Mun: Spectrum – Will its intrinsic value decrease in a small cell world?

An old investment adage tells us to “buy low, sell high.” This seems so obvious that it’s almost not worth stating. However, knowing what’s high or low requires knowing the standard with which to compare. Determining this standard, or intrinsic value, of an asset is not easy. Value today may not be the same as value tomorrow.

In wireless, one of the key strategic assets is spectrum. Determining the intrinsic value of this fundamental asset can be difficult – and the common metric of “$/MHz-Pop” is sometimes not a good indicator as each auction brings different competitive dynamics. At spectrum auctions, bidders often hire a team of economists and consultants to help determine this, or get close to it. In traditional wireless business, the licensed spectrum delivers value in many ways:

  1. Spectrum provides a fundamental asset by which wireless traffic can be delivered to subscribers without interference from others;
  2. The exclusive use of licensed spectrum provides a way to manage service quality over large areas;
  3. The exclusive use of licensed spectrum in perpetuity creates a barrier for new entrants.

In small cell deployment scenarios, however, these value attributes seem to diminish. On the item #1, the relative impact of spectrum weighs less in small cell deployments as compared to traditional macro cells. To illustrate this, we analyzed the unit economics of delivering a Gigabyte (GB) of data per month with certain cost assumptions about spectrum, equipment, site lease, and backhaul. 

In terms of cost, spectrum contributes about 23% of the unit cost of delivering a GB of data per month from a macro cell, and a much smaller portion, about 2%, of the unit cost for the outdoor small cell case. In both instances, site lease and backhaul constitute a majority of the costs. Relative to other cost elements, spectrum ($/MHz-Pop) is a small portion of the overall cost for the small cell case since the incremental cost of spectrum is directly proportional to the population covered by that small cell. In other words, on a variable basis, the spectrum cost in a small cell deployment is much less in comparison to the macro case since the cell coverage, thus population, is smaller.

On item #2, the exclusive use of licensed spectrum to deter interference, especially over a wide area, becomes less valuable in a smaller geographic area covered via a small cell. In fact, unlicensed spectrum uses via contention avoidance technology such as Wi-Fi has provided a meaningful substitute for cellular data in sedentary use. Inherently, the probability of interference is less for a small cell case than for a macro cell (assuming potential interferers are uniformly distributed). Now that the LTE-U/Wi-Fi coexistence test plan has been completed and agreed upon, we expect to see LTE-U and LAA small cell deployments to ramp up in the coming years. The unlicensed LTE technologies such as LTE-U, LAA, LWA, and MulteFire will likely further diminish the value of licensed spectrum use in small cell deployments as mobile operators and new entrants look to harness the free and cheaper spectrum bands.

Finally, on item #3, the spectrum value as a barrier to new entrants seem to be under attack. The shared spectrum uses in the 3.5GHz CBRS band, and also proposed for the 24 GHz band, effectively limits the long-term value by limiting the time frame of “use it or share it” rule. Instead of holding a licensed spectrum in perpetuity, a licensed tier of the three-tier licensing scheme has a term limit which effectively limits its value as a deterrent for new entrants. Also, large numbers of licenses covering smaller geographic “parcels” lowers a capital requirement to acquire those licenses. New entrants can cobble together multiple low-cost licenses to cover a given metro market. As a comparison, a bidder would have spent billions of dollars to purchase paired blocks of AWS-3 spectrum in a top metro market like New York.

Small cell architecture, unlicensed LTE and shared spectrum technologies are providing new flexibility and favorable unit economics. While the intrinsic value of licensed spectrum – i.e., fundamental asset of wireless capacity, managed service quality, and barrier to new entrants - still holds true for the current business model based on a traditional macro architecture covering a wide area, those values seem less critical in small cell deployments. Value judgment on spectrum is constantly changing and myriad players ranging from cable operators, over-the-top players like Google, and mobile operators appear to be adjusting their valuation calculus on spectrum. 

Spectrum will always be valuable, especially for expansive outdoor coverage and mobility. Indoors, the old formula for calculating spectrum value does not apply. New technologies leveraging unlicensed bands, in-building equipment, and new industry players change the equation. As we approach the 5G world, we need to sharpen our pencils.

Kyung Mun is a Senior Analyst at Mobile Experts LLC. Mobile Experts is a network of market and technology experts that provides market analysis on the mobile infrastructure and mobile handset markets. Over the course of his 20+ years in wireless and cable industries in a dynamic range of roles from engineering to product management and technology strategy, he has contributed to the advancement of mobile communication, while working at leading companies in the mobile value chain including Motorola, Texas Instruments, Alcatel-Lucent, and a few startups in between. He holds undergraduate and graduate degrees in electrical engineering from the University of Texas at Austin and Georgia Tech, and studied finance and strategy at Southern Methodist University.