%238%37%Source: CNC team inputs; foreign benchmarks; BCG analysis
    • 44. THE EMERGING SERVICES MARKET IS EXPERIENCING TREMENDOUS GROWTHMarket size (RMB Bn)Estimated emerging data services Portal Broadband - content Broadband - network applications IP VPN Web Hosting /Collocation(1)(1) Listed separately in CNC market analysis Source: BCG estimatesExamplesBackup
    • 45. Emerging servicesOpportunities and challenges for CNCExpected market size in 2004 (RMB BN)Web hosting / collocation(1)Fit with CNC’s strategy as a leading backbone provider Build relationship with large businesses Need to develop specific capabilities, e.g. e-commerce applications, security May lead to future application hosting opportunities5PortalB2B represents huge business values; opportunities to build relationship with large businesses; but first-mover advantage is important, competition may be fierce and market will eventually be dominated by a few big players B2C portals’ target are end consumers; may not be the most profitable option for CNC16 (B2B: 5 B2C: 11)Broadband contentContent development not our priority May form strategic partnerships with content providers; improve customer retention2Total emerging services market expected to reach RMB 32B by 2004, or more than 20% of the total data / Internet services marketIP VPNFit with CNC’s strategy as a leading backbone provider Part of a complete product bundle (including Internet access, web hosting); help to build relationship and reduce churn Need to address concerns including network security, end-to-end service1(1) Listed as “existing’ service in CNC market analysis Source: international benchmarks; Morgan Stanley; BCG analysisBroadband network applications (e.g. video telephony)Fit with CNC’s strategy as a leading backbone provider Need to address issues including network performance and QoS-level guarantee1A NUMBER OF EMERGING SERVICES WILL BECOME ATTRACTIVE OPTIONS OF PRODUCT OFFERINGS FOR CNCBackup
    • 46. CARRIER: EXPECTED SHARESCompetitorKey strengths / weaknesses20002004Expected shareCNCIP/DWDM backbone, bandwidth and cost leadership Strike for excellent service and value for money Weakness: late-starter<1%7% mobile bb 26% addr. ISP 15% China Telecom Voice services Satellite communicationsExtensive existing backbone Existing local access, captive demand Weakness: legacy system makes it difficult to cut down costs80%65%Existing coverage / user base Weakness: bandwidth constraint5%5%UnicomEarly mover to market Nearly complete coverage in 150 cities Captive mobile demand Weakness: In-fighting board, acrimonious relationship with CT10%20%JitongSolution primarily based on VSAT coverage, bandwidth constraint Existing bandwidth primarily occupied by own IP phone service Questionable excess bandwidth for resale5%3%Rationale / BenchmarkMobile operators A high percentage of the leased line demand in local lines (80%), but CNC will have a relatively low share in local lines (~1%). For backbone, expect by 2004 about 40% will be IP-based and CNC will have a 65% share (as a dominant IP-backbone provider) ISPs Expect by 2004 about 70% of the market will be addressable (instead of being captive by China Telecom). Since there will probably be at least 5 to 6 backbone operators at that time, we expect about a 15% share out of this. Paging operators Not CNC prioritySource: CNC team inputs; BCG analysis
    • 47. LARGE / MEDIUM BUSINESSES: EXPECTED SHARESCompetitorKey strengths / weaknesses20002004Expected shareCNCMultiple alternatives to ensure rollout/coverage in local access to targeted customers Able to capitalize on different local practice at CT , and partnership with MOR Need to build up capability & share quickly1%15%China Telecom Voice services Satellite communicationsIncumbent: existing customer base Extensive coverage in local loop Weakness: service back log, bureaucracy drags service quality Legacy system in backbone, hinders ability to offer lower price/high bandwidth services74%55%Coverage/bandwidth limited Weakness: bandwidth constraint in later years5%5%UnicomEarly to market, built local access in TJ/CQ/SC Weakness: continuous management problems10%15%JitongAggressively going after market from current ISP base Weakness: bandwidth constraint (VSAT-based solution)10%10%Rationale / BenchmarkIn 2000, CNC coverage and time-to-market limits share achievable By 2004, expect CNC coverage of over 80% of the business demand, with overall share of around 20% in the business leased lines and Internet access markets. Also expect to aggressively go into the web hosting/ coll, reaching 15% share by 2004. For other emerging data and Internet services, expect to get a 5% share due to the nature of more fragmented competition.Source: CNC team inputs; BCG analysis
    • 48. OFF-NET VOICE: EXPECTED SHARESCompetitorKey strengths / weaknessesCNCLow cost based on new technology Innovative & motivated management team Favorable regulatory backings Weakness: late to market, IP voice the only consumer oriented product (lacks synergy)China Telecom Voice servicesIncumbent / early entrant Existing coverage No interconnect hassle Weakness: - higher cost position based on existing infrastructure - cannibalize circuit voice servicesUnicomEarly to market Synergy with mobile business Weaknesses: - initial stage unclear management focus - Acrimonious relation with CTJitongQuick and successful launch in 1999 ,early to market Aggressive marketing, high discount to distributors Weakness: no long-term backbone solution - satellite network may affect voice quality (latency)20002004Expected share15%28%55%40%15%20%15%12%Rationale / BenchmarkIn 2000, CNC deployment plan will not be as aggressive as our competitors; only about 60% of the IP voice market will be addressable. CNC will have about a 25% market share in these (addressable) areas. By 2004, expect to cover all key areas. But CNC share will dropp from the peak of 33% in 2002 due to more competitors entering. Benchmark: In Hong Kong, 3 new entrants with call-back: 20-30% share in 2 years; i.e., incumbent <=70%; half a year after ISR, incumbent has <50% shareSource: CNC team inputs; BCG analysis
    • 49. SHARE ASSUMPTIONSBackup
    • 50. CNC REVENUE PROJECTIONBackup
    • 51. 20002004CNC shareCNC shareMarket growthMarket growthMarket size RMB 10BRELATIVE ATTRACTIVENESS OF DIFFERENT PRODUCT CATEGORIES CHANGES IN THE NEXT FEW YEARS80%40%0%0%15%30%0%15%30%3000%50%0%300%Emerging dataIP voiceExisting datacomOn-net voiceCarrierOn-net voiceCarrierEmerging dataExisting datacomIP voice
    • 52. CONTENTKey strategic principles Regulatory overview Market overview Demand forecast Geographic segmentation Customer segmentation Competition overview Business models Next steps
    • 53. CHINA TELECOM MARKET CONCENTRATED IN TOP CITIES Top 60 Cities Account For 64% of the Voice & Data Market1998 Accumulated Revenue (RMB Bn )(1) Excluding paging of RMB 14 BN Source: Pyramid Report; China Statistics Yearbook; BCG analysisBJGZSZNext 11 citiesNext 45 cities223SHChina Total(1)050100150200250010203040506070
