pproaches47637 F e b r u a r y 2 0 0 9 N o t e N u m b e r 2 4 Output-Based Aid in Telecommunications New Models for Universal Access in Latin America Arturo Muente Kunigami D uring the 1990s many Latin American countries developed universal access programs, financed through universal service funds, to increase telecommunica- tions access in rural and low-income areas. The mechanism most often used to allocate the funds was the minimum subsidy tender, with subsidy payments linked to predefined performance criteria (such as installation of telephones in rural areas) under an output-based aid (OBA) approach. After more than 10 years' experience, a study evaluated the performance of the universal service funds in Latin America, home to some of the first OBA Photo © World Bank programs in telecommunications. The aim was to identify lessons and best practices and help regulators define a new generation of tender, with subsidy payments linked to predefined universal access programs using OBA mechanisms. performance criteria (such as installation of telephones Until the early 1990s, providing access to basic tele- in rural areas)1--an output-based aid (OBA) approach. phone service in Latin America was usually the purview After more than 10 years' experience, a group of do- of a monopolistic state-owned operator. Coverage in nors and stakeholders decided to evaluate the perfor- rural and low-income areas, by at least a pay phone, mance of these programs through a two-year in-depth was often part of specific service obligations placed on study (Stern and Townsend 2006).2 The main objectives these (usually inefficient) incumbent operators. were to review the performance of telecommunications During the 1990s Latin America led a wave of funds, calculate the investment required to close the telecommunications reform that would later spread access gaps that remained, and provide recommenda- worldwide. Privatization of state-owned fixed-line op- tions to members of Regulatel (the Latin American fo- erators, along with gradual liberalization, brought ef- rum of telecommunications regulators) on developing fective competition and new services to the sector. But and implementing effective, targeted, and sustainable competition focused mostly on urban areas, leaving universal access programs, including OBA programs. rural and low-income areas behind. In response, many countries undertaking reforms developed universal access programs that focused on increasing access to 1Minimum subsidy tenders are open, competitive tenders voice service in these areas. Most of these programs that award public subsidies for a specific project to the have been financed through universal service funds company requesting the smallest amount. (USFs). These funds, usually financed by fees levied as 2The study was funded by GPOBA and the Public-Private a small percentage of revenue on all sector operators, Infrastructure Advisory Facility (PPIAF); the Economic are managed by the sector ministry or regulator. Commission for Latin America and the Caribbean (ECLAC); the Alliance for the Information Society of the European The main idea behind these funds is simple: since Commission (@lis); and Regulatel. This note consists mainly many rural and low-income areas are unattractive to of extracts from the report. private investors, the government would use smart subsidies to create enough incentives for a private op- Arturo Muente Kunigami is an ICT Policy Specialist in the World erator to serve them. The mechanism most often used Bank's Policy Division (part of the Global Information and to allocate the funds has been the minimum subsidy Communication Technologies Department). Supporting the delivery of basic services in developing countries pproaches Fund performance areas. By the end of 2005 they had financed more than 27,000 pay phones in nearly 13,000 rural and low- Telecommunications reform in the region produced income villages as well as 19,000 telecenters benefiting impressive results. Between 1995 and 2004 the num- nearly 10 million people. The funds allocated totaled ber of lines increased from 45 million to 365 million, almost US$300 million, leveraging up to four times this and teledensity (telephone subscribers per 100 people) amount in private investment. In some cases compe- from 10 to 50. New services, such as mobile telephony, tition among operators ended up in projects being covered most urban and even some rural areas. awarded with no subsidy requested. The main driver of this growth has been private Despite these impressive figures, universal service investment: between 1994 and 2005 the private sector funds showed some limitations, especially in more invested more than US$116 billion in telecommunica- complex projects. tions in Latin America. Most of this investment took place during and right after