We are developing the social individualist meta-context for the future. From the very serious to the extremely frivolous... lets see what is on the mind of the Samizdata people.

Samizdata, derived from Samizdat /n. - a system of clandestine publication of banned literature in the USSR [Russ.,= self-publishing house]

Some more thoughts on African (and other) mobile phone networks

Brian recently wrote a piece about the importance and usefulness of mobile phones in poor countries, particularly in Africa. I couldn’t agree with him more, but there is another interesting story in just where the expertise and money to build these African networks are coming on, and it is an oddly positive story.

But first, a seeming digression. When mobile phones came along in the early to mid 1980s, there were generally two patterns of licensing. Firstly, there were countries (eg Australia) where the incumbent telephone monopoly was given a monopoly on the new technology. Secondly, there were countries (eg Britain) where two mobile companies were licensed. One of these was usually the incumbent telephone monopoly, and the second was usually a new company that was brought into being to provide the new service. It is worth remembering that few people at this point expected that there would be a large market for mobile phones, so often these second players were small start up companies that paid very little for the licences. (The US issued no national licences and instituted called party pays pricing, and its market thus evolved differently from the rest of the world. Discussing this is an article in its own right, and I won’t go into it further here). In both instances additional networks were licensed when second generation digital services came into being in around 1993. Even in 1993 nobody realised what a big deal mobile phones would be, and although it generally took pre-existing companies to raise the necessary capital, it was still possible for relatively small players to enter the market at this stage. In some instances the companies that took out these licenses were mobile companies that already had networks in other countries, but usually these were companies new to mobile telephony.

However, the market share of companies in various markets seems to depend very much on which choice was made in the early 1980s. In cases where the existing telecoms monopoly was given a mobile monopoly, that company is usually to this day the dominant player in that national market. In most such instances, that player has a market share of around 50%. Other players can be more profitable, have most of the premium customers, or be perceived as providing better service, but it has been difficult in such markets for the later players to gain large amounts of market share. (Australia is a good example of such a market. Telstra (formerly Telecom Australia) has a market share of around 50%, and Optus and Vodafone (which entered the market with 2G licences in around 1992) have about 30% and 15% respectively).

In markets where there were two operators licensed from the start, the incumbent was usually unable to entrench a large market share in this way. Often, although the start up competitor had much less in terms of resources, it made up for this by having a nimbleness and a better cost structure than the incumbent. When additional competitors entered the market in the early 1990s with the introduction of 2G phones, they were often able to challenge the incumbents more effectively than was the cases in markets that were previously monopolies. The classic example of this is the British market, from which Vodafone initially became the strongest player, but in which the later entrants were able to grow to similar sizes to the existing players. There are four networks in the UK, and all four presently have about 25% market share. (Although a powerful but dominant player in the UK, Vodafone was able to expand internationally to become the biggest player in Europe and the world).

In about 1999-2001, the strongest players in the various European markets went on an acquisition binge, generally buying the weaker operators in other European countries (and further afield), leading to cross-border brands in Europe. The three dominant players that came out of this were Vodafone (originally the second operator in the UK), T-Mobile (former German telecoms monopoly) and Orange (former French telecoms monopoly), all three of who own networks in many European countries, and elsewhere. Sadly, these companies have not grown into pan-European networks in a way that would be good for consumers. Although there might be a Vodafone network in Germany, it still costs a huge amount for a Vodafone customer from Britain to use it. International roaming is so expensive, and such a profit source for the mobile networks, that it is not in their interests to break it down and leave us with networks that appear international in scope to their users. This will happen eventually, but not until cellular networks face competition from vastly more operators or from other technologies. This may be happening – I have a PDA that runs the Voice over IP client Skype and from which I can make free calls whenever I can find a WiFi hotspot in a foreign country – but it is going to take a few years to really happen.

So that is where we are. We have brands that are international, and it is a truly miraculous thing that I can turn my mobile phone on almost anywhere in the world and it will just work, but pricing mechanisms and a regulatory environment that is are sadly far too subject to national borders.

But what does this have to do with Africa? Well, as a general rule the big European mobile companies have not been terribly interested in African markets, due to the fact that African markets often have peculiar regulatory requirements (ie African governments and bureaucracies are horrible), the potential profits have been seen as small compared with developed world and Asian markets, and these companies just generally are not very imaginative. So where has the expertise and capital for these African mobile networks come from?

Well, there is one mobile phone market in Africa which did follow the European model, sort of. South Africa did not have analogue mobile phone services, due no doubt to a combination of the general incompetence of the apartheid regime and a lack of desire to introduce a technology that that very paranoid regime likely saw as potentially subversive. As it happened, the first mobile phone networks did not come into existence until 1993. The powers that be licensed two digital GSM licences – one was essentially given to the telecoms monopoly Telkom (which imported expertise and some capital from Vodafone, and ultimately launched a network under the name Vodacom), and one of those new startup mobile phone companies typical of Europe sprung up. This company was (and is) named MTN.

Like many of its counterparts, MTN was better run and nimbler than its monopoly owned counterpart, but in the South African market itself the monolithic advantages of Telkom won out in market share. Vodacom quickly gained and retains to this day the largest market share in South Africa – presently a little over 50 percent. MTN has a little under 40 percent, and a third operator that was licensed about ten years later now has around 10 percent.

While MTN was not quite as successful as Vodacom in the domestic market, where it did become aggressive was in other African markets. South African firms in consumer businesses usually do not try to compete in other developed markets- they are generally too small, too lacking in capital, and in truth levels of customer service in South Africa are often not high enough for the product to be good enough to succeed in America or Europe. There are exceptions (for instance the “Nandos” fast food chain) but not many.

