Viewpoint Note No. 132 November 1997 Has Price Cap Regulation of U.K. Utilities Been a Success? Richard Green Developing the approach-a proposal scheme would concentrate consumer protec- for BT tion where it was most needed while giving BT some freedom to change the balance of its The privatization of British Telecom (BT) in prices. The scheme would be simple for the 1984 reintroduced the regulation of private regulator to monitor, and it wvould minimize infrastructure companies in the United King- the risk of regulatory capture because it would dom, after nearly forty years of public owner- be nondiscretionary. Above all, Littlechild ship. Two regulatory approaches were initially hoped that the scheme would be temporary, proposed: controlling BT's prices by setting a for competition is by far the most effective maximum rate of return,' and imposing an means of protection against monopoly. He output-related profits levy on BT at a rate that therefore recommended that the scheme be would fall as the company's output rose, giv- adopted for five years, and reviewed after four. ing it an incentive to keep output high and prices low. A price cap for British Gas Professor Stephen Littlechild, an academic who A second price cap was adopted for British later became the first regulator of the electricity Gas when it was privatized in 1986. This price industry, was asked to choose between the two cap is based not on individual prices, but on schemes. But instead he proposed a third, a price the average revenue yield from sales to small cap system that the government adopted. consumers-the value of sales divided by the Littlechild argued that a maximum rate of return volume of gas sold. This formula gives the com- would limit BT's incentives to operate efficiently, pany more freedom to introduce new tariffs, while the effect of an output-related profits levy since there are no controls on individual prices remained uncertain. Expecting increased com- and no need to worry about the weights in a petition in telecommunications, he pointed out basket. (See tables 1 and 2 for a comparison of that if BT faced a levy that its competitors did the two types of price controls and traditional not, it could be at a significant disadvantage. rate-of-return regulation.) Littlechild proposed an explicit limit on the A second change is the pass-through of the prices that BT could charge in the areas where cost of gas purchases to consumers so that they, it wvas expected to retain some monopoly not British Gas, bear the risk of changes in the power-line rentals and local call charges. The wholesale gas price. This price is outside Brit- weighted average of a basket of these charges ish Gas's control. but accounts for nearly half should fall in real terms by a specified amotint its costs. The formula became known as RPI - each year. This was equivalent to saving that AT + i.A X denotes the expected productivity the charges should not increase by more than increase (2 percent a year), and Y the pass- RPI -X, where RPI is the increase in the retail through of the cost of gas. In the years after price index (a standard measure of general in- priv-atization, the cost of gas fell significantly, flation in the United Kingdom) and X is the and the benefits were passed straight on to real reduction in prices. He argued that this consumers. The World Bank Group . Finance, Private Sector, and infrastructure Network WNl Has Price Cap Regulation of U.K. Utilities Been a Success? A price cap for water seas practically impossible and a permanent control would be fieeded. Indirect competition Littlechild was consulted again when the water in the form of yardstick regulation (advocated industrE was prepared for privatization, in 1986. by Shleifer 1985) Eas possible, however, and The industry wtas required to complete a large Littlechild recommended using comparisons backlog of investments, and prices were ex- between companies ahen setting and resettinl pected to rise to finance them. Once again, price controls. Because prices 'ere expected Littlechild recommended a control on prices to rise, the scheme became known as RPI + K, rather than profits, although it was clear that although it was really RPI - X + Q, where Q is direct competition betpeen water companies the cost of investdgent to meet qualit targets. Revision of the first price control peal their proposals, the regulators have in- creased the amount of information they release In 1988, it became clear that BT did not face and generally argue that they follow the "MMC enough competition to abolish the price con- methodology." If a company knows that the trol, so the regulator proposed a new control. commission is likely to use the same method- For the first five years. BT had faced a control ology as the regulator. it also knows that it has of RPI - 3 on line rentals and directly dialed little to gain from an appeal. The commission's domestic calls, and a subsidiary price control role has brought a welcome increase in trans- that limited any increase in line rentals to 2 parency in the resetting of price controls. percent a year in real terms.2 The regulator proposed a new control, RPI - 4.5, which would One-time reductions also bring some operator senrices into the bas- ket of controlled prices. When the early price controls were reset, only the value of X was changed, so prices stayed The regulator gave little information on this on a smooth path. But in 1994, Littlechild faced decision at the time, but later explained how different needs in resetting the price controls the new value of X had been chosen. His staff for the distribution businesses owned by the had built a financial model of BT (in consulta- twelve regional electricity companies in England tion with the company) so that they could pre- and Wales. These companies, given RPI + X price dict BT's profits and rate of return on capital, controls wlhen they were privatized in 1990, had given a value of X. In 1988, BT earned more subsequently earned high profits, so a large re- than its perceived cost of capital, so the regu- duction in prices was clearlv required. But if lator chose a value of X predicted to eliminate only the value of X was to be changed, Littlechild the excess return by the end of the price con- faced a difficult choice. If he set X so that prices trol period (set at four years). This approach reached the "correct" level at the end of the has much in common with the calculations used period, the companies would continue to earn in rate-of-return regulation, although the fixed high profits for several years, an excessive re- review period and the regulator's ability to dis- ward for their earlier cost reductions. regard excessive costs give the company a greater incentive to be efficient and force it to The alternative was to set prices so that the com- bear more risks. BT accepted this control; if panies would receive an appropriate amount of the company had rejected