The answer, according to institutional shareholders Elliott International, is ‘yes’. The investment firm describes it as the logical next step following operational separation. Elliott’s proposes a “simple structural split” of Telecom’s retail business from its wholesale and access arms, with both under separate ownership.
Yet, the proposal has received a cool industry reception at best — and deservedly so. Structural separation is far from simple. It carries with it far-reaching industry and regulatory consequences, and the risk of unintended long-term outcomes, with uncertain benefits.
Generally, structural separation is mooted as a regulatory remedy — a final and extreme redressing of structural imbalances within a market aimed at driving competition.
As such, its popularity has waxed and waned. The classic case was the separation and divestiture of the Bell System in the US, in 1984. This was, however, a very different separation model, designed to separate out local and long-distance operations in the US market. It did result in a very competitive infrastructure — although merger and acquisition activity has reduced the separate entities to three today.
However, the true separation of an incumbent’s network and wholesale assets from its retail operations is unprecedented anywhere in the world (and, indeed, regarded as almost heretical by many incumbents). Yes, the concept is being raised again, driven in part by private equity investors who see it as a means of unlocking shareholder value — as in Ireland — or in the context of a move to a nationalised, open-access fibre network, as in Australia and Singapore.
But it has not been implemented, simply because of the complexity and the risk.
Ireland is an interesting case study. The former state telecomms company, Eircom, is negotiating with the regulator, ComReg, on the voluntary structural separation of its business into two companies. Under the plan, Eircom’s current owners, Australian investment fund Babcock & Brown, would continue to own and operate the network, while the retail and mobile customer base would be sold.
But this is a complex proposal. Babcock & Brown can push Eircom’s separation through without regulatory approval, but it needs assurance from ComReg that the pricing access regime will ensure the fund a strong enough return on investment. ComReg, for its part, is still considering the potential ramifications.
These include the following: Babcock & Brown may provide assurance of long-term network investment following separation, but how would this realistically be enforced? In addition, what type of access regime could be implemented and how would existing and future local loop unbundling be provisioned for?
And, if a service provider decides to invest in a separate and competing network, should this be allowed or banned, enshrining Eircom’s network monopoly?
Herein lies the challenge facing Elliott’s proposal in NZ. Any private equity firm driving a separation model will be looking to unlock value through separation, and will focus, justifiably enough, on the accompanying return on investment, and how asset values can be maximised for future sale.
But the incremental enhancements in shareholder value may come at the cost of long-term market development and competition — which is the regulator’s concern.
Structural separation is absolute. There is no easy way to return once ownership changes. It can simplify the regulatory framework, as it fundamentally changes the incumbent’s business model: there is no incentive for the incumbent to use its market power to discriminate against competitive access-seekers if it no longer owns the network.
Nevertheless, the complexity of managing cost-effective wholesale access, while ensuring ongoing investment incentives and continued innovation to meet market demand, still exists — just under new ownership.
The risk is that, following separation, wholesale and retail markets get out of sync and become disconnected. Distancing network investment from end-user influences means network development may come to embody a ‘lowest common denominator’ service approach, and so fail to keep pace with rapidly evolving market demand. Indeed, if the network becomes a price-regulated monopoly there may be very little incentive to invest or innovate.
This is a particular risk for telecommunications, in contrast to other utility sectors such as power and gas. Technology is at the core of telecommunications, and it is getting more and more complex.
The convergence of fixed and mobile networks and the convergence of telecoms with other industries are additional complicating factors. The introduction of next generation networks blurs the dividing line between network access and core, and will over time transform the industry.
A vertical separation, which may make sense in today’s telecoms landscape, could become a barrier to innovation in an all-IP world.
In short, structural separation carries a risk of unintended consequences that may not initially be clear but which could require further, and massive, structural change to redress later.
Hence the shift to ‘operational separation’ — the separation of bottleneck assets into a separately governed, regulated and independently audited unit.
This allows the incumbent to retain control of the asset and still integrate some group functions, for scope and scale, while providing transparency and an equivalent service to access-seekers. Just as critically, it allows for the flexibility to revisit and fine-tune the structure if, over time, it isn’t working.
On this basis, it is hard to see structural separation as the “logical next step” for Telecom, as Elliott claims. In fact, today it represents an onerous intervention that increases costs and reduces regulatory certainty at a time of both increased competition and investment.
The debate may be a worth raising again if, in time, we see significant market failure or if separation becomes a critical part of a long-term national fibre network rollout.
But, right now, it is too much, too late and has too little going for it.
Nelson is telecommunications research manager for IDC New Zealand