noTE no. 5 ­ may 2006GRIDLINES Sharing knowledge, experiences, and innovations in public-private partnerships in infrastructure Expanding the frontiers of telecom markets through PPP in Peru Lessons for pro-poor initiatives Robert Stephens, Jorge Bossio, and Jean-Christophe Ngo P eru introduced private participation in that uses competitive public bidding to award telecommunications in the early output-based aid (OBA) subsidies to private 1990s, along with wide-ranging operators willing to invest in underserved reforms fostering competition and inde- communities. The OBA subsidies, paid on the pendent regulation. These efforts attracted basis of performance (or output), are designed some US$8.3 billion in private investment in to make projects attractive to private operators. the sector in 1990­2004. To help bring tele- A recent study, supported by a PPIAF grant of phone service closer to Peru's poorest and US$318,640, evaluated the first generation of most isolated areas, where people still had Peru's OBA projects. This note reports its main to travel some 56 kilometers on average to findings.1 reach a pay phone, a pioneering fund offered subsidies to attract investment by private The goal: universal access operators. Initial efforts led to impressive Peru set out to liberalize telecommunications achievements, though slow implementation in the early 1990s as part of a broader program left room for improvement. A PPIAF-funded of infrastructure privatization. In 1994 a land- mark telecommunications law established a new assessment of the first projects helped regulatory agency for the sector, Osiptel, while design the next generation of initiatives-- assigning the Ministry of Transport and Commu- and pointed to lessons for other developing nications the power to grant concessions and countries. licenses. The law also created a rural telecommu- Peru has some 67,500 rural towns with fewer nications fund, Fitel. Operating under Osiptel, than 3,000 inhabitants--typically located in the fund collects a 1 percent assessment on the its highland mountains and jungle regions and gross revenues of all operators to finance univer- among the country's poorest and most isolated sal access programs. To ensure that isolated and rural communities. Until recently the people unserved rural communities would benefit from living in such communities had to travel an aver- the sector liberalization, Fitel set as a goal for age 56 kilometers to reach the nearest pay phone, 1999­2003 to install pay phones in 5,000 rural sometimes by foot. A series of public-private communities and introduce Internet access in initiatives reduced that distance to an average 500 district capitals. 5.7 kilometers, shortening the travel time from a day or more to at most several hours. Robert Stephens is a telecommunications specialist at the World Bank and was team leader of the study. Jorge Bossio PUBLIC-PRIVATE INFRASTRUCTURE ADVISORY FACILITY How did Peru bring telephone service closer to is a staff member of Fitel, Peru's universal access agency. its rural communities? Peru (along with Chile) Jean-Christophe Ngo is PPIAF's regional program leader for pioneered the concept of a universal access fund Latin America and the Caribbean. Helping to eliminate poverty and achieve sustainable development through public-private partnerships in infrastructure PUBLIC-PRIVATE INFRASTRUCTURE ADVISORY FACILITY Table 1 The first generation of Fitel projects: basic indicators and achievements Indicator Fitel I Fitel II Fitel III Fitel IV all projects Rural communities covered 213 2,170 2,520 1,614 6,517 Population benefiting 140,000 1,600,000 2,100,000 2,900,000 6,740,000 Total subsidy (US$ millions) 5.1 12.1 30.7 11.4 59.3 Subsidy per site (US$) 23,937 5,575 12,163 7,061 8,266 Year of tender 1998 1999 2000 2001 Year of full service 2000 2003 2003 2004 Pay phones installed 214 2,208 2,461 1,616 6,499 Telecenters installed 0 260 255 0 515 Average distance to pay phone in project communities (kilometers) Before project 90 54 24 60 56 After project 5 8 4 6 5.7 Source: Intelecon and apoyo Consultoría. What Fitel achieved fic, the average was lower than expected, in part In 1998­2001 Fitel held successful bidding because cellular networks encroached into some rounds, or tenders, for four OBA projects target- of the areas they served. These areas, previously ing different geographic areas. The bidding was deemed unattractive to cellular companies, based on the lowest subsidy operators would gained appeal with technological innovation, require to provide pay phones, limited Internet the falling costs of cellular networks, and the access, or both. The subsidies offset part of the demonstration by the Fitel pay phones that Peru's first up-front capital investment costs and, in a few some rural communities had enough demand generation cases, a small and declining share of the operating costs, and are paid as operators meet milestones to sustain commercial operations. of projects in construction and service. · Differences in traffic patterns. The subsidy brought These initial tenders mobilized on average more needs had been estimated on the basis of the pay phone than twice as much in private investment as traffic patterns of the dominant carrier's rural access Fitel paid in subsidies. They exceeded Fitel's pay phones. Located in less isolated communi- to almost initial targets, benefiting 6,517 rural communi- ties, these generated more than three times as ties and bringing access to pay phones to about much traffic as the Fitel phones. 