Competition policy and network regulation in changing energy markets

In 1993, the Hilmer Report presented a ground breaking review of national competition policy, establishing the framework for public utility regulation in Australia that has, in large part, defined today’s electricity industry structure. The emphasis of those reforms, particularly in respect of utilities, was to deliver better and lower cost services to customers. Hilmer’s principles have been a touchstone for Australian governments’ thinking in energy market reform, and continue to shape policy considerations on issues as diverse as ownership, market structure and the introduction of competitive forces to previous monopoly services.

Remarkable changes are taking place in the power sector in technology, demand and external, particularly environmental, pressures. These raise the possibility a range of valuable, innovative, even transformative services for customers.

In the face of this, the Energy Networks Association asked Euan Morton and Professor George Yarrow to go beyond the day to day regulatory shorthand about what the principles are said to mean, and more deeply assess how the underpinning approach and content of the Hilmer principles can be comprehensively applied to meet the long term interests of customers in this rapidly changing environment. A more detailed report is available here.


Applying Hilmer Principles in Changing Energy Markets

Professor George Yarrow
Mr Euan Morton, Principal, Synergies Economic Consulting

Historic context

The policy and regulatory regime for the electricity industry pre-Hilmer developed in a context of government owned, vertically integrated, self-regulated monopolies. The industry was inefficient, with no competition and modest interstate coordination and trade. Under the Hilmer reforms, regulated natural monopoly[1] networks or network service providers (‘NSPs’)[2] were separated out from more competitive generation and retail supply, which were largely unregulated. The judgement at the time was this approach best advanced the long term interest of customers.

Hilmer’s cost benefit test: balancing policing against greater competition

The Hilmer reforms were based on an implicit cost benefit test: regulation should be applied only when the costs of regulation (including costs from constraints on business form) were less than the benefits and were therefore in the long term interests of consumers.

The fundamental trade-off is the costs of policing business conduct against the benefits that can be expected to derive from greater competition. The preference for the electricity industry post-Hilmer was to select rules, including vertical separation, which would at least in principle not require large policing costs to foster a sufficient increase in competition.

Those rules made sense at that time. The belief, then, that policing costs of any structure other than vertical separation would be substantial, was held with strong justification but was contingent on three considerations that are harder to justify today:

  • the legacy of state ownership and limited private sector participation;
  • limited scope economies (between the stages of production), ‘vertical externalities’[3] and price distortions; and
  • alternative regulations suited to a structure with a greater degree of vertical integration would have been more costly.

But the costs and benefits are very likely to change as markets change.

A changing sector

Identifying where the balance lies depends on the relevant economic context. The objective of regulation, to foster the long term interests of customers, has not fundamentally changed, but technology changes in the industry can change the optimal structures to achieve this objective.

The power sector has rarely seen such rapid technological change (e.g. smart metering, storage, renewables, distributed generation, smart grids, microgrids, and a plethora of improvements in data collection, monitoring and control), business changes (such as the emergence of distribution system operators),[4] and external pressure (e.g. climate change).

These changes herald the prospect of a range of innovative services and opportunities to meet customer demands using innovative and lower-cost technological mixes. It is early days for many of the new technologies and services.

The process of competition will be critical in seeking out these innovations, determining which do and do not deliver value to customers, the speed with which they are taken up and the best ways in which the most effective innovations can be provided to customers. That, in turn, requires a careful assessment of the market and regulatory arrangements most likely to foster that process of competition.

Many of the emerging disruptive technologies operate across several levels of the electricity supply chain and not solely on one side of a clear, existing boundary between contestable and natural monopoly (network) levels, boundaries which themselves are defined by regulatory decisions taken in an earlier, rather different economic context.

Energy networks have a potentially important role to play in providing services to customers that make use of these innovations. They have a number of attributes that can make them effective contributors to innovation. Networks, for example, have:

  • a great deal of knowledge about their customers’ demand characteristics and preferences;
  • a comprehensive understanding of their network, and how and where to invest to deliver superior outcomes (such as exploiting opportunities for scale economies); and
  • the skills and resources to implement investments and design choices, including through pricing reforms.

It is therefore pertinent to ask whether the long term interests of customers could be better met by evolving current regulation to a model that might more effectively capture the increasingly significant efficiency benefits of scope economies and reduced vertical externalities, by allowing greater participation of NSPs in the competitive markets from which these new services will spring. At the same time, the long term interests of customers requires recognising that NSPs are likely to retain monopoly power over some network activities, and managing that concern in the least restrictive effective manner is necessary.

A possible approach — calibrated regulation

Hilmer, in essence, espoused the ‘necessity’ principle: competition restrictions should only be considered where there is no less restrictive way of ensuring network ownership does not adversely affect competition. The ‘no less restrictive’ criterion cannot be settled once and for all in relation to NSPs. It will depend on the specifics of the context at a particular time. The question, therefore, is whether it is satisfied by current arrangements in the emerging context.

