Wireless Spectrum Life Cycle – Not Exactly Like Real Estate
Spectrum license valuation is often analogized to real estate. It is physically bounded, effectively zoned by regulators for specific purposes, and often leased. The analogy of spectrum licenses to real estate is effective. However, almost all analogies eventually break down. The spectrum license/real estate analogy breaks down when considering long-term future use.
March to Higher Frequencies Leaves Low-Frequency Ghost Towns Behind
The history of wireless spectrum has been a slow march to higher and higher frequencies over time. In 1902, a team led by Guglielmo Marconi made the first successful transatlantic radio transmission from Poldhu, Cornwall to Sydney, Nova Scotia using 272 kilohertz (0.270 MHz), after experimenting with even lower frequencies.[1] The migration to higher frequencies has led to the confusing terminology we have today. High Frequency, for example, is 3 Mhz 30 MHz, relatively low-frequency today when most mobile phones operate between 600 MHz and 2600 MHz. Regulators have had to develop with new names for new higher frequency bands over time, from Very High Frequency (30 MHz to 299 MHz), Ultra High Frequency (300 MHz to 3,000 MHz), Super High Frequency (3,000 MHz to 30,000 Mhz), and Extremely High Frequency (30,000 MHz to 300,000 MHz).[2] Most recently, Tremendously High Frequency (300,000 MHz to 3,000,000 MHz) joins the list.[3]
The traditional real estate analogy would lead us to believe that the new higher frequencies are aking to real estate on the edge of town that, at first, are not particularly valuable but becomes more valuable as the town grows into the outskirts. This is where the real estate analogy often breaks down. In real estate, the center of town, the first area to develop, usually remains the most valuable as the town expands. However, the earlier developed lower frequencies are often left behind and fall into disuse with low value – much like ghost towns in former mining communities.
It would be difficult to imagine a commercially viable use-case today for the 270 kHz spectrum Marconi used.[4] Until recently, spectrum in the 600 MHz to 1,000 MHz was considered the most valuable waterfront property. But demand for higher data rates and technology improvements allowed for the use of higher frequencies. As a result, the prices in the FCC’s 2017 600 MHz Broadcast Incentive Auction drew lower prices than the 2008 Auction 73 for adjacent 700 MHz spectrum. Paging licenses in frequency ranges 35 MHz to 36 MHz, 43 MHz to 44 MHz, or 152 MHz to 159 MHz range find few buyers at any price. At the same time, the 2015 AWS-3 auction (FCC Auction 97), split between 1,700 MHz and 2,100 MHz, hit new heights on a $/MHz-PoP basis. The recent C-band auction, which set an FCC record for total proceeds, was for spectrum at 3,700 MHz to 3,980 MHz that was formerly considered unusable for wireless service. Now C-band is regarded as prime territory for the 5G rollout.[5]
Lower Frequency Ghost Towns are Sometimes Redeveloped
New wireless services sometimes use formerly abandoned spectrum bands and revive value – much like the redevelopment of abandoned industrial lofts for residential use. For example, 220 MHz spectrum and AMTS 216-218 MHz lay dormant for years. Many license holders struggled to justify keeping their licenses, and the FCC took many away as the license holders failed to meet buildout or substantial service requirements. Today Positive Train Control (“PTC”),[6] which requires long-distance coverage but minimal bandwidth to remotely track and stop errant trains, has targeted 220 MHz and AMTS 216-218 MHz bands, as have companies, such as utilities, seeking to monitor remote assets.
Satellite operators relied heavily on C-band in the 1980s and 1990s, but satellite applications slowly moved to the higher Ku- and Ka-bands. Value for C-band orbital slot authorizations had fallen considerably. However, demand for C-band for terrestrial wireless use increased as improvements in air interface, antenna technology, and computational power made the band viable for mobile applications. Harkening back to the real estate analogy, license restrictions, much like zoning, have prevented the satellite operators in the US and Canada from fully realizing value from the alternative use.
On the other hand, redevelopment of frequency bands does not always happen. AM Radio at 550 kHz to 1,550 kHz still exists, but AM licenses sell at values far below those in the heydays of the 1920s and 1930s.
Timing Is Everything
As with many things in life, projecting future spectrum value comes down to timing. As technology becomes more sophisticated and evolves more rapidly, it becomes harder to project future use scenarios for wireless spectrum bands. Who would have predicted, as recently as 15 years ago, the plethora of applications available on smartphones today? Or that high-throughput satellite technology would make satellite broadband viable for consumers? Sometimes companies and investors get the timing wrong. The FCC initially auctioned LMDS spectrum for considerable amounts in the 1990s. Investors believed they could use LMDS spectrum for wireless cable, but those efforts failed. LMDS spectrum largely lay fallow until, nearly 20 years later, the potential demand for 5G and the FCC’s Spectrum Frontiers proceeding[7] liberalized portion of the band for mobile applications. Given the carrying costs of investment, being 20 years early with an investment idea is being wrong.
More Volatility and Greater Complexity
The increasingly rapid rate of change in the communications sector will likely lead to more volatility in spectrum pricing. The industry is now more like the fast-moving Silicon Valley and less like the stable Ma Bell. It is increasingly uncertain how long a specific band of spectrum will be economically valuable Consolidation of the communications sector, from television to wireless to satellite, reduces the number of natural buyers for spectrum assets, further exacerbating volatility for spectrum license pricing. Regulatory trends, including spectrum sharing, add additional complexity. Simply put, spectrum licenses are no longer necessarily the stable long-term store of value they were once considered. Spectrum valuation is becoming more challenging, and responsible investment decisions will require greater thought.
[1] John S. Belrose, “Fessenden and Marconi: Their Differing Technologies and Transatlantic Experiments During the First Decade of this Century.” International Conference on 100 Years of Radio — 5-7 September 1995. https://ewh.ieee.org/reg/7/millennium/radio/radio_differences.html
[2] D. L. Sengupta, V. V. Liepa, “Applied Electromagnetics and Electromagnetic Computibiliy” (John Wiley & Sons, 2005) at 468. https://onlinelibrary.wiley.com/doi/pdf/10.1002/0471746231.app2
[3] Institute for Telecommunication Sciences. https://www.its.bldrdoc.gov/fs-1037/dir-038/_5606.htm
[4] Difficult does not mean impossible. Summit Ridge Group is open to considering creative use cases for valuing spectrum.
[5] “FCC Beging Major 5G Spectrum Auction: Auction of Mid-Band Spectrum Will Further American 5G Leadership” FCC Press Release (December 8, 2020). https://docs.fcc.gov/public/attachments/DOC-368618A1.pdf
[6] Positive Train Control, U.S. Department of Transportation, Federal Railroad Administration https://railroads.dot.gov/train-control/ptc/positive-train-control-ptc
[7] FCC, “Spectrum Frontiers R&O and FNPRM” (July 14, 2016). https://www.fcc.gov/document/spectrum-frontiers-ro-and-fnprm