Clash of the broadband saviours

It's pretty apparent to most that South African broadband pricing is out of kilter with other developing countries. Many believe wireless is the solution to the country's lack of wired connectivity competition; however, there is only so much performance one can extract out of the available (shared) radio spectrum.

While many applications are perfectly suited to wireless, the advanced media-rich communication of the future will demand dedicated capacity that only wired connectivity can provide.

Logically some form of regulatory intervention will be necessary to counteract the stranglehold the incumbent operator has on fixed-line infrastructure, in order to bring local broadband pricing closer to globally competitive norms.

I believe some form of pricing control will be necessary, but where in the service dependency chain this is applied will have far-reaching consequences for how long it will take for pricing benefits to reach end-users, the sophistication of future product offerings and the very nature of how fixed-line competitors will evolve.

For simplicity of comparison, I've grouped the two ends of the intervention spectrum into "infrastructure unbundling" and "regulated wholesale".

In this approach, suitably licensed competitors are entitled to lease, at cost-based pricing, the 'raw' communications infrastructure components and facilities from the incumbent operator. Using these foundation building blocks, together with self-provisioning where applicable, competitors will build their own communication networks and their own unique end-user services.

This approach should support the creation of truly independent competitors by opening up the market and at the same time mitigating much of the incumbent's historical advantages.

In theory, it sounds like the perfect free market solution for SA's broadband woes. However, if one starts delving into the more practical aspects such as geographical dispersion, technology evolution and basic economics things start to look a little less rosy.

Wide geographical dispersion and lower density at the local loop aggregation points (exchanges) mean greater infrastructure spend per subscriber in terms of access concentrators (DSLAMs), backhaul links and core networks. Together with lower economies of scale in terms of total potential subscribers in the country, it means new fixed-line competitors are going to need very, very deep pockets.

Lastly, the real 'spanner in the works' is the continuing progress of fixed-line broadband technology. For newer copper-based broadband technologies, such as ADSL2+, to reach their maximum speed potential (eg, 24Mbps), local loop lengths cannot exceed much over 1 000m, which is not the norm in SA. This has grave consequences for the competitiveness of new entrants.

What would be the point of a competitor launching ADSL type services (<8Mbps) with access to the traditional local loop and exchanges, when the incumbent has already moved the DSLAMs to the street corner from where they are able to offer ADSL2+ services?

To remain competitive, new entrants would also need to start deploying their DSLAMs at the street corner with the result that investment costs start to escalate exponentially.

Full, unfettered competition is the only solution to bring SA's broadband market in line with the rest of the world.

Roman Hogh is head of technology and product strategy at MWEB Business.

This approach aims to regulate the incumbent at more of a service offering level rather than a 'raw' infrastructure component and facility level. Generally, the wholesale service offering, referred to as a 'bitstream service', would consist of the bundling of the local loop, access concentration (DSLAMs wherever situated) and backhaul to major regional centres, from where handover to the competitor's network would occur.

By clearly defining these wholesale bitstream services and applying cost-based price caps to them, regulation would ensure new entrants could enter that market in a price competitive fashion while not being (initially) burdened with significant infrastructure investment costs required to match the incumbent, even partially, in new technology roll-out and geographical service coverage.

That's not to say this approach is not without its own pitfalls. The fact will remain that a single organisation will dictate fixed-line convergence, installation lead-times and future technology adoption in the country. Also, the regulator will have its work cut out in trying to understand the incumbent's accounts practices with respect to input costs and specifically interaction between wholesale and retail divisions.

That said, regulated wholesale should provide enough scope for competition and differentiation in the areas of retail pricing, value-added services, operational support, and hopefully when competition arises within the international cabling arena, bandwidth options and performance.

In my opinion, full, unfettered competition is the only solution to bring SA's broadband market in line with the rest of the world.

However, there's no denying it would be unreasonable to believe any single organisation could in a few short years match the incumbent's fixed-line dominance.

Therefore, I believe a two-pronged approach incorporating both regulated wholesale and infrastructure unbundling would be best.

New fixed-line entrants could use regulated wholesale products to compete effectively in the short- to medium-term, thereby generating the revenues required to take advantage of unbundled infrastructure in the longer term.

* Roman Hogh is head of technology and product strategy at MWEB Business.