Ministry of Economic Development Home| Contact MED|


 
 
 

Links to this page were:

Section Subnavigation Links:

Chapter 6: The Decision on Whether to Impose Regulation


Review of Regulatory Control Provisions Under the Commerce Act 1986: Discussion Document

Ministry of Economic Development
[ Last Updated 30 March 2007 ]


88. In addition to a regulatory-specific purpose statement, further guidance is likely to be desirable on when regulation should (and/or should not) be imposed on a firm or sector. Such guidance would improve certainty for decision-makers and businesses with respect to when regulation may (and may not) be imposed.

89. Note that in most jurisdictions overseas the decision on whether a firm or sector should be subject to economic regulation has already been determined and is set in legislation. Detailed cost-benefit analysis is not normally undertaken on whether control should be imposed. As far as the Ministry is aware, prices for basic infrastructural services with monopoly characteristics are controlled in some way in all other OECD countries.

90. While it appears that New Zealand is unusual in providing for a decision-making process, and in particular detailed cost-benefit analysis, on whether economic regulation should be imposed, the Ministry proposes that this should continue to be the case. Regulation incurs costs and may result in a net cost to the economy, and accordingly should be a last resort measure.

Current provisions

Under Part 4, the Minister of Commerce (or Energy in the case of energy matters) may make a recommendation to the Governor-General that control be imposed (by Order in Council) on a given good or service if two specific thresholds are met (s52). Specifically, s52 states that:

"Goods or services may be controlled if:

  • the goods or services are, or will be, supplied or acquired in a market in which competition is limited or is likely to be lessened; and
  • it is necessary or desirable for those goods or services to be controlled either:
    • in the interests of persons acquiring the goods or services (whether directly or indirectly), if the goods or services are acquired from a person who faces limited or lessened competition for the supply of those goods or services; or
    • in the interests of suppliers, if the goods or services are supplied to a person who faces limited or lessened competition for acquisition of those goods or services."

The competition test (whether competition is limited or likely to be lessened) is relatively low. It likely applies to many markets in New Zealand, including many where control is clearly undesirable, and thereby fails to provide certainty for businesses as to when regulation will likely be imposed.

The acquirers' benefit test (whether there are net benefits to acquirers from control) may not be appropriately targeted as it does not require explicit or separate consideration of efficiency effects. The role of economic efficiency was illustrated in the Airports inquiry, where Minister decided that control of Auckland International Airport was not desirable because, despite positive net benefits to acquirers from control, there was a relatively large net efficiency loss and the benefits to acquirers were not large enough to justify such a loss.

The Act also currently provides no specific guidance for when control should be imposed. This has resulted in debate regarding the legislative intent of:

  • whether the tests for when control should be recommended are different from the tests in s52;
  • the applicability of the overall purpose statement of the Act (and what that purpose statement means in the context of markets which are not subject to competition); and
  • the applicability of s3A of the Act, which requires the Commission, when considering whether or not conduct will result in a benefit to the public, to have regard to any efficiencies that will likely result from that conduct.

The Act also requires separate decisions on whether and how control should be imposed. This means that decisions on whether to control will be based on incomplete information. Furthermore, the separation is arguably time-consuming and costly as it requires two, largely duplicate, processes. As such the regulatory processes may not be cost-effective and timely.

Part 4A

Under Part 4A of the Act (the electricity lines threshold regime) a breach of a threshold triggers the Commerce Commission's consideration of whether control should be imposed.

6.1 When regulation may be imposed

91. This section considers the criteria to guide decision-makers on when a firm (in a firm-specific inquiry) or sector (in the case of a sector-wide inquiry) may be regulated.

Competition test or criterion

92. A competition criterion, covering the extent of, and scope for, competition, provides some guidance to business, the Commission and the courts as to one of the minimum necessary conditions for regulation. It is generally agreed that regulation should only be considered where market power is significant and sustained and cannot be more effectively addressed by other means.

93. One option is to test, as a prerequisite for regulation, whether the relevant goods or services are supplied by a natural monopoly. A natural monopoly is where the goods or services are most efficiently supplied by one supplier. However, the economic proof of the existence of a natural monopolist is highly complex and requires that there be subadditivity7 of costs across the relevant output range. A major disadvantage of this option is that it is not easily understood and, by virtue of being cost-based, would not have the predictability value of other competition tests.

94. Accordingly, it may be appropriate for the standard to be somewhere between the range of a "substantial degree of market power" to "natural monopoly". The term "substantial degree of power in a market" is used elsewhere in the Act (such as s36) so it has an established meaning.

95. Note that the proposed purpose statement (see Chapter 5) also includes reference to markets with little or no competition or prospect of competition. The wording in any competition criterion and in the purpose statement would need to be brought into alignment during the legislative drafting process.

96. Possible options include:

Competition criterion

[Goods or services may be regulated if:]

Option A

There is little or no competition or prospect of competition in the relevant market

Option B

The goods or services are supplied by a person or persons with a substantial degree of market power.

