Post-2015 Consensus: Infrastructure Assessment, Auriol Fanfalone

Assessment Paper

Summary of The Targets From The Paper

Infrastructure
Target	Discounted Benefits ($B) 2015-2030	Discounted Cost ($B) 2015-2030	Benefit For Every Dollar Spent
Increase World fixed broadband penetration by three-fold from 2014 levels (from 10% to 30% in 2030)	$35,930	$1,735	$21
Increase Developing countries’ fixed broadband penetration by approx. three-fold from 2014 levels (from 6% to 20% in 2030)	$21,279	$1,031	$21
Increase World mobile broadband penetration by approx. three-fold from 2014 levels (from 32% to 90% in 2030)	$37,659	$2,203	$17
Increase Developing countries’ mobile broadband penetration by approx. three-fold (from 21% in 2014 to 60% in 2030)	$21,578	$1,260	$17
Increase World penetration of Fixed+ Mobile Broadband from 42% in 2014 to 100% 2030 (assuming to reach the target with 1/3 of fixed lines and 2/3 of mobile connections)	$38,050	$3,161	$12
Increase Developing countries’ penetration of Fixed+ Mobile Broadband from 27% in 2014 to 80% in 2030 (reaching the target with 1/3 of fixed lines and 2/3 of mobile connections)	$21,891	$2,431	$9
Universal fixed broadband penetration by the year 2030	$38,103	$7,343	$5
Universal mobile broadband penetration by the year 2030	$38,072	$2,523	$15

Summary 

Introduction: The importance of information communication technologies (ICTs) for development

Expanding affordable access to ICT infrastructure has become a priority for policy makers, both in developing and developed countries, as ICTs are important enablers for social inclusion and economic development. They also have an impact on economic growth and have strategic spillover effects on other sectors by creating jobs and fostering innovation. 

The 2008 OECD Seoul Declaration of the Future of the Internet Economy stated how ICTs, and in particular the Internet, will strengthen the capacity of countries to improve the quality of life of their citizens. ICT deployment was embedded in the MDGs and is in the draft post-2015 agenda. However, the goals are expressed vaguely, so this paper considers some rather more specific ones.

Sustainable Development Goals and targets regarding ICTs

Millennium Development Goals and Post 2015 Agenda: where do the targets lie in terms of ICTs?

The International Telecommunications Union (ITU) uses two metrics – telephone lines and cellular subscribers per 100 inhabitants, and computers and Internet subscribers per 100 inhabitants – to track progress of MDG targets. As for the post-2015 agenda, the Centre for International Governance Innovation and Korea Development Institute has suggested three targets:

• Universal access to modern telecommunication networks to eliminate exclusion from the new global digital society.
• Universal connectivity in recognition of the growing risk of “digital divide” among and within nations.
• Affordable access to technology and knowledge for enhanced inter-businesses and inter-governmental cooperation.

More recently, the Open Working Group on Sustainable Development Goals has proposed Goal 9c: “significantly increase access to ICT and strive to provide universal and affordable access to Internet in LDCs by 2020.” The direction is clear, but the question remains as to what constitutes ‘significant’.

Reducing the scope of ICT targets and focusing on Broadband: Why Broadband?

To reduce the various proposals to a target which can be the subject of a cost benefit analysis, the suggestion is to focus on broadband indicators. This is firstly consistent with the latest (June 2014) draft of the Sustainable Development Goals and secondly is a transformative platform that impacts the economy beyond the ICT sector.

Broadband can contribute to GDP growth (a 2009 study showed a 10% increase in broadband penetration increased GDP growth by 1.34% in low to medium income countries), create jobs and increase consumer surplus. Despite considerable government investment and tremendous growth in broadband availability in recent years, major differences persist in the availability of broadband among developed and developing regions of the world, thus, this ‘digital divide’ remains a key impediment to development. To state an example, mobile broadband penetration was only 21.1% in developing countries in 2014 compared to 83.7% in the industrialised world.

What is Broadband and how can it be measured?

There are a number of definitions of broadband; for example, the World Bank defines it as an ecosystem which includes users and applications in an interconnected high capacity communications network. Services can be measured via three dimensions: availability, quality and affordability. Comparing affordability of broadband services among countries, that is, the pricing dimension, is made more difficult by the complexity of broadband service plans and the improving quality of services over time.

Existing Broadband Targets proposed by the ITU Broadband Commission in 2010 and where we stand in terms of achieving them by 2015

The Broadband Commission established the following targets to achieve Millennium Development Goal 8F (“In cooperation with the private sector, make available the benefits of new technologies, especially information and communications technologies”) by 2015:

• Target 1: Making broadband policy universal. By 2015, all countries should have a national broadband plan or strategy or include broadband in their Universal Access/ Service Definitions.
• Target 2: Making broadband affordable. By 2015, entry-level broadband services should be made affordable in developing countries through adequate regulation and market forces (amounting to less than 5% of average monthly income).
• Target 3: Connecting homes to broadband. By 2015, 40% of households in developing countries should have Internet access.
• Target 4: Getting people online. By 2015, Internet user penetration should reach 60% worldwide, 50% in developing countries and 15% in LDCs.

