Flexible prices make markets hum,
but discounts discriminate when monopolies rule.
Many expect that the electricity industry is moving inexorably toward a much-publicized "new competitive era." Companies, regulatory officials and experts all regard the momentum as powerful.
So far, the changes are just beginning, and there is a long way to go to reach fully effective competition. %n1%n Yet even at this early stage, the merger and pricing strategies adopted by the established electric firms may be threatening the prospects for competition.
Among these strategies, possibly the strongest weapon is strategic price discrimination for favored customers, which has already spread widely in the industry. The discounts are widespread, and often deep. Many also are secret and long-term. These conditions mean the discounts may well arrest or block the movement toward competition before it really gets going.
As a result, the shift toward effective competition may soon become stalled, even while the old monopolies still hold a high degree of market dominance (with a market share more than 50 percent). This dominance could prevent effective competition in many or most local power markets, and it could entrench many or most of the private utilities as near-monopolies. The hoped-for new competitive era may not, in fact, arrive. However, there is a solution (em regulatory action limiting use of such anti-competitive tactics.
Discounts are not always damaging
The incentives to discount are nearly universal. In many market settings, the discounting strongly promotes competition. Unfortunately, in other settings where there is market dominance, discounting by the dominant firm tends to suppress competition.
The issue is discriminatory pricing, particularly as it is done by dominant firms in a dynamic and strategic process. It is a complicated, and often subtle, phenomenon, which can pack a large anti-competitive punch. Price discrimination always has been a widespread and well-known business practice, in all kinds of industries and markets for more than a century in industrial experience, economic writings, and regulatory policies. %n2%n It has long been attacked for being unfair; "charging what the traffic will bear" does make one group pay more than another group.
However, a price discount is not automatically discriminatory. If the costs differ in just the same proportion as the prices, then the discounts to the "favored" buyers actually are "cost justified," (em not discriminatory.
Discrimination arises because demand elasticities differ among customers. The discrimination generally follows an "inverse-elasticity" pattern: Customers with good alternative sources get charged lower prices than do the captive customers.
In the extreme case, discrimination may be so complete and refined as to extract all consumer surplus from individual customers.
Price discrimination is found in nearly all market situations, and often, it is pervasive. The prices reflect the varying price-sensitivity (or demand elasticity) of the differing customers.
Large market share can mean anti-competitive pricing
In general, the extent of price discrimination varies directly with the firm's market share. The higher the market share, the more extensive and deep the discrimination will be. The competitive or anti-competitive impacts usually vary in line with the market share of the firm doing the discriminating.
The monopolies and dominant firms discriminate most deeply and systematically. Firms with large market shares, or other sources of market power, usually sell across widely varying types of customer groups, and often will attain deep, thorough price discrimination. These firms can reap large profits by this sharp discrimination, and so the discriminatory pricing becomes
an important profit-maximizing activity.
In contrast, firms with small market share and a minimum of market power usually will be able to discriminate only on a limited basis. These firms have customers that easily can shift to other suppliers. The competitive constraints of small-market positions set tight limits on how widely and deeply the small-share firm can successfully vary their price-cost ratios among customers.
In fully competitive markets, price discrimination is entirely beneficial. It is simply the stream of flexible discounting that all firms continually use in order to win customers and try to gain market share.
However, in a market where one firm dominates, price discrimination has different, more powerful impacts that tend to reduce competition, not promote it.
Discrimination functions as a dynamic competitive device. A dominant firm's pricing variations respond to (em and can anticipate (em the varying competitive pressures for its diverse customer groups. Dominant firms simply do what is natural: They cut specific prices so that the lowest price-cost ratios go to the customer groups where competition is sharpest, because that is where the demand is most elastic.
As economists and regulators have known for more than a century, discounting by dominant firms or monopolies will meet or undercut the rivals in the more intensely competitive parts of the market. At the same time, the firm skims off higher price-cost margins from its more "captive" customers, who have lower demand elasticities.
When small-share firms do it, such pricing is benign and pro-competitive. When monopolies or dominant firms do it, the whole effect is to suppress competition.
Although every firm tries to apply the discriminatory pricing as much as it can, the discrimination usually tends to suppress competition only when it is done by dominant firms.
This array of discriminatory, pin-point pricing (em adjusted and sharpened as events proceed (em tends to quell or kill specific rivals in market niches.