97. While the above options are "tougher" tests than the current test for when control may be imposed ("competition is limited or likely to be lessened"), it is expected that the decisions would not necessarily differ from those that have been made to date on whether control should be imposed. To date, control has only been imposed in markets with natural monopoly characteristics and it is likely that the tests proposed above would have been passed for those markets.

Efficiency and distributional (consumer welfare/protection) criteria

98. Following from the discussion in Chapter 4 on the policy objectives of economic regulation, the Ministry proposes both an efficiency and a consumer welfare/protection or distributional test for when control may be introduced. The first test is economic efficiency; while the second test (assuming that economic efficiency test is negative) is similar to the net acquirers' benefit test in the current legislation.

Efficiency and distributional tests or criteria

Goods or services may be regulated if economic regulation is necessary or desirable to:

(a) promote efficiencies in a market; or

(b) provide long term benefits to persons acquiring the goods or services that [substantially] [clearly] exceed the direct and indirect costs of regulation.

99. The second limb of the test (net acquirers' benefit test) offers two alternatives in square brackets as to the extent that long term benefits to acquirers exceed the direct and indirect costs of regulation, namely "substantially exceed" or "clearly exceed". The views of submitters are sought on the preferred weighting of consumers' or acquirers' benefits to any costs.

100. Potentially, further, more explicit guidance could be provided to assist the decision-maker in the weighing up of any direct and indirect costs against the consumer welfare benefits of limiting monopoly rents. If such guidance is provided, it may result in greater regulatory certainty for decision-makers and businesses as it would indicate an explicit minimum standard that must be passed for regulation to be imposed. However, such an explicit standard (e.g. a specific figure for how much economic efficiency loss is acceptable to transfer $1 from producers to consumers)8 is likely to lack flexibility to account for policy objectives of different governments and for industry/firm specific circumstances. It might also be impractical, as it would require an unattainable degree of precision around the appropriate weight. The Ministry therefore considers that such explicit trade-offs should not be set in legislation.

Other issues

101. The current legislation allows control to be introduced where it is necessary or desirable to benefit both acquirers and suppliers of goods and services. The latter applies to markets with monopsony characteristics, i.e. where there is only one buyer of goods and services. Such markets are relatively rare, and it is questionable whether the power to regulate is needed to address this situation. Submissions are sought on this point.

102. The "net acquirers' benefit" test in the current legislation also refers to persons acquiring goods or services whether "directly or indirectly"). Again, submissions are invited on whether the "directly or indirectly" provision should remain. The effect of the provision is likely to be that the Commerce Commission does not need to establish (as part of its cost-benefit analysis on whether the tests for when control may be imposed have been met) whether the benefits of any reductions in monopoly rent in an upstream market will be passed through to end-users. The assumption behind this provision is that competition in downstream markets will ensure that benefits flow to end-users, or alternatively that if there are enduring factors that preclude such pass-through (such as lack of competition due to inappropriate barriers to entry), that these matters should be addressed as a separate issue.

103. Submissions are sought on whether the Commission should be required to establish that any benefits from control will be passed on to end-users in New Zealand as a necessary pre-condition for control, or whether the present provisions should remain.

6.2 When regulation should be imposed

104. While the Ministry considers it is clearly desirable to state the purpose of regulation and when regulation may and may not be imposed, the Ministry does not propose that there be a legislative test or criterion that states when regulation should be imposed. That is, the Ministry proposes that the only test in the legislation for imposing regulation (in addition to having a regulatory-specific purpose statement) should remain a "may" test.9 The reasons for this is to retain flexibility for the Minister to take into account wider policy matters and to retain the presumption that regulation is a last resort.

6.3 Who should decide whether to impose regulation?

105. There are several options as to who should make the decision on whether or not to impose regulation on a given firm/industry:

  1. The Minister (by way of a recommendation to the Governor-General for an Order in Council) following a recommendation from the Commerce Commission or on his/her own volition;
  2. The Minister (as above) but only after receiving a recommendation from the Commerce Commission; or
  3. The Commerce Commission could have the power to declare that a given firm/industry will be regulated.

106. The main argument for having the decision made by the Minister is that the decision on whether or not to regulate, while being based on technical analysis, often comes down to a "policy-type" judgement. In particular, it often requires judgements on how to make trade-offs between conflicting policy objectives. For example, regulation may be expected to limit monopoly rents and hence improve or maintain consumer welfare, but may come at a net cost to the economy.

107. Furthermore, there may also be wider factors that need to be considered, such as effects on business confidence, effects on overseas investors, credibility of the overall regulatory regime and so forth.

108. Policy decisions of this nature are usually considered to be the prerogative of government where political accountability mechanisms are designed to ensure a robust decision-making process.