    • 54. IP / OFF-NET VOICE REPRESENTS SUBSTANTIAL OPPORTUNITY IN NEAR-TO-MEDIUM TERMIP Voice Market Size (RMB Bn)CNC’s Expected Share (%)Estimated CNC Revenue and Profit (RMB Bn)RevenueLikely Profit(1)YearYearYearMarketShareRevenue / ProfitCNC’s share drops after 2002 when the IP voice market becomes more competitive(1) Estimated using a profit margin of 30% in 1999 down to 10% in 2008; based on foreign benchmarks Source: CNC team inputs; BCG analysisProfit margin drops as competition intensifies and price decreasesIP voice continues to replace traditional circuit voice in DLD & ILD
    • 55. FOR VOICE MARKET: IDD/MOBILE CONSUMPTION MAJOR SOURCE OF REVENUE IN TOP CITIESMobileSHBJGZSZOther 11 cities45 other citiesRest of ChinaSource: China Telecom; BCG estimate36%1998 voice market size (RMB Bn)26%11%5%7%7%8%LocalDLDIDD3.12.31.6719.4302.92.42.30.4 0.34.413.8233.219.32.715.11.87.82.767.42.47.62.85.42.86.8BackupTotal: 199
    • 56. GEOGRAPHICALLY EVEN MORE CONCENTRATED FOR DATACOM 46% Revenue in Top 4 Cities1998 Data Communications Revenue (RMB Bn )Source: Pyramid Report; China Statistics Yearbook; BCG analysisBackupNext 11 citiesNext 45 citiesTop 4 citiesChina Total24600+
    • 57. GEOGRAPHIC SEGMENTATIONTop 4Next 11(1)45 others(2)Beijing, Shanghai, Guangzhou, ShenzhenFuzhou, Xiamen, Shijiazhuang, Zhengzhou, Wuhan, Changsha, Nanjing, Xuzhou, Jinan, Tianjin, Hangzhou Hefei, Chongqing, Foshan, Jiangmen, Shantou, lanzhou, Guilin, Quanzhou Nanning, Haikou, Baoding, Daqing, Harbin, Qiqihar, Guiyang, Handan, Tangshan, Luoyang, Xiangfan, Changchun, Jilin, Nantong, Suzhou, Nanyang, Changzhou, Wuxi, Yangzhou, Nanchang, Anshan, Dalian, Shenyang, Chengdu, Jining, Taizhou, Qingdao, Weifang, Weihai, Yantai, Taiyuan, Xian, Urumqi, Kunming, Ningbo, Shaoxing, WenzhouBackupSample list: actual names of city may differ(1) According to CNC plan (on CNC backbone) (2) Ranked by total GDP and GDP/cap, expected to correlate with telecom spending level
    • 58. TOP CITIES BOAST HIGHEST TELECOM SPENDING PER CAPITA 1998 Annual Telecom Spending Per CapitaRMB /capSH BJ GZ SZ (1) 11 cities 45 citiesMobile525 444 1135 6721 117 60BackupIDDDLDLocalROC22233 256 397 2351 30 124176 200 348(2) 348(2) 66 5526240 187 239(2) 239(2) 105 7734(1) Large non-local/business traveler population base (2) GZ, SZ estimated under same assumption
    • 59. CONTENTKey strategic principles Regulatory overview Market overview Demand forecast Geographic segmentation Customer segmentation Competition overview Business models Next steps
    • 60. BUSINESS CUSTOMERS REPRESENT 54% OF VOICE MARKET, 97% OF DATA MARKETLocal1998 Telecom revenue (RMB Bn )ResidentialBusinessDLDIDDMobilePagingLeased linesInternetVoiceData35282227315323778230.80.26349186915231Source: China Telecom; Pyramid researchTotal 237
    • 61. 4 BUSINESS CUSTOMER TIERS ARE IDENTIFIED ACCORDING TO MONTHLY TELECOM SPENDINGTier rankingTypical company profileCompany examplesTypical monthly telecom spending per office (RMB K)1Top high tech firm Top international trade Top financial institutes Top news agency and ministry info centerEricsson, Motorola, Sinochem, Bank of China, XingHua News Agent500-10002Financial services companies (regional HO, foreign banks) FIE/JV professional services companies (e.g. consulting firms, market research firms, law firms) Medium size international trade firms and advanced general manufacturing companies (e.g. active application of ERP to improve efficiency, several locations in China) ABN AMRO, Pudong Development Bank, BCG, Xi'an Janssen, Tung Tai Food100-5003Local high tech firms (usu. 100-500 employees) JV professional services companies (usu. Around 50-100 employees) General manufacturing firms ( typically. 100-500 employees) SH Hua Teng, Nestle, Zhen Fong Electronics, Gillette 10-100Small companies42-10
    • 62. EACH TIER HAS DISTINCT NEEDS AND EXPECTATIONSTier RankingMost Important telecom needs Increasing bandwidth Increasing network coverage Voice/data system integration Improving the ability to using the internet for eCommerce Increasing bandwidth Increasing network coverage Introducing new services to improve efficiency Cost management Increasing bandwidth Handling growing demand for internet access Installing and managing LAN/WAN Cost management Increasing bandwidth Products/services needed Video conferencing lower cost VPN better link of wireless and fixed line phones(1) VPN Video conferencing High speed data access to the office from home Redundancy, to reduce dependence on China Telecom( esp. FS firms) Lower cost, more reliable access to the foreign countries Easy adjustment of telecom usage High speed internet access Telecom cost tracking system High speed internet accessService Level expectations Extensive access network coverage One stop, flexible services Detailed service level agreement (e.g. monthly downtime limit, on-site support time) More responsive support services (e.g. quick fix of leased line problems) Less lead time/ response time for telecom services (e.g. adding lines or bandwidth) Less down time for reliable email/data transmission1 2 3 4 (1) Product/services CNC might not provide in near term
    • 63. INDUSTRY TYPE AND CONSUMER TIER ARE HIGHLY CORRELATEDFinancial Services General Manufacturing High Tech Professional Services International Trade Other industry local voice, DLD local data, international data, most of which are real time Potential needs from ecommerce business involvement local voice, DLD local data for management systems, international leased line (for FIE) Local voice, DLD, IDD (FIE) local data transmission between HO and branches, international leased line (for FIE) Extranet, VPN local voice, DLD, IDD (FIE) local data, international data(FIE) local voice, DLD, IDD local data between HO and branches Internet, Extranet for ecommerce local voice, (DLD) dial up internet no dedicated leased lineTier 1 Tier 2 Tier 2 (most FIE) Tier 3 (other advanced) Tier 1,2 (top FIE) Tier 3 (local/medium) Tier 2 (FIE) Tier 3 (local/medium) Tier 1 (largest) Tier 2 Tier 3 or belowIndustry TypesMain Telecom NeedsTypical Tier Ranking
    • 64. LARGE AND MEDIUM ENTERPRISES REPRESENT 62% OF TELECOM SPENDING IN BUSINESS SECTOR Over 95% For Data Services48431616Small(1) (Tier 4 and below)Voice (107)Data (23)1998 Revenue (RMB Bn )(1) Size definition according to China Statistics Yearbook Source: China Telecom, Pyramid Research, China Statistics Consultants Co., China Statistics Yearbook, BCG estimate61Large(1) (Tier 1,2,3)Medium(1) (Tier 3, Tier 4)4,6002042# of companies (‘000)Total: 130
    • 65. EXAMPLE: KEY INDUSTRIES AND COMPANY PROFILECompany nameABN amro bankGangao SecurityPudong development bankXi’an JanssenNestleAdvance ElectricsEricsson (SH)Ericsson (BJ)Hua TengBCGHorizonOverseas tradeCompany nameSHBJBJSinochemEast NetGubei Real EstateBJSHSHSHSHSHSHSHSHSHSHSHTier ranking2222332132421N.A.N.A.Telecom spending RMB per m2/month1472506756202332515033677050N.A.N.A.N.A.Telecom spending RMB per/cap/month 31432000100055640175024024004172000538167N.A.N.A.N.A.OtherTradePSHTGMFSBackupInfo. high wayBJN.A.N.A.N.A.
