ICASA's spectrum assignment plans welcomed by industry

I think ICASA might (and that's a big might) be on the right track here. Looks good in principal - hopefully the smaller issues can be ironed out, especially those that are yet to be discovered.

I don't think the operators will want to take drag this to court, as the entire industry will lose out due to the delays.
 
A single wireless wholesale network sounds like a good idea until the following is taken into consideration:

1) A new telecommunication monopoly is created. We are still trying to survive the last one. Competition is good for customers. Even if the wholesale access network is ran as a not-for-profit then there still won't be pressure to obtain inefficiencies. The single network also creates other monopolies, especially in terms of infrastructure provisioning.
2) Networks won't compete on factors like coverage, throughput speed etc.

I wonder if it is possible to stop this nonsense through the competition commission.
 
"30% historically disadvantaged individual (HDI) shareholding requirement."

Could anyone explain to me exactly what this means ?
 
"30% historically disadvantaged individual (HDI) shareholding requirement."

Could anyone explain to me exactly what this means ?

not easy to answer on a detailed level but, simply put: HDI is essentially BEE + white women (anyone who was discriminated against prior to 1994 on the basis of race, gender etc etc). so if you want to play in the process you must prove to ICASA that you have at least 30% HDI equity ownership. This is simple for most of the smaller guys (who had to have this in place when they were VANS licence holders prior to 2009) but there is no clear mechanism for calculating it for listed companies and some of the larger operators do not have it.
 
A single wireless wholesale network sounds like a good idea until the following is taken into consideration:

1) A new telecommunication monopoly is created. We are still trying to survive the last one. Competition is good for customers. Even if the wholesale access network is ran as a not-for-profit then there still won't be pressure to obtain inefficiencies. The single network also creates other monopolies, especially in terms of infrastructure provisioning.
2) Networks won't compete on factors like coverage, throughput speed etc.

I wonder if it is possible to stop this nonsense through the competition commission.

How would the new wholesale wireless network be a monopoly? There are already at least 6 big existing wireless networks in operation (Vodacom, MTN, Cell C, 8.ta, Neotel, iBurst, several Wimax networks, and loads more WiFi and Microwave networks). Your arguments are poor and ill-conceived.

The only monopoly existing in the telecoms industry is Telkom, as they are the only ones with a monopoly on the last mile fixed line copper infrastructure. NOBODY else has access to it, except Telkom. There are no competing networks that could even come close to 5% of the scale of Telkom's network. Now that is what you call a monopoly.
 
How would the new wholesale wireless network be a monopoly? There are already at least 6 big existing wireless networks in operation (Vodacom, MTN, Cell C, 8.ta, Neotel, iBurst, several Wimax networks, and loads more WiFi and Microwave networks). Your arguments are poor and ill-conceived.

The only monopoly existing in the telecoms industry is Telkom, as they are the only ones with a monopoly on the last mile fixed line copper infrastructure. NOBODY else has access to it, except Telkom. There are no competing networks that could even come close to 5% of the scale of Telkom's network. Now that is what you call a monopoly.

The spectrum is for LTE deployment not just another GSM/UTMS network. Other technologies simply won't keep up with LTE. This would create a monopoly on the next generation technology, not the current.

And yes, Telkom has a monopoly. But this doesn't justify another one. Besides cellular networks are not known to be competition friendly to begin with. Just take the six big networks you mentioned and plot them by subscriber numbers/revenue/data transmitted then you will see an interesting picture. Vodacom & MTN is orders of magnitude bigger than the rest. I would definitely not call Neotel/iBurst/WISP's big networks.

Just to add: If we had a regulator that could actually regulate - then this might work. But ICASA has failed from the word go in promoting competition. If we had a strong regulator that could actually set the principles for an open access/whole sale network and effective regulate it then it would be practical and I will support it. Until ICASA get their act together we can only rely on the brute force of competition all along the value chain.
 
Last edited:
The spectrum is for LTE deployment not just another GSM/UTMS network. Other technologies simply won't keep up with LTE. This would create a monopoly on the next generation technology, not the current.

