Jan 2012 – The Urban-Rural Digital Divide...Wider and Wider?
The UK’s policy objectives for superfast broadband speeds and coverage are too modest. If they’re achieved, the gap between urban and rural bandwidths will continue to increase, rather than decrease, as innovation and competition will drive further improvements for urban broadband users over the next decade. This matters, because superfast broadband will become a key enabler for future local and national economic growth.
Over the last 15 years, the headline downstream bandwidths of mass market broadband services have increased exponentially, at about 75% per annum, for high-end users in the UK’s urban areas (somewhat faster than the 50% annual growth predicted by ‘Nielsen’s Law’). Vigorous competition between BT, Virgin Media (and its predecessors NTL and Telewest) and Local Loop Unbundlers has led to the introduction of ever-higher bandwidths in those areas where there is a high density of demand.
Meanwhile, rural areas have typically been the last to get new broadband services, as their lower population densities (and lack of infrastructure-level competition) make rural broadband investments less commercially attractive for operators.
Policy-makers have sought to address the emerging urban-rural digital divide. Initially, the aim was for basic broadband (taken, then, to be 0.5Mbit/s) to be ubiquitous by 2005. More recently, the objective has been set of having the best superfast broadband in Europe, with targets of 2Mbit/s being available to all and superfast broadband (over 24Mbit/s) being available to 90% of the population by 2015, and ‘recognising’ the European target of 30Mbit/s for all by 2020. Hence, the policy targets have also increased exponentially over time, as shown in the chart below (note the logarithmic vertical axis).
Figure 1: High-end urban bandwidths and ‘ubiquitous bandwidth’ in the UK
Source: SQW
The problem lies in the difference in the growth rates of urban bandwidths and of policy targets for ‘ubiquitous bandwidths’, and the implications of these two curves for the digital divide are rather profound: the most rural parts of the UK would progressively fall further and further behind their urban counterparts (if current targets are achieved). So, whereas first-generation broadband interventions ensured near-ubiquitous 0.5Mbit/s connectivity by the end of 2005, five years after this first became available in urban areas (by which time urban areas had moved on to having speeds 20 times faster), achieving near-ubiquitous 30Mbit/s connectivity by 2020 would get such services out to the most rural areas 12 years after they first became available in urban centres (by which time, urban areas are likely to have speeds available in the order of 1,000 times faster).
Some will say that it is fanciful to expect affordable urban bandwidths of 10Gbit/s or more by 2020, pointing to a lack of current demand for such speeds. But the experience of the last 15 years has been remarkably consistent: the exponential improvements in the underlying technologies of electronics (Moore’s Law observes that the density of transistors on integrated circuits doubles every two years) and photonics (Butter’s Law suggests that the amount of data coming out of an optical fibre doubles every nine months), combined with intense competitive pressures, have led to the introduction of higher and higher bandwidths, the prices for which rapidly reduce to levels that are affordable for most households and businesses. We also note that new entrants such as Hyperoptic, CityFibre Holdings and Fluidata/Independent Fibre Networks Limited have recently announced launches of 1Gbit/s services in the UK (Hyperoptic, for example, offering a 1Gbit/s service for £50 per month to users in multi-occupancy buildings). And back in 2010, Verizon trialled a 10Gbit/s service over its FiOS network in the US, using a GPON system developed by Alcatel-Lucent.
It may be that this increasing divergence is simply unavoidable, given the differing dynamics of cost and competition in urban and rural areas, and it may not present a sufficiently large socio-economic problem to justify more ambitious intervention. On the other hand, we note that cloud-based computing is on course to become increasingly mainstream for micro, small and medium enterprises, and this could be a ‘killer application’ for much higher business bandwidths, with significant economic consequences for those areas lagging behind in broadband speeds.
Local authorities, Local Enterprise Partnerships, devolved administrations and central government need to consider whether their objectives for superfast broadband are genuinely aligned with their rationale for intervention (i.e. addressing the digital divide and stimulating local economic growth), given the ongoing market developments in urban areas. In SQW’s experience of developing, appraising and evaluating broadband interventions, and assessing their economic impact, we find that the best value for taxpayers’ money comes from those interventions that have the strongest links between their rationale and their specific objectives, and that take explicit account of the fact that the broadband market is – and will continue to be – highly dynamic.