Competition "vs." Regulation: Have We Achieved Conversational Clarity? (Part I)

The "central, continuing responsibility of legislatures and regulatory commissions [is] finding the best possible mix of inevitably imperfect regulation and inevitably imperfect competition."

— A. Kahn, The Economics of Regulation: Principles and Institutions, Vol. I, Introduction at xxxvii; Volume II at 114 (1970; 1988 edition)

*   *   *

Professor Kahn said it all.  Resist either/or, expect imperfection, show curiosity continuously.  Skip the ideology; use facts, intellect, integrity, and humility.  These are the paths toward "the best possible mix."

To what end?  Missing from Dr. Kahn's quote (but omnipresent in his ensuing 600 pages) is this answer:  Competition and regulation share a common purpose—to align private behavior with the public interest.  Effective competition induces competitors toward efficiency, customer service, and reliability.  Effective regulation does the same.  Together, they cause accountability—to the consumers, investors, and the public.

Do our market structure debates serve this purpose?  Do we emulate Kahn's insistence on facts, intellect, integrity, humility, and the absence of ideology?  How well do our words assist our aims?

The Stakes

At stake are billions of investor and consumer dollars, our infrastructure’s reliability, and our economic progress.  In the electric industry, FERC and the states struggle to determine the mix, at wholesale and retail.  Their efforts intertwine.  Wholesale markets need wholesale buyers; but there will not be wholesale buyers if states, distrusting wholesale markets, tell their utilities to build rather than buy.  Effective wholesale competition requires efficient retail demand signals, but those signals remain blurred when retail prices reflect embedded cost rather than current production cost.  Effective retail competition requires multiple, viable entrants with economic access to bottleneck facilities; active shoppers; and the absence of unearned incumbent advantages.  But the politics of retail competition often demand deviations from these principles.

In the telecommunications industry, laden with a century of societal expectations, we struggle with similar questions:  In the many markets for telecommunications services, when are competitive forces sufficiently vigorous and customer-responsive so that traditional policy purposes like "carrier of last resort," “universal service,” service quality, and intercarrier cooperation no longer require regulatory mandates?

Semantic Suboptimality:  Three Examples

Asked at dinner parties "What do you do?" we mumble abbreviated answers.  Some of these words, oversimplified to the point of meaninglessness, invade discussions among policymakers who should know better.  Consider these examples.

"Deregulation":  This term, to the experienced practitioner, means “The statute authorizes entry by multiple competitors.”  But the term is hopelessly ambiguous because (a) "authorized" competition is not "effective" competition; (b) authorized competition, after a century of monopoly, still requires “regulation” for licensing, fraud prevention, access to bottleneck facilities, prevention of affiliate abuse, and assurance of last-resort service; (c) the term literally could mean either "elimination" or "reduction" of regulation (e.g., the term "decelerate" means reduce speed, not eliminate all motion); and (d) if effected incorrectly, the result of “deregulation” is still "regulation," except that it is regulation of the market by the incumbent, to protect its position; rather than regulation of the incumbent by the commission, to protect consumers.

Proponents of "deregulation" intend the prefix "de " to replace a negative (regulation) with a positive (elimination of regulation).  But the phrase cannily avoids accountability; for if they labeled their goal "effective competition" rather than "deregulation" they'd have to show evidence in place of rhetoric.  The converse applies as well:  Critics of "deregulation" imply that "regulation" creates benefits, but they do not always identify (and guarantee) those benefits.  In short, the bipolarity of the term "deregulation" makes discussions more stick figured than sophisticated.

"Competition works":  I once shared a panel with someone who said, platitudinously, "Competition works."  How much ambiguity can two words hold?  "Competition" in which geographic markets?  Which product markets?  For which consumer segments?  During which time periods?  "Works" when?  Overnight?  After years of investment by newcomers, during which time period incumbents enjoy "first mover advantage" without facing competitive pressure?  "Works" how well?  For whom?  For the incumbent?  For the newcomer?  For some customers?  For all customers?  How does someone say "Competition works," without being "hooked" off the stage, vaudeville style?  Why does our community tolerate such mental muddiness?

"Market share":  We often say that "generation is competitive.”  Competitiveness refers to a market.  A market has a geographic component and a product component, and sometimes a temporal.  "Generation" is not a market.  "Generating capacity serving Maryland’s Eastern Shore on August afternoons” is a market.  A company can own a generating unit constituting 1 percent of the generating capacity in the PJM region (Ohio to Virginia), suggesting, to a layperson, no market power.  But due to locational luck, those same generating units could constitute 90 percent of the capacity available to a transmission-constrained subregion on a hot afternoon.  "Generation" is not a market.

Further, "market share" does not readily translate into "market influence.”  The reason lies in the distinction between "market share" and "pivotality."  Both are necessary to measure competitiveness.  A company can have a small market share (say, 1 percent), yet in certain time periods be "pivotal."  A pivotal supplier is an indispensable supplier—her supply is necessary to fill the demand at the time.  If the total capacity in a market is 100 MW, and demand is 95 MW, then any supplier with more than 5 MW of capacity (a mere 5 percent of the total) is indispensable, because if she withdraws her supply then demand is not served.  At the other end of the market share spectrum, a company can have a high market share (say, 75 percent), but if there are potential entrants poised at the perimeter, their entry threat can discipline even a monopolist.  Facts matter.

A Sequence of Questions

In these three examples, efforts at semantic simplicity yield words with multiple meanings.  Confusion replaces comprehension.  Finding the "best possible mix" is complicated.  Our language should expose that complexity, not hide it.  Dinner guests will wish they'd never asked (happens to me all the time), but at least they'll understand why.  Instead of opining on "competition versus regulation," consider framing a sequence of questions:

  1. Which geographic and product markets do we wish to discuss?
  2. Is competition economically desirable, i.e., are economies of scale and scope sufficiently low that competition will not damage "static efficiency"?  (See the work of Dr. John Kwoka of Northeastern University.)  And if there is potential loss of "static efficiency," will this loss be offset by gains in "dynamic efficiency" as the rivals pressure each other?  (Thanks to Dr. Kenneth Rose for his explanation of the difference between "static" and "dynamic" efficiency.)
  3. Is competition technically feasible?  Can the physical and communications networks accommodate the new traffic stimulated by competition?
  4. Are consumers ready and willing to shop?  Theories must confront practice.  If consumers are too busy going to work, attending soccer games, and cooking meals to compare prices and offerings, competition will not work.

Read Part II