Efficiency targets – What are reasonable efficiency targets?

You're in the section: Frequently Asked Questions -> Price Level and Tariff Design -> Efficiency targets – What are reasonable efficiency targets?

[Response by Sophie Trémolet and Diane Binder, November 2010]

Regulators need to set efficiency targets in order to perform core regulatory functions, including setting fair and reasonable tariffs for the next price control period – and therefore incentivizing the utility to control its costs – and specifying and enforcing performance requirements, as described below.

Defining the efficiency factor in a price cap regulatory regime (referred as “X” in the RPI-X formula):

The X factor is a target productivity change factor. The calculation of the X factor is derived from the business plan submitted by the company, including its historical and projected costs and adjusted in the light of comparisons with other companies. According to Green and Pardina (1999), the target should be set to give the company a demanding, but not punishing target. If the firm is more efficient than the target, it can retain the benefit of such efficiency gains until the resetting of tariff levels at the start of the next price-setting period, when such gains are distributed between the firm and consumers.

Defining the X factor is critical for the long-term viability of any price cap regulation plan. If too small an X factor, the firm will easily overshoot the targets, realize excessively high profits and therefore risk losing public and political support. This is what happened in England and Wales in the early years following the privatization of water utilities, when efficiency targets were not sufficiently challenging and therefore led to “windfall” profits (such profits were later taxed by the Labour government through a windfall tax). If the X factor is too ambitious (i.e. too high), efficiency targets will be unachievable and the financial equilibrium of the firm can be threatened (Bernstein et al., 2000). In England and Wales and in some other countries, statistical benchmarking methods are used to help determine the relative efficiency of individual firms’ operating costs and service quality compared to their peers. This information can then be used as an input to setting values for X (Jamasb and Pollitt, 2001).

Setting targets for performance improvements :

Thee are set with the goal of increasing service standards and coverage. As mentioned in another FAQ[1], incentive regulation that provides incentives for cost reduction also potentially creates an incentive to reduce service quality.

The determination of “reasonable” efficiency targets is a complex issue. Some “common sense” rules can be defined, such as establishing binding [2] targets only for those costs that are controllable and focusing on important items such as technical losses, capex, etc. In addition, reasonable efficiency targets can be set by using the following methods:

  • Benchmarking of the utilities based on their relative efficiency. The aim of performance comparisons is to set a target path for the company’s costs. Countries such as the Netherlands, the United Kingdom and Norway have all adopted benchmarking as part of the process for setting the X-factor. Benchmarking identifies the most efficient firms in the sector and measures the relative performance of less efficient firms against these. Individual X-factors are then assigned to utilities based on their relative efficiency. Generally, the more inefficient a utility, the higher the X-factor assigned to that firm. The aim is to provide the firms with an incentive to close their efficiency gap with the frontier firms. However, in a number of countries, the lack of reliable data makes benchmarking challenging and to set efficiency targets on the basis of insufficient results is not recommended.
  • Consulting stakeholders about achievable efficiency targets. It is not possible to be rigidly prescriptive about how efficiency targets can be established. Instead, Shugart and Alexander (2009) set out a process for determining those targets. This is an area where the role of an external advisor can be vital if credibility is to be established. In Great Britain, the electricity sector regulator, Ofgem, sets targets through a consultation process with major stakeholders. For its price control review in 2004, Ofgem published initial proposals of what it deemed to be “reasonable” efficiency targets. The regulator then probed these targets by consulting broadly and received comments from distribution companies, consumers’ associations and various interest groups. Ofgem took into account the comments in the final proposal, which was published six months later. After publication of the final price control determinations, the companies are supposed to form their own view in deciding whether to accept the proposals – if they do not, Ofgem has to refer the matter to the Competition Commission.
  • Evaluating consumer’s willingness to pay for quality improvements. For output targets (quality of service), the optimal amount of quality provided is at the point where the consumer’s marginal willingness to pay equals the distributor’s marginal cost to improve service quality. In non-economic terms, it means that quality targets must be set in a way that the premium that customers are ready to pay for improved quality should match the increased cost (additional capital expenditure, increased maintenance cost, etc.) of providing this additional quality.

Because reasonable targets are difficult to define, an innovative approach consists of giving more leeway to operators in achieving these targets. In Great Britain, the electricity regulator adopted a “menu” of sliding scale mechanisms to deal with uncertainties about future capital investment requirements to meet specified targets. The sliding scale menu allows firms to choose between getting a lower capital expenditure allowance but a higher powered incentive (and a higher expected return on investment) that allows them to retain more of the cost reduction if they can beat the target expenditure levels or a higher capital expenditure allowance combined with a lower powered sliding scale mechanism and lower expected return (Ofgem, 2004).
 

Footnotes

  1. See related question of FAQ: “What are the key challenges that need to be addressed when introducing incentives?“
  2. In this context, “binding” for Shugart and Alexander (2009) means that the company will bear the losses (or take the gains) if actual costs exceed (or are less than) the targets.