8. Market models for extending access

- Free ISPs – sharing call revenues
- IXPs, RXPs and international peering
- VOIP – Challenging the incumbents
- Cyber-cafes: Access without owning a PC

This section looks at how different types of business practice and regulatory models can affect internet growth. It takes four examples that affect different aspects of infrastructure: free ISPs, internet exchange points, Voice Over IP (VOIP) and the use of cyber-cafes where individual computer purchase is too expensive.

Free ISPs – sharing call revenues

The free ISP model originated in the UK, when the regulator OFTEL decided that the incumbent telecommunications company BT had to share revenues with ISPs.

The basis for free ISPS is simple: the ISP and the company split the revenue from the call made to access the internet, at an agreed rate. It allows the ISP to offer its service for free or nearly free and the company gets a percentage of the large amount of extra traffic generated. In the UK free services attracted millions of users and therefore pulled in a considerable amount of new traffic. The largest of these – Freeserve – is now one of the largest ISPs in the UK with 2.6 million subscribers.

When Egypt’s Ministry of Communications and Information Technology (MCIT) wanted to spread internet technology across the country, it announced a free internet access initiative based on the revenue-sharing model. In partnership with licensed ISPs, the government-owned Telecom Egypt has set up an estimated 15,000 ports, capable of serving 2 million internet users. Those taking up the service have to make a local phone call for access; no subscription is needed to access the net.

The user only dials his or her favourite ISP and gets access immediately. The cost of the call is shared between the ISP and the telecom operator, 70% and 30% respectively. The cost of one hour’s connection is one Egyptian pound (about US$ 0.22). There is a wide range of different ISPs who cover 90% of Egypt’s inhabited areas and the number of users is around 1 million, with most concentrated in the cities of Alexandria and Cairo.

Kenyan internet service provider Swift Global has launched a free ISP in conjunction with fixed-line operator Telkom (Kenya) and Interactive Media Services. Branded ‘Internet Direct’, the service allows users to access the internet without having to subscribe to an ISP. The revenue generated through this premium rate telephone number is shared between the three partners.

Another variant is the free ISP service from MTN in Uganda. Its fixed line customers can now dial up to the ISP of their choice and but do not need to pay monthly ISP charges or any initial connection fees. According to Erik van Veen its Chief Marketing Officer: “The basic premise of the product is that, in collaboration with ISPs, MTN has bundled the telecoms and ISP costs into a ‘per minute’ dial-up rate.”

Sharing of revenues in this way is one of a number of regulatory approaches that can be used to encourage the growth of internet usage. Another approach is unmetered usage. In 2001 UK regulator Oftel insisted that the incumbent telecommunications company BT offer unmetered access to the internet at a fixed price. It noted at the time that unmetered call durations are, on average, four times longer than metered call durations. Unmetered access requires available capacity but a fixed price, and it encourages greater usage.

In the developed world, users are accustomed to ringing a nationally available number at local call rates. However in many developing countries (particularly in Africa) numbers of this kind have not been introduced. In these circumstances, the unfortunate user has to pay long-distance rates to connect to the internet. A simple change in the regulatory framework can bring about its introduction.

IXPs, RXPs and international peering

The issue of the cost of international connections has already been raised. For example, in most African countries if you send email across town it makes a long and circuitous journey to North America or Europe before going back to its intended recipient. It costs money in international connectivity charges and gives latency problems (a fractional but sometimes problem-causing delay).

Local internet exchange points (IXPs) allow one country to route all (or most of) its internal internet traffic at a national level, thus saving money and adding speed to the connection. IXPs are the keystone of the entire internet economy: they interconnect different parts of the internet and they allow different ISPs to connect with each other, creating in effect a clearing house. Routing traffic the long way around is not an efficient way to use the network and thus the IXP mantra ‘keep local traffic local’ developed.

Local IXPs offer both internet users and ISPs a number of distinct advantages:

• They improve quality by speeding up connection times: there is a 200-900 millisecond delay in each hop the message makes across the system, compared to 5-20 milliseconds locally.

• They save money because the calling costs are all at a local level.

• They create new revenue opportunities because for example local content providers can start creating things like locally hosted web sites, a range of e-services and streaming. The latter would simply not be feasible if an international connection had to be made.

When Pacific Rim countries found in the 1990s they were paying far more than they were happy with for international connectivity, their approach was to say: Why do we need to get to the USA anyway? Most of our trade is national or regional. If we all peer our traffic within our countries and then within our regions we can dramatically reduce our connectivity costs.

Consequently, local and regional connectivity increased, international connectivity decreased and costs came down. In the process, the internet backbone providers found that the quality of connectivity that they were offering their customers in domestic markets was being reduced. The only way for them to maintain the quality was to establish Points-of-Presence (POPs) at the national and regional peering points in the Asia Pacific region. The internet backbone providers now bear the international connectivity costs, not the Asia Pacific ISPs (it is interesting to note that the Korean Internet Exchange Point is today the largest in the world). This shows how this approach can address international cost issues.

Africa is at a much earlier point in this cycle. The Kenyan IXP went online fully in 2002 with initially four ISPs but is now used by 10 of them. On an uncongested link, the latency is now 30-60 milliseconds. One rather conservative ISP decided that it would only require a 64k circuit to handle likely traffic and within two hours it was so packed that it got congested. Before it was established, international connectivity charges were nine times more expensive than local costs. Within a very short period of time Telkom Kenya had slashed its international call rates in half.

There are now six IXPs in Africa: South Africa, Zimbabwe, Nigeria (Ibadan with only two ISPs), Mozambique, Egypt, and the Democratic Republic of Congo (the last three opened recently). More are promising to open in the near future.

