• Home
• Regulation
• Competition Policy
• Commerce Act Pts 4/4A/5 Review
• Regulatory Control Provisions
• Discussion Document
• Chapter 9

• Competition Policy

• Commerce Act Pts 4/4A/5 Review

• Regulatory Control Provisions

• Commerce Amendment Act 2008
• Draft Legislation
• Discussion Paper
• Submissions
• Supplementary Submissions
• Cabinet Papers
• RIS/BCCS
• Media Statements
• Submissions
• Discussion Document
• Terms of Reference

145. This section considers issues relating to the design of a regulatory control regime including processes for setting control terms (i.e. price/revenue paths and quality requirements). In particular it considers:

1. The merits of using comparative benchmarking versus customised "building blocks" analysis to set thresholds and control terms
2. Whether provision should be made to regulate sectors as well as individual firms, and if so, what criteria or test should apply
3. The role of firms in the process of setting control terms, and in particular, whether a "propose/respond" model should be provided for.

9.1 Methodology for setting control terms

146. In setting regulatory control terms (i.e. price/revenue path and quality requirements), the regulator may employ comparative benchmarking and/or firm-specific analysis.

147. Comparative benchmarking uses a high degree of assumption, approximation and statistical averaging. The main benefit is that it is relatively non-intrusive into business activities and less resource intensive for the regulator. However, for this approach to be cost-effective there may need to be several firms in a controlled industry (e.g. >5).

148. Applying comparative benchmarking may, however, be problematic as companies are often very different from each other and their capital expenditure profiles differ too. For example, in a sector characterised by very lumpy investment, if one company is just starting a major capital investment cycle, its prices may need to rise so that it can earn a return on the new investment. Applying sector-wide regulatory parameters to this company could discourage it from carrying out essential capital expenditure and/or put it at risk of making non-commercial returns. If another company in the sector has no requirement for significant capital expenditure, its prices may be able to fall. Applying industry parameters to this company may result in it earning a substantial margin above its costs of capital.

149. A customised approach avoids the problems with comparative analysis by taking into account firm-specific investment plans, cost structures, demand growth, and so forth. The main disadvantage of this approach is that it is likely to be more intrusive for businesses and very resource intensive for the regulator, especially if applied to a sector comprising a large number of firms.

150. Note that these methodologies are not mutually exclusive and it may be possible to use them in combination with each other.

151. As outlined above, there are pros and cons of either using comparative benchmarking or setting customised control terms. It is likely that, for a given firm or sector at a given time, one methodology will be more cost-effective and appropriate relative to the others.

152. For example, comparative benchmarking is likely to be a more cost-effective methodology to set control terms for a sector comprising many firms, relative to setting customised control terms for every firm in that sector. Note that employing comparative benchmarking may mean foregoing some level of allocative efficiency that might otherwise be achievable if the Commission was to undertake detailed firm-specific analysis. However, this allocative inefficiency may be acceptable given the reduced costs of setting control terms and given that the most significant abuses of market power would be constrained.

153. It is worth noting here that benchmarking does allow for pressure to be put on a sector to improve its average efficiency over time. This is illustrated in the figure below:

→ Larger version of figure [11 kB JPG]

154. The Ministry therefore considers that the Commerce Commission should have the ability to either use comparative benchmarking or a customised approach, depending on which option provides the most cost-effective and efficient outcomes. The Ministry proposes that the legislation is clarified so that either approach, or a combination of the approaches, can be availed.

9.2 Sector regulation

155. The Act is currently unclear as to whether the Commission may recommend that a sector (with natural monopoly characteristics) be placed under economic regulation without needing to establish that regulation is justified for each and every firm individually in the sector.

156. Several considerations are relevant:

• As noted earlier, the convention has developed in New Zealand that comprehensive, quantitative cost-benefit analysis should be undertaken for each inquiry on whether control may and should be introduced, and that analysis is undertaken in detail for each firm in the sector. This is time-consuming¹⁴ and resource-intensive and appears to be unique to New Zealand.
• As noted in the previous section, there are some cost-effectiveness advantages in using comparative benchmarking. However, the detailed firm-by-firm cost-benefit analysis currently undertaken effectively loses these advantages since the analysis is of similar scope to customised building-blocks analysis.
• Again related to the advantages of comparative benchmarking, benchmarking requires all firms in a sector (above a de minimus size) to be included in the regime, since, by definition, it is based on comparing the performance of firms in a sector in order to put pressure on less efficient firms to become as efficient as industry norms. This process becomes less effective if a number of key firms are excluded from the regime.

