By Ryan Ketchum and Chris Flavin. Chris is Head of Business Development at Gridworks and Ryan is a partner at Hunton Andrews Kurth LLP.
Previous articles in this series have:
- discussed the pressing need for increasing levels of investment in electricity transmission systems in Africa to reduce costs, facilitate the transition to energy systems that are less carbon intensive, increase system stability, and reduce the level of generation reserves that are required to maintain system stability;
- identified four business models that can be used to unlock new sources of capital by facilitating private investment in transmission infrastructure (whole of network concessions, independent transmission projects (“ITPs”), privatizations, and merchant lines);
- described ITPs in more detail; and
- introduced whole of network concessions, described the enabling environment required for whole of network concessions, and the contractual structure of a typical whole of network concessions.
In this article, we will examine the alternatives for economic regulation of whole of network concessions and introduce some regulatory considerations that are specific to concessions.
Economic regulation – a brief overview
The central problem that economic regulation must solve is to ensure consumers of power are protected from the ability of a monopoly to charge prices that are not reasonable, while assuring investors that their long term investment will be fairly rewarded and that they will be protected from populist pressure to reduce prices to a level which does not allow for this.
As a general rule, legislative frameworks that govern electricity sectors establish an independent regulator – a separate and independent legal entity that is responsible for technical and economic regulation. Although the government may establish policy objectives for the sector, the regulator is responsible for ensuring efficiency, transparency, and fairness in the management of the electricity sector and benefits from the discretion that is required to achieve those objectives and to balance the interests of investors and consumers.
As discussed in previous articles in this series, the role of the regulator in a typical ITP Project is likely to be limited to reviewing a project prior to financial close, licensing it, and ensuring that any licensing conditions or KPIs are adhered to. In contrast, the role of an electricity regulator in a sector with a whole of network transmission concession is much more substantial. Whole of network concessions are a more complex business model. The concessionaire will be responsible for operating, maintaining, and usually also expanding the network to meet the transmission needs of customers in the concession area over a long period of time. The costs associated with this (including operating costs, capital investments and financing costs) are dynamic over that period of time, and tariffs will need to be adjusted to recognize changes in these costs. Tariff guidelines will typically be in place when a concession company makes investments in the network and the regulator will be responsible for applying those guidelines, approving operating costs and capital investment plans, and monitoring the transmission utility’s performance. The concept of regulatory independence and discretion mean that a regulator may also be permitted by law to modify its tariff guidelines at any time.
Risks around regulatory discretion and the track record and experience of the relevant regulator are a major factor for investors in deciding whether they can fund a transmission concession, and if so, what the risk premium applied to calculate their returns should be. As a result, a government support agreement is usually entered into in relation to a whole of network concession, and it usually containing a change in law clause which provides that if (i) the regulator modifies the tariff guidelines, fails to apply the tariff guidelines, or issues decisions that are contrary to the tariff guidelines, and (ii) the action (or inaction) of the regulator decrease the revenues earned by the concessionaire or increase the costs incurred by the concessionaire without affording the concessionaire a reasonable opportunity to recover those increased costs, then the host country will compensate the concessionaire. That compensation may take the form of a one-time payment or an ongoing subsidy to the concessionaire, depending on the nature of the action taken by the regulator.
The frameworks that are used to regulate network industries can be classified into two general approaches – the cost-of-service approach and performance-based regulation. Although many of the concepts involved in these approaches are similar, there are some key differences that are worth highlighting as we explore these two approaches.