7 million 6.74 million people (table 1). Today 82 percent people of Peru's population is estimated to live within · Unforeseen regulatory changes. Enacted after 5.7 kilometers of a pay phone. Another notable the tenders, new interconnection and tariff achievement: the winners of the tenders were not regulations had the unintended effect of further the largest telecommunications operators but new, undermining the financial plans of the OBA smaller operators using satellite technology. operators by raising the interconnection charges Where Fitel fell short the Fitel operators paid to larger operators for Alongside the achievements, Fitel's first genera- terminating (receiving) telephone calls. tion of projects also had several shortcomings. Second, designing, approving, tendering, and First, at least one operator proposed a subsidy that implementing Fitel projects took a long time, turned out to be too low, for several reasons: sometimes up to four years. The longest delays · Traffic volumes lower than expected. were due to the requirement that the projects be Although Fitel networks carried growing traf- approved by multiple government agencies. Expanding the frontiers of telecom markets through PPP in Peru Third, Peru still has among the lowest teledensity · Fitel began to allow rural operators to relo- (fixed and mobile lines relative to population) cate some of their satellite pay phones from in Latin America, in large part because of the communities where cellular networks had been country's challenging geography and low rural installed to other communities. incomes. But even taking into account household incomes and ability to pay, conservative estimates · To speed the preparation of tenders and the suggest that teledensity could be raised to at least use of OBA funds, the Ministry of Transport 25 phones for every 100 people in a couple of and Communications in December 2004 began years, with much of this increase in rural areas. allowing private operators to propose universal Satellite-based pay phones proved effective in access projects to Fitel. Proposals could request improving access during the first generation of Fitel subsidies, with the cap for a project set by Fitel projects. But a second generation could help the 1 percent annual contribution the sponsoring stimulate the expansion of cellular networks into company makes to Fitel. If Fitel finds a project Cellular rural communities, with an even greater impact to be cost-effective, it will authorize implementa-networks on teledensity. With technological and market tion and allocate the company's tax contributions to help offset the investment costs. innovations, cellular phones have become the could have an most widely used communications device for · Fitel published a proposal for an OBA tender even greater poor people and could be commercially sustain- aimed at expanding cellular coverage in 18 rural impact in able in a growing number of rural communities. corridors that border existing cellular networks rural Peru by establishing 81 new base stations--a proj- Finally, the lack of terrestrial fiber-optic cables ect that would double the number of rural in Peru's interior poses one of the biggest chal- communities benefiting from Fitel initiatives.2 lenges in developing rural telecommunications. A Investment was estimated at US$20.3 million, terrestrial backbone to the interior is too expen- including a subsidy of up to US$11.6 million sive for a single company to finance on its own, to make the project commercially viable and and a perception of insufficient demand further attractive to private operators. This project discourages private operators from pursuing such marks a shift in Fitel's philosophy, from a sole an investment. As a result, operators wishing to focus on increasing community access (univer- provide service in rural areas must use satellite- sal access) to a broader approach also seeking to based networks, which are expensive and provide increase phone subscribers (universal service). limited connectivity. · The legislature adopted a new law requiring New initiatives launched that infrastructure using public rights of way During and shortly after the PPIAF study, be shared. The aim is to encourage electric- the government of Peru and private operators ity companies building transmission lines, for launched new initiatives linked to the study's example, to allow telecommunications compa- findings and recommendations: nies to use these lines (subject to reasonable · To help offset the effects of the unexpected commercial arrangements). The price would traffic patterns and changes in interconnec- be a fraction of the cost of building single-use tion tariffs on the financial viability of Fitel infrastructure. This is an important first step in operators, Osiptel in 2004 allowed asymmet- creating incentives for constructing a backbone to the interior. Fitel