157. The issue arises therefore as to whether provision should be made in the legislation to allow for a sector to be placed under some form of economic control involving a benchmarking process. This could take the form of providing for:

• The costs and benefits of regulating to be considered for a sector as a whole (rather than each individual firm within the sector, with the implication that the tests must be fully met for each firm), and, possibly
• The costs and benefits to be determined in qualitative terms rather than quantitative terms. (Quantitative analysis is likely to require detailed assessment on a firm-by-firm basis).

158. Submissions are sought on the above issues and options. In the meantime, options identified later in this document assume that it is possible to regulate a sector and not just individual firms within a sector.

9.3 Role of firm in proposing regulatory control terms

159. Irrespective of whether control terms are set on a customised basis or largely using comparative benchmarking, there are options around whether the controlled firm has a role in proposing its own regulatory control terms. That is, the control terms may:

1. be set by the Commerce Commission, with no ability for firms to propose alternative terms; or
2. be proposed by a firm to the Commerce Commission, which would then have some degree of discretion in accepting/rejecting the proposal (see following section).

160. There is significant information asymmetry between a firm and a regulator. Firms are likely to be best placed to provide firm-specific information, make firm-specific trade-offs, and understand their customers' needs. A more efficient outcome may therefore be achieved if the firm is initially able to propose its own control terms. For example, one option is for a controlled firm to be given the opportunity to propose control terms and submit them to the Commission for its consideration.

161. The information asymmetry argument only applies when control terms are set using firm-specific information. In cases where control terms are to apply to a number of firms in a sector, it is likely to be more appropriate (and cost-effective) for the Commission initially to propose the control terms. For example, the Commission could propose sector-wide price-caps and associated quality standards based on industry averages.

162. However, benchmarking has the potential disadvantage of being too tough for some firms, and therefore may discourage such firms from carrying out essential capital expenditure. Therefore, if the Commission proposes sector-wide control terms that would disincentivise investment in particular instances, it may be desirable to allow that company to seek customised control terms if it can demonstrate that the benchmark is not suitable in its case.

163. In allowing such companies to propose alternative terms, there is a potential risk that a large proportion of firms within the controlled sector would propose customised terms, thereby undoing the cost-effectiveness benefits of employing benchmarking. It would be important, therefore, to ensure that firms were disincentivised from applying for customised control terms unless they had a strong case for such terms (discussed further below).

164. Therefore, when imposing control on a sector using benchmarking, it may be desirable for the Commerce Commission to be required to set initial control terms with an ability for firms to propose customised control terms within a statutory timeframe (e.g. 90 days). If the firm decided not to propose customised terms, the initial control terms would become binding.

165. A time limit of 90 days to apply for customised control terms may seem too demanding. However, firms would have had lengthy advance notice of pending decisions on initial control terms so could start their preparation early. Moreover, firms should know their business sufficiently well to mount a case in a timely manner.

Discretion of the Commerce Commission in accepting or rejecting firms' proposals

166. If firms have the ability to propose control terms as discussed above, then there is a question of how much discretion the Commission should have in accepting/rejecting proposals. That is:

1. the Commission could be required to accept proposals that meet specified criteria; or
2. the Commission could be able to exercise more discretion in deciding whether or not to accept proposals.

167. There are pros and cons associated with each option. If the Commission is required to accept proposals that meet specified criteria, then there is a likelihood that the resulting control terms will be "looser" than would be the case if control terms were determined by the Commerce Commission. That is, there may be an upward bias in price paths.

168. However, this upward bias needs to be weighed against several risks associated with increasing the Commission's discretion in deciding whether or not to accept proposals. In particular, the costs of regulatory error tend to be asymmetric: errors leading to too-tough control terms and resulting in under-investment are likely to outweigh the costs of errors of higher than normal rates of return being allowed. Also, firms are in the best position to make firm-specific trade-offs and to make investment decisions. It may also be more cost-effective for the Commission to assess proposals against broad criteria.

169. On balance, under this option we propose that the Commission should be required to accept proposals that meet specified criteria for a "reasonable" proposal. These criteria should be set by the Commission as one of the key input methodologies. When the criteria are not met the Commission should be able to determine the appropriate control terms.

What might "reasonable" criteria look like?

170. Setting the criteria for what constitutes a reasonable proposal would require the Commission to consider (and consult on) what type of information and evidence, and to what standard, the firm needs to provide in order to persuade the Commission that its proposal is reasonable (and meets the requirements of the Act).

