(keitai-l) Re: Peter and Paul and Roaming

From: Benjamin Kowarsch <benjk_at_mac.com>
Date: 10/12/01
Message-Id: <5AFEF2D2-BF0A-11D5-A3BC-003065501888@mac.com>
On Friday, October 12, 2001, at 05:15 , Michael Turner wrote:

> You must also win.  And in any robbing-peter-to-pay-paul
> scenario, governments do the popular thing, not necessarily
> the thing that plays well with all industry segments.  Intercontinental
> roaming just doesn't buy you many votes.

Well, "visited network charges" doesn't rob Peter to pay Paul.

Most people make the mistake to think of a visited network as the one 
guys (Peter) and of a home network as the other guys (Paul) but that is 
incorrect. There is no GSM network that is purely a visited network and 
there is none which is purely a home network. They are all both Peter 
and Paul.

Therefore the question is will there be more money for them under the 
bottom line. Any network that can make more money on incoming than on 
outgoing roamers by using the "visited network charges" model, has a 
strong incentive to go for it.

And in Europe, where almost all countries have large inbound tourism, 
the value proposition represented by the "VN charges" model to most 
operators is very interesting.

>> The other major obstacle for CAMEL is that operators have to spend all
> that
>> money on the technology but it only benefits other networks not
>> themselves - they rely on others to deploy in order to benefit. Again,
>> something that favours cartelisation and hence has a tariff increasing
>> potential.
>
> Yes, what economists call the Free Rider problem.  Now try to
> scale it up to global roaming, and you have an even bigger problem.

Correct. That problem is hitting the vendors who have invested in CAMEL 
right now as the industry is reluctant to deploy and nobody has yet been 
able to come up with a viable business case for CAMEL.

BTW, IBC asked me to speak at CAMEL 2001 this year about "ZEBRA 
outstripes CAMEL", a title which had been proposed by the organisers not 
in an apparent attempt to have me lynched at the conference ;-) but 
because many operators are looking for alternatives.

>> The reason why the EU and also the mobile operators are interested in
>> the "visited network charges" model (i.e.ZEBRA) is because it is the
>> first time that there is a proposed solution that leaves something on
>> the table for everyone.
>
> Except that the money has to come out of somebody's pocket,
> and unless it drastically reduces operating costs,

Exactly that is what it does. It avoids entirely the enormous overhead 
of the existing model.

> If it's prepaid, how can it be "instant"?

In a nutshell it works like this ...

1) tourist arrives in a foreign country and switches on the mobile phone
2) network is chosen either manually or automatically depending on phone 
setting
3) request for service is relayed to home network
4) home network confirms authenticity but rejects or restricts roaming 
(i.e. incoming only) or
roaming centre notifies of user's expressed preference for the 
alternative
5) visited network creates instant local prepaid account with any of
- zero credit
- welcome credit, i.e. 5$ to lure customers
- credit transfer from home network (if HN supports it)
6) user receives welcome message with local phone number and advice 
where to buy local credit
7) voicemail greeting in HN is changed to announce local phone number 
(via voice circuit) or
if HN supports it, local phone number is sent to HN to allow them to 
activate unified messaging/forwarding

> And what in this ZEBRA model
> guarantees that it'll be cheap for tourists

Networks who will deploy this will (at least initially) mainly be 
targeting tourists. Initially there will be only a few networks offering 
the service and therefore it will likely be run in non-cooperative mode 
which means the only facility that allows to stay in contact with 
callers is the change of voicemail greeting (which can be done via voice 
circuit without deploying any interface into the home network). This is 
satisfactory to tourists but not to business users.

Those tourists will only use the service if they can be convinced that 
the tariffs are reasonable. In respect of the tourist service, networks 
will need to overcome the public perception that roaming is outrageously 
expensive or they will not get a lot of business out of it. However, as 
they do not have to share the revenue with the home networks they can 
afford to offer the service at tariffs that are roughly comparable to 
tariffs on their local prepaid service.

