You may have recently noticed construction workers from different businesses digging up the roads/pavements near where you live. You may also have noticed them laying fibre optic cables. Why has this been happening? Does it make economic sense for different companies to dig up the same stretch of pavement and lay similar cables next to one another?
For many years the UK had one national fixed communication network that was owned by British Telecom (BT) – the traditional phone landline made from copper wire. This is now operated by OpenReach – part of the BT group but a legally separate division. In addition to this national infrastructure, Virgin Media (formed in 2007 from the merged cable operators, Telewest and NTL) has gradually built up a rival fixed broadband network that now covers just over 50 per cent of the country.
Although customers have only had very limited choice over which fixed communication network to use, they have had far greater choice over which Internet service provider (ISP) to sign up for. This has been possible as the industry regulator, Ofcom, forces OpenReach to provide rival ISPs such as Sky Broadband, TalkTalk and Zen with access to its network.
Expansion of the fibre optic network
Recent government policy has tried to encourage and incentivise the replacement of the copper wire network with one that is fully fibre. This is often referred to as Fibre to the Premises (FTTP) or Fibre to the Home (FTTH). A fixed network of fully fibre broadband enables much faster download speeds and many argue that it is vital for the future competitiveness of the UK economy.
Replacing the existing fixed communication network with fibre optic cables is expensive. It can involve major civil works: i.e. the digging up of roads and pavements to install new ducts to lay the fibre optic cables inside.
Over a hundred companies, that are not part of either OpenReach or Virgin Media O2 (the parent company of Virgin Media), have recently been digging up pavements/roads and laying new fibre optic cables. Known as alternative network providers (altnets) or independent networks, these businesses vary in size, with many of them securing large loans from banks and private investors. By the middle of 2023, 2.5 million premises in the UK had access to at least two or more of these independent networks.
After a slow initial response to the altnets, OpenReach has recently responded by rapidly installing FTTP. The business is currently building 62 000 connections every week and plans to have 25 million premises connected by the end of 2026. In July 2022, Virgin Media O2 announced that it was establishing a new joint venture with InfraVia Capital Partners. Called Nexfibre, this business aims to connect 5 million premises to FTTP by 2026.
Is the fibre optic network a natural monopoly?
Some people argue that the fixed communication network is an example of a natural monopoly – an industry where a single firm can supply the whole market at a lower average cost than two or more firms. To what extent is this true?
An industry is a natural monopoly where the minimum efficient scale of production (MES) is larger than the market demand for the good/service. This is more likely to occur where there are significant economies of scale. Digging up roads/pavements, installing new ducts and laying fibre optic cable are clear examples of fixed costs. Once the network is built, the marginal cost of supplying customers is relatively small. Therefore, this industry has significant economies of scale and a relatively large MES. This has led many people to argue that building rival fixed communication networks is wasteful duplication and will lead to higher costs and prices.
However, when judging if a sector is a natural monopoly, it is always important to remember that a comparison needs to be made between the MES and the size of the market. An industry could have significant economies of scale, but not be an example of a natural monopoly if the market demand is significantly larger than the MES.
In the case of the fixed communication network, the size of the market will vary significantly between different regions of the country. In densely populated urban areas, such as large towns and cities, the demand for services provided via these networks is likely to be relatively large. Therefore, the MES could be smaller than the size of the market, making competition between network suppliers both possible and desirable. For example, competition may incentivise firms to innovate, become more efficient and reduce costs.
Research undertaken for the government by the consultancy business, Frontier Economics, found that at least a third of UK households live in areas where competition between three or more different networks is economically desirable.
By contrast, in more sparsely populated rural areas, demand for the services provided by these networks will be smaller. The fixed costs per household of installing the network over longer distances will also be larger. Therefore, the MES is more likely to be greater than the size of the market.
The same research undertaken by Frontier Economics found that around 10 per cent of households live in areas where the fixed communication network is a natural monopoly. The demand and cost conditions for another 10 per cent of households meant it is not commercially viable to have any suppliers.
