Posted by: Joan Marsh on March 14, 2014 at 2:16 pm
When the Commission’s mobile spectrum holdings proceeding was initiated over a year ago, AT&T argued that the Commission’s basic spectrum aggregation test – as originally conceived – remained a sound approach. The test seeks to strike a balance between regulatory certainty, by assuring licensees that spectrum accumulations within a safe harbor will be approved, and regulatory flexibility, by giving the Commission a focused tool to assess whether proposals that exceed the safe harbor screen will foreclose competition.
The benefits of this balanced, consumer-focused approach have been extraordinary. Competition has enabled the U.S. to become the world leader in deploying the next generation 4G LTE technologies and mobile innovation and investment continues to create U.S. jobs, spur U.S. investment and power U.S. growth.
To be sure, modest steps are still needed to update the screen and restore its validity. For one thing, the screen continues to exclude a substantial amount of spectrum that the Commission’s own reports to Congress recognize as usable for mobile wireless service and that, in fact, is being used today. Most prominently, the Commission should correct a current glaring omission by including in the screen the entire 194 MHz of BRS and EBS spectrum held almost entirely by Sprint/Clearwire, rather than the mere 55.5 MHz the Commission has included to date.
Moreover, some recent decisions have departed from longstanding precedent by no longer treating the safe harbor as “safe,” requiring divestitures even where the screen has not been exceeded. These ad hoc departures from the Commission’s framework undermine the predictability that is critical to business planning. While ad hoc review of spectrum holdings in excess of the screen is both expected and appropriate, extending that process to transactions that do not trip the screen unnecessarily adds uncertainty to business planning. The Commission should make clear that its case-by-case analysis will be reserved for proposals to exceed the threshold level in any local market and that this review will be properly focused on the potential for actual foreclosure.
These two steps are all that is necessary to restore the screen to its proper and intended function of addressing the potential for market foreclosure. But as is invariably the case, some parties want to game the regulatory process to their own advantage and, to that end, have sought radical changes to the time-tested framework. Arguments for caps were advanced, but appear to now have faded from favor. There have also been a string of complicated proposals involving different weightings which would do little more than create uncertainty and lead to arbitrary results that would likely reduce investment and innovation.
Sprint has recently returned to the weightings fray with a new proposal. On the plus side, Sprint appears to now recognize that its significant BRS/EBS spectrum holdings will finally be counted in some meaningful way. But to counteract that, Sprint now proposes a new “weighted” screen test where its 2.5 GHz spectrum would be discounted to a small fraction of its rivals’ holdings in low frequency bands. It should surprise no one that this approach would basically relieve Sprint from almost any meaningful spectrum aggregation constraints while effectively foreclosing AT&T from acquiring additional spectrum it needs to meet customers’ needs. Indeed, Sprint’s proposal is so divorced from reality that a combined T-Mobile and Sprint, with spectrum holdings that would dwarf any other carrier, would not trip the screen in any market.
It should thus come as no surprise that, like all of the other weightings proposals, this one is fatally flawed for many of the same reasons, which are enumerated in a filing by AT&T today.
First, it flies in the face of basic economics. The basic argument is that low frequency spectrum is more valuable because its propagation characteristics support larger maximum cell sizes and thus lower deployment costs. Even if this was always true, expected deployment costs are reflected in spectrum costs – if some spectrum is less valuable because it will cost more to deploy, then, all else equal, it will trade for less in the market, which is certainly true for the BRS/EBS bands.
Second, the proposals treat the propagation potential of spectrum as the only determinant of spectrum value, when, in fact, this is neither the only nor, in many cases, even the most relevant driver of spectrum value. Indeed, there are many factors that cut in favor of one band over another for any specific carrier, including channel sizes (larger is better in a broadband world), compatibility with existing holdings (consider the recent wars over AWS spectrum), susceptibility to interference, and standardization/international harmonization, to name just a few.
Third, the proposal attempts to measure purported propagation advantages in a laboratory construct that bears no resemblance to the real world. In the real world, in the suburban and urban areas where the screen matters most, capacity – not coverage – is now king because of soaring wireless demand. Thus, despite its low band holdings, AT&T must continually “densify” its network, just like any other carrier, which is why it is adding 10,000 macrocells, 40,000 small cells and over a thousand distributed antenna systems. Simply put, in capacity-driven areas cell sizes are already small and only getting smaller, and any propagation “advantage” of low-frequency spectrum no longer exists. Indeed, it is a disadvantage because of the greater potential for interference between base stations.
In addition to ignoring the densification of networks that is erasing any theoretical propagation advantage of low band spectrum, the weighting proposals also ignore the fact that all carriers use a mix of high band and low band spectrum and therefore have designed their cell grids to account for the lower propagation potential of the high band spectrum they are using. Weighting factors that assume the use only of low band spectrum are thus wholly unrealistic.
Of course, the proof is always in the pudding, and if it were really the case that a provider using high-frequency spectrum faces higher network deployment costs than a provider using low-frequency spectrum, one would expect to see that T-Mobile and Sprint have much denser networks than does AT&T. Yet empirical data confirms that AT&T, Sprint and T-Mobile each have cell densities that are roughly equal – and not just in urban areas, but throughout entire CMAs of greatly varying population density. In fact, AT&T has more cell sites than Sprint in almost all of the top 100 CMAs and more than T-Mobile in about half of those. Hard to see the alleged deployment advantages there.
These and other arguments are laid out in detail in AT&T’s filing. AT&T continues to believe that, rather than dramatically changing the spectrum screen, the Commission should focus instead on updating and refining the screen to make it clearer and more reliable and predictable. These adjustments would establish the type of fair and reliable spectrum framework necessary for continued U.S. wireless leadership in the 21st Century.