The foreign markets South African companies can succeed in are other African markets. South Africans have experience in dealing with other African countries, customers are not as sophisticated as in America or Europe, and foreign competitors with better customer service are largely absent. In addition, there is one other crucial skill that South African companies have that is quite important and valuable. In the days of apartheid, many foreign companies would not do business in South Africa. Some South African businessmen became quite skilled at looking at successful businesses in the developed world, cloning them reasonably well, and then selling the resultant product in South Africa. In those days the resultant businesses could not be exported to the rest of Africa, but since the end of apartheid they have been able to do so. And in many cases they have done so.

Mobile telephony is not perhaps a perfect example, but it is one. MTN has expanded into various other African markets – Nigeria, Rwanda, Swaziland, Cameroon and Uganda – and has been very successful in these markets. Nigeria and Uganda I am told are particularly profitable. The company has at times been a darling of the Johannesburg stock market, and it is for its African exposure that it gains a lot of its stock price. MTN has become sort of a pan-African Vodafone – powerful but not dominant at home, but the biggest player in Africa as a whole. Its domestic competitor Vodacom has been far less aggressive at expanding into Africa, being more content to take rich pickings in its home market. Vodacom does have one important network elsewhere in Africa, in Tanzania. (This does have the interesting distinction of providing the highest mobile phone service in the world. Alex will no doubt be delighted to know that his Blackberry will work on top of Mt. Kilimanjaro).

South Africa is a conduit for developed world skills and technologies to find their way into Africa, and this is good. South Africa is also a conduit for developed world market structures and business models to work their way into Africa, and this is also good. The present South African government is not a good government – it has an urge to tax and regulate everything, and puts far too strong an ideological bent on what it says and does – but there is quite a bit of entrepreneurial spirit of the right kind in the country. (The apartheid governments of South Africa were also dreadful, vile, protectionist things too, to be fair). This is an example. It is not the only South African company expanding into other African markets, but more are needed.

Obscure correction: A commenter on this article nitpicked and stated that South Africa did have a very small scale and unsuccessful analogue mobile phone system prior to 1993. This gave me enough information to do some research, in which I discovered that South Africa did have a small network using a technical standard that was unique to South Africa called C-450. (When I was researching the article I couldn’t find South Africa on lists of countries that used any of the main analogue standards, which is why I assumed there was no network). This is not as unusual as it sounds – there were a multitude of different analogue mobile phone standards in use in the world, and one reason why Europe came up with a singe 2G standard (GSM) was because analogue mobile telephony in Europe had been a disaster due to there being a large number of incompatible standards in use – but clearly the reason that first generation analogue mobile phones failed in South Africa was because of a terrible technology choice, mixed in with a questionable regulatory choice in using the 450MHz band rather than the 850MHz band used in most places. (That said, quite a few other countries – many in Scandinavia and Eastern Europe – allocated 450MHz and produced networks that succeeded). The legacy of such a total failure when going it alone might explain why Telkom partnered with Vodafone for 2G, too.

So in fact South Africa corresponded to the Australian model – a monopoly in analogue technology, and then multiple licences for digital from 1993 – but the incumbent telco was even worse, which didn’t stop it from still being able to use its size to gain a majority market share in 2G. Interesting.

Tweet about this on TwitterShare on FacebookShare on LinkedInShare on TumblrShare on RedditShare on Google+Share on VK

6 comments to Some more thoughts on African (and other) mobile phone networks

  • Excellent write up.

    Nit picking:

    It’s Telkom not Telcom as our morbid fixed line monoploy is named.

    There are 3 cellphone players in South Africa now. The 3rd one is called CellC.

  • Yes, my mistake. Spelling Telkom with a k but Vodacom with a c is a wonderful encouragement to make that mistake (which I have fixed in the text).

    (And I did allude to Cell C when talking about marketshare – I just didn’t name it. That fits the global pattern too. Many developed countries licenced additional operators in the late 1990s / early 2000s. These have generally not made money – Cell C has done better than most, probably because most other markets had more than two incumbents by that point).

  • Further nitpicking: Telkom did in fact have a small mobile phone network all its own before the advent of the GSM networks; I can’t remember if it was analogue or digital, but it used those “small-briefcase” type of phones seen in the first Lethal Weapon movie. They were mostly used as car phones. It was an object lesson in market economics. Telkom’s little network attracted a few thousand subscribers in all its years of operation, covered only Johannesburg and Cape Town, the phones cost R15,000 (about £1,500 at the time) and call charges were about R10 (£1) a minute. Within a year of the GSM launch subscribers were in the hundreds of thousands, the network spanned every city and all the major highways, you could get a phone for free, and call charges were about R1.50 (15p) per minute.

  • Stephen: That is intriguing. Please see my addendum to the post.

  • Motorola’s first Electronic Mobile Exchange (EMX) was built using standard (for the day) mobile technology. Compared to today’s systems (small cell based) it was quite limited. It was based on the rules in force at the time. It was about automating what had been a manual operation.

    However, all the basics were there. Frequency reuse, hand off when crossing cell boundaries etc.

    A really amusing technological detail is that it used hundreds aof Z-80 processors per exchange because the Motorola 6800 was not fast enough at the time. There was a Z-80 in every line card.

  • I forgot to mention: mobile in the day of the first EMX meant automobile. It took up quite a bit of trunk space.