Michal Frankl (CETIN): We will build a million optical connections. We want to compete with O2 in the auction for 5G frequencies.
We want to take part in the auction for 5G frequencies, build a national mobile network and lease it on a wholesale basis, says the Strategy and Business Support Director of CETIN.
We are beginning to discontinue the acceleration of metallic lines, and our new priority is on building optical lines, says Michal Frankl, Strategy and Business Support Director of the infrastructure business CETIN. The company, which is part of Petr Kellner’s PPF Group, currently has about half a million optical lines at various stages of progress and wants to build a million lines within seven years.
At the same time, CETIN is looking at the IPTV market where it wants to launch its own platform for the provision of Internet TV. This January, the company published a new wholesale offer and started offering operators the Terminátor, an end-point device for customers’ households. “The Terminátor [connects to] a gigabit LAN port, so this is a certain shift from the standard RJ11 telephone socket. Customers will connect their routers downstream to it,” Frankl explains.
However, CETIN’s plans do not end there. In the upcoming auction for 5G frequencies, the firm wants to bid for the 700 MHz block, which would enable to build a national mobile network, and for the 3.5 GHz band frequencies where high-capacity data services can be operated. “To be clear, PPF will not seek a disproportionately large part of the spectrum and we respect the spectrum limits for the Group,” says Frankl.
The terms of the auction, published last year by the Czech Telecommunications Office (“CTO”) when the former Chairman of the Board Jaromír Novák was still in office, rule out CETIN’s involvement. This means that CTO would have to rework them significantly into a new version. In an interview with Lupa.cz, Frankl describes the obstacles to the development of broadband in the Czech Republic.
Fibre optics to your door
2019 brought at least one major change for CETIN – the move of the headquarters into new buildings. How was last year in terms of business results?
The initial figures tell us that we have met our commercial objectives under the business plan, which is important. What is also important is that we posted a year-on-year growth for the first time since the inception of CETIN. Our portfolio includes declining services such as fixed voice. This is the first time that we managed to compensate for a decline in historic services with new services. This applies primarily for Internet connections, which is where we grew significantly from our viewpoint.
We also generated higher income through the roll-out and operation of the access mobile network for O2 and Nordic Telecom. This allowed us to meet our investment plan, which is ambitious in the long run – as we invest about CZK 4 billion on investments annually and we will continue this in 2020.
At this point, we are finishing works on the FTTC service, which is the acceleration through optical networks in remote DSLAMs. This project is about to be completed as we have somehow activated 92% of our lines, which resulted in the situation when 100 Mb/s transmission rate or higher is currently available to 60% of lines in our network. 22% of our lines are faster than 50 Mb/s. For 2020, our focus is on the construction of optical lines to households, or FTTH.
Does this mean that you have reached the technological limits of metallic lines?
No; there are technologies that allow further acceleration, but we consider FTTH to be a right solution for the coming decades. Of course, metallic lines will serve for a long time; our current project means shortening the metallic loops and taking optical cables to RDSLAM; the great majority of local metallic loops are now reasonably short and we have deployed the VDSL 2 or VDSL 3 technologies on them.
Rates of around 100 Mb/s are sufficient now and there is currently no relevant technology that could be implemented on the metallic lines in the network. This is not to say that there never will be one; we just don’t see anything like that now. What we see is that we need to go optical. We have a relatively small amount of optical connections today, about 33,000, but since we started deploying them last year, we have around half a million households in a certain stage of progress. This does not mean that we can complete them all now, but we see them as our main investment programme. Our goal is to build one million FTTH connections in the next seven years.
We also evaluated the Gfast technology, which falls somewhere between optical and VDSL 3 in terms of speed, but we don’t see a lot of use for it because it is relatively costly. We find it more efficient to build FTTH; however, Gfast is too expensive for locations that are not suitable from FTTH point of view, like rural areas. In addition, the demand for 100+ Mb/s rates is not very high in such areas. In terms of our network potential, it is really utilised on less than 50% of the lines meaning that less than half of our customers have ordered the maximum available speed.
This is related to the price for such a connection, I guess.
Certainly – faster lines cost more. But it’s also a matter of how people use the Internet. Some find 50 Mb/s enough and at a fair cost. But when you have two kids, each of whom plays something on their tablets, you will likely opt for a higher speed and capacity.
You said you have half a million optical lines are at various stages of deployment. Where will you build them? In areas without fast metallic lines, or in parallel with accelerated metallic lines?
We will build them in areas where we know it does not make sense to build metallic lines, for example if a competitor’s cable is there. In this instance, accelerating metallic lines makes no sense, but we can compete with them by building our optical lines. In addition, there are locations where we did accelerate in the past, say, in 2015, and it paid off; so today, we will cover that with optical, which will be ready in 2021. Of course, we will avoid situations where we would be building a RDSLAM today and an optical line tomorrow – we make sure we are coordinated on this. Basically, we are trying to estimate where the demand for optical is and where the investment will return.