Therefore, policies towards the promotion of competition, regulation, and government support for the fixed communication network might have to be adjusted depending on the specific demand and cost conditions in a particular region.
Articles
Review
Questions
- Explain the difference between fixed and wireless communication networks.
- Draw a diagram to illustrate a profit-maximising natural monopoly. Outline some of the implications for allocative efficiency.
- Discuss some of the issues with regulating natural monopolies, paying particular attention to price regulation.
- The term ‘overbuild’ is often used to describe a situation where more than one fibre broadband network is being constructed in the same place. Some people argue that incumbent network suppliers deliberately choose to use this term to imply that the outcome is harmful for society. Discuss this argument.
- An important part of government policy in this sector has been the Duct and Pole Access Strategy (DPA). Illustrate the impact of this strategy on the average cost curve and the minimum efficient scale of production for fibre broadband networks.
- Draw a diagram to illustrate a region where (a) it is economically viable to have two or more fibre optic broadband network suppliers and (b) where it is commercially unviable to have any broadband network suppliers without government support.
- Some people argue that network competition provides strong incentives for firms to innovate, to become more efficient and reduce costs. Draw a diagram to illustrate this argument.
- Explain why many ‘altnets’ are so opposed to OpenReach’s new ‘Equinox 2’ pricing scheme for its fibre network.
Concerns have been expressed about the UK’s relatively poor record of upgrading broadband services so that households can receive ultrafast connectivity. Some commenters have argued that future economic growth prospects will be harmed if the UK continues to lag behind its leading rivals.
Much of the fixed line system that allows people to connect to broadband was originally installed many years ago for the land-line telephone network. The so called ‘final mile’ consists of copper-based wiring that is carried from street cabinets to the premises of the end-user. This wiring is transported via a huge network of telegraph poles and cable ducts (small underground tunnels).
In order for people to gain connectivity to ultrafast broadband this copper based wiring needs to be replaced by fibre optic cables. This is commonly referred to as Fibre to the Premises (FTTP). Unfortunately, the UK has a relatively poor record of installing FTTP. Japan and Korea were forecast to have 70% and 63% coverage by the end of 2015 as opposed to just 2% in the UK.
Why is the UK’s record so poor? Many observers blame it on the structure of the industry. In other network industries, such as those for gas pipeline and electricity grids, the business responsible for managing the infrastructure, National Grid, is a regulated monopoly. This company does not directly supply services to consumers using the network it is responsible for maintaining. Instead, customers are supplied by the retail sector of the industry, where firms compete for their business. This sector includes the so-called ‘big six’ (British Gas; npower; SSE; Scottish Power; EDF; E.On) and a number of smaller suppliers such as Ovo Energy and Ebico.
The structure of the fixed line telecommunications sector is very different. The company that manages the ‘final mile’, Openreach, is a subsidiary of BT. BT also competes with other Internet Service Providers (ISPs), such as TalkTalk and Sky, to supply broadband to customers using this network. Its market share of 32 per cent makes it the largest player in the broadband market. Sky and TalkTalk have market shares of 22 per cent and 14 per cent respectively. Virgin Media also supplies 20 per cent of this market using its own network of ducts and cables.
Given that in most cases ISPs such as Sky and TalkTalk are stuck with the network Openreach provides, BT may have limited incentives to invest. It can still earn a good return from its infrastructure of copper-based wiring and avoid installing expensive FTTP. Dido Harding, the chief executive of TalkTalk, argued that:
“We need to separate Openreach from the rest of BT to create a more competitive, pro-investment market”
Ofcom, in its recent review of the market, has taken a different approach. Rather than creating an entirely separate monopoly business to manage the network (i.e. splitting Openreach from BT), the regulator instead opted for a policy of encouraging competition between different suppliers that deploy fibre optic cables. It states in the report that:
“We believe competition between different networks is the best way to drive investment in high-quality, innovative services for customers.”