    • 66. A TYPICAL BUSINESS BUILDING GENERATES ~4.5 MM RMB TELECOM REVENUE PER MONTHBackupTypical Business Building# Company (by type)1~2 Tier 1 3~5 Tier 2 50-60 Tier 3 ~30 Tier 45000m2 15000m2 4000m2 750 250 50 4Typical Spending Per Month (K RMB)1,500 3,000 120===(1) Gross Floor Space Source: BCG interview, BCG analysisGFS(1): 30,000m2 Occupancy: 80%Average in SH: 3,000 (RMB K)Total: 4,620 (RMB K)either x or x x x
    • 67. INTERVIEWS CONDUCTED TO ASSESS CUSTOMER NEEDS AND TRENDS, TO COMPLEMENT AND VERIFY SECONDARY DATAIdentify key customer needs, ensuring CNC supplies what our customers need Telecom needs in the coming 1 to 3 years Expected growth in telecom spending Verify assumptions on key customer segments Typical customers’ profile (industry type, office size, # of employees, avg. monthly spending, FIE/SOE, etc.) Typical purchasing decision process/patterns Test possibility and criteria for customers switching to CNC, hence CNC addressable market and plausible share in a building Main dissatisfactions toward current telecom service providers What will make customers be willing to switch? What are their key concerns regarding switching? Estimate demand for typical business buildings, to enable local access economic modeling
    • 68. IN DEPTH INTERVIEWS REVEAL SIGNIFICANT OPPORTUNITY FOR NEW ENTRANT TELCOSLong process for CT to meet customer requirements, which is fast-growing On average, expecting 5-10%per account yearly growth in the expenditure on data services; key growth driver in data usage penetration in enterprises Demand for data services driven by the increasing needs for management tools (e.g. ERP), Internet access, eCommerce capability, etc. Bandwidth usage for most large companies still relatively low A few leased lines; mostly 64-128kbps, only a few at 512kbps Would welcome higher bandwidth at reasonable cost Major discontent is in response time and service quality Long lead time to install service, e.g. more than half a year for an international leased line Never inform consumers of service down-time, take an ‘intolerably long time’ to fix Big users don’t feel favorably treated when their inquiries processed with residential and small business users’ through a common service number 112 Hence price is not the primary factor for switching to CNC Quantified service level guarantee and excellent service quality are the key switching factors for big and medium size business usersBackup
    • 69. SERVICE QUALITY & SERVICE LEVEL ARE THE KEY SWITCHING FACTORS Especially for Tier I & 2 CustomersSwitching CriteriaProduct/ Service quality Service level Price Product offering Relative maturity as telecom customer Tier 1Tier 2Tier 3Tier 4Source: BCG interviewImpact on switching decision weak strong
    • 70. CUSTOMER EXPECTING GROWTH IN DATA SERVICES EXPENDITURE“We expect 5-10% increase each year in data services spending mainly because of the using of ERP and the pursuit of ecommerce opportunities.” - Head of Communications, Sinochem “Many things will drive up our telecom needs rapidly in the near future: the growth of the company size, the increasing traffic between HQ and branches, the increasing needs for product testing and development, and the development of ecommerce capability. I think our company’s data services expenditure will continue to grow at 5-10% per year.” - Head of Communications &Network, Shanghai Hua Teng “We are seeing our trader group expanding at very fast pace. Our telecom spending, especially data services spending, will definitely go up fast (at least more than 15% per year) if our business continues to grow.” - Branch Manager, GangAo Securities “Our current MIS system requires more frequent data transmission between sub-branches, branches and the HQ. We are replacing all dial-up with leased line and adding bandwidth. Telecom spending will thus increase.” - IT DIV. Mgr., Pudong Development BankBackup
    • 71. CUSTOMERS CAN BE WON OVER BY MEETING GROWTH NEED China Telecom’s Slow Pace and Long Cycle is Major Area of DiscontentISPs seeking alternatives to CTBusiness Clients also frustrated“ As we build up strong momentum in customer base growth, having the right capacity at the right time is critical. China Telecom, with their own network development plans, tends to starve us on capacity or to force us to pay in advance for excess capacity.” - Founder, EastNet “We want to build our own national brand, but the fact that it takes China Telecom two months every time we need an extra line makes us very difficult to have our own customers. We want another operator who can get us leased lines fast.” - Manager, Div. Of Customer Develop, Information Highway (IHW)“We’re are fretting over our customers’ complaint about increasing difficulty to access time critical information and perform trades. Our request for bandwidth upgrade were submitted to China Telecom months ago, nothing’s happening.” - Branch Manager, GangAo Securities “We need help to easily adjust our usage of telecom services, such as adding lines or bandwidth. CT’s not very responsive to our requests.” -Head of Communications &Network, Shanghai Hua TengBackup
    • 72. RESPONSE TIME AND MEASURABLE SERVICE QUALITY ARE DIFFERENTIATION FACTORSCustomers need quantifiable metrics“We are not satisfied if the carrier simply gives us qualitative reports about downtime in general - We are interested in quantifiable, detailed MTBF(Mean Time Between Failure) and recovery time reports as measurement of service quality. So far China Telecom failed to deliver on it.” - IT Manager, Ericsson “When our Leased Line link is down, we need an expert at the other side to answer our specific questions. The 112 Customer Service number China Telecom provides for both residential and business customers merely adds frustration and delay.” - IT Manager, Ericsson Beijing HQ “As one of the biggest telecom users, we will certainly require favorable treatments, like bandwidth guarantee and priority in IP phone processing.” - IT Manager, SinoChemKey customers require specialized careBackup
    • 73. TELECOM SERVICE QUALITY AND SERVICE LEVELS ARE THE KEY PURCHASING CRITERIA AMONG KEY CUSTOMERSOne-stop, end-to-end solution desiredPrice is important, but not predominant“Multiple carriers had approached us with various service propositions. Our answer is: We’re committed to sign up if it can give us one contact point for a total solution and service agreement.” - IT Manager, Ericsson “P&G has clear initiatives in building world wide VPN and driving eCommerce growth. When it comes to data infrastructure, what intolerable is the delay in contract arrangement for each of the building blocks.” - Manager, Data & Computing Services, P&G “Lower price may appear attractive. But we won’t consider using services any less reliable than CT’s services - even if they are of much lower price.” - IT Mgr, Pudong Development Bank “We are definitely more concerned with the overall telecom service performance and reliability. I would say price is not the primary concern for any financial institutes.” -MIS Manager, ABN AMRO Bank “As a trade company, we welcome lower price service providers. But we are not going to make compromise on service levels and qualities. International call over IP is much cheaper than IDD, but is inconvenient to use and the current voice quality is unsatisfactory.” - IT personnel, Overseas TradeBackup
    • 74. ISP AND MOBILE ARE THE MAJOR CARRIER SEGMENTS(1) Customers include MOR, CATV and other large volume users Source: Pyramid research; BCG analysisDark Fiber, Conduits & Access Ports(1) Paging ISP MobileN/A 0.2% 46% 5%CAGR (98-05)Carriers’ market size (RMB Bn )142080.60.60.61.1010203019982005 (projected)0
    • 75. BOTH CHALLENGES AND OPPORTUNITIES FOR CNC TO ADDRESS CARRIERS’ NEEDSISPs looking for alternatives to CT ... More reliable services (backbone, international gateway) Broader bandwidth at a reasonable price Less lead time for adjusting info needs (e.g. adding line/bandwidth) … also worry about carriers compete directly with ISPs CNC best positioned to address such needs But need to address ISP’s concern of CNC’s 171 as competitionISPMobile carriers still concerned with IP voice quality More lead time expected before toll quality will be achieved and carriers will be willing to lease IP backbone Tradition link of Mobile Operators with CT weakening over time Opportunity for CNC, but CNC needs to provide value-for-moneyMobile
    • 76. CONTENTKey strategic principles Regulatory overview Market overview Competition overview Business models Next steps
    • 77. COMPETITIVE SUMMARYRegulatory response to WTO development will lead to more open market in general More emerging (Type II) players in the Value Added Service areas Emerging players more likely to become CNC customers than competitors China Telecom’s split along service lines will create cooperation opportunities to CNC Geographical break-up of China Telecom fixed line services uncertain, opportunity of cooperation for CNC Break-up may not be complete, as former China Telecom constituents still hold commanding share at least till year 2005 Unicom might emerge as CNC’s arch-rival if internal conflict ends Jitong has built quick customer response capability based on VSAT platform, but would largely remain as a minor competitive force due to bandwidth constraint Not excluding cooperation opportunity Therefore, Initial assessment asserts CNC to target about 20%+ long term share in the Carrier, Large/Medium Enterprise Solutions markets
