In his recent keynote speech at the CTIA show in New Orleans, Federal Communications Commission Chairman Julius Genachowski reiterated his (and the industry’s) concern that the “demand for mobile services is on pace to exceed the capacity of our mobile networks” and, therefore, we must “tackle the capacity challenge.” The Chairman has previously foretold of a future where spectrum exhaust could make “consumers […] face slower speeds, more dropped connections, and higher prices.” Plainly, spectrum exhaust remains a key challenge for both mobile service providers and policymakers.

The Chairman also took the chance in his CTIA speech to challenge what we and others have said about the peculiar effects of industry concentration under a spectrum shortage.  That is, when spectrum capacity is exhausted, an increase in the number of competitors may lead to higher, not lower, prices. We provided the first theoretical analysis of this important issue in our paper entitled Wireless Competition After Spectrum Exhaust.  (See my non-technical explanation of the paper here.)  In this paper, we demonstrate, using a standard model of competition (that is, price falls and the number of competitors rises), that when spectrum capacity is exhausted and there is a type of scale economy in spectrum holdings (e.g., if you double the amount of spectrum a firm has, it more than doubles its capacity), then fewer—not more—firms produces lower prices and more innovation.  The economic logic is undeniable; when capacity is exhausted, quantity cannot rise, and if quantity cannot rise, price cannot fall.  Other papers have found similar results using different modeling assumptions.  Moving out of the journals and into the boardroom, AT&T Chairman Randall Stephenson recently made a similar argument, observing: “The more competitors you have, the less efficient the allocation of spectrum will be. … It’s got to change. I don’t think the market’s going to accommodate the number of competitors there are in the landscape.”

Chairman Genachowski apparently doesn’t care for this model of competition for the wireless industry.  In his speech, the Chairman interprets the theory as implying that “competition is bad for consumers.”  This statement indicates that the Chairman misses the point entirely.  The theory does not say that competition is bad for consumers.  Rather, the theory says that spectrum exhaust is bad for consumers, and it’s unnecessarily bad for consumers if spectrum is inefficiently allocated among competitors.  If there are too many competitors, each with too little spectrum to operate efficiently, then more competitors leads to higher prices and less innovation.  Since the Chairman claims to want “lower prices and more valuable services,” perhaps the FCC should be more open to the economic realities of the industry it regulates.

The Chairman also contends that our theoretical findings are “at odds with basic free-market principles.”  In fact, there is really only one free-market principle, and that is that market exchange is preferred to regulatory manipulation.  What the free-market principle says about spectrum exhaust is that market participants will seek out ways to reallocate the scare spectrum resource to eliminate inefficiencies and, consequently, permit more output to be produced (and thus permitting prices to fall).  In the business world, such reallocations are termed mergers and acquisitions.  Standing in the way of this free-market driven reallocation is the license transfer authority of the FCC, an agency that appears determined to impose its view of efficient industry structure on the mobile sector come hell or high water.  Unfortunately, as we have shown, the agency remains dependent on a flawed economic model of the industry.  What is most at odds with “free-market principles” is not the idea that competition may lead to higher prices (economic theory provides numerous examples of the paradox), but rather the idea that the regulator knows better how to allocate resources than do buyers and sellers.  For confirmation that regulators are out of step with the wireless broadband market, I point you to the LightSquared debacle.

In the same vein, the Chairman falls back on ideology in proclaiming, “we’ve staked our economy on the proposition that competition doesn’t lead to higher prices for the same product, but to lower prices and more valuable services.”  Certainly, it is true that in pursuit of low prices for consumers, policymakers often turn to competition, and for good reason.  Competition has both theoretical and empirical support for its price cutting effects.  But competition is not the goal; it is a means to an end.  The desire for competition derives from the expectation that more firms leads to lower prices (among other things).  With spectrum exhaust, however, that expectation is no longer valid.  In fact, under plausible conditions, more competitors could lead to higher prices, which means more competitors is no longer a legitimate means to the desired end.  Competition is a wonderful slogan, but under spectrum exhaust, competition may be more effective at lowering prices with fewer rather than more firms.  It’s an admittedly odd result, but that’s the nature of capacity limitations.

The Chairman also interprets our argument as implying that “competition drives spectrum inefficiency.”  Again, he misses the point.  The very nature of market forces is to eliminate inefficiency, and transferring assets from lower-valued (inefficient) to higher-valued (efficient) uses is a manifestation of that process.  Such transfers may result in industry consolidation—but even so, lead to lower prices.  For spectrum, the inefficiency arises not from competition but from the technological nature of the resource and its use.  Using the Chairman’s logic, we could divide the 500 or so megahertz of spectrum we have into a one megahertz slices and have 500 competitors.  Such a proposal is ridiculous, of course.  Service quality and innovation are enhanced if firms possess large quantities of the spectrum resource, and if that resource is finite, which it is, then only a few firms can have large assignments of spectrum.   If the capacity from the resource is exhausted, then the influence of the number of firms using it is altered.  The tradeoffs are apparent (and are discussed in my blogs here and here).  The question is how many is the right number?  And the question is how long the regulator will interfere with the market’s attempt to reach that number before yielding to forces beyond its control?

One other comment made by the Chairman is worth evaluating.  The Chairman suggests that some claim the agency’s blocking of secondary market transactions is “causing a shortage of spectrum.”  Yet, no one has made such a claim.  The spectrum shortage exists, and the Chairman has said as much (indeed, it is a major theme of the National Broadband Plan).  So, the agency’s blocking of transactions is not creating the shortage; it is only interfering with the efforts of mobile providers to attenuate the shortage so that they may (arguably) offer better service at lower prices to their customers.