Finding a cure for Malaysia’s broadband illness

 

From Mohamed Awang Lah

There are a few critical parts behind any internet or broadband service.

Starting from an end-user located anywhere in the country, there is last mile connectivity that links the user to a point of presence (POP) of a service provider, which could be in a building, telco room, a tower, or just a street cabinet.

The technology could be wired (such copper for xDSL, fibre for FTTx) or wireless (such 4G, 5G and WiFi). From a POP there is usually a long distance backhaul link to a central control centre operated by the same service provider. It could be a series of microwave or fibre links.

Traffic from the users of a service provider needs to be exchanged with traffic from other users served by different providers. For domestic traffic, this is done through MyIX (Malaysia Internet Exchange). For international traffic, a service provider may have its own links to foreign exchanges in other countries.

We used to have KRS (Konsortium Rangkaian Serantau) to act as an international exchange for Malaysia. But it has been closed due to conflict among its members – a sad story.

Before MyIX was established, local traffic was exchanged in other countries such as Singapore or Hong Kong, thus incurring additional costs. This issue was resolved with MyIX, which is operated by a service providers’ association and promoted by the Malaysian Communications and Multimedia Commission (MCMC)

During my tenure as the first chairman of MyIX, we created an automatic formula to match cost and revenue. With higher traffic, unit cost would be lower, which then translates to a lower leasing rate, which benefits service providers and end-users. The rates were reviewed every quarter.

Currently, the biggest challenge is the backhaul. We still do not have a mechanism to share this high-cost infrastructure. It is expensive due to the need for right-of-way (RoW) to dig up roads for cable installation.

Certain large service providers are proud to have their own exclusive backhaul as a competitive advantage, while small providers must depend on and compete against them at the same time.

The “ego” mentality that existed before MyIX now persists for backhaul. Sometimes, we see a few providers digging the same road, multiplying the costs which eventually must be paid by the end-users.

The issue is not new, but we have not been able to overcome this “ego” mentality. It seems that MCMC is also not helpful. Previously, we had NFCP (National Fiberisation and Connectivity Plan), which has been replaced by the National Digital Network (Jendela). Both do not have any meaningful initiatives to share backhaul infrastructure.

Instead, they are focused on the last mile portion which, by right, should be open for retail service providers to compete in serving the end-users. By awarding contracts for the last mile portion, monopolistic behaviour would prevail.

Yes, there might be “open access” clauses in the contract, but how can we expect fair play when an infrastructure provider is also forced to compete against its own customers?

Jendela’s latest quarterly report updates were almost all about last mile “success” stories – fibre home-passed, 4G coverage, 5G initiative, fixed-wireless, etc. It is a service provider-centric report, not end-users’.

Previously, we had programmes such as “Kampung Tanpa Wayar”, which may be considered as a failed last mile initiative due to the lack of backhaul connectivity. In general, performance was measured based on service providers’ achievements in completing projects rather than the resulting customer satisfaction.

The reported number of complaints received by service providers reflect their “unsatisfaction”. We do not have any benchmark on what acceptable numbers should be. There should be independent test servers located at MyIX for users to test speeds and latencies of broadband connectivity from anywhere in the country. Independent parties should be regularly engaged to conduct all tests.

In my view, the absence of a concrete programme for the sharing of fibre backhaul is the main cause of our lack of progress in broadband expansion. We don’t seem to understand the need for competition in the last mile to improve quality of service.

We also do not understand the need to share and reduce backhaul cost for the benefit of end-users. We ignore the need to separate wholesalers (lessors) and retailers (lessees). KRS should be replaced.

The latest demand by the four big mobile network operators (MNOs) related to DNB shareholding is a recipe for failure.

Initially, DNB was supposed to be a single wholesale provider. However, due to objections from some MNOs, the government agreed to allocate 70% shares of DNB to MNOs.

But now, the four MNOs are demanding majority shares within the 70%. If we are not careful, DNB could become a retail consortium, and fights between large and small MNOs might occur, like the defunct KRS. Worst still, it could become a cartel. Who will be protecting the end-users?

5G is a last mile service, similar to 4G or FTTx. We should let the MNOs compete to serve the end-users. The government, through DNB or other entities, should focus on the provision of shared passive infrastructure such as fibre backhaul (and towers or poles), to enable affordable and competitive last mile services in all populated areas.

We should not marginalise people in rural areas by providing substandard services. Everyone should have the chance to participate in digital economy activities. Then, we can have real and meaningful socio-economic transformation.

Some people might think that fibre backhaul is expensive. Yes, it is expensive if it is done by a single operator for its own exclusive use. But, through sharing and operation by an independent third party on a not-for-profit basis, the average cost would be very much lower.

Overall cost could be less than the amount allocated for DNB or Jendela. However, the government must intervene.

A few years ago, we wasted billions of ringgit in providing netbooks without network; we wasted more billions in providing broadband for schools believing that wireless can stand without fibre; we spent a lot of resources in setting up “Kampung Tanpa Wayar” without sufficient wired backhauls.

Now we are spending billions more in providing lip service “open access” last mile infrastructure without segregating lessors and lessees. Lessors who are given RoW to build infrastructure (which is a national resource) should not use it for competitive advantage. We should stop creating little and large napoleons with monopolistic behaviour.

Perhaps it is time to review the Communications and Multimedia Act 1998 (CMA’98) to strengthen anti-competitive provisions. It has been more than 20 years. Compared to our neighbours, we continue to lag.

We need to focus on protecting the end-users, rather than service providers. At the same time, we need to reduce infrastructure costs for service providers and enable them to compete effectively on a level playing field. Small and innovative new players should be encouraged to play their roles in the development of the country.

Let’s hope that our illness can be cured. - FMT

Mohamed Awang Lah has been involved in internet services since its inception in the mid-1980s. Prior to 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 MMKtT