CETIN succeeded with projects to cover so-called “white/blind spots” in the first round of the second grant call of the Ministry of Industry and Trade. What do you plan to build for those small communities?
About 50% of the projects that we submitted for the grant call are FTTH (5,500 lines), the other 5,500 are FTTC which are genuinely so-called “super-white/blind spots” (literally, without a proper Internet connection).
Does that mean that it would not pay to build a network there without a subsidy?
It does. Even with FTTC, we would not make ends meet without a subsidy in those areas. The [population] is so sparse there and deploying fibre optics for RDSLAM is so costly that it would not make commercial sense.
Building networks: problems with municipalities and authorities
Have the conditions for building broadband infrastructure improved in recent years? When I talk to operators, I hear all the same complaints about bureaucracy and problems with easements. The Czech Republic enacted legislation to facilitate building networks in recent years. Did that help?
I see both positive and negative changes. We definitely welcome the Action Plan 2.0 made at the Ministry of Industry and Trade (MIT) in cooperation with operators, which summarises the most important matters that should facilitate the deployment of typically linear optical projects. Most of those changes remain on paper, however, and have not been materialised.
What helps us a lot is that we do not need a zoning permit for our lines. We can arrange this directly with the owner, taking care of the private law side of things and then notifying the authority of the project. That is an amazing change because zoning permits take months to obtain. Also, they used to be incredibly expensive. We used to have to pay a thousand crowns per line and three years ago, it was as high as twenty thousand and they are currently considering hiking the fee back up – those are the opposing tendencies [laughs]. The ideal scenario is when you do not need to pay the fee at all when you are just digging into a garden or a pavement to install an individual line.
What certainly helped is that the Electronic Communications Act caps the price of easement by the expert valuation. In theory, this prevents situations where land owners claim excessive amounts. Of course, municipalities or authorities will still ask excessive amounts sometimes. Easements can even be expropriated, but that is pure theory since the expropriation process takes so long that it is totally unusable for our type of projects.
So what do you do when someone insists on an exaggerated amount?
You have to make the decision on whether you will complete the project at the cost of an excessive rate per square metre, or pass on it. Some projects work around the issue by taking a different route instead of the optimum one. Respect for private property is essential, but claiming loss of value and asking excessive easement payments is nonsense especially when we install networks within pavements.
There are two sides to every story. If a municipality acts this way, they may eventually end up with no infrastructure. Frankly speaking, we have more investment opportunities in the Czech Republic than we can deliver in the given amount of time. So, we naturally tend to work with communities that have a sound and open attitude to us – by pricing easements fairly as well as by taking a flexible stance on administrative issues and not asking excessive charges for occupying public space.
However, municipalities are not the only source of problems. Many government agencies such as the Office for Government Representation in Property Affairs and Správa železnic [ed. note: Railway Administration, formerly known as SŽDC] own a lot of land, some of which is irreplaceable. For example, in relation to constructing mobile towers intended to cover railway corridors, you have virtually no chance of doing this, without reaching an agreement with Správa železnic.
Are there any problems in that respect at the moment?
The prices are typically very high. The Association of Mobile Network Operators (APMS) has a task force that liaises with SŽ and will hopefully reach a comprehensive agreement that would pave the way to covering the corridors efficiently. It is similar to the Prague Metro where the coverage project stalled for years due to the unrealistic expectations as to how much operators should pay as rent for installing leaky cables, and we only saw a breakthrough on this recently. The deployment is in progress now and it is clear that the Metro will get its coverage.
Coming back to municipalities, can you provide a specific example of where things worked well in your opinion, and where you thought the municipality’s approach was wrong?
An example of where it worked well was Teplice city. We wanted to build remote DSLAMs and fibre optics in mobile base stations, but the original requirements were quite high and unrealistic. We visited the then-Mayor, the late Jaroslav Kubera and explained that the networks had a certain value for the town. After some discussion he said that he understood and that it all made sense to him. As a result, we agreed to having two expert opinions in order to determine a realistic price and also to use the opinions as a form of benchmark, to indicate a customary price, to avoid needing to commission a new opinion for each individual cable route.
I terms of where it hasn’t worked well, I don’t want to name any names. But we face the biggest problems in the towns that tend to build their own infrastructure or in towns that are allied with a local operator to some extent.
Construction authorities avoid settling disputes
I have been hearing complaints like that from operators of all sizes for years; nothing has changed.
The principal problem is the Construction Act. Every simplification of the Construction Act in recent years has resulted in the protraction of the project approval procedure [laughs]. The relationship between the public interest in building something and the legitimate protection of other interests – not just ownership rights but environmental protection and so on – is in a state of imbalance.