Unfortunately, it seems that none of these targets will have been achieved by the deadline.

Proposed Broadband Targets for the Post-2015 Agenda

Based on existing and proposed ICT targets, and focussing specifically on broadband, this paper examines the following specific targets for 2030:

1. Increase world fixed broadband penetration by three-fold from 2014 levels (from 10% to 30% in 2030)
2. Increase fixed broadband penetration around three-fold in developing regions (from 6% to 20%)
3. Increase world mobile broadband penetration around three-fold (from 32% to 90%)
4. Increase mobile broadband around three-fold in the developing regions (from 21% to 60%)
5. 100% Fixed + Mobile broadband penetration (World and developing regions)
6. Increase penetration of Fixed + Mobile Broadband from 27% to 80% in developing regions
7. Universal Fixed Broadband penetration
8. Universal Mobile broadband penetration

Cost Benefit Analysis of deploying broadband infrastructure

This is a challenging exercise, firstly because it is extremely difficult to calculate global costs of broadband deployment, given the range of technological options available, and secondly measuring the benefit is a difficult task because broadband has become a General Purpose Technology whose effect on growth has been the subject of an entire research agenda over the past few years. Also, given the rapid rate of technological change, there is an inherent degree of uncertainty to ICT infrastructure investments, which may further complicate the cost benefit analysis.

With these caveats in mind, a cost benefit analysis is attempted which aims to illustrate its benefits, in a form to make it comparable with other development goals.

Measuring the Benefits of Broadband Deployment

A recent report for the European Commission estimated that the Digital Agenda for Europe should create around two million jobs and €28.6 billion in terms of consumer benefits. In particular, the study showed that the 27 Member States can expect to receive cumulative benefits of between €200 billion and €600 billion in the period 2012 to 2020, representing a benefit cost ratio of 2.7-2.9.

The present study focusses on the impact of broadband infrastructure on GDP growth, which occurs via three main channels. First, the deployment of broadband technology across firms improves their productivity given the adoption of more efficient business processes. Second, extensive broadband deployment increases innovation through new applications and services. Third, broadband leads to more efficient functioning of firms by maximizing their reach to labor pools or access to raw materials or consumers.

A World Bank study showed that broadband had the largest impact on growth out of a range of telecommunication investments and that low and medium income countries are in the best position to take advantage of this. However, more recent work suggests that there is a certain critical threshold (e.g. 30%) above which the economic impact is larger.

There is a general consensus that there is a positive relationship between broadband and growth, although there has been some debate about causality. From the evidence of a number of studies, the impact of each 10% increase in broadband penetration on GDP growth varies from 0.08 to 1.5%.

Measuring the costs of broadband deployment

There are two primary components of a broadband network: the access network connecting users to the nearest network node and the core and backhaul network, making connections over large distances. User access may be via copper, fibre, cable or wireless. The core network is a national backbone, usually fibre, and the backhaul network connects the local access network to the national backbone.

The cost of broadband deployment depends on the speed targets. Costs will vary both with technology and population density, with fibre being more cost effective where population density exceeds 3,000 inhabitants per square kilometer, and wireless being less expensive for lower densities. A 2010 study found the cost per line to reach 50% of German households with ultra-fast broadband was €1150-1425, while providing a new 1 Mbps line in Brazil would cost $450.

There are many challenges to establishing a cost per line deployment at a global level, given that there is highly unlikely to be a “one size fits all” infrastructure cost. Despite these difficulties, our analysis uses as a benchmark the cost per line calculated in studies related to network deployment investments within the context of National Broadband plans and digital agendas.

A simplified methodology to assess the Cost Benefit Ratio

In order to calculate the stream of costs and benefits and to bring them to present value, the first step is to calculate the expected change in terms of subscriptions that are embedded in each of the eight targets. This ranges from just over one billion new subscribers for target 2 (threefold increase in fixed broadband in developing regions) to nearly 7.5 billion to provide universal fixed broadband (target 7).

The incremental benefits of these increases in provision are mapped to the potential GDP growth impact using the elasticities from the literature covering three scenarios: conservative, medium and optimistic. The baseline is a world GDP of $74.9 trillion and a developing region GDP of $24.5 trillion in 2013 (World Bank data). The cost of providing lines is also covered in three scenarios. In the conservative scenario, the cost per fixed line is taken as €1500 ($1950) and in the medium case this is €1150 ($1460). In both cases, the cost per wireless connection is taken to be €500 ($640). For the optimistic scenario, the cost of a fixed line is taken as €600 ($770) and a wireless connection €400 ($510).

Results of the Cost Benefit Analysis

The benefit-cost ratios for the various scenarios are summarized in the table.

Given the rapid rate of change in the ICT sector, a 3% or 5% discount rate is low compared to what telecommunications companies take into account when they make investment decisions, which is why we also include an 8.8% rate more typical of the industry. However, for comparison with other studies in this series, the lower discount rates should be used.