In this light, anyone can assess the impacts on the market by considering just two relatively direct and objective conditions: market share and thoroughness of discrimination (see figure 1). Discrimination becomes anti-competitive when it is done by firms with high market shares, and is thorough and systematic. In contrast (em and this is important (em discrimination is usually pro-competitive when it is done by firms with small market shares, sporadically and briefly.
A dominant firms's resort to dynamic strategic price discrimination always will pose dangers to moving the firm toward effective competition. As its high market share faces competitive attacks, the firm seeks systematically to apply strategic discrimination in order to retain or even increase its dominance. It is all quite natural, predictable and inevitable. The action proceeds over time, often in a dynamic, dramatic sequence. As part of its whole strategy, the firm may cut prices particularly deeply in some parts of the market, sometimes even down to "predatory" levels, which are below cost. This action can send signals of "credible threats" of future decisive price cuts, which deters future potential competitors. By making extra money in the future from deterring many other rivals, the dominant firm can gain profit even when its prices against one rival seem to lose money. Deep price cutting can be entirely rational.
Discrimination can block the emergence of competition
During the attempt to deregulate a former monopoly and move it toward effective competition, the monopoly's resort to strategic pricing may be especially strong and frustrating. In its previous franchised-monopoly period, the established firm has usually adopted discrimination reflecting the variations in demand elasticities. Even (or especially) if the regulators have let the utility firm adopt such discriminatory prices already, the firm usually has an extensive knowledge of the elasticities and the customers' response patterns.
As soon as public policy shifts from natural-monopoly regulation to open access and promoting competition, the same monopoly firm's pricing structure is transformed from static efficiency-promoting to competition-blocking. Even if the price structure may have promoted efficient allocation under natural monopoly, it becomes a deadly weapon for containing and fighting little rivals. The regulators must shift their thinking and their policies diametrically: The discrimination has changed from a largely favorable pattern to an anti-competitive weapon, which may prevent effective competition.
Remember, strategic price discrimination can be quite healthy and pro-competitive, but only if the situation already is strongly competitive. Indeed, the discounting is a common, crucially valuable part of the competitive process; flexible price discounting can be the life-blood of competition.
But that benign effect occurs only where the firms doing the discounting have only small market shares and where there is no substantial market power. In such a situation, discounting and discrimination is just another way to win business, and no firm can use it widely and deeply enough to suppress healthy competition.
The applications to the electricity industry are clear and important. The former franchised-monopoly utilities naturally will resort to dynamic discrimination with the onset of competition. Left unchecked, discrimination may be used to stop or slow down the rise of competition. Therefore, regulators need to take special care to prevent firms from using this pricing weapon before and during the transition. When and if competition is fully established (em and only then (em discrimination will pose no anti-competitive threat.
There is widespread evidence that electric utilities are already deeply involved in this type of strategic pricing, which started several years ago in some states. A newspaper account in April 1996 noted that in more than four states (em Massachusetts, New York, Michigan and California (em large-customer discounts had been applied already (see table 1).
In some cases, the discounts were explained as being merely a response to the customers' threat to move elsewhere. Those are now known as "business-retention" or "load-retention" discounts (or to "prevent uneconomic by-pass"). In other cases, the discounts apparently are done at the initiative of the utility firm. Such discounts often are called "business-enhancement" or "economic-development" discounts.
In either case, the anti-competitive effects are the same: The best and largest customers tend to be favored with the low prices. The utility chooses to favor precisely those customers that it wants to keep away from new competitors.
Two main features make discounts anti-competitive. The first is long-duration agreements. Some recent contracts are 10 years long, such as Detroit Edison's discounts to the Big Three automobile producers (see table 1). By locking companies in with discounts, the utility makes those customers difficult, if not impossible, for any new competitors to attract. Discounts contracted for long periods merely extend the lock-out of new competition.
Of course, long-term contracts usually are harmless or even efficiency-promoting in a competitive market. But when a powerful dominant firm uses them effectively to block out future competitors, the effect is anti-competitive.
The second feature making discounts anti-competitive is that the discounts often are held secret by the firms and the public regulatory agencies. %n3%n Typically, even the name of the favored customer is kept secret, as well as the extent of the discount. Massachusetts, for example, now has scores of secret rates. Each customer is listed by the Department of Public Utilities by an anonymous code number with no pricing information.
By allowing secrecy, the regulators are assisting the monopoly in maximizing the anti-competitive effect of the discounting. Secrecy lets the dominant firm confine its pin-point price reductions to the minimum extent. This maximizes the harm to emerging competition and minimizes the firms' own sacrifice of revenues.