Also, the construction authorities’ attitude is often not helpful when it comes to resolving disputes. They tend to say things such as, ‘there will be no project unless and until you have obtained all consents’ instead of saying, ‘here is one argument, there is a counter-argument, and this is my decision’. By all accounts, they do not even have this type of authority under the current legal concept – although even lawyers do not always agree on this.
Does it work that you are informed when a City is planning an extensive street repair or their own excavations, so that you could bury your optical cables in the process?
Operators still have little chance to find out that such projects are underway. For instance, Brno city publishes an overview of the streets where excavations are planned on its website. This is good, but the overview is posted as a .PDF file. You need to know that Brno publishes it somewhere, then go searching for it, and then retype it into your own system.
Obtaining information is not easy; we often learn it too late and do not manage to coordinate the projects. Our experience shows that digging a hole is not the most expensive part – it is restoring the surfaces to the original conditions. Also, municipalities often want the new pavements to be better than the original ones when the project is completed. That can be quite costly. In some cities, there are almost “monopolies” for who can build pavements, which leads to unrealistic prices.
An amendment to the Linear Structures Act is currently being prepared, and it should permit adding cable installation to projects. This would mean that when a private law agreement is made, the cable can be simply added under the zoning permit that was issued to the first party to apply. That used to be the biggest problem: someone obtained the permit and was about to start digging; we wanted to join them but we did not have our zoning permit, and they would not wait for six months until we obtained it. So, this could help a lot. Generally speaking, building infrastructure requires a whole lot of patience [laughs].
How do you think the new construction bill could change the current status quo?
The controversy of the new construction bill is in that it approaches things very differently. For example, it discontinued the veto power of the individual parties to the construction procedure. It assumes approval if a party does not comment within a specified period. We live in the Czech Republic and we are not used to this. But things won’t work if things don’t change. For example, the law in Poland – which is a country we look up to for affordable mobile data – rules that if you obtain a public law consent and the land owner disagrees, you become a temporary holder. With that, you can start building; the land owner gets an upfront payment and a procedure for expropriation runs in parallel, and it is de facto all about a financial compensation.
That sounds like an extreme solution.
I don’t think it is extreme because it does not apply to just any project – it only applies to projects in the public interest that have been discussed under public law.
This January, CETIN came up with a new wholesale bid, offering new options for connection that does not differentiate between VDSL and fibre optics, with your Terminátor end device included.
We innovated our reference offer for the MIT with effect from 2 January. In addition to the previous portfolio, we made a new portfolio with the basic transmission rates, and the Terminátor is the end point of the network. It is in fact a modem that we commissioned and tested in cooperation with the manufacturer to ensure that it is optimally tuned. We invest a lot of money into improving the line to the customer, but if the modem at the end of the line is not tuned right, it means that one half of the investment is thrown out the window because customer experience of the device can be seriously impaired. The Terminátor also enables VDSL 3. VDSL 3 models available on the market are expensive. Also, the Terminátor is required for bonded lines. Modems for this technology are not commonly available in the market.
The Terminátor units were actually used in the optical network before, albeit not under this name. The Terminátor terminates in a gigabit LAN port, so this is a certain mental shift from the standard RJ11 telephone socket. Customers will connect their routers downstream of it, and they may prefer models offering high-quality Wi-Fi. In addition, we will supervise the network segment all the way to the Terminátor, which has advantages in terms of detecting defects. The price of the service includes repairing and replacing the Terminátor.
The purpose of the service is to free our 19 partners from issues such as this: one customer has a simple VDSL, so they need this modem; another customer has 250 Mb/s, so they need a different modem; the next customer has fibre optics, which means yet another modem. It may seem like a small matter, but it’s not. There is a fair number of situations where a technician arrives with the wrong end device because it was incorrectly specified in the order.
The new portfolio is not mandatory – the earlier one works alongside it, and you can still order services without the Terminátor.
What is the demand for the Terminátor among your partners?
It is only six weeks after the launch, so it is too early to say, especially when it comes to larger partners who communicate with us through their B2B gateways and need to make adjustments on their side. The orders are just starting.
I assume that installing the Terminátor will apply primarily to new lines and new clients.
Yes, it will – new lines and upselling.
Proprietary IPTV platform
You also announced that you were preparing your own IPTV platform to offer to your partners. What is it?
It is our proprietary technological platform for the provision of Internet TV. Simply put, the platform is on our premises and it sends selected TV content to the customer’s set-top box. There are such platforms on the market already – T-Mobile has one; O2 uses Nangu.TV. Our platform will be new and robust, and as such it will not be impacted by the historic development that some other platforms have been through. The basic business plan includes the Czech Republic as well as clients in the Balkans countries where we have our operators as part of the Group. We will try to offer it in a wholesale form to partners who find developing and operating their own platforms too difficult.
