1、Internet pricing and price comparisons,Dr Tim Kelly, ITU ITU/TOT Workshop on Trends in Telecom Prices and Costing in Developing Economies of the Asia Pacific Region Bangkok, 26-29 November 2001,The views expressed in this presentation are those of the author and do not necessarily reflect the opinio
2、ns of the ITU or its membership. The author can be contacted at tim.kellyitu.int.,Agenda,Internet around the world Still growing worldwide, even after the D bubble burst Doing even better in Asia Internet pricing Cost components Pricing components Input costs and prices Internet price comparisons OE
3、CD price comparison methodology ASEAN price comparisons Choosing the most appropriate model,Bursting the Telecom Bubble,Total market value of telecom operators down from US$6.3 trill. to US$3.8 trill. More than 400000 redundancies announced in telecoms since Oct 00 On average, a major telephone oper
4、ator goes bust once every six daysSource: ,Share price trends in the US “Technology Media and Telecoms (TMT)” sector,Nevertheless, the Internet continues to grow Internet users, million, and growth rate in %,Source: ITU.,How is ASEAN doing? Growth rate in Internet users,Source: ITU World Telecommuni
5、cation Indicators Database.,0%,50%,100%,150%,200%,250%,1992,1993,1994,1995,1996,1997,1998,1999,2000,ASEAN average,growth rate,Global average,growth rate,The SE Asia Digital Divide,Source: ITU.,Can and ISP control its costs? Regulatory bottlenecks,Wireless Access,ISP POP or IXP,International connecti
6、vity,ADSL?,Cable TV,?,?,?,?,?,?,?,ISP,Dial-up,Leased line,National connectivity,Internet pricing: Cost components,International connectivity Near-end and far-end half-circuits Termination and routing services National connectivity Distribution of points of presence (PoPs) National peering? National
7、dial-up number? Commercial and operational costs Costs of staff recruitment, training and retaining Marketing and customer acquisition costs Billing and customer service,Typical ISP cost comparisons,ASEAN countries,OECD countries ,International component of ISP costs,Source: Study carried out by Ant
8、elope Consulting for DFID, available at: www.antelope.org.uk.,International Gateway,Cost issues Near-end and far-end costs Backhaul costs Unit costs related to economies of scale Is connectivity bundled in or separate from infrastructure costs? Regulatory issues What is the typical waiting period to
9、 obtain bandwidth How much choice for ISPs? Can ISPs own their own gateway (e.g., using VSAT technology)?,Wholesale Price per 64 kbit/s of international bandwidth, US$,Source: ITU Case Studies,The critical factor: IP connectivity,1.1,0.2,0.5,1.0,0.6,0.1,0,500,1000,1500,2000,2500,3000,Viet Nam,Indone
10、sia,Thailand,Malaysia,Philippines,Singapore,Intl traffic (bn mins) and Intl bandwidth,(Mbit/s),0.0,0.2,0.4,0.6,0.8,1.0,1.2,Bit-Minute Index,Intl traffic mins,Intl bandwidth,Bit-Minute Index,Note: International traffic (in billion mins) includes both outgoing and incoming. Data is for 1999 except Ind
11、onesia & Philippines (2000). International IP connectivity is in Mbit/s (Sept. 01).,Source: ITU/ TeleGeography Inc.,The bit-minute index is calculated as Mbit/s of intl bandwidth divided by billions of mins of intl traffic,National connectivity: Leased line price comparisons,Huge differences in leas
12、ed line prices across the region Big differences between urban and rural areas in some countries Infrastructure competition is critical factor in achieving lower prices Leased line prices affect competitiveness of ISPs,Typical prices for a 64 kbit/s leased line, per month (US$) Source: ITU ASEAN Int
13、ernet diffusion case studies.,Retail pricing models,Nationwide at local call One number: Malaysia, Singapore, Indonesia, Vietnam, Thailand Internet on demand No pre- registration required in Viet Nam, Indonesia Prepaid cards Thailand, Philippines, Viet Nam Telephone charges Usage: Singapore, Vietnam
14、, Laos, Cambodia No charge: Philippines Lower rate: Malaysia, Indonesia Flat rate: Thailand,OECD pricing comparisons: Methodological issues,Wide variety of pricing options make direct comparisons difficult E.g., pre-paid/post-paid; bundled/unbundled telephony charges; usage limits/always-on; broadba
15、nd/narrowband What is an “average” usage basket? Average hours per month of AOL users has risen from 10 in 1996 to around 35 in 2000 OECD defines baskets at 10, 20, 30 hours per month plus “always-on” For broadband, cost per 64 kbit/s per month How to handle telephone charges? Include PSTN dial-up c
16、harges Include PSTN line rental charge?,Internet retail pricing,30 hours of Internet access, US$, October 2001,Source: ITU adapted from ISPs / PTOs,Competition among ISPs drives down prices,No. of ISPs licensed (operational),3 (3),2 (2),5 (4),150 (50),150 (60),18 (18),18 (?),44 (?),Source: ITU Inter
17、net diffusion case studies.,Internet strategies: What works where?,Prepaid Internet cards In the Philippines, more than 40% of use is pre-paid, and growing fast Internet cafs / Public access Only a quarter of Thai users use the Internet from home In Indonesia, there are more than 2500 Warnets provid
18、ing public Internet access Nationwide dial-codes In Viet Nam, around 40% of users dial-up via 1268 and 1269 numbers, without pre-registration Low infrastructure costs, through competition In Singapore, a 64 kbit/s leased line costs as little as US$30 per month and international bandwidth is relative
19、ly plentiful,Conclusion,Internet price comparisons are difficult to do but necessary For benchmarking For policy evaluation For encouraging consumer choice Higher prices reflect limitations on market entry and/or higher input prices (esp. international connectivity) Different PSTN pricing strategies
20、 have impact of Internet take-up Consumers want “choice”,For more information ,ITU Internet diffusion case studies 2000: Bolivia, Egypt, Hungary, Nepal, Singapore, Uganda 2001: ASEAN region, Cambodia, Indonesia, Laos, Malaysia, Philippines, Thailand, Viet Nam OECD Internet price comparisons www.oecd.org/sti/telecom DFID study of Internet costs www.antelope.co.uk,www.itu.int/ti/casestudies,
