Broadband tariffs in Latin America: Benchmarking and analysis - PowerPoint PPT Presentation

Broadband tariffs in Latin America: Benchmarking and analysis Hernan Galperin, Ph.D. Universidad de San Andrés and CONICET (Argentina) Christian Ruzzier, Ph.D. Universidad de San Andrés (Argentina) Regional Dialogue on Information Society

Goals 1. Develop meaningful indicators of broadband performance (penetration, quality, price, affordability) for Latin America 2. Benchmark against developed countries (OECD) 3. Support national broadband policy initiatives by estimating price elasticity and affordability

How to measure broadband penetration? • The most common approach: connections per 100 people or households • The results are very predictable  strong correlation with wealth indicators (eg, GDP per capita) • Our approach: compare actual vs. expected penetration based on economic and social endowments  Broadband Performance Index • PENETi = f (PBICAPi, DENSHHi, POPURBi, AGEi, EDUCi) +  i

LAC countries generally perform below expectations in fixed broadband penetration Performance below Performance above expectations expectations Belize (15) 0.17 Nicaragua (20) 0.05 Guatemala (21) 0.05 Mexico (2) -0.04 Brazil (5) -0.04 Paraguay (17) -0.05 Jamaica (11) -0.05 Costa Rica (8) -0.06 Trinidad & Tobago (4) -0.09 Colombia (10) -0.10 El Salvador (16) -0.10 Bolivia (18) -0.11 Chile (1) -0.13 Surinam (19) -0.14 Guyana (22) -0.15 Dominican Rep. (12) -0.15 Panama (9) -0.19 Peru (14) -0.23 Venezuela (7) -0.25 Ecuador (18) -0.25 Argentina (3) -0.34 Uruguay (6) -0.45 Source: ITU (2010) and authors

Methodology for price and quality indicators • Survey of all fixed broadband plans offered by main DSL and cable operators in LAC, 2Q 2010 • 323 plans offered by 54 operators in 23 countries • Largest city in each country  best prices available • List price, discounts, advertised speed, data cap

Mbps) is well below OECD average Average QoS in LAC (advertised speed in 10 12 14 16 18 0 2 4 6 8 Average OCDE Chile Jamaica Brazil Trinidad & … Argentina Colombia Mexico Panama Average LATAM Dominican … Costa Rica Peru Ecuador Paraguay Uruguay Venezuela Honduras El Salvador Nicaragua Bolivia Guatemala Surinam Belize Source: authors Guyana

PPP) are 2.5 times higher than OECD Average prices in LAC (list price in USD $100 $150 $200 $250 $300 $350 $400 $50 $0 Jamaica Uruguay Average OCDE Mexico Chile Trinidad & Tobago Venezuela El Salvador Guatemala Colombia Costa Rica Brazil Argentina Paraguay Average LATAM Panama Nicaragua Honduras Peru Ecuador Guyana Dominican Rep. Surinam Belize Source: authors Bolivia

PPP/Mbps) is 15 times higher in LAC Average cost of Mbps across plans (USD $100 $200 $300 $400 $500 $600 $0 Average OCDE Jamaica Chile Mexico Trinidad & Tobago Colombia Brazil Argentina Panama Uruguay Costa Rica Paraguay Ecuador Dominican Rep. Peru El Salvador Guatemala Average LATAM Honduras Venezuela Nicaragua Surinam Guyana Bolivia Source: authors Belize

plan (in USD PPP) is comparable to OECD In some cases cost of least expensive $100 $120 $140 $160 $20 $40 $60 $80 $0 Minimum OCDE Venezuela Uruguay Trinidad & Tobago Brazil Mexico Costa Rica Paraguay Jamaica Panama Guatemala Chile Colombia Ecuador El Salvador Argentina Dominican Rep. Peru Bolivia Nicaragua Guyana Surinam Honduras Source: authors Belize

Key results • LAC countries generally perform below expectations in fixed broadband penetration • Higher tariffs and lower quality services than OECD • Very high dispersion within region suggests opportunities for regulatory action • Proactive policy to reduce cost of entry-level plans helps reduce gap with OECD prices

Correlation between price and penetration suggests association, but not enough to estimate effect 90 80 Penetration (per 100 HH) 70 60 50 R² = 0.623 40 30 20 10 0 $0 $100 $200 $300 $400 $500 $600 Price (US$ per Mbps) Source: authors and ITU

Methodology to estimate the effect of price on broadband demand • Step 1. Regression model (OLS) for broadband demand:  PENETi =  0 +  1 PRICEi +  2 PBICAPi +  3 AGEi +  4 EDUCi + ui  Results are significant and model is good (R 2 =0.84) but price effect is small (  1= -0.029) • Step 2. Identify a factor that affects price but not demand for broadband (IV)  Household density per km 2 • Results for the IV model reveal a larger effect of price on broadband demand (  1= -0.19)

Results show that broadband demand is relatively elastic to price, particularly in Latin America • Price elasticity for LATAM is 1.9 vs. 0.53 for OECD. Consistent with lower penetration and higher tariffs in LATAM • Different for every country  price effect is larger in countries with lower penetration and higher tariffs • LATAM: price reduction of 10% would result in increase of 19% in penetration (4.7M additional subscriptions) • Brazil example: to double penetration rate (7 to 14%) a 50% price cut is needed. But to triple rate (PNBL goal) a 85% cut would be needed.

HH) is much lower (8x) in LATAM (average broadband price as % of GDP per On average broadband affordability 10% 15% 20% 25% 0% 5% Bolivia Guyana Belize Nicaragua Honduras Dominican Rep. Surinam Paraguay Average LATAM Ecuador Guatemala Peru El Salvador Brazil Panama Turkey Colombia Argentina Costa Rica Jamaica Uruguay Venezuela Portugal Chile Slovakia Czech Rep. Mexico Poland New Zealand Hungary Trinidad & Tobago Australia Average OCDE Austria South Korea Spain Canada Germany Denmark Belgium Norway Luxembourg Netherlands Sweden Iceland France Finland USA Japan Italy UK Switzerland Greece Ireland

OECD levels the least expensive plan, but still far from The picture improves when considering 0.0% 1.0% 2.0% 3.0% 4.0% 5.0% 6.0% 7.0% 8.0% 9.0% Guyana Honduras Nicaragua Belize Bolivia Surinam Average LATAM Guatemala Paraguay El Salvador Dominican Rep. Peru Jamaica Ecuador Colombia Argentina Chile Panama Brazil Costa Rica Uruguay Portugal South Korea Poland Mexico New Zealand Hungary Slovakia France Spain Italy Belgium Czech Rep. Germany Average OCDE Greece Denmark Japan Canada Netherlands Australia Iceland Finland Austria USA Trinidad & Tobago Norway Switzerland Sweden Venezuela Luxembourg Ireland Turkey UK

Key results • Relatively elastic demand provides opportunity to increase broadband penetration through regulatory initiatives • Significant price reductions will be needed to achieve policy goals • LAC households must make a much greater effort to afford broadband • Highly unequal income distribution means that subsidies and/or public access will still be needed to achieve universal service

Hernan Galperin hgalperin@udesa.edu.ar More information: www.dirsi.net Diálogo Regional sobre Sociedad de la Información