it, the issue would revenue over the period as a whole. If the prices have been decided by the Monopolies and declined evenly from the present level, how- Mergers Commission (MMC), the United ever, they would have to fall well below their Kingdom's competition tribunal. long-term level to bring down total revenue, creating future problems. Littlechild therefore A check on the regulator's discretion decided to implement a one-time cut in prices followed by a control of RPI - 2, in order to The Telecommunications Act of 1984, which combine an acceptable total revenue with a sen- set up the system of regulation, gave the Mo- sible price level at the end of the period. Simi- nopolies and Mergers Commission the role of lar one-time cuts have since been made in the 'court of appeal," acknowledging the need for price controls of British Gas Transco (gas trans- a check on the regulator's discretion. The com- mission and distribution) and the National Grid mission has had to rule on six price cap dis- Company (electricity transmission). putes since 1992. These rulings have gradually built up some case law, for while each case is An assessment decided individually, the commission has rec- ognized the value of developing a consistent Price controls in the United Kingdom have been methodology. Not wanting companies to ap- controversial. The initial price controls for the Has Price Cap Regulation of U.K. Utilities Been a Success? electricity companies and the water industry, and returns the gains to consumers after a short based on underestimates of the companies' time. The high profits of the early 1990s were scope for reducing costs, turned out to be over- due largely to unanticipated, one-time produc- generous, allowing them high profits. The 1994 tivity gains following privatization, which are review of the electricity price controls prom- unlikely to be repeated. The established method ised to reduce the companies' revenue by more for resetting price controls makes further ''mis- than a sixth over the period, which Littlechild takes" unlikely. In the future, price controls thought would give them no more than an ad- can be expected to give companies an incen- equate return. But Northern Electric, threatened tive for efficiency without allowing them to earn with a takeover, was still able to promise to excessive profits. make a one-time payment to shareholders in cash and shares valued at twice the price for l Rate-of-return regulationi, the traditional method of utilitv regula- which it had been sold four years earlier and tion, is widely used in the United States. The regulator allows the to increase future dividends. This caused a company to charge the prices expected to produce profits equal major political row, and Littlechild decided to to a fair rate of return on the fair value of the capital invested in the company. If profits fall helow this level, the comnpany can reopen the review in March 1995.3 Three request approval for a new set of prices. Investors therefore know months later, he announced additional price that profits should not be too low, and consumers know that they cuts of about 10 percent, supported largely by should not he too high. The prohiem with this system is that it gives the company little incentive to improve efficiency andi may a different treatment of the companies' asset even encourage inefficient behavior that raises the company's al- values at flotation. Even these reductions were lowable revenues. Viewpoint is an open 2 Like many other telecomamunications companies, BT had set line forum intended o less than some people had expected, and share rentals below, cost and call charges shove cost. To prepare for encourage dissemina- prices later rose, partly because of takeover competition, the company needed to rehalance its prices, raising tion of and debate on bids. Littlechild received heavy criticism for his rentals and reducing call charges, but that would increase the bills ideas, innovations, and paid by small consumers, who made relatively few calls. The RPI best practIces for ex- handling of thesituation. + 2 constraint svas imposed for political reasons, to limit the speed panding the private of these increases. sector. The views pub- Following these events, some commentators It had been generally assumed that the controls were finalized, linhed are tboae of the h ut hecause a final stage of formnal consultation remained hefore authod and ahosd nofth (such as Burns, Turvey, and Wkeyman-Jones they could be implemented, Littlechild was legally allowed to be attributed to the 1995) suggested forms of profit sharing regula- change his proposals. World Bank or any of tion. Under such schemes, the regulator would It has been suggested that the interval hetsveen main reviews its affiiiated orga d zan continue to set a price control for several years couild he lengthened, to restore incentives. But h, the end of the tions. Nor do any of the ~~~~~~~~~~~~~~~~period, prices (even aifter profit sharing) could diverge just as far conclusions represent at a time, but would also monitor the company's from costs as they would under a pure price cap and a shorter official policy of the costs and profits from year to year. If profits review period, negating the main advantage claimed for profit World Bank or of its sharig Executive Directors rose above a trigger level, the company would ring. orthe countries they have to reduce its prices, returning some of the represent. 'excess" profits to consumers. These suggestions References To order additional have gained political support, and some regu- Burns, P., R. Turvey, and TG. WVeymanJones. 1995. "Sliding Scale copies please call lators have considered implementing them, al- Regulation of Monopoly Enterprises."'Discussion Paper 11. Centre 202-458-1 1 11 or contact though they have not yet done so. The practical for he Stuy of Regulated ndustries, London. Suzanne Smith, editor, thuhpatcl Littlechild, Stephen. 1983. Regu/at/on of Britisb Telecommtrunicatiotas Room F6P-188, problem is that profit sharing would require au- Profitability. London: HMSO. The World Bank, dited cost information everyv year for calculating Shleifer, Andrei. 1985. "A Theory of Yardstick Competition." Rand 1818 H Street, NW, the allowable prices, and obtaining this infor- Journal ofEconoetica 16(3): 319-27 Woah IntontD.C 20433 mation would place a heavy burden on both ssmithl©worldbank.org. firms and regulators. Profit sharing would also Richard Green, Department of Applied The series is also weaken companies' incentives to reduce costs.4 Economics, University of Cambridge, and available on-line Visiting Fellow to the Economic Development (www.woridbank.org/ html/fpd/notes/ The utilities remain unpopular in the United Institute of the World Bank notelist.html). Kingdom, but most experts would be willing e Printed on recycled to defend the price control system as one that paper. gives companies an incentive to cut their costs