    • 78. China UnicomCOMPETITIVE LANDSCAPE TODAY: 1999 (I)ResidentialBusinessSegmentsFull-service providersValue Added Service ProvidersCellular Internet access Broadband applications On-net voice Off-net voice IP phone On-net voice Off-net voice Internet access Leased line VPN / other broadband applicationsMultiple ISP’sLocal MSOsSatelliteCNCChina UnicomPagingChina TelecomSatellite Communications CorpMobile Communications CorpFormer China TelecomFormer CT pagingHold license / Deploy ?JitongJitongJitongCNCMultiple ISPsCNC
    • 79. COMPETITIVE LANDSCAPE TODAY: 1999 (II) Competitors In Telecom MarketsTargeted offeringsCarrier’s CarrierSwitchedVoiceAllDLDILDIPLarge/medium business customersChina TelecomChina Telecom SatelliteUnicomJitongCompetitor33333(3)(3)3TJ, CQ, Sichuan333333CNC3333
    • 80. 2 SCENARIOS WILL EMERGE IN COMPETITIVE LANDSCAPE Depend on Interaction of Pro-competitive regulation and FDIPro-competitive Regulation ImplementationActiveness of FDI Number of FDI backed players Intensity of FDI fundingLandscape today Few basic carriers Limited number of type II carriers Low FDI activity Funding depend on internal resourcesScenario 2: Open Competition In the VAS area Controlled number of basic carriers Numerous active FDI backed regional/Type II playersOpen CompetitionClosed CompetitionLatentActiveLikely in 2004Less likely in 2004 Scenario 1: Open in all fronts Multiple CLECs and IXCs Numerous regional/Type II players up to 50% FDI
    • 81. MOST LIKELY COMPETITIVE LANDSCAPE IN 2003/4ResidentialBusinessSegmentsFull-service providersValue Added Service ProvidersCellular Internet access Broadband applications On-net voice Off-net voice IP phone On-net voice Off-net voice Internet access Leased line VPN / other broadband applicationsMultiple ISP’sProvincial MSOsSatelliteCNCChina UnicomPagingCT broken into regions Six regional operating companies for NLD HQ controls international and data Satellite Communications CorpMobile Communications CorpFormer China TelecomFormer CT pagingJitongJitongJitongMultiple ISR playersMultiple I-Phone playersMultiple ISR playersMultiple I-Phone playersUtilitiesCATVsMultiple Regional PlayersMultiple VANsMultiple ISP’sMultiple PlayersCNC4-5 more national playersCATVs
    • 82. LIKELY SCENARIO MEANS MORE OPPORTUNITY THAN THREAT1-2 type II new entrants in each locality likely allowed by regulation CNC low cost IP based backbone provide national/international access and transport Initial market research point to CNC’s aspired one contact services attractive to type II playersNew entrant ISPs, MSOs and Wireless Carriers as potential customerLocal autonomy and separate P&L lead to flexibility in partnership CT local bandwidth requirement not easily accommodated by national plan due to separate investment ownershipPaging, Wireless, ISP services separated from incumbent as potential customerBecoming more responsive to key customer needs Starting to form allegiances with new entrants (e.g. CT with Unicom in wireless) Increasingly focus on building datacoms capabilities in anticipation of demand growth Building capability to bring quick solutions to key customers (e.g. JT with VSAT)Incumbents building flexibilityOpportunityThreat
    • 83. CHINA TELECOM (I)99%Recent split along service lines, national paging service folded into UnicomBut each former China Telecom constituent still holds dominating share in respective market *:China Mobile Communications Corp.China Telecom (Voice and Data Services)China Satellite Communications Corp.National Long Distance ServicesInternational Long Distance ServicesBroad- band Data ServicesISP service (163, 169 China-Net)National / International LevelRegional LevelMunicipality / City LevelProvincial LevelLocal Voice and Data ServicesLikely to further break down into 6 regional companiesLikely to remain as one national entity VSAT based backbones and local base stationsLikely to remain as one national entity Holding local GSM networks and frequenciesLikely to remain as one national entityLikely to further break down into 6 regional companies~100%~100%90%80%+95%100%* As of 3Q 1999
    • 84. CHINA TELECOM (II)The fixed line services of China Telecom will likely to be broken up into 6 regional companiesGenerating significant competitive impact to CNCPotential threat: Separate P&L improve ability to have business focus and cohesive control FDI cooperation become more feasible Resulting in overall stronger competitor CNC expect 4-6 years’ window of opportunity to establish and preempt competitive disadvantage CT’s huge amount of complexity in control and asset will delay separation process CT’s management in-flights need to be resolved before significant progress made Central regulatory maintain tight control on FDI in basic telecom services Expect only some baby CT could establish successful coop with strong FDIChina TelecomWest China TelecomSouthwest China TelecomNorth China TelecomEast China TelecomSouth China TelecomCentral China TelecomBroken upLocal networks will belong to the region Uncertain if there will be a separate national long distance carrier Further breakdown along geographical lines possible
    • 85. CHINA TELECOM (III)Key strength Unmatched existing infrastructure: - international gateway in BJ, SH, GZ and SZ; - 18 International SOC(1); “8X8” national FO backbone of 173,000 km; - ChinaDDN, ChinaPAC, ChinaFRN, ChinaX.25, ChinaNET comprise national datacom coverage Holds critical resources in local networks and largest business and residential customer base Key weakness Management team and organization not competition ready Slow to provide total solution in product offering and support, due to multilevel/department control Competitive move Reducing tariffs (latest in 3/99) and installation fees (latest reduction up to 50%), led to subscriber base boom; Offer 50% - 70% discount on DLD, eliminated administrative fees Increasingly try to raise value of each subscriber, e.g. installation of 2nd phone line, VAS Aggressively investment in datacom: 1998-1999 growth 24% Improving customer service to key accounts (e.g. Ericsson) Threat / Opportunities presented to CNC Before complete and transparent regulatory implementation, strong ties with local telecom administration will remain powerful leverage against new entrants Currently holds ~100% business customer base, network effect and CT’s initiative to improved service will add difficulty for CNC to persuade switch After CT split, mobile and ISP service have potential to become customers or partners to CNC Uncertainty around regional breakup of fixed-line services, if split may present cooperation opportunities(1) Submarine Optical Cables, CT SOC connects to Japan, Korea, Europe (including land cable) and US
    • 86. UNICOMEntrenched new player The only other GSM carrier besides CT, the only CDMA carrier. Holds ~5% share wireless market Licensed to operate local service in TJ, CQ and Sichuan province, TJ service rolled out with 10’000 users Inherited CT’s profitable paging service with RMB14bn in asset value, and 41M subscribers Received MOR’s telecom assets worth RMB10bn, with 42,000km SDH FO links covering 500 L/M cities Key strength Backbone coverage to all major cities nationwide, with self-built intra and inter provincial FO backbones covering 25 provincial capitals, plus the MOR backbone (one of the largest reseller of capacity) Own access networks in some key cities (e.g Tianjin) enables short time-to-market Key weakness Lack of cohesiveness in management and organization Lack of focus in business strategy and product offering Competitive move Sets priority on Datacoms and Mobile network development, and long distance services Won approval to offer direct Internet access and VPN services to business customers, covering 100 major cities and most of China’s local nets Recently started wireless local loop trial in CD, targeting commercial service by mid-2000, also begin LMDS trial by EOY 1999 Threat / Opportunities presented to CNC Emerge as strong competitor to CNC’s targeted business customer base Uncertainties around possibility for CNC to resell capacity to its mobile services
    • 87. JITONGPrimarily an ISP, with aggressive targets in business market Holds one of the two existing Internet backbones National VSAT covering 14 provincial capitals Key strength VSAT based solution provides flexibility and quick response to ensure rollout/coverage in local access to targeted customers International gateway presence in BJ, SH, and SZ Holds large business and residential datacom user base Key weakness Bandwidth with existing primarily VSAT based backbone primarily occupied by its ISP service Weak management team, lacks clear vision in network building and service offering Competitive move Aggressively targeting business customers in key geographies Aggressively seeking funds for network buildout, planning IPO in early 2000 Threat / Opportunities presented to CNC Has aggressive plan targeting business customers in key cities, e.g. SH, BJ Cooperation opportunity likely to exist for CNC IP/DWDM backbone to offer transport capacity