171. Such criteria might include matters such as the following:

• Compliance with the input methodologies (that is, an application would not be an opportunity to re-open issues relating to input methodologies, such as how to calculate WACC);
• Evidence that significant capital expenditure (that is, of a different level and character than in the recent, say 10 years, past) is required in order to meet customers' quality expectations and requirements;
• Evidence that full and proper consultation with customers on the requirements for the significant capital expenditure has been undertaken and that customers (or a reasonable proportion of consulted customers) agree with the company's proposals and price/quality trade-offs¹⁵;
• Evidence that the necessary capital expenditure cannot be financed within the proposed price/revenue path even with reasonable levels of efficiency gains;
• Evidence that the firm is efficient and that the scope for significant efficiency gains is limited;
• Evidence that the proposed quality terms are too stringent;
• A requirement for evidence to be certified by independent experts approved by the Commission (prior to the 90 day timeframe commencing); and
• Evidence of materiality.

172. Several additional provisions are suggested to remove or mitigate any incentives for firms to seek customised control terms for gaming reasons (e.g. to delay control decisions or overload the Commission). It is important that firms are only motivated to submit proposals if they clearly meet "reasonableness" criteria. Proposed additional provisions include:

• A requirement that the firm fund the Commission's reasonable costs in considering the firm's proposal;
• Power for the Commission to set control terms at an appropriate level different from (and if appropriate more stringent than) the initial benchmark control terms where the firm's proposals do not meet the Commission's pre-set criteria for a reasonable proposal; and
• That the Commission's initial proposed (benchmarked) control terms would apply from the time of a control decision, and pending the outcome of any application for customised terms.

173. Firms may be concerned about the level of discretion the Commission may have in assessing the firm's application against the "reasonableness" criteria. This paper proposes that this risk may be mitigated by the ability to take merits review of the Commission's decisions on final control terms (refer Chapter 12).

Order of decision-making (if firms have ability to propose control terms)

174. If firms have the ability to propose control terms as discussed above, then there is a question of whether the proposal and consequent consideration by the Commerce Commission should occur before or after the decision to declare control.¹⁶ That is, either:

1. Declaration of control first: If a declaration of control is made (based on the Commission's initial control terms), then firms may propose different control terms. Provisional control terms would apply in the interim;
   
2. Firm proposal first: If the Commerce Commission issues a draft recommendation to control, then the firm may propose different control terms. Only after the proposal and consideration process has occurred, is control declared and hence control terms imposed.
   

175. Each option has pros and cons. If control is declared first (and hence control terms apply before the firm has a chance to propose different control terms), then there is a possibility that firms will suffer losses in the interim. For example, this may occur if terms are set for a sector using benchmarking and the price path is too low for a specific firm. Another potential issue with this approach is that the decision to declare control would not be based on complete information. That is, the final control terms may differ from those on which the decision to control was based.

176. However, if control is declared after the proposal and consideration process, a long drawn-out process could occur. If a firm is making excessive profits, it will take longer for its market power to be constrained.

177. It also carries a risk of gaming by firms. In particular, given that control terms would not apply throughout the proposal and evaluation process, firms would have an incentive to overload the Commission with proposals and draw out the process to extend the period of no-control.

178. In the case of a sector inquiry, we consider that there may be value in providing for the proposal and evaluation process after a declaration of control. This is because the decision to control a sector is based on the characteristics of the sector in aggregate. The final terms for individual firms within the sector are not crucial to the decision to declare control over the sector.

179. Furthermore, we consider that the risk of the initial benchmarked control terms causing financial distress is very low. Benchmarked terms only (normally) set the X in a CPI-X formula and do not make a P_o (start price) adjustment. The reason for this is that a P_o adjustment requires a full building-blocks analysis. Thus benchmarked prices will only be substantially lower than the most efficient price if major capital expenditure is required. Such expenditure typically has significant lead time, and could be postponed during the proposal and consideration process.

180. The risk of a "too tough" price path could also be mitigated through an ability for a firm to recoup losses if the firm-specific control terms are materially different to those set initially (provided that any recovery is done in such a way as to minimise any price shock). In addition, the Commerce Commission could have the ability to prioritise proposals, and thereby consider as a priority proposals of companies who could show that they would make substantial losses.

181. On balance we propose that for firm-specific inquiries, the proposal and evaluation process should occur before a declaration of control. This is because, unlike a sector inquiry, the final control terms for the firm under investigation are crucial to the decision on whether or not to control that firm. It is unnecessary to wear the risk of the decision to control being based on incomplete information.