This review looks at the local economic impacts of broadband. Broadband infrastructure and services are fundamental components of today’s internet. ‘Broadband’ is a generic term to describe always-on access to an internet service, and there are many varieties of broadband technology in use around the world today.
In the UK, the government would like universal coverage of ADSL by 2015, and – while it remains technology-neutral – is focusing its ‘superfast broadband’ efforts on ‘fibre to the node’ technologies. Even though ‘superfast’ in the UK is rather slower than is standard in countries like Korea, the government hopes that the economic impacts of faster broadband speeds will be substantial. Research commissioned by DCMS projects that fast broadband could add £17bn to the UK’s annual GVA by 2024.
To get a sense of where these economic benefits might come from, we need to look at the wider role of the internet, and information and communication technologies (ICTs) in general. Broadband internet, like many other ICTs, is generally considered to be a ‘general purpose technology’ in that it functions across many areas of economic and social life, and is an enabler of further innovation in those fields. Economic analysis of these ICTs looks at both direct effects (such as the growth of the digital economy and online industries) and indirect effects (on GDP, employment and wages, industry structure and the organization of work). This review focuses on these wider, indirect economic effects, which are of most interest to local policy makers concerned with whether or not to intervene to improve broadband provision in their areas. We also focus on studies of past impact rather than projections of possible future impact.
There are differing views of how ICTs and the internet shape economic progress. Enthusiasts argue that ICT adoption helps explain growth in productivity (particularly of labour) in developed countries. They point to US experience with computerisation in the 1980s as an illustration of what ICT can deliver. By contrast, sceptics suggest that these economic effects are overstated, and that the internet – in particular – is far less significant than is often assumed. These critics also emphasise the potential for ICTs to increase inequality and the economic cost of technological disruption.
There is a middle camp between these groups, which assumes that the internet and ICTs contribute to economic change, but that the extent of this change depends on how people and firms adapt to and innovate around technology. For example, a number of studies suggest that ICT investment only delivers productivity gains for firms who also introduce training for staff, and new ways of working.
In turn, this perspective suggests that takeup and use of broadband may be more important than simply availability of the technology. It also suggests that the firms most likely to gain from broadband might be the most likely to adopt the technology, which creates a challenge for evaluators (see discussion below).
What economic impacts of broadband availability might we expect to see? For firms and their workers, broadband should allow for efficiencies in production, both by lowering costs (for data storage, advertising or working with suppliers) and by enabling innovation (reaching new customers online, for instance, or employing big data analytics). Those productivity gains could translate into higher wages, and possibly higher levels of employment (although firms might well shed staff in response to technological change). At the same time, broadband may allow for more flexible patterns of work, including working at home or on the move. For some groups of people, such as those with caring responsibilities, more flexibility may increase labour force participation, which could in turn raise employment. More broadly, broadband may lower the barriers to starting a business, particularly in sectors like retail.
It is also important to recognise that there may be winners and losers from these changes. If broadband makes industries more competitive, some firms will lose staff or go out of business altogether. ICTs like broadband are complementary to human capital, so we might also expect skilled workers to gain more (in terms of wages). Broadband might also help accelerate automation, which penalises less skilled workers and those doing routine tasks. Increased labour force participation might raise overall employment levels, but that increase in labour supply might depress wages – or leave employment rates unchanged (if changes in participation outweigh numbers of people moving into work).
We should also expect to see broadband having different economic impacts in different types of places. Specifically, the academic literature suggests that the economic effects of broadband (and ICTs) in general may be bigger in urban areas. This is because, as discussed above, broadband and ICTs enable ‘production complementarities’, especially for skilled workers and knowledge-intensive firms. Both of these groups are – by and large – urban-orientated. However, for sparser, rural areas, broadband might still have some economic impact, and could deliver social gains. It is also important to note that while these predictions on likely relative benefits are evidence based it would be very helpful to know whether they hold in practice for broadband.
Given broadband’s general purpose nature, policymakers need to set these economic factors alongside social and environmental issues when deciding what their broadband strategy should be. In most countries, broadband infrastructure and service provision is market-led: the state’s role is to ensure a competitive market, and to regulate service levels and coverage. Critically, this means we have relatively few examples of explicit broadband delivery programmes, and thus, relatively few policy evaluations to draw on. In the EU, state aid rules may further limit national government’s freedom of manoeuvre. As we explain in Section 6, this also makes evaluation of broadband’s economic impacts more challenging than evaluation in some other areas of economic development policy.
Where governments do take a more active role, this takes two broad forms. The first is indirect provision that subsidises connections to existing networks, either through voucher schemes or loans. Related to this, governments may part-fund infrastructure in locations where there is market failure (such as some rural areas), or provide basic funding to incentivise research and development.
The second form of active policy is direct provision of broadband infrastructure or services. Some city governments have partnered with the private sector on projects such as Google Fiber, the YorkCORE project or public wifi systems. In other cases, such as Norway’s Public Broadband programme, a state-owned utility has rolled out a network across the country; in a few examples, such as Munich, city governments own telecoms companies that provide the network and services. In the UK, Hull had a wholly publicly owned telecoms company until 1997, when the council sold the majority of its shares; the rest were sold in 2007.
In this review we cover the literature on the economic effects of broadband, and look at policy evaluations where these exist. Because ICTs like broadband have a very wide set of impacts, we cluster these together, and focus on: productivity; firm entry and business numbers; employment and other labour market outcomes; income and wages; and other outcomes (e.g. sales and property prices).
We also discuss studies that compare impacts in different places, typically urban and rural locations. These are the effects that we think are most likely to be of interest to local policy makers when thinking about the potential impact of broadband on local economic performance.
Unfortunately we are not able to say as much as we would like about policy design issues such as the relative effects of indirect versus direct provision; ‘superfast’ versus ‘ultrafast’ technologies; policies that target SMEs or other types of firms; or scheme costs. This is because not many explicit policy evaluations exist, and not all those that do exist are robust enough to pass our quality filters. As recent NESTA research argues, further broadband demonstrator / local evaluation projects are essential to understand the economic and social impacts both of the technologies and how they are made available.
Definition of Broadband
Broadband and “High-speed internet” are generic terms used to describe the nature and the speed of access to an internet service. As we set out in the Introduction, broadband is always on and faster than dial-up; in practice, there are multiple broadband technologies (such as DSL, fibre and wireless) offering a range of speeds and modes of access. We used this definition to search for articles on the topic. We searched for articles on ‘broadband’ in order to capture studies evaluating fast internet services, in addition to searches for ‘wi-fi’ and ‘mobile internet’, in order to capture newer high-speed internet technologies which rely less upon extensive cabling and physical infrastructure.
Ultimately, we consider studies covering four types of broadband projects:
- Private provision: Our searches turned up a lot of studies looking at private sector broadband provision, for reasons we discuss in the Introduction.
- Direct public provision: for example, where government provides local councils with funding so that they can hire private companies to extend infrastructure, or where broadband infrastructure is state owned.
- Indirect public provision: typically, grants, loans or vouchers to connect users to existing networks.
- Demonstration projects: where a private company provides fast internet infrastructure and/or service, such as Netville or Google Fiber; or public-private partnerships, such as the free wi-fi scheme on the London Underground.