This competition could come from ISPs such as TalkTalk and Sky or other smaller network providers such as CityFibre and Gigaclear.
One major problem with this approach is that potential new entrants might be deterred from entering the market because of the very high initial costs involved in building a new network in order to deploy FTTP. In particular, the costs of digging up the roads and laying the ducts are considerable. Matt Yardley, author of a study on the industry, said:
“It is widely accepted that civil works such as digging trenches account for up to 80% of broadband deployment costs.”
One way of reducing these costs and encouraging more competition is to allow rival firms access to the existing ducts and poles that are currently managed by Openreach. Once access has been obtained, these firms could effectively rent space inside the ducts and lay fibre optic cables alongside the existing copper-based wiring. Vodafone reported that a similar policy in Spain had reduced its capital expenditure of building FTTP by 40 per cent compared with constructing its own network of ducts and poles.
Ofcom first introduced this type of policy in 2010 when it launched its Physical Infrastructure Access (PIA) initiative. Unfortunately it has proved to be relatively unsuccessful with very little demand for PIA from rival firms. The success of this type of policy will depend on a number of factors including (1) the prices charged by Openreach to access and rent space inside the ducts; (2) the simplicity of any relevant administration; and (3) the availability and reliability of information about the ducts. With this last point, key issues include:
|
|
| • |
Where they are located . |
| • |
How much space is available: i.e. is there enough space for firms to lay fibre optic cables alongside the existing wiring? |
| • |
What condition they are in: i.e. are they flooded or clogged up with sand and mud, which will involve expensive work to make them usable again?
|
Firms did complain about the pricing structures and bureaucratic nature of the administration process under the PIA scheme. However, their most significant concerns were about the uncertainty that was created by the lack of information about the ducts and poles. For example, analysts from the consultancy firm, Reburn, argued that if a firm contacted Openreach to try to obtain access to the network it was informed that:
“We don’t know what condition the ducts and poles are in. Please pay £10 000 for a survey. Also unfortunately we are rather busy and we can only start in six weeks.”
Matthew Hare, the chief executive of Gigaclear, argued that it was like going to a shop where the assistant says:
“Give me some money, and I’ll tell you whether you can have it or not.”
In response to these criticisms Ofcom has introduced a number of changes to PIA, which has been re-named Duct and Pole Access (DPA). In particular, it has imposed a new requirement on Openreach to create a database that provides information on the location, condition and capacity of its ducts and poles. The database must be made available to rival ISPs and network providers. DPA must also be provided on the same timescales, terms and conditions to all businesses including other parts of BT – this is referred to as ‘equivalence of inputs’.
The first big test of this policy is in Southend where City Fibre is hoping to deploy 50km of fibre optic cables using DPA. However, reports in the media have suggested that the initial surveys have found very limited capacity in some of the ducts, which would make DPA impossible.
It will be interesting to see how the trial in Southend progresses. If it is successful, then DPA may be viable for about 40 per cent of premises in the UK. If it fails, then Ofcom might ultimately have to force Openreach to be completely separated from BT.
Articles
How the gothic city of York became a broadband battleground The Telegraph, Kate Palmer (22/5/16)
City Fibre first to mount BT challenge after Openreach is told to share network The Telegraph, Kate Palmer (1/3/16)
Challenges as CityFibre Moot Using BT Cable Ducts in Southend-on-Sea ISPreview, Mark Jackson (2/5/16)
CityFibre to build pure fibre infrastructure for Southend Networking (5/4/16)
Ofcom tells BT to open up infrastructure to rivals The Guardian, Rob Davies (26/2/16)
Questions
- Draw an average total cost curve to illustrate the economics of building a network of ducts and poles. Label the minimum efficient scale.
- To what extent does DPA create a contestable market?
- For DPA to deliver productive efficiency, what must be true about the economies of scale of laying fibre optic cables?
- In the run-up to Ofcom’s review of the telecoms industry, many commentators described Openreach as being a natural monopoly. To what extent do you agree with this argument?