The major issue is one of trust. ISPs need to be able to work with their competitors and in some countries this level of trust has not yet been established. As Brian Longwe of the African ISP Association told a workshop at the Southern African Internet Forum: “Getting any IX/ peering arrangement off the ground is 10% technical work and 90% socio-political engineering.” He also pointed out the importance of getting (“written”) regulatory support. Setting up a local IXP is neither costly nor difficult.

Once there is enough traffic, continental exchange points develop. The more you aggregate traffic, the better the deal you will get. For example, if Africa were to develop a continental peering point it would not have to pay to interconnect to the rest of the internet. It would be of a similar standing in the network hierarchy to a Tier 2 ISP or Tier 1 internet backbone provider and would therefore ‘peer’ with other equivalents.

VOIP – Challenging the incumbents

The convergence between voice and data is a development that is having a particularly significant effect upon the relationship between ISPs and incumbent telecommunications companies, particularly in Africa.

VOIP stands for Voice Over Internet Protocol and is sometimes used as a shorthand for internet telephony: in other words calls made over the internet. International VOIP minutes are estimated to have tripled in 2002. International Data Corporation (IDC), a market research firm, says that by 2004, VOIP minutes (retail and wholesale) and revenues will grow to approximately 135 billion minutes and approximately $20.7 billion, respectively, representing estimated compound annual growth rates of over 100%. Beyond cost savings, IP technologies will further the potential for the internet to become the preferred medium for both communications and commerce.

As a result, the internet is expected to carry a growing volume of US and international long distance voice traffic says IDC. Other analysts forecast that, by 2004, IP telephony will account for between 25% (Analysys) and 40%
(Tarifica) of the global international voice traffic (compared to an estimated 3% of the total in 2000).

In Africa, VOIP is used ‘illegally’ by ISPs and cyber-cafes to offer international calls to people at much cheaper rates than charged by telephone companies. Because there is an enormous gap between the cost charged for international calls and what it actually costs to buy the connection, ‘grey market’ operators are exploiting the price difference.

As this takes calls away from national companies, most governments and regulators try to police the market. For example there are periodic police raids and the confiscation of equipment in Kenya and Ethiopia. In Ghana at one point the government even jailed some ISP owners for a short period of time. VOIP calls are difficult (but not impossible) to detect unless there is a large volume of calls. As a result the grey market often makes up 10-15% of the market in most African countries. In the case of Ghana Telecom, the company has estimated that it is losing somewhere between US$15-25 million a year in international call revenues to the grey market.

Why does this matter? Because the internet is creating new (often not currently legal) ways of doing business in developing countries. Regulators will need to consider whether they open up the monopoly on international call termination, and whether they involve those currently in the internet business. In the long-term there will be a shift from analogue to digital calling that will in large part use the internet. Developing country telecommunications companies need to plan for this transition now. It brings some advantages: for example, the cost of switching equipment is cheaper and it may help ease network congestion.

Cyber-cafes: Access without owning a PC

As we will explore in chapter 9, the internet is a relatively expensive service to access for those with low incomes. In most developed world countries, large numbers of people have access to a laptop or PC, either at home or at work. But for those with low incomes or people on the move, cyber cafes that offered relatively cheap access were a natural extension of the industry. Places like Easyeverything offer internet access at US$1 per hour. Tourists and students formed a substantial part of the initial users attracted.

In countries where the cheapest PC (US$200-300) can be a substantial portion of the average annual earned in-come, the role of cyber cafes is particularly important. Without them, the number of people using the internet would be substantially smaller. Almost all African cities now have a wide range of internet cafes that offer access for between US$1 and US$5 an hour.

The cyber café allows users to control their own costs and enables them to walk in straight off the street without having to pay for their own machine. A similar approach to cell phones is the selling of pre-paid scratch cards for the internet. The user scratches to obtain a number, which he or she then dials. No dial-up account is necessary and the number allows the user a pre-allocated amount of time online.

In developing countries, the users of these cyber cafes tend to be young and are often former students who had access to the internet at university. They are mainly used for email: many users communicate with friends and relatives in the growing African diaspora worldwide. However there is increasing use of the internet, which may challenge existing values: young women in some African countries search for husbands in the developed world. Other uses are illegal: there is a growing level of internet card fraud where buyers use fake credit card numbers and get goods delivered to accomplices in the United States.

Cyber cafes are an excellent way of creating access in large population centres and are spreading to smaller towns, particularly in tourist areas. However in rural areas there are two particular problems. First, it is hard to get the density of people required to create a sustainable user base. Second, if it is a poor rural area, people there may not be able to afford using even quite low-cost facilities. These issues are explored in chapter 9.

The recurrent issue in policy terms is: who pays for the infrastructure and how? A company invests in the infrastructure and has to be able to make a return on its investment that will cover the capital invested and provide it with a reasonable rate of return.

Any investment will need to make this return over a particular period of time. So, for example, in former days a telecom company would look to make a return on its infrastructure investment in 10-15 years. In more recent times, these investment cycles are much shorter: often companies look to make a full return in five years or sometimes even less.

Often these calculations will not be completely transparent and it will be the regulator’s role to clarify with those seeking licences what the expected investment return cycle is. For example, it would be no good the regulator granting a licence for five years when the investment was only to be recouped in the seventh year.

All the models above apply different approaches to making the market deliver high levels of internet use. However, the interests of internet users and private companies are often different, and whilst the market can extend internet access to many people, it is clear that it is not enough on its own. When access is not profitable, it is not provided by the market, and social mechanisms are needed to extend access to the most disadvantaged sectors of society.