The Federal Government released the Productivity Commission's (PC) final report (Report) on 11 February 2014. The Report proposes yet another test for the controversial 'criterion (b)'. However it is not clear that the new test would have the effect that the PC envisages in circumstances such as those in the Fortescue cases. The Federal Government has said that it will respond to the Report following the upcoming 'root and branch' review of competition policy.

So, you may ask: where is the National Access Regime headed? How will State based access regimes be affected in the meantime?

Our views in short are:

  • The private test for whether a facility is uneconomical to duplicate (laid down by the High Court in Fortescue) will remain law for the foreseeable future.
  • Amending criterion (b) in the manner proposed by the PC would:
    • Reintroduce a variant of the natural monopoly test to the National Access Regime; and
    • Reopen export infrastructure to the possibility of declaration.
  • Many State based access regimes which were imposed pre-Fortescue, under a natural monopoly test, will remain in legal limbo. This may result in:
    • Infrastructure owners seeking review of State based regimes;
    • States imposing access by fiat rather than under rule of reason declaration criteria.


The National Access Regime is contained in Part IIIA of the Competition and Consumer Act 2010 (CCA) (previously the Trade Practices Act 1974).

The National Access Regime provides a number of mechanisms by which third parties can gain access to certain nationally significant infrastructure facilities. The most notorious of these mechanisms is declaration:

  • If the infrastructure and associated markets meet a number of criteria then the service provided by the infrastructure is 'declared'; and
  • If the infrastructure is 'declared' then an access seeker can force the infrastructure owner into negotiation with recourse to binding arbitration by the Australian Competition and Consumer Commission (ACCC).

The High Court ruled on the declaration mechanism in the National Access Regime in 2012 in Fortescue (The Pilbara Infrastructure Pty Ltd v Australian Competition Tribunal [2012] HCA 36 (14 September 2012)).

Variants of the declaration mechanism, using very similar criteria, are used in a number of State based access regimes (and the National Gas Law). State based access regimes may be (and most are) certified as effective under the National Access Regime. That certification precludes the operation of the declaration mechanism contained in Part IIIA, although a variant of the declaration mechanism in State legislation may still apply. For example, the State based access regime applying to the Queensland coal rail network is certified as effective under the National Access Regime (and hence cannot be declared under Part IIIA) but is subject to the declaration provisions in the Queensland Competition Authority Act 1997.

Another mechanism in the National Access Regime is for infrastructure owners to give an access undertaking to the ACCC under Div 6, Part IIIA of the CCA. These are sometimes called 'voluntary' undertakings however there has not to date been a Part IIIA undertaking given voluntarily. Rather, these have generally been required under separate legislation (this is the case for the wheat export undertakings which are required under section 7 of the Wheat Export Marketing Act 2008).


The most controversial aspect of declaration under the National Access Regime is criterion (b). Criterion (b) makes access conditional on (in part) whether it would be 'uneconomical for anyone to develop another facility to provide the service.' Competing interpretations of criterion (b) go to the heart of one (often claimed) intention of the National Access Regime in relation to productive efficiency: is the National Access Regime intended to:

Policy option 1: Intervene to address inefficient duplication of significant infrastructure; or

Policy option 2: Let the market address inefficient duplication of significant infrastructure.

If policy option 1 is the intended purpose of the National Access Regime then criterion (b) should be a test comparing the costs of meeting demand by sharing the existing facility to the cost of meeting demand by duplicating the facility.

If policy option 2 is the intended purpose of the National Access Regime then criterion (b) should be a test of whether it would be profitable for another person to provide duplicate infrastructure. If it is profitable to duplicate then parties, realising that, would make a commercial deal to share if that was profitable; if it was not profitable to share then duplicate infrastructure would be developed. The underlying intention of policy option 2 is that access should only be given if no potential third party user could profitably develop a duplicate facility.


The Federal Government started the PC's inquiry into the National Access Regime in October 2012. Following initial public consultation the PC released its draft report in May 2013. The PC undertook further public consultation before providing its final report to the Federal Government in October 2013. The Federal Government released the final report on 11 February 2014.

The PC's key points in its final report are:

  • The National Access Regime should be retained, although some amendments are proposed.
  • Governments considering whether to regulate access should demonstrate that there is a lack of effective competition that is best addressed by access regulation.
  • The ACCC should have the power to direct infrastructure expansions and extensions but this power should be exercised subject to safeguards and ACCC guidelines.

The PC proposes two significant changes to the declaration criteria. The first significant change is to criterion (b) which the PC proposes should be a variant of a natural monopoly test (Report, p160). The differences between the PC's proposed test for criterion (b) and the natural monopoly test used by the Competition Tribunal in Fortescue are: that potential demand should be measured by reference to the market demand for infrastructure services rather than demand for the particular facility; and that coordination costs should be included.

The second significant change proposed by the PC is that criterion (f) become a full blown net public benefit test. Under the PC's proposal the public benefit would have to be positively established. This is in contrast to the, negatively framed, criterion (f) which requires that access 'would not be contrary to the public interest.'