- What are the advantages of marginal cost pricing? What issues might a regulator face if it tried to impose marginal cost pricing on a natural monopoly?
- Using a diagram, explain how the network of ducts and poles might be a natural monopoly in rural areas but not in densely populated urban areas.
- Discuss how Ofcom has tried to increase the level of separation between Openreach and BT.
“There are ‘incredible economies of scale in cloud computing’ that make it a compelling alternative to traditional enterprise data centers.” According to the first article below, cloud computing represents a step change in the way businesses are likely to handle data or use software. Rather than having their own servers with their own programs, they use a centralised service or ‘public cloud’, provided by a company such as Microsoft, Google or Amazon Web Services. The cloud is accessed via the Internet or a dedicated network. It can thus be accessed not only from company premises but by mobile workers using tablets or other devices and thus makes telecommuting more cost effective.
There are considerable economies of scale in providing these computing services, with the minimum efficient scale considerably above the output of individual users. By accessing the cloud, individual users can benefit from the low average costs achieved by the cloud provider without having to invest in, and frequently update, the hardware and software themselves.
In the case of large companies, rather than using a public cloud, they can use a ‘private cloud’. This is hosted by the IT department in the company and achieves economies of scale at this level by removing the need for individual departments to purchase their own software and servers. Of course, the costs of providing the cloud is borne by the company itself and thus the benefits of lower up-front IT capital costs are reduced. This is clearly a less radical development and is really only an extension of the policy of many companies over the years of having centralised servers holding data and various software packages.
In autumn 2010, EMC Computer Systems commissioned economists at the Centre for Economics and Business Research (cebr) to quantify the full impact that cloud computing will have over the years ahead. According to the report, The Cloud Dividend:
The Cebr’s research calculates that €177.3 billion per year will be generated by 2015, if companies across Europe’s five largest economies continue to adopt cloud technology as expected.
The Cebr found that the annual economic benefit of cloud computing, by 2015, will be:
• France – €37.4 billion
• Germany – €49.6 billion
• Italy – €35.1 billion
• Spain – €25.2 billion
• UK – €30.0 billion
Will the ability of cloud computing to drive down the costs of IT mean that a new revolution is underway? Just how significant are the economies of scale and are they likely to grow as cloud providers themselves grow in size and experience? The following articles look at some of the issues.
Articles
Reports and information
Questions
- What specific economies of scale are achieved through cloud computing?
- Why might the minimum efficient scale of cloud computing services be above the level of output of many companies?
- What are the downsides to cloud computing?
- How would you set about assessing the statement that we are on the brink of a fundamental revolution in business computing?
- Why are customer-heavy sectors, such as financial services, utilities, governments, leisure and retail, expected to buy into the concept fastest?
- How can product life cycle analysis help to understand the stages in the adoption of cloud computing?
You might think that small environmentally-friendly companies would be moving into the green energy market: that setting up a wind farm, for example, would be a perfect business opportunity for a small company. In fact, the big companies are taking over this market. As the Der Spiegel article below states:
Europe’s wind energy sector is currently experiencing a major transformation. New massive offshore wind parks are soon expected to crop up off Europe’s coastline. Big companies like Siemens and General Electrics are increasing their stakes in a market worth billions. But experts warn that a new energy oligopoly may soon emerge.
So what is it about the wind energy market that makes it suitable for an oligopoly to develop? The two articles explore this question.
Winds of Change Der Spiegel, Nils-Viktor Sorge (1/11/10)
GE and Siemens Outpacing Wind Pioneers, Becoming Clean Energy’s “New Oligopoly” Fast Company, David Zax (2/11/10)
Questions
- What market failures are there in the wind energy market?
- What barriers to entry are there in the wind energy market?
- What economies of scale are there in this market?
- How are changes in this market affecting the minimum efficient scale of companies?
- Would there be room in the market for enough competitors to prevent collusion?
- How might the authorities prevent (a) open and (b) tacit collusion in the wind energy market?
- Do small wind energy companies have any market advantages?