109. On the other hand, the decision does require extensive informational and technical input, such as calculation of the costs and benefits of regulation versus no regulation. Such expertise usually resides with a specialist regulator, such as the Commerce Commission. The Ministry proposes that the Minister should make the decision on whether or not regulation should be imposed on a particular firm/sector. However, the Minister should be required to seek the advice of the Commerce Commission. For example, the Commerce Commission would be required to advise on whether or not the purpose of the regulatory provisions and the tests or criteria for when regulation may and/or may not be imposed, as set out in the Act, have been met.

110. In other words, the current decision-making powers should be retained, except that the Minister's discretion to recommend regulation (to the Governor General) without seeking a recommendation from the Commerce Commission would be removed.

111. For the avoidance of doubt, if the Part 4A regime is retained, it is proposed that the Commerce Commission retain its current powers to decide on whether to control a firm in the event of a threshold breach.

6.4 Sequence of decisions on whether and how to regulate

112. As noted earlier, the Act currently requires decisions on whether and how to impose regulation to be made sequentially. The alternative would be to require that the decisions be undertaken at the same time. That is, the Commerce Commission, in estimating the costs and benefits of regulation versus no regulation, would be required to recommend how regulation would be imposed (if the Minister were to decide that it should be imposed).

113. The main implication of the Commission's inability to consider how regulation will be imposed when undertaking the "whether to regulate" analysis, is that the Minister must rely on incomplete information in deciding whether or not to impose regulation. Furthermore, in the event that the Minister recommends that regulation be declared, a second inquiry is necessary to determine the terms of control. This inquiry is potentially just as significant in complexity and length as the first one.

114. A possible disadvantage of compressing the current two-step process into one stage is that the decision on whether to impose regulation may take longer. This may increase uncertainty for companies as to whether they will be subject to regulation while the decision is being made. The main gain from concurrent decisions, however, is a better informed, and potentially timelier decision-making process (by avoiding two separate inquiries) with a saving in resources and greater certainty for interested parties. Note however, that a further proposal in this document related to setting input methodologies (outlined in Chapter 8) may extend the overall timeframes somewhat.

115. The Ministry proposes that, on balance, it is desirable that the decisions on whether and how to regulate be undertaken simultaneously to the extent possible.10 The Ministry considers it is important that the decision on whether to regulate a given firm or sector is based on as complete an information set as possible, given the significant implications of a decision to regulate.

Questions for submitters: Chapter 6

Q1. Do you agree with the proposed criteria for deciding on whether regulation may be imposed?

Q2. If you agree that one of the tests for whether control may be imposed should be where the long term benefits to acquirers exceed direct and indirect costs, do you consider that such benefits should (a) "substantially" or (b) "clearly" exceed costs, or should there be some other guidance on weighting?

Q3. If you agree that one of the tests for whether control may be imposed should be where the long term benefits to acquirers exceed direct and indirect costs, should those benefits be considered regardless of whether acquirers acquire the goods and services directly or indirectly, or should it be necessary to establish that benefits will be passed on to end users (or consumers or end-acquirers)?

Q4. Should the current provisions in the Act allowing control to be imposed in the interests of suppliers (to a monopsonist) be retained?

Q5. Do you agree that there should not be a legislative test for when regulation should be imposed?

Q6. Do you agree that the Minister should remain the decision-maker on whether control should be imposed under Part 4, but that that the Minister must receive a report and recommendation from the Commerce Commission before making a decision?

Q7. Do you agree that the decisions on whether and, if so, how to regulate should be undertaken simultaneously rather than sequentially?


7 Subadditivity is a mathematical concept which, in this context, means that output is produced at least cost by one firm.

8 For each gas pipeline company the Commerce Commission calculated the efficiency costs of transferring $1 of excess returns from suppliers to acquirers. The results were as follows: NGCT cost of $0.30 to transfer $1 to acquirers; NGCD $0.37 to transfer $1 to acquirers; Powerco cost of $0.17 to transfer $1 to acquirers; Vector cost of $0.09 to transfer $1 to acquirers; and WGL $0.71 to transfer $1 to acquirers. On the basis of these calculations the Commission recommended that control should be declared on Powerco and Vector, but not on the other firms. The Minister agreed with the Commission's recommendation. This implies that an efficiency loss of somewhere between $0.17 and $0.30 was considered acceptable in this instance.

9 For the avoidance of doubt, it may be desirable to make it explicit in the legislation that there is no separate or specific test for when control should be imposed. The issue of the nature of any "should" test (separate from or additional to the "may" test) has arisen in litigation on control decisions in the context of s56 ("the Commission may report to the Minister on whether or not an Order in Council under s53 should be made, amended or revoked").

10 Note that Chapter 9 considers whether a controlled firm should have a role in proposing its own regulatory control terms. If this approach is adopted, there may be merit in providing for the proposal of the firm to be considered after the declaration of control. Refer to section 9.2 for further discussion.



Back to Top