Focusing just on the conservative scenario, which underestimates the impact of broadband on GDP and overestimates the costs of deployment, we still find that benefit-cost ratios are quite high. In all three scenarios, the targets related to the expansion of broadband penetration (either mobile or fixed) by three-fold in developing regions of the world exhibit the highest Benefit-Cost ratios (Targets 2 and Target 4). Targets 3 and 1 (expanding world mobile and fixed broadband threefold) also have high BCRs.

A valuable takeaway from these results is that, given the cost advantages that wireless technologies can have over fixed network deployment, universal penetration goals seem to be better reached through wireless deployment. The costs savings of deploying such a network are quite significant, with some studies suggesting that the cost per user of a wireless network may be one third that of a fixed network. The lack of existing fixed infrastructure also makes a strong case for wireless broadband as a cost-effective solution in India and other countries where mobile phone use is increasingly rapidly.

A McKinsey report for the World Economic Forum in 2010 estimated that bringing mobile broadband penetration in emerging markets to the levels of penetration of mature markets could add USD 300-400 billion to world GDP and translate into 10-14 million direct jobs in emerging regions. Many users in Africa and Asia are already accessing the Internet through a mobile device. In China, for example, three quarters of Internet users accessed it using their mobile phones.

A sensitivity analysis on the conservative scenario shows that assuming a higher elasticity of impact on economic growth improves the BCRs slightly, while assuming a medium growth rate for the economy or a lower cost of delivery can improve the ratio quite significantly.

Institutional framework as the foundation to reap the benefits of ICT

Some key enablers for ICT infrastructure development as a way to promote efficiency and extend ICT coverage are:

• Competition in the telecommunications service provider market
• Policies that foster the access and adoption of ICTs
• Development of applications and local content
• Government aid to extend ICT network coverage and provide services in underserved areas (i.e. investment in strategic parts of the network)

The main question is then how can governments help drive the cost of networks down while providing incentives for broadband deployment?

Specific policy and regulatory interventions to foster network deployment

Deployment of broadband infrastructure is an expensive undertaking given the large fixed costs that it entails, so any measures that reduce the cost of deployment should help public funds have a greater impact as well as rendering investment by private operators more attractive. In general, any policy or regulation that reduces deployment costs should maximize the social impact of the investment. These can include a regulatory framework which discourages anticompetitive behavior, efficient policies for interconnection and spectrum management efficient use of existing infrastructure.

For fixed networks, ensuring ease of access for new entrants opens up competition and reduces the cost of deployment. However, wireless networks are likely to be particularly important for extending coverage in the developing world, and there are specific factors which can keep the costs down. The first is efficient management and allocation of the wireless spectrum to increase competition. The second is sharing of infrastructure, particularly the passive sharing of towers, masts and cabinets, which is common across Europe.

Equally important is an institutional and regulatory framework to tackle anticompetitive behavior, with a well-designed independent regulator and Competition Authority. Both should have enforcement powers. Finally, although there is a place for active government involvement in the rollout of broadband coverage, there is always a risk of crowding out private investment. A framework is needed to prevent this; the European Commission put in place state aid guidelines for broadband investment in 2009 and there have since been a number of broadband support packages approved since then.

What type of broadband infrastructure should governments fund in order to drive network deployment costs?

An important role of government is to ensure that high-speed robust core backbone networks are available across the country. Many governments have opted to provide funds for international connectivity (i.e. submarine cables or satellite links), the core and backhaul network and internet exchange points, as well as investing in the access network in remote areas. International organizations can be platforms for sharing information and best practices regarding regulation and policies that work to provide incentives for network deployment. The donor community can also help with aid for investment in key parts of the broadband infrastructure.

Conclusions

The Cost Benefit Analysis conducted in this paper should be interpreted with caution as it is based in a series of over-simplifying assumptions regarding the cost per line per type of technology to achieve network coverage. With this caveat in mind, the targets related to the expansion of mobile broadband penetration by threefold at a global level or in developing regions exhibit the largest Benefit-Cost Ratio. In particular, the highest B/C ratio that is robust to different methodologies of assessing the benefits are related to expanding mobile broadband in developing regions by three-fold (i.e. Target 4, with a B/C ratio of 14.41 and 21.74, depending on the way the benefits are assessed). Secondly, expanding world mobile broadband penetration threefold (from 32% to 90% as expressed in Target 3) also exhibits a large B/C ratio of 13 and 29.42 (according to both methodologies to assess the benefits).

Given the cost advantages that wireless technologies have over fixed technology, targets aimed to achieve the universal availability of broadband seemed to be better off if they are reached through wireless technology, (at least in the developing regions of the world.) However, these results depend on the cost assumptions and the available indicators chosen to measure these targets. It should be stressed that any network deployment targets should be met with the principle of technology neutrality.

Finally, there are some actions governments can take in order to drive the cost of deployment down as well as fostering incentives for network deployment such as putting in place policies and regulations that improve market competition and provide incentives to invest in network deployment by the private sector. The combination of institutional conditions in conjunction with any type of state aid will help achieve ICT infrastructure objectives by the year 2030. In this sense, institutional and regulatory framework conditions provide fertile ground for ICT infrastructure deployment.