privatization, as operators Slow disbursement. By the end of 2005, 12 percent of sought to meet pent-up demand and, in some cases, the funds raised had been committed to projects fulfill build-out obligations. (42 percent excluding countries with a disburse- The USFs, created in 14 of the 19 countries stud- ment ratio of 0) (table 1). Of the 14 countries ied, helped mobilize investment in previously unserved with a universal service fund, 3 used or allocated Table 1. Disbursement ratio of universal access programs in Latin America as of December 31, 2005 Amount raised Committed or used Disbursement Country Fund (US$ millions) (US$ millions) ratio (%) Argentina FFSU -- n.a. 0 Bolivia FNDR 43.5 0.0 0 Brazil FUST 1,680.8 0.0 0 Chile FDT 30.0 30.0 100 Colombia FCM 448.6 166.0 37 Costa Ricaa None n.a. n.a. n.a. Cuba None n.a. n.a. n.a. Dominican Republic FDT 65.7 10.8 16 Ecuador FODETEL 1.0 0.0 0 El Salvador FINET 32.7 0.0 0 Guatemala FONDETEL 17.9 7.8 43 Honduras None n.a. n.a. n.a. Mexico FCST 25.3 25.3 100 Nicaragua FITEL 4.0 0.8 19 Panama None n.a. n.a. n.a. Paraguay FSU 13.0 12.5 96 Peru FITEL 143.1 45.1 32 Uruguay None n.a. n.a. n.a. Venezuela, R.B. de FSU 113.2 0.0 0 Total 2,585.0 298.1 12 Excluding countries with 713.8 298.1 42 disbursement ratio of 0 -- Not available. n.a. Not applicable. a. Costa Rica created a fund in 2008 (FONATEL) but it is not yet active. Source: Stern and Townsend 2006. Supporting the delivery of basic services in developing countries pproaches 90 percent or more of their fund and 11 used Lagging regulation. Since the sector reforms of the or allocated 45 percent or less, including 6 that early 1990s, new trends and new technologies used none. Slow or zero disbursement was due to have emerged. Rules and regulations have not problems such as jurisdictional or legal disputes kept pace. USFs are no substitute for periodically between fund administrators and other ministries updating sector regulations to promote invest- (mostly regarding the scope of activities the fund ment in rural areas. The analysis showed that as could cover), diversion of the funds to uses other long as the right regulation is put into place, part than those for which they were intended, and, in of the market could be served by private operators some cases, use of the funds as a contingency to without public funds. improve fiscal balances. Mixed track record on telecenters. Projects involving All these shortcomings stem from slow adjustment telecenters and Internet access need complementa- to an ever-changing environment. Most first-generation ry programs to ensure sustainability of the telecen- programs had clear, measurable, and achievable goals. ters and effective use of the Internet. The most suc- But as projects and demand have evolved, many have cessful such projects have been those linked from failed to update their objectives and lack a cohesive the outset to specific goals, such as e-education strategy with consistent goals and metrics. and e-government, and those involving beneficiary communities from early on. One example is the Recommendations from the study Informatics Training Centers project in the Do- minican Republic, led by the telecommunications The evaluation of the universal service programs led to regulator. Focused mainly on education, it signs a range of recommendations for stakeholders, which partnership agreements with private schools and can be grouped under three main headings.3 nongovernmental organizations that commit to operating a telecenter according to the regulator's Update and redefine universal access rules and providing classes during part of the day. The new generation of programs will need to take Lack of sufficient backbones. For projects to provide market and technology developments into account in bandwidth-intensive applications and integrated their design and implementation and to focus on both voice-data services, the lack of sufficient domestic supply-side and demand-side stimulation. They also backbones in rural areas becomes a critical bottle- need to sharpen their focus on improving access to neck. But the region has almost no experience with and productive use of the Internet. Key success factors backbone projects financed through USFs. Peru's will be clear policy goals, cross-sector coordination, government, through its fund, is trying to develop and active involvement of beneficiary communities. a tender for a rural backbone project that would Programs also need to add new objectives, includ- cover the Andean region, providing access to ing universal geographic coverage, allowing poor people broadband at relatively low cost and allowing the who can afford private phones (entrepreneurs, local emergence of smaller operators or more ambitious farmers) to get them, and universal community access. last-mile projects. Top-down