    • 88. HIGH POSSIBILITY FOR COOPERATION AT LOCAL LEVELCompetitorChina Telecom fixed line ServicesChina Telecom Wireless Unicom (include former CT paging services)MethodProbabilityMethodCNC’s shareRationale/commentsPotential partnershipPotential as carrier customerInterconnect Leased line for local access100% for VoIP 1999-2000: 50% of targeted cities 2001-2004: 60% of targeted citiesInter-city traffic diverted to CNC backboneInter-city traffic through CNC backbone as backup20% of total CT wireless traffic goes through CNC backboneAt most 5% of inter-city traffic goes through CNC backboneRequired by government, but expect some delay in implementationCT backbone is assumed to be CT wireless’ primary carrier CNC IP/DWDM back bone as complementary carrier with lower costCNC only complements areas where Unicom cannot cover CNC as low cost backup solutionJitongCNC backbone provide back-up bandwidth20% of total Jitong trafficJitong’s primarily VSAT based backbone will not meet expansion needCATVHFC as local access partner in major citiesOver 50% possibility in CNC target cities, if fiber availablecarrier customer and fiber leasing 2000-2002: 5 key cities 2003-2004: 15 major provincial capitalsCNC fiber only complements areas SARFT lacks coverage ISPscarrier customer1999-2000 10% 2000-2004 40%CNC high speed back bone is effective to drive up end user demand
    • 89. CARRIER: EXPECTED SHARESCompetitorKey strengths / weaknesses20002004Expected shareCNCIP/DWDM backbone, bandwidth and cost leadership Strike for excellent service and value for money Weakness: late-starter<1%7% mobile bb 26% addr. ISP 15% China Telecom Voice services Satellite communicationsExtensive existing backbone Existing local access, captive demand Weakness: legacy system makes it difficult to cut down costs80%65%Existing coverage / user base Weakness: bandwidth constraint5%5%UnicomEarly mover to market Nearly complete coverage in 150 cities Captive mobile demand Weakness: In-fighting board, acrimonious relationship with CT10%20%JitongSolution primarily based on VSAT coverage, bandwidth constraint Existing bandwidth primarily occupied by own IP phone service Questionable excess bandwidth for resale5%3%Rationale / BenchmarkMobile operators A high percentage of the leased line demand in local lines (80%), but CNC will have a relatively low share in local lines (~1%). For backbone, expect by 2004 about 40% will be IP-based and CNC will have a 65% share (as a dominant IP-backbone provider) ISPs Expect by 2004 about 70% of the market will be addressable (instead of being captive by China Telecom). Since there will probably be at least 5 to 6 backbone operators at that time, we expect about a 15% share out of this. Paging operators Not CNC prioritySource: CNC team inputs; BCG analysis
    • 90. LARGE / MEDIUM BUSINESSES: EXPECTED SHARESCompetitorKey strengths / weaknesses20002004Expected shareCNCMultiple alternatives to ensure rollout/coverage in local access to targeted customers Able to capitalize on different local practice at CT , and partnership with MOR Need to build up capability & share quickly1%15%China Telecom Voice services Satellite communicationsIncumbent: existing customer base Extensive coverage in local loop Weakness: service back log, bureaucracy drags service quality Legacy system in backbone, hinders ability to offer lower price/high bandwidth services74%55%Coverage/bandwidth limited Weakness: bandwidth constraint in later years5%5%UnicomEarly to market, built local access in TJ/CQ/SC Weakness: continuous management problems10%15%JitongAggressively going after market from current ISP base Weakness: bandwidth constraint (VSAT-based solution)10%10%Rationale / BenchmarkIn 2000, CNC coverage and time-to-market limits share achievable By 2004, expect CNC coverage of over 80% of the business demand, with overall share of around 20% in the business leased lines and Internet access markets. Also expect to aggressively go into the web hosting/ coll, reaching 15% share by 2004. For other emerging data and Internet services, expect to get a 5% share due to the nature of more fragmented competition.Source: CNC team inputs; BCG analysis
    • 91. OFF-NET VOICE: EXPECTED SHARESCompetitorKey strengths / weaknessesCNCLow cost based on new technology Innovative & motivated management team Favorable regulatory backings Weakness: late to market, IP voice the only consumer oriented product (lacks synergy)China Telecom Voice servicesIncumbent / early entrant Existing coverage No interconnect hassle Weakness: - higher cost position based on existing infrastructure - cannibalize circuit voice servicesUnicomEarly to market Synergy with mobile business Weaknesses: - initial stage unclear management focus - Acrimonious relation with CTJitongQuick and successful launch in 1999 ,early to market Aggressive marketing, high discount to distributors Weakness: no long-term backbone solution - satellite network may affect voice quality (latency)20002004Expected share15%28%55%40%15%20%15%12%Rationale / BenchmarkIn 2000, CNC deployment plan will not be as aggressive as our competitors; only about 60% of the IP voice market will be addressable. CNC will have about a 25% market share in these (addressable) areas. By 2004, expect to cover all key areas. But CNC share will dropp from the peak of 33% in 2002 due to more competitors entering. Benchmark: In Hong Kong, 3 new entrants with call-back: 20-30% share in 2 years; i.e., incumbent <=70%; half a year after ISR, incumbent has <50% shareSource: CNC team inputs; BCG analysis
    • 92. CONTENTKey strategic principles Regulatory overview Market overview Competition overview Business models Next steps
    • 93. AGENDABusiness Models Domestic and international long-distance voice Wholesale/carrier Enterprise solutions Issues going forward
    • 94. Economic modelingChoice of business models to pursue and timingRoadmap for pursuing business models and expected financial performance Integrated approach to CNC business planMajor issuesExpected output How do the market sizing and share assumptions translate into overall top line revenue for CNC? What capital investments will be necessary to build out metro and long haul fiber networks? Predicted cash flow profile by business model and selected scenariosAssessment of market opportunities How will China datacom market develop? How large is the overall opportunity for a new entrant? Overall market sizing and revenue forecast by product areaRegulatory and competitive analysis What type of regulatory environment will evolve in China? Will equal access for voice and data be granted and when? What effect will WTO have? Regulatory mapping and CNC share predictions across scenarios Strategic implications and capabilities assessmentMETHODOLOGY BEING USED TO DEVELOP BUSINESS MODELS AND OVERALL STRATEGY- Current areas of focus
    • 95. CNC AT A CRITICAL STRATEGIC CROSSROADSPreliminary ConclusionsRecommendations/Decisions to be MadeOffnet VOIP predicted to generate to provide breakeven economics for building backbone1 Wholesale revenue provides significant upside potential Majority of wholesale revenue relies on access to mobile carriers Enterprise solutions economics very attractive, but substantial complexity and resources involved High-bandwidth international gateway critical to success in both wholesale and enterprise Economic predictions highly sensitive to a set of key assumptions Accelerate vendor selection and backbone construction; time to market critical Commitment to utilizing IP/DWDM invlolves risk to mobile carrier business RFP to vendors should be based on product requirements vs. technology Staging of investments and service launch must consider tradeoff between quality of service and coverage Preliminary talks with international carriers should begin ASAP Scenario modeling will help us decide where to focusOverarching question: Can CNC successfully pursue all opportunities outlined in the short/medium term? (1) Assuming settlement fees of 10% of revenue
    • 96. Overall approach Where How BUSINESS MODELS SUMMARY: THREE CORE ELEMENTSEnterprise SolutionsCapture datacom growth in key business centers with leading-edge products and superior customer service Top business districts in major urban areas; only the most dense areas in short term Focused Deployment Leverage existing conduits to lay in major urban areas Superior service and bandwidth Target CT’s weakness in service and bandwidth Utilize LMDS in intermediate cities and areas where time to market is critical Wholesale/CarrierTarget mobile carriers and ISPs with backbone transport; consider supplying fixed-line incumbents Cover POPs in all major calling zones; develop local leased lines network in key locations Aggressive deployment of backbone infrastructure to provide unparalleled bandwidth Establish high bandwidth international gateway to differentiate internet access Superior service with clear positioning “The clear alternative to CT” ISPLong distance voiceCapture early revenue from to fund development of subsequent business models Top 60 POPs by end of year 2000 utilizing mix of leased lines from CT and CNC network Position offnet voice as first product from “China’s first datacom carrier” Do not overextend resources in VOIP as it does not fit CNC’s long term strategy Create “dial-around” solutions for business and interconnect terms- 17930 -