Cost of service regulation
The traditional cost-of-service approach to regulation was developed in the U.S. at the beginning of the 20th century. The first step in determining rates using the cost-of-service approach is to determine the annual revenue requirement for the utility being regulated. The annual revenue requirement is the total amount of revenues that the utility must earn to recover its costs and earn a reasonable return on its investments. The basic formula for establishing the annual revenue requirement is as follows:
ARRy = (RateBasey x WACCy) + Depreciationy + O&My + Taxy
|ARRy||means the annual revenue requirement for year ‘y’;|
|RateBasey||means the value of the assets of the utility that are useful in delivering the service provided by the utility and are used by the utility for that purpose at the beginning of year ‘y’;|
|WACCy||means the weighted average cost of capital approved by the regulator for use during year ‘y’;|
|Depreciationy||means the amount of depreciation that the utility will recognize during year ‘y’;|
|O&My||means the expenses that an efficient utility would incur to operate and maintain the assets in the rate base and otherwise perform the function of delivering the utility’s services to its customers during year ‘y’; and|
|Taxy||means all of the taxes incurred by the utility during year ‘y’, including ad valorem taxes, corporate income taxes, and other taxes.|
These terms are further explored below.
The Rate Base
As a general rule, at least in the context of cost-of-service regulation, the rate base is determined by using the historic acquisition cost of each asset within the rate base and subtracting the depreciation that has accumulated since the asset was placed into service, usually using straight line depreciation.
The weighted average cost of capital
The weighted average cost of capital may be determined by the regulator using the following process.
- First, the regulator establishes a target debt to equity ratio for the utility, which may be expressed as X%:Y%. When expressed in that form, X is the total debt of the utility divided by the total capitalization of the utility (the sum of debt and equity) and Y is the equity of the utility divided by the total capitalization of the utility.
- Second, the regulator determines a cost of equity for the utility. The cost of equity may be determined by using the capital asset pricing model, which describes the relationship between the risk of investing in an enterprise and the expected returns. The capital asset pricing model starts with a risk-free rate of return and adds a risk premium (which is based on the beta of investments in that sector, which is a measure of the volatility of investments in the sector compared to the volatility of investments in a market generally) and, for investments that are not liquid (such as an investment in a closely held utility, as opposed to a publicly held utility), a liquidity premium, to estimate the returns the investment must generate to incentivize investors to invest in the enterprise.
- Third, the regulator determines the cost of debt for the utility. This may be determined by benchmarking the cost of debt for similar utilities or the cost of debt for large corporate borrowers generally, which can be estimated by drawing comparisons to an index of yields on bonds issued by corporate borrowers (for example).
- Finally, the cost of equity and the cost of debt are weighted by X and Y to determine a weighted average cost of capital.
The steps described above are regularly used in mature regulated electricity markets with a history of privately operated utilities such as those found in North America, Western Europe, Australia, and New Zealand to name just a few. The set of laws, rules, caselaw, and normative expectations that makes the level of discretion described above possible is generally referred to as the “regulatory compact”. In those countries, the regulatory compact has evolved and stabilized over the course of 100 plus years. In markets which may be putting a whole of network concession in place for the first time (as would be the case in most countries in Africa) it is likely that neither investors nor lenders would be be able to bear the risks that would be created by granting that level of discretion to a regulator without the same long-term track record. There is also the added complication that debt markets are likely to be less liquid and will provide fewer obvious reference points. As a result, countries that are seeking to implement a whole of network concession for the first time may need to reduce those risks in order to incentivize investment. This could be achieved by (i) allowing bidders to bid the return on equity, which would remain constant over the term of the concession, and (ii) allowing the concessionaire to pass through the actual cost of debt available to the utility (as opposed to the regulator setting the expected pricing). These are just two examples of the types of changes that could be made to reduce the risks borne by investors and lenders. Additional steps may be required.
The depreciation is calculated by applying the depreciation methodology established by the regulator for that sector to the assets that constitute the rate base. Straight-line depreciation is often used to calculate the depreciation component of the annual revenue requirement. To take a simple example, a regulator may establish a depreciation period of 30 years for an asset with a long service life, such as a transformer. In this example, a utility would recognize depreciation equal to 3.33% of the historic acquisition cost of the transformer each year over 30 years. Utilities maintain a register of all of their assets, including the historic acquisition cost of each asset and the depreciation it has recognized since the asset was placed in service so that it can perform these calculations.