So, rather than targeting T-Mobile or O2, you go for operators who do not have their own platforms.
Yes, and for those who find the service life and further development of their platforms not efficient.
How will it work in terms of technology? Will it really be IPTV?
It will be both IPTV and OTT.
Do you also plan on delivering content in one way or another?
The content will typically be up to the operators.
We are interested in the auction for the 5G frequencies
The Czech Telecommunications Office and the Ministry of Industry and Trade are discussing a change in the terms and conditions of the 5G frequencies auction. You voiced your concerns about the existing version last year. Your complaint was that the terms and conditions effectively put you out of the auction – they ruled that, within an existing group, only an existing operator may take part in the auction of the frequencies – which is O2 in your case. Would you be interested in participating in the auction if this condition changed?
Yes, we would – of course only after we got to know the final terms. We think it should also include a model for a wholesale mobile service offer. We do not mean a wholesale offer required by the regulatory conditions, like the ones that frequency holders still have from the previous auction. We mean a retail concept based on the fact that we are a wholesale-only operator and that we do not have to address the conflict between selling the service to end customers on our own and having to offer it on the wholesale market, in effect creating our own competitors.
Our business model is based on the idea that our wholesale partners have to be successful to be able to buy data from us. Of course, we are not ready to sell our services cheaply or give them away, but the success of our partners will always matter to us, because that’s the only way for us to generate sufficient revenue.
The wholesale model in the fixed network has proven its value. After years of disputes over wholesale DSL offers, not one dispute has occurred during CETIN’s existence. We managed to reverse a trend – the share of DSL was decreasing for years, but it is growing now thanks to investments. We believe that this model is also viable in the mobile network.
So yes, we would like to take part in the auction. To be clear, we would not seek a disproportionate size of the spectrum, we respect the limits for the Group. We believe that the wholesale model can function well even under that scenario.
I think it is worth debating the restriction that we can only compete for 2 × 10 MHz of the spectrum. Of course, it would be more attractive for us to compete for 2 × 20 MHz, so that our offer could be full in terms of bandwidth. Obviously, we are talking national coverage in the 700 MHz band; high capacity parts of the network could be covered in the 3.5 or 3.7 GHz bands where, in turn, national coverage is unrealistic, but they can serve well as capacity enhancement.
Does that mean that you want to vie for the block intended for the ‘new operator’, and if you are successful you will go on to build a new mobile network?
We have an access network over which O2 provides services, as does T-Mobile as part of sharing, so we can build it relatively quickly and without having to start from scratch. The return on investment can work quite well. If a complete newcomer was in the same situation, their capital expenditures would likely be higher than ours.
Apparently, the problem with Nordic Telecom is that there are rumours that it could be linked to PPF, even though both the Group and the operator deny that. Do you think that the CTO and the MIT will be helpful towards CETIN, the ownership of which PPF admits quite openly?
There is no rule saying PPF cannot take part in the auction [laughs]. This is why I mentioned the spectrum limit. We do not aspire to get twice as much as anyone else. We understand that CETIN and O2 cannot compete for the same thing at the same time and win both.
What would your wholesale offer be like then?
We would proceed the same way as in the fixed network. We would publish a commercial offer built on the principle of non-discrimination. This means that it would contain terms and conditions that we would apply to all market players the same way.
Of course we are interested in taking part in the announced tender for data services as part of PPDR (Public Protection and Disaster Relief), which we expect the Ministry of the Interior to call at some point. We think it is quite interesting, and we also believe that we can offer infrastructure with an above-standard coverage, which is what the ministry requires – for example, on the borders or border crossings. We also believe that we can provide a level of cybernetic security that is not commonplace and will satisfy the requirements for critical state infrastructure. These are our strengths. This may be why we are more expensive than our competitors sometimes, but security affairs are in our DNA, and in order to maintain our high standards, we are not always able to make things simpler or more affordable.
Also, let us be frank in that the PPDR could stabilise the entire business model. Not all of the virtual operators will survive, and you cannot bet on all of them. This contract, however, would allow for better return on investment from the very beginning.
Have you discussed the plan for CETIN to take part in the auction and respect the spectrum limit for the Group with O2 in any way?
By participating in the auction – we are a competitor vis-à-vis O2 today.
But does that not mean a potential loss of spectrum for O2.
O2 would not be excluded from our wholesale offer. Therefore they could access the spectrum in one way or another, although just not on an exclusive basis like they would be able to if they bought the frequency bands directly.
All that’s left to do, then, is wait until the regulator publishes the new version of the conditions.
According to the information from the parliamentary sub-committee for ICT, the new public consultation should take place at the end of March.
Author | David Slížek