The best form of regulation going forward will not be resolved easily, but it is possible to set out some alternatives that merit serious consideration. Some may require evolving away from the current model of ‘strict’ structural separation and the presumption that NSPs should be excluded from (or have little involvement) in contestable markets.

The fallacy of the ‘one-size’ fits all

There is nothing novel in seeking to adapt rules to context; one obvious example is when businesses below a threshold size are exempt from particular regulations, because compliance costs are judged excessive relative to the benefits.

In electricity, there is a wider range of possibilities than the near blanket prohibition of NSP participation in contestable activities; for example, a distributor that separates out its network assets to form a DSO, which then tenders for inputs (such as wires and storage services) to meet defined customer needs, is a very different regulatory proposition from today’s distribution networks.

The process of competition, particularly in the development of innovative/ disruptive services, depends upon participants in those markets being able to adopt business forms and relationships that better allow them to manage risk and incentives. At times of rapid change, when the nature and value of innovations is uncertain, selecting appropriate business structures can be critical: indeed it is part of the competitive discovery process itself.

Where innovations affect multiple levels of the supply chain — for example electricity storage which can affect the provision of network, wholesale power and retail power services — vertical business relations can be critical, and vertical integration is often an appropriate business form for managing the risks of delivering innovative services (addressing vertical externalities) and exploiting scope economies. Regulatory constraints on business form necessarily impact competition, in turn affecting the extent to which customers benefit from innovation.

Differentiated rule-books

This raises the intriguing prospect that, rather than regulators defining one set of rules proscribing certain business relations, regulators could offer a ‘menu’ of rule books suited to the business choices of the market participants. That is, different regulatory rules could be applied to the various business forms NSPs adopt. NSPs that sought to compete in identifying, developing and supplying innovative services to customers might operate under a different ‘rule book’ from NSPs that were content not to do so.

The regulator has several tools at its disposal (e.g. ring-fencing, accounting separation, information disclosure, standardised costs, pricing rules, prescription and proscription). The extent to which a particular set of rules — or combination of these tools — supports competitive neutrality and non-discrimination might then depend upon, or be calibrated against, the extent to which an NSP’s business’s choices affect policing costs and the risks of anti-competitive behaviour.

The balance inherent in the Hilmer principles then comes down to whether the set of rules is well calibrated (i.e. appropriate and proportionate) to preferred business choices, particularly in relation to their likely implications for the long term interests of customers.[5]

Could calibrated regulation operate in Australia?

In Australian it might be feasible, for example, to adopt a rule-book that provided for:

  • NSPs that did not participate (or had little involvement) in contestable segments  might remain under broadly current arrangements;
  • NSPs that adopted a split structure comprising, say, a distribution network assets business and a distribution system operation business (a DSO) might operate under a regulatory model more akin to telecommunications; or
  • NSPs that participated in contestable markets under their current structure might be offered a set of rules with more onerous accounting separation, information disclosure and ring fencing, with attendant higher compliance costs.


Competitive outcomes, particularly regarding disruptive innovations, are uncertain and unpredictable, so there is little reason to believe that regulators are well placed to predict them. Markets rely upon the process of competition to discover and provide the best mix of services, technologies and innovations to customers. Given rapid technological change, it is preferable to rely upon competition rather than the dictates of regulators to deliver superior outcomes to customers.

Diversity is generally good for innovation.  So long as proportionate regulatory responses to clearly identifiable problems are possible (via calibrated regulatory measures), removing incentives for NSPs to compete imposes avoidable costs that policy-makers and regulators should recognise and re-assess when there are material changes in circumstances.



[1]     We note in passing that what constitutes a ‘naturally’ monopolistic activity in modern classification schemes – which themselves differ from those used in classical economics – is itself context-dependent; the boundaries can shift with changes in technology, input prices and demand, for example.

[2]     Also system control and market operation.

[3]     A vertical externality exists where expansion of a business activity at one level of a supply chain confers benefits (e.g. via lower costs or increased demand) on a separate business operating at a different level. Vertical integration is one, but only one, way to ‘internalise the externalities’ and increase the payoffs from expansion, including by innovation. Video games are an oft-cited example of the problems that can arise in the development stage: absent software that can run on it, there is no demand for an innovative console; absent an innovative console, there is no demand for software than can run on it.

[4]     DSOs refer to distribution system operations, a form of business structure emerging in parts of the US. The main paper provides a brief summary of all these changes impinging on the power sector.

[5]     A flavour of this approach is to be found in the policies of Clare Spottiswoode, Director General of Gas Supply in the UK in the 1990s. For a more detailed description of Ms Spottiswoode’s approach see the main paper.