The expenses that an efficient utility would incur to operate and maintain the rate base (the assets used to provide the service) and otherwise operate as a business can be determined by reviewing the expenses incurred to determine whether they were “prudently incurred”. Prudently incurred costs can be described as those costs that are actually incurred and that could reasonably be expected to be incurred by a qualified, experienced, responsible and financially sound utility, acting reasonably, prudently, fairly and in good faith.
Stepping back for just a moment, it is easy to see the underlying rationale for the formula set out above. The component (RateBasey x WACCy) provides a utility with a return on its investment. The component Depreciationyprovides a utility with the return of its investment. The components O&My and Taxy simply pass through costs incurred by the utility at the utility’s cost. This in turn means that the only return on the investments made by the utility comes from the component (RateBasey x WACCy).
Allocating the annual revenue requirement to consumers
After the annual revenue requirement has been established, it is allocated to consumers through end user tariffs which will typically be collected by a distribution utility and paid to the transmission concessionaire pursuant to a transmission service agreement or similar arrangement. The annual revenue requirement may be allocated to consumers by the quantity of the service supplied to the consumer (by the amount of energy consumed or transmitted for example) or, in some cases, by a measure of the value of the assets that are dedicated to serving that consumer (in the case of charges that are based on the peak demand of a consumer for example). In practice, the annual revenue requirement is typically divided into charges and rates that are established using a mixture of these concepts.
In a classic cost-of-service system, a utility files for a change to its rates when it would like to change the rates it is authorized to charge. In such a system, a utility’s rates remain in effect until they are changed by the filing of a rate case and the issuance of a decision by the regulator that authorizes the utility to charge new rates. In practice, this expensive and time-consuming process often occurs annually.
Performance based regulation
The cost-of-service approach is vulnerable to problems caused by information asymmetry. Information asymmetry is a reference to the fact that the utility will always have better and more current information about its business than the regulator. A utility can use this information asymmetry to find ways to earn returns that exceed the returns it should earn.
Performance-based regulation addresses this and related problems by creating an incentive for a utility to become more efficient and thereby outperform the regulator’s expectations. It works by establishing an annual revenue requirement for a period that is longer than one year. Such a period is known as the control period. Control periods generally fall within a range between three years and seven years. The annual revenue requirements for each year during a control period are established by the regulator in advance of the control period. If the utility incurs costs that are lower than the annual revenue requirements approved by the regulator, it can retain the difference as increased earnings. Although the utility may retain those earnings, the additional earnings come at a cost, at least when viewed from the perspective of the utility – the utility will have revealed to the regulator that it is capable of operating more efficiently and will have established a new benchmark for efficiency that the regulator is unlikely to ignore when it approves annual revenue requirements for the next control period. Conversely, if the utility incurs costs that are higher than the annual revenue requirements approved by the regulator, the utility’s earnings will decrease. This arrangement effectively requires a utility to compete against itself and rewards a utility for operating efficiently.
A regulatory regime that uses performance-based ratemaking could involve the following series of steps.
1. Business plan
The utility submits a business plan to the regulator that:
- identifies the outputs the utility will be expected to deliver during the regulatory control period (including such outputs as safe, reliable and efficient transmission service to its existing customers, the connection of new customers in a non-discriminatory and timely manner, the expansion of the system where necessary, environmental improvements, security improvements and other outputs);
- reflects the views of stakeholders, as determined by a consultative process undertaken by the utility and the regulator; and
- contains a program of capital expenditures that sets out the capital expenditures the utility plans to make to deliver the outputs.
2. Regulated asset base
The regulator establishes the regulated asset base (the rate base) for the first year in the regulatory control period. The initial rate base may be established by privatization or by the award of a concession (depending on the structure of the concession). The regulated asset base is then (i) increased by the investments made by the utility, and (ii) reduced by depreciation. It is carried forward into each successive regulatory control period.
3. WACC, O&M, Taxes
The regulator establishes the weighted average cost of capital the utility is permitted to earn, the operations and maintenance costs that an efficiently operated utility would incur to operate and maintain the regulated asset base and otherwise perform its functions and a projection of the utility’s tax liabilities.