We are talking about spectrum, not a specific technology. Spectrum can always be refarmed to be used for newer technologies (think DTT, HSPA 900, etc). There is no monopoly on spectrum. There are plenty of companies with access to spectrum, as well as the capital/funding to make good use of it. It is up to these companies to find the right technology to fit the spectrum, and find a business model that works.

And yes, Telkom has a monopoly. But this doesn't justify another one. Besides cellular networks are not known to be competition friendly to begin with. Just take the six big networks you mentioned and plot them by subscriber numbers/revenue/data transmitted then you will see an interesting picture. Vodacom & MTN is orders of magnitude bigger than the rest. I would definitely not call Neotel/iBurst/WISP's big networks.

Remember the spectrum they are talking about is NOT going to a SINGLE entity. It will be going to multiple entities, one of which will be an open-access wholesale network which will be made available to all (valid licensees). I fail to see where the word monopoly fits into all this.

Also, iBurst and Neotel might have small networks now, but in terms of the spectrum they are currently allocated, they have a potentially big future ahead of them. They've just got to find a business model that works.

You are trying to make it sound like there will only be one LTE / Advanced wireless network in the country, and it will be a monopoly, along the same scale as Telkom. ICASA for once have got things right (in principle). Yes they are useless (especially on implementation), but there is definitely no monopoly on wireless. Eventually, newer technologies greater than LTE and LTE-advanced will emerge, and the old-school GSM frequencies (currently used extremely inefficiently compared to newer technologies) will be refarmed to make use of these newer technologies.
[/QUOTE]

Just to add: If we had a regulator that could actually regulate - then this might work. But ICASA has failed from the word go in promoting competition. If we had a strong regulator that could actually set the principles for an open access/whole sale network and effective regulate it then it would be practical and I will support it.

This is actually a rare occasion that ICASA did get the principles right, and industry seems to agree (for now, until the finer details emerge). Time will tell.
 
We are talking about spectrum, not a specific technology. Spectrum can always be refarmed to be used for newer technologies (think DTT, HSPA 900, etc). There is no monopoly on spectrum. There are plenty of companies with access to spectrum, as well as the capital/funding to make good use of it. It is up to these companies to find the right technology to fit the spectrum, and find a business model that works.

Nice theory. Last time I checked there are still GSM devices being sold. You cannot refarm spectrum if you actually need all of your existing spectrum to serve your current customer base. Sure you might reorganize it a bit, but nowhere near enough to actually make LTE happen. Otherwise LTE networks would have been launch already.


Remember the spectrum they are talking about is NOT going to a SINGLE entity. It will be going to multiple entities, one of which will be an open-access wholesale network which will be made available to all (valid licensees). I fail to see where the word monopoly fits into all this.

Open access just refers to that there will be competition on the consumer package side. It is like saying that there is a lot of competition in the ADSL ISP space. And it is true - there is a lot of competition and innovation in the ADSL ISP space. But they are all constrained by the wholesale network they use. The network forces a non-optimal design (IPC) and very bad commercial arrangement. The sad thing is that Telkom doesn't really make too much of a profit. They are just super inefficient. So if we get open access networks that are operated inefficiently and have a bad design we will be exactly where ADSL is right now.


Also, iBurst and Neotel might have small networks now, but in terms of the spectrum they are currently allocated, they have a potentially big future ahead of them. They've just got to find a business model that works.

I stopped holding my breath for those two networks about two years ago. The only way they will grow big in the short to medium term is if they get massive outside capital injections. Building networks are an expensive capital intensive business.

You are trying to make it sound like there will only be one LTE / Advanced wireless network in the country, and it will be a monopoly, along the same scale as Telkom. ICASA for once have got things right (in principle). Yes they are useless (especially on implementation), but there is definitely no monopoly on wireless. Eventually, newer technologies greater than LTE and LTE-advanced will emerge, and the old-school GSM frequencies (currently used extremely inefficiently compared to newer technologies) will be refarmed to make use of these newer technologies.

The time-line for LTE roll-out is about 3 years max. After that you have to start thinking about roll-out of the next-gen technology. GSM is not going to go away in the next 3 years. And there is no current monopoly in wireless, but my issue is not "general wireless". It is specific monopoly for next generation LTE networks. These networks are important for two reasons:
1) They support higher throughput - making LTE a more competition to some fixed-line applications.
2) They support more users -which is important for growing subscriber numbers.