    • 97. Enterprise solutions POTENTIAL BUSINESS MODELS COVER WIDE RANGE OF PRODUCT/MARKET ALTERNATIVESOpportunity for growthCurrent market sizeWholesale/ carrierConsumer ISP?Domestic and International Long Distance VoiceResidentialMed/large enterprise customersCarriersPotential traffic per consumerProductsEmerging datacomDataVoiceEmerging data niche
    • 98. AGENDABusiness Models Domestic and international long-distance voice Wholesale/carrier Enterprise solutions Issues going forward
    • 99. APPROACH TO DOMESTIC AND INTERNATIONAL VOICE BUSINESS MODELObjectivesHypothesized ApproachCapture early revenue from launch of prepaid IP calling cards “Cash cow” for funding other business model development Pursue prefix and equal access long distance for business customers as soon as possible to begin establishing relationships Manage pricing and product life cycle effectively to maximize total margin and avoid investing in declining products Do not overextend ourselves nor blur our “datacom” imageFight the regulatory battle to ensure favorable approaches to equal access and interconnect Market calling cards to business customers in the short term for travelling personnel Emphasize quality image/brand to distinguish from CT and Unicom--position calling card as first step in becoming a next generation full services provider Establish mechanisms to link marketing expenditures with revenue and margin growth by product to ensure effective investment Emphasize low cost targeted marketing and loyalty programs Do not overextend Always emphasize advanced technology and evolution to full services provision
    • 100. DOMESTIC AND INTERNATIONAL VOICE SUMMARYPreliminary EconomicsPhase I CapEx (2000,2001): ~3.8 B RMB Fiber/construction: ~2.4 B RMB IP/DWDM equipment: ~720 M RMB POP/VOIP: ~530 M RMB OSS/Network management system: ~100 M RMB OpEx expected to be ~30% of revenue by 2002 Market share and revenue estimates - 2002 Offnet DLD: 30% Offnet ILD: 30% ~ $2 B RMB IP Intl. termination: 27% 5 year NPV: Essentially breakeven considering offnet voice alone1 Key Issues to be AddressedInterconnect agreements with local PTA’s; attempt to obtain blanket policy from MII International gateway license and connectivity Settlement charges commensurate with VOIP pricing Development of business offnet strategy Scalable “dial-around” solutions in short term Equal access longer term Quality of service for voice, must approach switched quality rapidly Point of diminishing returns for adding VOIP gateways vs. strategic value of providing coverage(1) Highly sensitive to settlement fees
    • 101. LARGE MARKET WITH POTENTIAL TO GAIN SHARE QUICKLYNew entrants typically gain share quicklyExample: IDD and DLD servicesMarket sizeable - off-net traffic accounts for ~15% of total DLD/ILD revenue by 2004Source: China Telecom annual reports; CNC’s team inputs; BCG surveys, analysis & benchmarkingYear after entryOptus (DLD)Tele 2 (DLD)Mercury (DLD)Hong Kong (IDD)Japan (IDD)US (IDD)(RMB BN)DLD On-NetDLD Off-NetILD On-NetILD Off-Net(%)
    • 102. CAREFUL MANAGEMENT OF PRE-PAID CALLING CARD BUSINESS NECESSARY TO ALIGN WITH LONG TERM STRATEGY Prepaid calling cards call for different capability set and target customers than longer term business models Focus on consumers will not complement long term vision of providing enterprise solutions Mass advertising and marketing around a low-cost position may not fit image required for future needs Three factors important to consider in managing prepaid calling card business Attempt to position cards in marketing messages as the first product from a company that is building the most advanced network in PRC Consider selling cards to businesses for their traveling personnel to begin establishing enterprise relationships Carefully manage product life cycle to begin pulling back marketing investment as wholesale and enterprise business models grow
    • 103. CNC VOIP REVENUE AND MARKET SHARE EXPECTATIONS Source: CNC’s team inputs; various benchmarks; BCG analysisVoice revenueShare assumptionsCNC Revenue (RMB BN)Off-Net DLD% of total CNC revenue100%76%37%30%23%16%Off-Net ILDInternational Termination1999 3% 80% 1% 3% 80% 1% 5% 80% 0%Off-Net DLD Off-net share of total DLD Geographic coverage of CNC CNC share within coverage Off-Net ILD Off-net share of total ILD Geographic coverage of CNC CNC share within coverage International Termination IP share of total Geographic coverage of CNC CNC share within coverage2000 10% 60% 25% 11% 60% 25% 11% 60% 20%2001 17% 75% 29% 18% 75% 29% 18% 75% 25%2002 23% 90% 33% 26% 90% 33% 24% 90% 30%2003 30% 100% 30% 33% 100% 30% 30% 100% 30%2004 32% 100% 28% 35% 100% 28% 35% 100% 31%
    • 104. PRELIMINARY ECONOMICS FOR LONG DISTANCE VOICE MODEL (PHASE I BUILDOUT) VOIP Revenue Alone Justifies Building Backbone5 year PV(1) (M RMB)CapEXOpExRevenue5 year NPV@15%:~ -500M RMB5 year IRR:~12%Essentially breakeven economics for operating backbone for VOIP onlyVOIP(2)Backbone constructionBackboneInternational termination IDD DLDIP POP/Access platform OSSPresent value of cash flows(1) Assuming 15% cost of capital (2) Including settlement charges estimated at 10% of VOIP revenue, and marketing/sales at 10% of revenue (3) Backbone OpEx charges allocated 1/3 each to VOIP, wholesale, and enterprise business model economics Source: BCG benchmark database; industry interviews; BCG analysis
    • 105. AGENDABusiness Models Domestic and international long-distance voice Wholesale/carrier Enterprise solutions Issues going forward
    • 106. APPROACH TO WHOLESALE/CARRIER BUSINESS MODELObjectivesHypothesized ApproachDevelop wholesale business as traffic generator to improve economics of backbone through higher utilization Become the wholesale carrier of choice with technologically superior service offerings including high bandwidth international gateway connectivity Consider wholesaling access to CT, Unicom, and Jitong depending on competitive implications Superior customer service with clear positioning “The clear alternative to CT” Aggressive deployment of backbone infrastructure Connecting top 15 cities by end of 2000 and expanding to top 50 cities by 2002 Seek partnerships to establish high bandwidth international gateway connectivity--absolutely essential for differentiating CNC offering Develop interconnection capabilities in all major POPs and mobile basestations in key geographical locations Wholesale access to incumbent providers where feasible, but do not wholesale sources of competitive advantage (e.g., enhanced data services such as IP VPNs) Rollout product offering in staged manner to ensure quality of service Internet connectivity Mobile interconnect Access ports to backbone
    • 107. WHOLESALE/CARRIER SUMMARYPreliminary EconomicsPhase I CapEx (2000,2001): ~100 M RMB ISP access platform: ~50 M RMB OSS/Provisioning systems: ~50 M RMB OpEx expected to be ~ 10% of revenue by 20021 Market share and revenue estimates - 2002 Mobile (backbone): 15% ISPs: 9% ~ 1.1 B RMB Access ports: 100%2 Dark fiber: 100%2 5 year NPV: ~ $2.1 B RMB Assumes launch date of 3Q 2000 for leased lines and relatively aggressive mobile shares Potentially too optimistic Key Issues to be AddressedBackbone technology platform - QOS for voice vs. lower cost deployment? High-bandwidth international gateway paramount to differentiating ISP access Favorable regulatory backing for courting regional CT mobile carriers Ensuring existing VOIP gateways can serve wholesale needs Revenue opportunity of wholesaling dark fiber vs. enabling competition Organizational challenges(1) Including allocation of backbone OpEx (2) Market estimates based on revenue generation by CNC alone
    • 108. POTENTIAL WHOLESALE CUSTOMERS INCLUDE ISPs, MOBILE OPERATORS, AND FIXED LINE CARRIERS ISPs offer significant potential if CNC can provide superior bandwidth access and to international gateway Current satisfaction among regional ISPs very low International gateway license in conjunction with high bandwidth trans-oceanic carrier alliance could provide vastly superior service Mobile carriers will be searching for lower cost alternatives to carry long distance traffic due to intensifying competition CNC’s new high capacity VoIP network and international gateway likely to yield lower costs Fixed line carriers potentially looking for alternatives Existing long-haul transport infrastructure limited China Telecom could even be a possible customer given current focus on increasing residential teledensity
    • 109. OVERALL WHOLESALE MARKET SIZE IS SUBSTANTIAL AND GROWING AT A MODEST RATE Anticipated Price Decrease in Leased Lines Limits Overall Revenue GrowthLeased Lines - ISPMarket Size (RMB B)15TotalLeased Lines - Mobile99’-04’ CAGRAccess PortsLeased Lines - Paging1818202226284%41%1%152%(1)9%(1) 01’-04’ CAGR Source: CNC team inputs; foreign benchmarks; BCG analysisDark Fiber68%(1)
    • 110. CARRIERS SEEKING ALTERNATIVES...ISPs definitely seeking alternatives to CTMobile carriers likely to follow“We need a telecom service provider that is not our competitor.” - Founder, Eastnet “China Telecom, with their own network development plans, tends to starve us on capacity or to force us to pay in advance for excess capacity.” - Manager, Infohighway “The fact that it takes China Telecom two months every time we need an extra line makes it very difficult to have our own customers. We want another operator who can get us leased lines fast.” - Manager, InfohighwayMobile carriers might consider diverting part of their traffic to alternative service providers with: more attractive pricing higher quality service higher bandwidth Interviews with regional mobile carriers and Unicom need to be conducted to verify potential