The secrecy and rapid spread of discrimination makes it impossible to accurately assess the effects of anti-competitive tactics. Many commissions have complied with the utility and large-customer pressures for approval of these discounts (see table 2).
But, although secrecy is widespread, there are exceptions.
There has been resistance in some states (notably California, Florida, New Hampshire, and Ohio), both to the discounts and especially to the secrecy. In those states, the possibility of the large-customer discounts has stirred sharp debate.
Some of the resistance comes from a sense of unfairness that big companies are getting benefits before small buyers. Some of the resistance also recognizes the competition-blocking effects of the discounts.
The discounting is unfair and anti-competitive only during the critical, early phase of the transition to competition, when there is still market dominance. If and when fully effective competition is established, the restraints on
discounting can be dropped entirely. And smaller firms should always be permitted to offer discounting, since their gains will only promote competition.
In addition, if effective competition does not develop, then a "select" or "privileged" group of industrial customers may reap most or all of the benefits from the aborted attempt at competition. Even if some competition does develop, the sequence of discounting (em to biggest customers first (em leaves the smaller customers out in the cold. In fact, if the large-customer discounts are large enough, then common costs will be shifted to the later, less-lucky small customers. The small customers' prices will end up being higher, not lower.
are needed now?
First of all, a moratorium is needed to prevent anti-competitive price discrimination. In principle, the ideal regulatory action would be to void all previous, large-firm discounts. It would also set a moratorium on all new, large-customer discounting by utility monopolies to firms in their service area until the conditions of effective competition have been attained. Next, if discounts are permitted, the impact may be reduced by requiring full disclosure. This would increase the pressures to let all customers share the discounts and minimize the anti-competitive effects.
Finally, and perhaps even more important in practical terms, the discounts should be limited in duration. Preferably, discounts should extend no more than one year at a time. Otherwise, the discounts will lock up the best customers, so that new competition will have no attractive customers to begin with.
Any constraints on the discounting will be criticized by the monopolies and the favored customers. These companies will clothe resistance in favorable-sounding terms, claiming that the regulators are blocking desirable price cuts. The utilities and customers also will deny the discounts have any anti-competitive effects, which could place the regulators in an awkward position, seeming to be opposed to beneficial price cutting.
Nonetheless, the regulators need to explain vigorously why the discounts have unfair and anti-competitive effects, and they need to prevent or restrain the discounts during this critical phase. Otherwise, effective competition may never evolve. t
William G. Shepherd is Professor of Economics, University of Massachusetts, Amherst, and general editor of the Review of Industrial Organization, and The Economics of Industrial Organization, 4th ed., Prentice-Hall, 1997. Shepherd has been involved in electricity and telecommunications policy issues from the beginning. John Kelly and Diane Moody assisted in the compilation of this article.
Discount or Discrimination?
It depends on your definition
Price discrimination has had a lot of names through the decades, including: demand-based pricing, usage-sensitive pricing, charging what the traffic will bear, selective pricing, sharp-shooting pin-point pricing, picking the eyes out of the market (a British phrase), discount pricing, Ramsey pricing, etc.
The economic definition of price discrimination actually is rather specialized and technical. Discrimination exists when there is a difference among the price-cost ratios for related goods. In the purest, simplest case, the same good (produced at the same cost for everybody) is sold to different buyers at different prices. Discrimination also can occur when different (but related) goods with different costs are sold at prices which differ even more sharply (or instead, less sharply) than the costs differ. Price discrimination is found in nearly all market situations, and often is pervasive. Following are some examples:
• Pharmaceuticals. Pricing of medical drugs is an example of discrimination due to a difference in price-cost ratios for related goods. For instance, the price for small-scale patients through local drug stores is often five times more than the price for big hospital and HMO buyers.
• Computers. In the 1960s IBM set very low (even loss-causing) price-cost margins on its large machines, where Control Data was a particularly stiff competitor. The move nearly killed off Control Data (em as IBM hoped. But IBM also set much higher price- cost margins to small, uninformed buyers, where the competition was weak. These small buyers virtually were "captive" customers, who would buy "an IBM machine" at almost any price.
• Airlines. Another, well-known current example of pricing that extracts all consumer surplus from individual customers is airline pricing. Anti-competitive pricing in the airline industry has become perhaps more extensive and refined than any other instance in business history, and has been going on since 1978.
• Automobiles. Most automobiles are well-known to be sold often at low price-cost margins, while the spare parts are sold at much higher mark-ups over cost.
Antitrust impacts have also resulted from such companies as United Shoe Machinery, Xerox, and now Microsoft, for that matter.
