From Mohamed Awang Lah
In June last year, I wrote about five policy flaws related to telecommunications and broadband since the corporatisation of Telekom Malaysia (TM) in the mid-eighties which, in my view, have caused the country to slip to become a laggard.
Recently, we may have encountered the sixth policy flaw in dealing with 5G services which has slowed down our progress in rolling out this new technology.
We must remember that 5G was launched in a big event about five years ago. It is a natural progression for 4G players to move to 5G as technology progresses, to maintain competition using the same spectrum.
Spectrum should be “technology neutral”. We cannot expect 4G players to remain in 4G services only. However, we created a new monopoly, Digital Nasional Berhad (DNB), for just 5G wholesale services, and we expected the existing 4G players would be happy to subscribe to the new 5G services.
As we have seen, there has been resistance. Initially, the government compromised by agreeing for retail telcos to have shares in DNB, which could actually destroy the wholesale-retail ecosystem.
The wholesaler should be an independent entity. After further review, the single wholesale model has now been changed to a dual wholesale model. But is it really the right model? In my view, it is not.
Many questions remain unanswered. Why wait until 80% coverage of populated areas before starting the new model? What exactly is the size of the 80% coverage area in sq km? Are we concerned about the low 5G adoption rate by consumers? Have we done comprehensive testing for completed sites? Equipment was procured via a tender process, but was the specification biased to favour a specific brand?
When DNB announced the wholesale leasing rate of RM30,000 per Gbps per month for mobile network operators (MNOs) to subscribe to 5G, it was strange. It means higher cost for higher bandwidth. If dark fibre network is used, the leasing rate could be on a per link basis. It is up to the MNO to generate its own bandwidth – 1Gbps or 1Tbps or whatever – to future proof its cost. With economies of scale, the rate could be independent of distance, like bandwidth rate.
Allow me to offer some views:
- DNB should stop rolling out new sites immediately. There is no need to wait for 80% coverage. We need to cut the losses as early as possible. All sites must be reviewed for their suitability based on the new business model.
- MNOs should immediately be allowed to roll out their own 5G services using their existing spectrums. DNB can keep the 700MHz spectrum, which is suitable for rural areas due to the larger coverage area per base-station. Other spectrums (3.5GHz and others) can be unlocked and farmed out to MNOs. Some MNOs may consider combining their spectrums for better speed and coverage.
- Any gap in coverage, especially in rural areas, can be filled up by DNB, using Universal Service Provision (USP) or other funds. DNB can become a domestic roaming 5G service provider to enable customers of any MNO to get 5G services whenever they are in the area. MNOs do not need to have their own infrastructure. DNB should not have its own end-user customers. This is like international roaming.
- DNB can also build dark fibre networks and/or towers/monopoles in areas where these passive infrastructures are not available. Dark fibre can be leased by telcos (based on rates to be fixed by the government) for backhaul. Duplication of passive infrastructure can be avoided, and any cost-saving can be passed on to the end-users by lowering retail rates. A dark fibre network can be designed to avoid single point of failures. Last mile retail services for consumers – 4G, 5G, Fibre-To-The-Home (FTTH), WiFi, etc. – must be open for full competition.
- Monopoles already built can still be operated by DNB for roaming services, sold to MNOs or relocated to more suitable locations.
Proponents of 5G normally use high speed and low latency as justification for advanced applications such as remote surgery and driverless vehicles. However, these are special applications which are not expected to be for general use in the near future. Let innovation evolve at its own pace. For now, we should focus on solving basic problems – lack of coverage in rural areas, drop calls on some roads and highways, and other issues.
We must make sure services in cities and rural areas are of similar quality to promote digital economy activities irrespective of geographical locations. For indoor applications, new WiFi standards (WiFi6, WiFi6E and WiFi7) may be more suitable provided fibre backhauls are available, similar to 4G and 5G. For very remote areas, different solutions may be required, such as satellites. These are exceptions and must be addressed separately.
It has been reported that the take-up of 5G in Malaysia is only around 1% although more than 50% of the so-called populated areas have been covered. This is a serious issue. It indicates that planning was not carefully thought through. Some of DNB’s monopole base-stations are very near to the existing 4G towers. Demand analysis should have been done first.
The business model must be clear. If it was to be an exclusive monopoly, the buy-in from MNOs should have been sought first. Construction of base-stations should have been done after all the basic issues have been addressed, starting in areas where demand is high.
Unfortunately, these basic business principles have apparently been ignored. It is puzzling that a multi-billion ringgit project has been done in such a manner. Perhaps this is a manifestation of a single-track mindset, which is very common. When we do 5G, we forget about 4G, backhaul, WiFi, etc. We don’t consider the need to have open competition in retail services. We tend to view a project on a standalone basis. We forget about the common denominator – dark fibre – which is required and can be shared for 5G, 4G, WiFi, FTTH and other services.
Mohamed Awang Lah has been involved in internet services since its inception in the mid-1980s. Before his retirement, he was the CEO of Jaring Communications Sdn Bhd.
The views expressed are those of the writer and do not necessarily reflect those of FMT.