    • 111. …BUT TECHNOLOGICAL LIMITATIONS AND COMPETITIVE CHALLENGE ARE IMPORTANT FACETS TO MANAGECompetitive ChallengeTechnological LimitationsMobile carriers may be hesitant to use VOIP technology for primary applications Carriers currently addressing sound quality as a major improvement initiative VOIP has yet to deliver toll-quality voice transmission, even on landline Mobile carriers may take view that VOIP could further degrade voice quality China Telecom likely to have advantage in competing share of CT Mobile’s business Strong former intra-CT connection even after split Extensive backbone coverage and large TDM based capacity Unicom Mobile Services’ business as tough targetBuild presence by offering low-cost trials and backup capacityLobby for clear regulation from MII on freedom of choice for carriers
    • 112. PREDICTED WHOLESALE ECONOMICS ADD SIGNIFICANT VALUE TO VOIP BUSINESS MODEL Additional CapEx and OpEx Minimal5 year PV(1) (M RMB)CapEXOpExRevenue5 year NPV@15%:~2.1B RMB5 year IRR:~ 40%Wholesale critical to enhancing profitability of backboneVOIPBackbone allocationDark Fiber Access ports ISPs Mobile operators VOIPVOIP ISP access OSS(1) Assuming 15% cost of capital Source: BCG benchmark database; industry interviews; BCG analysisOSS/provisioningPresent value of cash flows8,0007,0005,0001,0006,000-3,000-4,000
    • 113. AGENDABusiness Models Domestic and international long-distance voice Wholesale/carrier Enterprise solutions Issues going forward
    • 114. APPROACH TO ENTERPRISE SOLUTIONS BUSINESS MODELObjectivesHypothesized ApproachCapture strong share among medium/large business by offering enhanced datacom solutions Goal to establish clear position as best service/quality provider in major markets Utilize most cost effective deployment technologies to cover major metro areas Minimize head-to-head competition by offering differentiated, data-centric products--attempt to drive datacom market Develop image as fast, responsive solutions provider Enable competitive advantage for business customers through datacom For building managers: make their buildings more attractive to tenants Deployment to target key buildings in major metropolitan areas Four cities by year 2000/01 Top 15 cities by 2001 Fiber in most dense urban hi-rise areas and LMDS to complement and serve less dense areas Initial lead products will be low cost voice over IP and high bandwidth internet access Migration to full datacom solutions as customer base and capabilities grow Quality customer service more important short term than full product offering Emphasis on ease of use and fast provisioning versus competitors--exploit CT’s weaknesses Education of customers on use of datacom products as competitive weapons Marketing partner with key building managers
    • 115. ENTERPRISE SOLUTIONS SUMMARYPreliminary EconomicsPhase I Capital Investment (2000,2001): ~1.1 B RMB First four cities (assuming fiber): ~650 M RMB Additional 11 cities1: ~400 M RMB OpEx expected to be ~ 40% of revenue by 2002 Market share and revenue estimates - 2002 Offnet Voice2: ~20% Existing data: ~10% ~ 1.1 B RMB Emerging data: ~ 5% 5 year NPV: Roughly 1.2 B RMB Assumes launch date of 3Q 2000 for data services Likely too optimistic Key Issues to be AddressedRight of way for existing ducts and digging Partnership strategy for high bandwidth IGW Rights to LMDS frequency spectrum Are the 15 cities designated for Phase I buildout the right 15 cities for local access? Tradeoff between pure economics by city vs. strategic value of providing end-to-end connectivity What is a realistic time frame for launch? Magnitude of organizational and human resource requirements(1) Assuming LMDS capital and revenue 2 times Fuzhou estimate for cities 11-15; 3 times Fuzhou estimate for cities 5-10 (2) Assumes no local voice revenue through 2004
    • 116. ENTERPRISE SOLUTIONS BUSINESS MODEL MOST COMPLEX WITH HIGH CAPEX AND OPEX REQUIREMENTS...Building metropolitan fiber rings to offer access to medium and large businesses poses significant challenge Operating expenses required dwarfs long haul network costs on a per city basis Complexity in obtaining night-of-way varies by districts within each city Designing fiber route and network configuration requires significant experience Converting customers to full CNC service may not be as easy as it seems on surface Initial risk for companies utilizing new entrant Coverage issues for offering service to all business locations Experienced sales-force with established relationships a must
    • 117. …BUT COMPRISES TREMENDOUS UPSIDE POTENTIALInternet AccessLeased LinesILDDLDLocalOtherPortalBroadband ContentWeb Hosting/ CollocationExisting ServicesEmerging Services98-04 CAGR124%40%-10%7%11%IP VPNBroadband Network Applications71%83%186%98%111%286%00-04 CAGRMarket size (RMB BN)Market size (RMB BN)7896187Overall16%Overall86%332Source: CNC team inputs; foreign benchmarks; BCG analysis
    • 118. CUSTOMER NEEDS EXIST THRUGHOUT THE VALUE CHAIN CT’s Offering Yields Significant GapsLearnBuyGetUse/supportPayCustomer value chainNeeds identifiedHow datacom services can help their business Assistance deploying solutionsQuicker and more convenient application channelsRapid and reliable provisioningFaster, reliable repair servicesPrompt, customized billing Lack customer focus Solutions virtually non-existent Mostly one-way product marketingNo customer input for provisioning CT determines queue without specific timing No penalties for missed appointmentsNegligent repair service Slow fulfillment & technical support responsePoor response to customers Limited bill customizationCT approachSignificant opportunity for CNC establish position in customer solutions, ease of use, and responsive serviceAreas highlighted most in customer interviewsSource: Customer interviews; BCG analysis
    • 119. STAGED PRODUCT INTRODUCTION NECESSARY TO MANAGE QUALITY Hypothesized Product IntroductionsVoiceData200020012002200320042008Phase I & IIPhase IIIPhase IV Leased Lines Mobile carriers Eneterprise High bandwidth dedicated internet access for businesses and ISPs Access Ports VPN Basic enterprise Broadband applications Platforms to enable VOD, etc... Web Hosting Dark Fiber Residential ISP?Advanced VPN Extranet Voice, video QoS guarantees Industry specific offerings Other emerging services (e.g. e-Commerce)IP-phone voice Prefix “Dial-around” solutionsEqual access voice?On-net voice?Focus on quality over quantityMarket as integrated productsStress CNC role as integrated datacom playerPositioning Issues
    • 120. CNC SHOULD USE CREATIVE WAYS TO GROW THE MARKET, AVOID HEAD TO HEAD COMPETITIONCompetitors all focus on the same demand from the same customers Savagely compete on price Reactively wait for customers to identify need, request productsPursue growing customers - data intensive enterprises Fresh opportunities for sales, instead of competing over existing business Approach proactively with new products, instead of waiting for customers to initiate sales process Creatively identify new markets I.e. market for companies requiring very rapid provisioning or customized products Become the only choice for these companies Lobby the government to support more e-commerce and information industries growthCTCNCUnicomJitongCT is likely to be unable to meet growing demand aloneTraditional, narrow viewNew approaches to growing sales
    • 121. TOTAL GDP VS. GDP/CAPITA IS A PRELIMINARY INDICATOR FOR ATTRACTIVE LOCAL ACCESS MARKETS Example: 15 Initial Cities with POPs on CNC NetworkShanghaiBeijingGuangzhouTianjinHangzhouWuhanNanjingJinanShijiazhuangZhenzhouChangshaXuzhouXiamenFuzhouShenzhenExamples of additional cities not in initial 15 designated for CNC POPsHypothetical breakeven curve for building local accessSuzhouWuxiTotal GDP (k RMB)GDP/cap (k RMB)
    • 122. VERY DENSE MARKETS SUCH AS SHANGHAI BEST SERVED BY METRO FIBER RINGSMPOP/rail stationHong Qiao Development AreaPeople’s parkHuai Hai RoadXujiahuiPudong Financial ZoneFiber length: Along subway: Along elevated ring road: Along road:17000m 9000m 8000m# of major business buildings accessed:~200Fibers alongSubway in construction Subway Elevated ring road Road (dig required) Dense Business Area served Metro stationsMMM
    • 123. SHANGHAI ECONOMICS VERY ATTRACTIVE DUE TO HIGH BUILDING AND BUSINESS DENSITY5 year PV(1) (M RMB)CapEXOpExRevenue5 year NPV@15%:~310M RMB5 year IRR:~70%Shanghai presents most attractive enterprise solutions marketSG & AAllocated backboneIDD Emerging data(3) DLD Data servicesSONET/switching equipment OSS Allocated backbone(2)(1) Assuming 15% cost of capital (2) Cost of providing F/R and ATM service on backbone allocated over first 4 cities (3) Web hosting IP VPNS, etc. Source: BCG benchmark database; industry interviews; BCG analysis; real estate agency interviews; field analysisPower provisioning, etc.Fiber constructionRight of wayPresent value of cash flows