4. Annual revenue requirement
The regulator sets the annual revenue requirement for each year during the regulatory control period by multiplying the regulated asset value for that year by the WACC and adding the efficient operations and maintenance costs and a projection of the taxes the utility will incur. Note that the regulated asset value for each year is set based on the then-current regulated asset value, the expected depreciation, and the investments carried out that have been approved by the regulator and will increase the rate base, as outlined in the approved business plan.
The annual revenue requirement is used to establish rates and charges in the manner described above in the section on cost-of-service regulation.
Rates are then smoothed from year to year, resulting in a constant increase (or decrease) to rates over the regulatory control period. These smoothed rates include an adjustment for projected inflation rates and account for the time value of money. They may also include an adjustment for projected changes to foreign exchange rates.
7. Inflation, foreign exchange adjustments
The projected inflation rates and foreign exchange rates are replaced by actual inflation rates and foreign exchange rates during periodic interim adjustments that occur at regular intervals during the control period. This is important because currency risks represent a major challenge for investors in African utilities where tariffs are collected in local currency, but financing is provided in hard currencies.
Options for establishing the regulated asset base
In many performance-based ratemaking systems, the regulated asset base is established based on the actual historic cost incurred minus accumulated depreciation, as is the case with traditional cost-of-service regulation. In other systems, the regulated asset base is revalued at the end of each control period to account for the inflation incurred during that control period. In these systems, the weighted average cost of capital is calculated in real terms, meaning that it does not include a component for inflation expectations. In other systems, the regulated asset base is adjusted at the end of each control period based on an estimate of the costs an efficient utility would incur to construct its facilities at the beginning of the control period, with an adjustment for the actual condition of those facilities.
In the context of a concession for a utility located in an emerging market, establishing the regulated asset base based on the actual historic cost incurred minus accumulated depreciation eliminates a few difficult problems that would be created by the other two systems (inflating the regulated asset base or revaluing the regulated asset base based on estimates of the then-current cost of construction). The most significant of these problems is that the latter two systems tend to increase the value of the regulated asset value over time. As we will see in the article on buy-out payments, the undepreciated value of the regulated asset base is used to calculate the buy-out payment a grantor must pay upon the expiration or termination of a concession. As a result, increasing the value of the regulated asset base increases the amount of the buy-out payment. A further problem is that the latter two systems increase the level of discretion granted to the regulator in ways that tend to reduce investor interest and impair the bankability of concessions.
Regardless of whether a regulator intends to regulate using cost-of-service or performance-based regulation concepts, the methodology it intends to be used should be clearly articulated in a set of tariff guidelines or a tariff methodology. In some systems, it may be possible for the tariff guidelines or tariff methodology to be set out in a schedule to the government support agreement or implementation agreement. However, in some jurisdictions such an arrangement is not possible because it would contravene the legal framework that governs the sector by impairing the independence of the regulator in a manner that is not consistent with that framework. In these systems, the tariff guidelines or tariff methodology should be articulated in a decision issued by the regulator or in a license granted by the regulator. The government support agreement should include a change in law clause in which the host country agrees that if (i) the regulator modifies the tariff guidelines, fails to apply the tariff guidelines, or issues decisions that are contrary to the tariff guidelines, and (ii) the actions (or inaction) of the regulator decrease the revenues earned by the concessionaire or increases the costs incurred by the concessionaire without affording the concessionaire a reasonable opportunity to recover those increased costs, then the host government will compensate the concessionaire. That compensation may take the form of a one-time payment or an ongoing subsidy to the concessionaire, depending on the nature of the action taken by the regulator.
The requirement to file a business plan with the regulator is particularly helpful in the context of a transmission concession. The rationale for implementing a transmission concession may include using private capital to finance significant improvements to, or significant expansions of, a transmission system. Many African countries have very low grid access and limited fiscal headroom to use public finances to expand their networks. A whole of network concession over all or part of a country could be a good way of using private capital to unlock service provision and increase energy access. Having the concessionaire submit a business plan to the regulator is useful because it facilitates a discussion around system planning, which impacts the capital expenses that will be incorporated into the regulated asset base during the next control period.