For a comprehensive discussion of price discrimination, including some of the static profit-maximizing features and as some of the dynamic competition-affecting aspects, read The Political Economy of Monopoly, by Fritz Machlup, Industrial Market Structure and Economic Performance, by F.M. Sherer and David N. Ross, and The Economics of Industrial Organization, by William G. Shepherd.
Table 1. Discount Prices to Large Electricity Customers, by Five Selected Utility Firms (as of April 1996)
Utility BOSTON NIAGARA DETROIT CONSUMERS PACIFIC GAS
Firms EDISON MOHAWK EDISON POWER & ELECTRIC
Customers Any of the largest To 72 large GENERAL GENERAL Some of the
12 customers if companies MOTORS, MOTORS 100 largest
they show an including: FORD, DOW Several
intent to move LOCKHEED, CHRYSLER CORNING, customers,
out of state CHURCH & UPJOHN including:
Already includes: DWIGHT EXXON,
POLAROID, HEWLETT- DIGITAL PACKARD
Discounts 20 percent, for up $66 million 10-20 per- 10 percent, $70 million to 4 years per year cent at 54 for 5-10 per year plants for years 10 years
Source: Agis Salpukas, "Utilities Rewrite the Rate Card,"
New York Times, April 5, 1996, pp. D1, D6.
Table 2. Selected Additional Permissive Regulatory Rules or Instances of Price Discrimination (as of September 1996)
STATE ARIZONA CALIFORNIA ILLINOIS INDIANA MASSA- MISS- NEVADA
and Mar. 1996 Oct. 1995 Oct. 1995 Sep. 1995 to CHUSETTS ISSIPPI by 1996
Date Jan. 1996 July 1996 by 1996
Conditions Pricing Flexibil- The Public There are special Approved at Scores of secret Permits secret Permits
Imposed, ity Rider Rate Utilities Com- contracts with least three discounts have electric utility discount
If Any No. 79. Permits mission at least 11 special dis- been approved discounts. rates but not "negotiated approved a large customers counts for by the regulatory secrecy. prices" under set of "pre- Secrecy was large customers agency a "comparative approved gen- barred in 1995 on a business- tariff. eric discount" by the Illinois retention basis. "Formally contract-rate Appellate Court, limited to options for but legislation business- large industrial to allow secret retention and commer- rateswas passed situations. cial customers. by legislature in July 1996.
Formally, special rates can be offered only to prevent "uneco- nomic bypass."
NEW NEW NEW OHIO OKLAHOMA PENNSYL- UTAH WASHINGTON
HAMPSHIRE JERSEY MEXICO Spring 1995 June 1995 VANIA since 1992 by 1996
by Jul. 96 Oct. 95 as of 1996
Has allowed Secret "off-tariff" In 1992, secret Regulators ap- The Public Cor- Special secret Permits secret Puget Sound
special dis- discounts of up load-retention proved a secret poration Com- discounts have discounts, Power &
counts for to 7 years are contracts were discount for mission permits been permitted including Light has had
large industrial permitted under permitted. Now American Steel the two utilities since 1992. incentive rates. at least two
customers, new rules of the all rates are on & Wire Corp. to offer special confidential-but not New Jersey the public record. rates, but only discount
secrecy. But Board of Public for 10 customers contracts
discounts Utilities. during a test with large
have been period of two customers,
resisted and years. but has not
the issue is fought publi-under hearing cation when
and it was consideration. requested.
1Effective competition requires competitive parity among numerous rivals, so that they compete hard and effectively, forcing each other to be efficient and innovative in order to survive. In practical terms, there need to be at least 5 major competitors in each market, with no one firm holding dominance. There also needs to be relatively easy entry for new competitors. See, The Economics of Industrial Organization, and "Deregulation: From Monopoly Only to Dominance? Telecommunications, Railroads, and Electricity," Quarterly Bulletin, National Regulatory Research Institute, Summer 1996, pp. 149-61.
2See J.M. Clark, Studies in the Economics of Overhead Costs, 1924, Joan Robinson, The Economics of Imperfect Competition, 1932, James C. Bonbright, Principles of Public Utility Rates, 1962, and Alfred E. Kahn, The Economics of Regulation, 2 vols., 1971.
3In some cases the commission staff and a few intervenors may be permitted to inspect rates, but public disclosure is still barred.
Articles found on this page are available to Internet subscribers only. For more information about obtaining a username and password, please call our Customer Service Department at 1-800-368-5001.