Criterion (b) proposal: a logical disconnect?

So far so good. However there appears to be a logical disconnect between the PC's recommendation for amending criterion (b) and the PC's view that the purpose of access regulation is not 'to improve productive efficiency through avoiding wasteful duplication' of infrastructure (Report, p86).

The PC is at pains to state that the purpose of the National Access Regime is to address a lack of competition (allocative efficiency), rather than productive efficiency that might arise from duplicating infrastructure. In fact the PC approach proposes that access regulation should address competition issues in two distinct notional areas:

  1. Markets for infrastructure services where there is an enduring lack of competition due to natural monopoly; and
  2. Markets where competition is dependent on third parties gaining access to the above mentioned infrastructure services.

The problem with this approach is that the PC's proposed change to criterion (b) is a test for productive efficiency in the market for infrastructure services. The PC's proposed criterion (b) does not address expressly the state of competition for the infrastructure services.

The educated reader might respond to this by saying that a test for natural monopoly infrastructure implies that there will be no competition for infrastructure services because it is generally not profitable for anyone to develop a second natural monopoly facility.

The problem with that response is that the facts in Fortescue had natural monopoly rail infrastructure that was profitable to duplicate. It was profitable to duplicate because world prices for iron ore were well above the costs of production in the Pilbara (prices were set by the costs of marginal producers Pilbara producers are 'inframarginal' producers). This arguably allowed Pilbara producers the luxury of some inefficiency in the form of duplicated natural monopoly infrastructure.

The PC is of the firm view that, on facts such as those in Fortescue, access should not be granted. The Report's 'stylised example' (Report at Box 2, repeated at Box 3.8) describes a vertically integrated miner who owns a rail line and produces iron ore where prices are set in world markets. The PC states that access regulation is not warranted in that circumstance. The problem is that it is not clear that the PC's proposed criterion (b) test, a variant of the natural monopoly test, would have stopped access being granted in that case (see Heydon's dissenting judgment in Fortescue which would have declared the Hamersley and Robe lines on the basis of a natural monopoly test). It is clear however that the private test stopped Fortescue getting access (see the Fortescue decisions in the Full Court and Tribunal).

The private test may have its problems as the PC notes (Report pp157-158). But the PC wants an access regime that picks up natural monopolies but carves out export facilities such as those in the Pilbara and other circumstances where natural monopoly characteristics only mean that duplication would be productively inefficient (but otherwise profitable). The private test neatly achieves that carve out in practice.

Perhaps what troubled the PC is that the private test worked in practice but not in theory or that the PC sought to narrow declaration still further. We suggest that a better way to achieve that may be to leave criterion (b) unchanged (and carrying the High Court's private interpretation) but add its proposed natural monopoly variant as an additional criterion.

Criterion (f) proposal: a do-all criterion?

We consider that the PC's proposed full public benefit test for criterion (f) has problems.

The current criteria apart from criterion (f) require benefits of certain kinds (for example, promotion of a material increase in competition in at least one market). Current criterion (f) allows a decision maker to check that, in spite of benefits accruing from the other criteria being satisfied, there is not some other matter that will weigh against the public interest.

Under the PC's proposal, criterion (f) would become the main game before decision makers. Anyone who has been involved in the full public benefit test elsewhere in the CCA can attest that it is a significant burden for an applicant to bear and a substantial task for decision makers to assess claimed benefits and counter claims.


The Federal Government's response to the Report has been to put the National Access Regime on the list of matters to be considered in the upcoming 'root and branch' review of Australian competition policy. The root and branch review is due to be completed in December 2014. The Government is, in effect, getting a second opinion and giving itself some further thinking time.

This and the Government's favourable stance toward the big miners suggests that the current private test (which the big miners favour) will remain law for the foreseeable future.


Given the issues with the Report that we have outlined above we consider that the Government is well advised to take the time and opportunity afforded by the root and branch review to give the National Access Regime some further thought.

However in the meantime there are State based access regimes in legal limbo. This is because the High Court's 2012 ruling that criterion (b) is a private test remains the law but a number of facilities declared under State law were declared earlier when criterion (b) was interpreted as a natural monopoly test. We consider that some of those facilities would not satisfy the current private test and the owners could seek to have declaration revoked. Putting off a decision on the National Access Regime makes it more likely that some owners will seek revocation of declaration.

In addition, the States are likely to want a say in what happens to third party access law in Australia. The National Access Regime was enacted in 1995 following agreement between Federal and State governments (see clause 6, Competition Principles Agreement, 1995). The States, particularly Western Australia and Queensland, have taken a close interest in third party access law as it applies in their respective States. Western Australia and Queensland have both sought to impose State based access regimes by executive decision and not subject to prolonged review by the judiciary (as happened under the National Access Regime in Fortescue). Rather than follow the PC's advice, which would weaken the ability to bring or keep facilities under access regulation, the States might in future simply impose third party access by fiat rather than the rule of reason, criteria based approach in the National Access Regime (and currently in State based regimes).