approach. Traditional projects financed by USFs are supply centric and designed under a Improve the use of funds top-down approach. These projects rarely offer There are many entrepreneurial activities and pilot the flexibility required to satisfy multiple needs projects across the region. But because of regulatory from different communities: as projects become and financial bottlenecks, they find it hard to scale more application centric, buy-in from local com- up. To support these local initiatives, the USFs need to munities becomes essential. Leveraging local broaden the range of mechanisms for disbursement. initiatives calls for OBA-based mechanisms other Possible options include financing bottom-up and than minimum subsidy allocation. For example, demand-oriented initiatives, creating special proce- fund administrators could assign a share of their dures for operator-designed projects and pilots, and annual collection to projects that are presented even giving grants on a first-come, first-served basis, or endorsed by community leaders on a first- through streamlined procedures, to partially cover in- come, first-served basis, as long as they meet preset criteria. 3For full results, see Stern and Townsend (2006, annex 1). Supporting the delivery of basic services in developing countries pproaches Box 1. Performance indicators for OBA schemes Putting in place service targets for existing opera- tors in areas considered to be commercially viable. To be effective, performance indicators to trigger payments under OBA schemes in rural telecommunications should: Reducing regulatory taxes and fees, especially spectrum fees in rural areas. · Focus on the needs of users in rural and remote regions. Lifting restrictions on the use of certain radio · Be quantifiable and calculated according to a clearly spectrum bands in rural areas, deregulating voice- defined formula. over-Internet protocol (VoIP), and simplifying · Avoid creating excessive administrative and financial licensing processes for rural operators. burdens for the operator. · Have penalties in proportion to the cost and inconvenienc- Facilitating interconnection and asymmetric tariffs es suffered by users. for small and rural operators. · Encourage the operator to improve quality and to invest. Helping small and rural operators obtain rights- · Take into account challenges involved in serving remote of-way for network facilities. and difficult locations. Requiring and enforcing infrastructure sharing be- tween telecommunications operators and electric- ity, gas, water, and road infrastructure providers. vestments. OBA schemes should be in place to ensure Impact and next steps service delivery, with carefully designed performance indicators for triggering subsidy payments (box 1). The study was presented in Lima, Peru, in November 2006, to an audience that included policy makers, reg- Implement reforms ulators, and fund administrators from the 19 Regulatel countries as well as representatives from companies As noted, legal and regulatory frameworks have not and governments outside the region. Published in early kept pace with technological and market innovations. 2007, the study has been widely used by the region's The study estimates a 10 percent market gap--that governments in reviewing their universal access policies. penetration could be increased by another 10 percent It has also provided insights for the World Bank's work with the right legal and regulatory framework. with countries in other regions. Some important bottlenecks are not directly under the control of regulators or sector ministries. Eliminat- Reference ing these will require strong leadership and effective coordination between ministries and between national, Stern, Peter A., and David N. Townsend. 2006. "New regional, and local governments. Key reforms include Models for Universal Access to Telecommunica- the following (not in order of priority): tions Services in Latin America." http://www. worldbank.org/gict/publications. About OBApproaches OBApproaches is a forum for discussing and dis- approaches. The case studies have been chosen seminating recent experiences and innovations and presented by the authors in agreement with in supporting the delivery of basic services to the the GPOBA management team and are not to be poor. The series focuses on the provision of water, attributed to GPOBA's donors, the World Bank, energy, telecommunications, transport, health, or any other affiliated organizations. Nor do any of and education in developing countries, in par- the conclusions represent official policy of GPOBA, ticular through output-, or performance-, based the World Bank, or the countries they represent. To find out more, visit www.gpoba.org The Global Partnership on Output-Based Aid Supporting the delivery of basic services in developing countries