and the Ministry of Trans- ric interconnection and tariff arrangements port and Communications are also designing between the Fitel operators and others. These new initiatives aimed at stimulating the devel- enable Fitel operators to pay less to terminate opment of backbones to the interior. their outgoing calls on the networks of larger (fixed line and cellular) operators and to charge lessons for pro-poor initiatives more to terminate incoming calls from those Developing effective ways to extend access to operators on their own networks. In addition, telecommunications to poor and remote commu- the Fitel operators took measures to stimulate nities is the biggest challenge Fitel confronts. The traffic and reduce costs. low incomes, the small size of communities, and the large investments required because of the · Governments should find ways to ensure that inhospitable terrain are all key issues in designing universal access funds are used in a timely way. viable options. Fitel's experience in dealing with While universal access programs should maintain these issues points to lessons for other countries financial discipline, they should avoid long delays designing pro-poor initiatives or seeking to reach in the review process, which can undermine the isolated communities: rationale and support for OBA programs. · Governments should set measurable and Notes achievable goals for a time frame of two to 1 For details on the design of Peru's first two OBA projects, see four years for their universal access programs, Cannock (2001); Navas-Sabater, Dymond, and Juntunen (2002); periodically updating the goals as they are met and Wellenius (2000). and as markets and technology evolve. 2 Shortly after the publication of the project proposal, the Ministry · Universal access funds can be most effective and of Transport and Communications allowed Telefónica of Peru sustainable if they create incentives for private to acquire BellSouth (the country's second largest operator) provision of services on a commercial basis. and permitted Telmex to enter the market as a new mobile · Imposing a 1­2 percent assessment on operators' operator. Both companies committed to expand cellular coverage revenues is an effective and transparent mecha- to many of the communities targeted by the new Fitel project. nism for financing universal access funds. The ministry's decision was based on findings by the PPIAF- funded study and the Fitel expansion project. As a result of this · Making the universal access program part of development, Fitel has not tendered the proposed expansion the sector regulator, rather than a stand-alone project and is evaluating similar tenders for other areas. agency or a line ministry, reduces political interference in the use of funds and makes it References easier to introduce critical regulatory changes Cannock, Geoffrey. 2001. "Telecom Subsidies: Output-Based to support the program. Contracts for Rural Services in Peru." Viewpoint series, Note · Sound regulatory measures can support univer- 234. World Bank Group, Private Sector Development Vice sal access. Allowing rural operators asymmetric Presidency, Washington, D.C. interconnection arrangements and greater Navas-Sabater, Juan, Andrew Dymond, and Niina Juntunen. 2002. flexibility in setting prices can sharply reduce Telecommunications and Information Services for the Poor: Toward a subsidy needs and improve the operators' Strategy for Universal Access. World Bank Discussion Paper 432. financial viability. Washington, D.C. · OBA subsidies are an effective use of universal Wellenius, Bjorn. 2000. "Extending Telecommunications beyond the access funds because they attract significant Market: Toward Universal Service in Competitive Environments." additional private investment. But adequate Viewpoint series, Note 206. World Bank Group, Private Sector safeguards are needed to ensure that bid Development Vice Presidency, Washington, D.C. GRIDLINES proposals for subsidies are based on realistic and commercially viable Gridlines share emerging knowledge on PPP and give an overview of a wide assumptions. If unforeseen market selection of projects from various regions of or regulatory developments the world. Past notes can be found at www. occur, the universal access ppiaf.org/gridlines. Gridlines are a publication of PPIAF (Public-Private Infrastructure Advisory agency should be willing to Facility), a multidonor technical assistance consider reasonable propos- PUBLIC-PRIVATE INFRASTRUCTURE ADVISORY FACILITY facility. Through technical assistance and als from operators to adjust knowledge dissemination PPIAF supports the efforts of policymakers, nongovernmental organizations, contractual requirements. research institutions, and others in designing and implementing strategies to tap the full potential of private involvement in infrastructure. The views are those of the authors and do not necessarily reflect c/o The World Bank, 1818 H St., N.W., Washington, DC 20433, USA the views or the policy of PPIAF,the World Bank, the World Bank, Phone (+1) 202 458 5588 FAX (+1) 202 522 7466 or any other affiliated organization. PUBLIC-PRIVATE INFRASTRUCTURE ADVISORY FACILITY generAl eMAIl ppiaf@ppiaf.org web www.ppiaf.org