Even if they charge 10% or 20% premium over what their local prepaid 
tariffs are, that would still be far less than roaming charges on the 
traditional model and certainly attractive to tourists. Also, incoming 
calls on the local number are free of charge to the roamer (with a few 
exceptions in countries where incoming calls are charged, i.e. Hong 
Kong).

Calls made to the home network phone number can be forwarded at lower 
cost because the facility used is an old fashioned vanilla call 
forwarding service instead of the complex and expensive GSM/MAP delivery.

The advantage of the GSM/MAP delivery is that a handover to another 
network will not interrupt a call in progress, where with an ordinary 
forwarding service a call in progress will drop if the roamer moves out 
of coverage of the visited network. This may be an issue for business 
travellers but certainly not for tourists.

Using ordinary forwarding however means a significant cost reduction.

Pressure for reasonable forwarding costs will further come from 
independent unified messaging providers. For example, www.yac.com offer 
personal phone numbers free of charge to users, and they forward calls 
abroad where the forwarding cost is paid for by the caller through 
premium rate calls. Such services could link up to a ZEBRA service and 
update the forwarding automatically to where their customers roam, 
thereby allowing them to receive calls free of charge abroad using a 
personal number in their home country with all the convenience of the 
traditional roaming service.

Now, even in the role as a supporting home network, there is incremental 
revenue to be earned. Most of those tourists would not have used their 
mobile phone overseas before. Now, as they do, their networks have the 
opportunity to support them while abroad with unified messaging 
services, such as

- forwarding calls selectively
- forwarding voicemail
- forwarding SMS

Again, if this is not competitively priced, those tourists are unlikely 
to use the services and independent unified messaging providers may fill 
the void. Cost for providing such services is however reasonably low to 
mobile operators.

Also there is no clearing required in this model. Each network charges 
for their own services directly. Yes, if credit transfer facilities are 
offered, then those transfers will need settlement. However, most users 
will use credit transfers only as a fallback or to establish an initial 
credit upon arrival (i.e. 5$ upon arrival in a foreign network). Thus 
the number of credit transfers will be limited and it will be a one time 
chunk instead of zillions of micro-transactions. Think of it as 
traveller cheque vs credit card. Less complex, which in this context 
means significantly lower cost and that cost will be levied based on 
transfers not on calls.

> you just don't have a big constituency that would benefit,
> and there are too many software providers who would
> actually lose out.

Mobile operators in countries like France, Spain and Italy earning an 
extra 1-2bn USD per annum by providing tourist services may not be a big 
constituency in terms of quantity but a significant constituency 
nevertheless.

Software vendors don't loose out either. It would not make sense to 
build a "VN charges" roaming solution into a prepaid or billing system 
because that can lead to compatibility issues, such as tourists from A 
can only roam with company X in country b etc etc.

In fact ZEBRA is designed not to sit inside any charging/billing 
platform but in an external roaming centre to which an open interface 
allows each vendor of prepaid or billing systems to connect to for the 
exchange of messages (i.e. "can you please create an account for this 
user", "here is the local number for this user" etc).

As a result equipment manufacturers can benefit too, as many prepaid and 
billing systems are licensed on a volume or user basis. More users (due 
to tourists obtaining service) means more license fees for them.

Even the clearing houses don't have to loose out, as it might seem at 
first sight. They can run those roaming centres.

Hotels, however may loose some of their phone related revenue. Also, 
operators of public phone boxes in tourist spots may see some decrease 
in volume.

Yes, Nokia and Ericsson might have to write off their CAMEL development, 
but that they may have to do anyway as things stand right now. I know of 
at least one company that has tried to obtain the patent rights to ZEBRA 
in order to let it disappear in a vault never to surface again, but they 
failed at the inventor's veto ;-) On the other hand there is no reason 
why the two models can't coexist. As I said before the formula is 
Business class *plus* Economy class not either one or the other.

Taking your Peter and Paul analogy, I guess there are too many Peters 
and too few Pauls in this case.

rgds
benjamin


[ Need archives? How to unsubscribe? http://www.appelsiini.net/keitai-l/ ]
Received on Fri Oct 12 15:00:57 2001