This is actually a rare occasion that ICASA did get the principles right, and industry seems to agree (for now, until the finer details emerge). Time will tell.

1) As far as I know ICASA published a slide-show. I have not seen any real commercial or network design principles to make sure there are sufficient competition on the network provision side. (Please note I say competition on "network provisioning", not "consumer packages"). The first draft ITA was supposed to be ready on Thursday, but there is still some process before it is done. So you cannot say ICASA got the principles right because they have not yet finalized the principles. Their little slideshow gave very little on the actual principles of operation of the wholesale network. They described what a wholesale network is and that they would like one.
2) Please qualify "industry seems to agree". If you read the article you will note that the three biggest players in wireless data have all politely refused to comment at this stage. None of the industry players who responded actually run a nationwide wireless data network.
 
I havent seen the full release yet, but from what i have seen so far I would think ICASA is trying to push those wanting the spectrums available to cater for a broader environment and subscriber base. Essentially it looks like ICASA is pushing LTE to be used as a form of rural broadband, similar to the NBN in Aus. In my view this is actually a good step.

As was pointed out, LTE and LTE Advanced is not limited to 2600, MTN is using 1800 and there are a few other networks in SA using other bands. There is also TDD to go with, where existing spectrum can be reused (with some decrease in speeds) So what ICASA is doing here is ensuring that the 800mhz range will go to rural areas where it makes sense. I good move in my view, however limited at this time.
 
@JohanD - interesting thank you. Assuming that ICASA proceeds as it has (roughly) indicated, is there a way in which this can be structured so that competitive forces can play a role in price-setting on the wholesale network (rather than waiting for ICASA to get its act together)? Will be very interesting to see who bids to be the wholesale operator...
 
@JohanD - interesting thank you. Assuming that ICASA proceeds as it has (roughly) indicated, is there a way in which this can be structured so that competitive forces can play a role in price-setting on the wholesale network (rather than waiting for ICASA to get its act together)? Will be very interesting to see who bids to be the wholesale operator...

Don't know. The structure makes it difficult because you have to artificially apply pressure and that has effect as competition and there are always going to be some hole that gets exploited. I have some ideas but what would actually be required is a case study of what worked overseas and what did not.

I think the important thing is to differentiate who owns the "logical country-wide network" and who owns the underlying physical towers. I don't thing the new entities need to necessarily own the physical infrastructure. Some quick idea of a structure within the current scope that might work (pretty much armchair stuff though):

1) New operators most like likely do not have the capital/network infrastructure/land-use rights to start from scratch. I think the regulations must be written in such a way that existing operators can actually build the network components leveraging their current infrastructure and capital.
2) The new wholesale entities do a couple of things:
a) They coordinate the build of the network using their spectrum.
b) They manage the network operations.
c) They facilitate and manage offset agreements (for entities that helped build parts of the network), and provide a single commercial platform for wholesale access to the logical network (for entities that only want to use the network).
d) Their income is provided by margin of the sale of access to the network, fees for managing the network. The management fees are based on the fees that current operators charge for maintaining the network.

The objectives would be:
1) Fast network deployment. Current operators try to build the wholesale network as fast as possible because offset agreements.
2) But they build it as cheaply as possible, because their cost of using the total network depends on the cost of their portion of the network.
3) Providers without their own infrastructure can negotiate a reasonable deal with the single wholesale operator without negotiating directly with competitors.
4) The profit of the wholesale operators are based on their efficiency (how cost effective they can manage the network), and how much wholesale access they can sell (number of operators on their network).

What is regulated:
1) How much infrastructure an operator can own within the wholesale network. E.g. saying for example "Vodacom can only own a maximum of 30% of Wholesale Operator A Network's infrastructure". This will ensure fairness of offset agreements, because it puts a cap on how much can be offset.
2) The management fee of the wholesale operator is determined by the management cost of the current cellular's operators network, plus a fixed margin.

But frankly, given that, I would still prefer existing operators compete head to head on all aspects of the network. They have the capital, the know-how and they are quite efficient at rolling out networks. The only thing I would require that is not in place is better regulations on wholesale access.
 
Top
Sign up to the MyBroadband newsletter
X