    • 124. MEDIUM SIZED CITIES WITH HIGH GDP/CAPITA SUCH AS FUZHOU BEST SERVED WITH LMDS SOLUTIONFiber length: Along road: # LMDS station: 5000M 1Area served(1): # of major business buildings covered:~40km2 ~40(2)Lake Fiber along road Railway Railway station Road LMDS station Major buildings Area servedRS(1) More detailed field analysis required to map buildings and most effective LMDS positioning; assumes 3.5 km radius for LMDS area (2) Assume half of the buildings are prime business buildings for economic modelingSR
    • 125. PRELIMINARY FUZHOU ECONOMICS UTILIZING LMDS VERY FAVORABLE5 year PV(1) (M RMB)CapEXOpExRevenue5 year NPV@15%:~20M RMB5 year IRR:~60%LMDS solutions for medium sized cities will play a critical role in developing enterprise solutionsSG & APower, provisioning, etc.IDD Emerging data services DLD Data servicesBase station CPE(1) Assuming 15% cost of capital Source: BCG benchmark database; industry interviews; BCG analysis; real estate agency provided data Bldg equipmentSpectrum fees Fiber to POPPresent value of cash flows
    • 126. OTHER CITIES OF SIZEABLE POPULATION BUT RELATIVELY LOW GDP/CAPITA MAY NOT WARRANT INVESTMENT Example: ShijiazhuangNo fiber connection required, assume rail station is LMDS base 1Area served(1): # of major business buildings covered:~40km2 ~20(2)Railway Road LMDS station Major buildings Area servedSS(1) More detailed field analysis required to map buildings and most effective LMDS positioning; assumes 3.5 km radius for LMDS area (2) Assume half of the buildings are prime business buildings for economic modeling
    • 127. OVERALL ECONOMICS APPEAR MARGINAL FOR SHIJIAZHUANG5 year PV(1) (M RMB)CapEXOpExRevenue5 year NPV@15%:~ -3M RMB5 year IRR:~5%Cities without dense business districts may not be attractive enterprise markets regard less of overall sizeSG&APower, provisioning, etc.IDD Emerging data services DLD Data servicesBase station(1) Assuming 15% cost of capital Source: BCG benchmark database; industry interviews; BCG analysis; real estate agency provided data Spectrum CPE Bldg. equipmentPresent value of cash flows
    • 128. OVERALL REVENUE AND MARKET SHARE EXPECTATIONS: ENTERPRISE SOLUTIONSPredominately voice (DLD / ILD) revenue in the first two years; migrating to data and Internet services by 2002CNC revenue (RMB BN)Internet AccessLeased LinesILDDLDShare assumptionsOther Emerging ServicesWeb Hosting / CollocationSource: CNC team inputs; foreign benchmarks; BCG analysis1999 3% 80% 1% 3% 80% 1% 0% 0% 0% 0% 0% 0%DLD Off-net share of total DLD Geographic coverage of CNC CNC share within coverage ILD Off-net share of total ILD Geographic coverage of CNC CNC share within coverage Leased Lines Geographic coverage of CNC CNC share within coverage Internet Access Broadband Narrowband / Dial-Up Web Hosting / Collocation Other Emerging Services2000 10% 60% 25% 11% 60% 25% 30% 2% 2% 0% 0% 0%2001 17% 75% 29% 18% 75% 29% 47% 10% 10% 0% 5% 1%2002 23% 90% 33% 26% 90% 33% 63% 14% 14% 0% 8% 2%2003 30% 100% 30% 33% 100% 30% 80% 18% 18% 0% 10% 4%2004 32% 100% 28% 35% 100% 28% 82% 21% 21% 0% 13% 6%
    • 129. OVERALL ECONOMICS ATTRACTIVE FOR PHASE I BUILDOUT IN TOP FOUR CITIES5 year PVs (M RMB)CapExOpExRevenue5 year city NPV5 year IRRShanghai Beijing Guangzhou Shenzhen Total145 149 157 95 546504 447 400 332 1683960 768 640 480 2848311 172 83 5370% 45% 30% 28%Overall Phase I 5 year NPV: ~620M RMB 5 year IRR: ~44%Source: BCG benchmark database; industry interviews; BCG analysis
    • 130. GOING FORWARD, CITIES WILL NEED TO BE ANALYZED ON A CASE BY CASE BASISDeployment technology will play a critical roleMethodology/criteria for choosing local access marketsInitial prioritization by total GDP vs. GDP/capita Relevance and proximity to CNC network Map urban area by biggest buildings and availability of existing conduits Determine if density falls into one of the three density categories Estimate costs and revenue associated with buildout1. 2. 3. 4. 5.Go or no go!Cost to deliver enterprise solutionsIncreasing business density and areaIntermediate density best served by LMDSVery dense and large area best served by fiberToo costly to serveFiberLMDS
    • 131. RECAP OF OVERALL PHASE I ECONOMICS AND KEY ISSUESEnterprise Solutions2 1.1 B 3.2 B (38%) 1.2 B Businesses/km2 Data growth IGW/connectivity Deployment cities Realistic rollout Buildout economics for 15 cities vs. connectivity valueWholesale/Carrier 100 M 2.3 B (27%) 2.1 B Mobile share Internet growth IGW/connectivity Voice QOS Regulatory backing Cost of deployment vs. service flexibilityLong distance voice 3.8 B 3 B (35%) -- Settlement charges Pricing/competition IGW/connectivity Managing life cycle Business solutions Economics of 60 VOIP cities vs. strategic value of coverage Economics1 CapEx 2004 Revenue (% of total) 5 YR NPV3 Key Issues Sensitivity Top issues to address Tradeoffs(1) All values in RMB and based on preliminary inputs to be refined (2) Includes only Phase I buildout to top 15 cities (3) Assuming 15% cost of capital
    • 132. AGENDABusiness Models Domestic and international long-distance voice Wholesale/carrier Enterprise solutions Issues going forward
    • 133. EMERGING OPPORTUNITIES EXIST FOR CONSUMER ISP AND ADVANCED DATA SERVICESResidential ISP CNC may be in unique position to offer true broadband internet access due to IP network and international gateway What strategies will capture significant value while minimizing development time and necessity for brand equity? Web Hosting/Data Centers Niche competitors such as Exodus in the U.S. aggressively pursuing this business model Will opportunities develop in the near term for China and could CNC supply resources necessary? E-commerce solutions Many telcos in Europe and U.S. offering e-commerce solutions Will opportunities develop in the near term for China and could CNC supply resources necessary?
    • 134. Datacom services explosion Slowed adoptionStable competition with CT, Unicom, and Jitong; clear regulationsIncreased competition with limited cooperation from CTFragmented, unstable competition; multiple entrants and unclear regulationsBase caseBestWorstDemand sideCompetitive/Regulatory Situation (Supply side)ENVIRONMENTAL UNCERTAINTIES LEAD TO THREE BASIC SCENARIOS TO BE ANALYZEDGrowth in existing data and and emerging datacom services (as expected)
    • 135. MONTE CARLO SIMULATION WILL ALLOW FOR GAUGING THE EFFECTS OF UNCERTAINTIES IMPLIED IN SCENARIOS Example: Effect of Average Building Size, Tier 1 Business Proportion & Spending on Shanghai Metro Access EconomicsEstablish Input Variables25,000.0028,750.0032,500.0036,250.0040,000.00Average building size (m2)MinMost likelyMax7.5%11.3%15.0%18.8%22.5%% Tier 1 businessesMinMost likelyMax300.00350.00400.00450.00500.00Average spending/sq m for Tier 1 (RMB)MinMost likelyMaxProbabilityMonte Carlo simulation of 1000 trials with 5 year NPV as forecast outputFrequency ChartCertainty is 95.00% from $238,500 to $558,816 000 RMBMean = $381,016.000.023.046.068.091022.7545.568.2591$150,000$262,500$375,000$487,500$600,0001,000 Trials 6 OutliersForecast: 5 YR NPVResulting forecast
    • 136. MONTE CARLO SIMULATION PROVIDES VERY USEFUL OUTPUT95% certainty that with the inputs provided, 5 year NPV will be greater than 239M RMB Statistically impossible for the project to have a negative NPV based on provided inputs Forecast mean is actually higher than our base case value (381M vs. 318M) due to effects of variable inputs A sensitivity report generated from the simulation tells us that average building size contributes the most to output variance, therefore it is the most important input to focus on clarifying
    • 137. CONTENTKey strategic principles Regulatory overview Market overview Competition overview Business models Next steps
    • 138. NEXT STEPSAgree on business mix assumptions Develop organization structure Agree on key principles and objectives Timeline, key milestones Develop implementation action plan High level plan (by quarters) till Jan 2002 Detailed plan (by month) Jan-Jul 2000 Key milestones, process control Integrate and refine overall CNC financial Objectives and assumptions for financial model Scenarios / sensitivities
    • 139. ORGANIZATION DESIGN METHODOLOGYDefine principles & objectivesKey criteriaDesign organization structure Possibility of combination/hybrid modelsDefine job responsibility/accountability corresponding to organization structure Example for process flow (how each function interacts) Example for KPI, principles for KPISelect models to test
    • 140. MAJOR OUTPUTS: ORGANIZATION DESIGNKey principles for KPI With examplesDivision/accountability “Internal market” as cross-dept incentiveOrganization chart Dotted line vs. solid line Reporting/accountability Rationale/methodology
    • 141. IMPLEMENTATION PLAN DESIGN METHODOLOGYObjective/target scheduleMajor function/module working backwardDetailed work and cycle time for module/sub-module ® refine timelineDraft implementation timelineKey checklist & mile stones per divisionKey process for rollout (across divisions) Product rollout Geographic rolloutRevise target if necessaryIdentify bottleneck, refine scheduleSegmented view of implementation
    • 142. Key cross-functional coordination process flow ExamplesMAJOR OUTPUTS: IMPLEMENTATION PLANKey rollout plan example Product CityMS project time-line/milestones Overall By divisionHighlights Major milestones resource planning
    • 143. WORKPLAN(1)(1) Refer to CNC team schedule for details. TDC off Dec 22-30; TC off Dec 20-24; Rest off Dec 27, 31 for holiday