By Zachary Thompson, Director, Critical Infrastructure — Select Spectrum LLC
April 9th 2026
Recent activity in the 2.5 GHz Educational Broadband Service (EBS) space has drawn welcome attention to a topic we have been actively engaged in for over fifteen years: the significant, and often underappreciated, value of EBS spectrum licenses.
The case that the educational institutions that hold these licenses should obtain professional valuations, and even representation, before entering into sale or lease negotiations is straightforward and widely recognized. For valuations, it is essential that public institutions have both an informed and defensible perspective on the value of their strategic assets before making an executive decision to dispose of it through a sale transaction, or before engaging in a long-term lease (perhaps up to thirty years) that is difficult or impossible to renegotiate the terms of.
The real question, therefore, is who is best positioned to provide the suite of services required. Not all appraisals are created equal, and the difference between a well-supported and poorly supported valuation can translate into either:
- Accepting a poor deal
- A failure to recognize a good deal
In either case, the lack of proper and relevant methodology that drives unrealistic expectations, in either direction, results in leaving significant sums of money on the table, often in the millions of dollars. Thus, accuracy is paramount.
The Limitations of Publicly Available Data
Any credible 2.5 GHz valuation will rely in part on publicly available information including FCC auction results, reported secondary market transactions, regulatory filings, etc. While these are important inputs, relying exclusively on public data leaves significant gaps in the analysis because the information is only loosely comparable and often outdated. Moreover, the limited amount of useful information that is available publicly can be grossly mishandled through fallacious comparisons.
Applying the national average $ / MHz POP from an FCC auction to a specific license tells you very little about what that license is actually worth and instead leads to unrealistic expectations of value. For example, a license covering a mid-sized suburban market has almost nothing in common with the auction-wide average, which blends together everything from dense urban centers to sparsely populated rural counties. What is actually required is to identify the specific licenses within the auction that correspond to similar market areas based on a constellation of relevant factors and then benchmark against those. Failing to do so is the equivalent of appraising a house in the suburbs by averaging every home sale in the state.
The 2.5 GHz secondary market has been shaped by a largely single-buyer environment for more than a decade. In that context, publicly reported transaction prices often reflect the monopsony dynamics at play rather than the true economic value of the spectrum. A valuation that simply averages publicly available prices without accounting for the competitive conditions under which those prices were established will almost certainly understate the fair market value of a given license.
At Select Spectrum, we bring the correct set of tools for this particular job:
- Proprietary transaction data from having directly advised, negotiated and facilitated the sale or lease of hundreds of 2.5 GHz EBS licenses, representing over $500 million in aggregate consideration across the firm’s history. A sizeable sample of these transactions occurred since the 2019 commercialization of the EBS band. For scale, that body of work is roughly equivalent to a quarter of all MHz POPs transacted in FCC Auction 108, the most recent public auction conducted for this band.
- Robust and salient methodology, supported by careful reverse engineering, has granted us direct visibility into how prices, both publicly available and proprietary, vary across market types, competitive conditions and time periods, and our models are calibrated against actual outcomes that are simply not available in the public domain.
This depth of comparable transaction data is critical. A valuation grounded in a handful of publicly reported figures is fundamentally different from one informed by hundreds of data points spanning urban, suburban and rural markets across more than a decade of deal activity. The former may produce a mathematically valid estimate, but the latter produces one that is a sound estimate.
Methodology Matters More Than Most Licensees Realize
Beyond the data itself, the analytical methodology applied to a 2.5 GHz valuation matters enormously, and this is an area where we see meaningful differences between how various firms approach the problem. Let us dive a little bit deeper into the weeds on this.
Consider the basic question of determining the population covered by an EBS license. EBS licenses are defined by circular geographic service areas with a 35-mile radius[1] from a fixed point. EBS license areas, therefore, rarely align with standard geographic boundaries. They do not follow county lines, census tract boundaries, or any other convenient administrative unit. This creates a real challenge: how do you accurately determine the population within an irregularly shaped polygon that cuts across multiple jurisdictions? Moreover, how do you categorize what type of market area a polygon belongs to?
A less rigorous approach might simply categorize the entire license area as “rural” or “urban”, based on the location of the transmitter site or the largest city within the circle, and then apply a single benchmark rate across the full area. This can produce material errors. An EBS license centered near a metropolitan area may cover significant suburban and urban population on one side while extending deep into agricultural or undeveloped land on the other. Treating the entire license uniformly — in either direction — ultimately misstates the value.
Our approach, which dissects each license into individual sub-polygons, allows for independent categorizations to occur based on our proprietary evaluation criteria. The key takeaway is that the rural portion of a license is compared against rural comparable transactions, the suburban portion is compared against suburban comparables, and so on. The results are then summed to produce a composite value that reflects the actual demographic profile of the license as opposed to a simplified approximation of it.
Precision in Population Estimation
Accurate population estimation within these irregular license boundaries is another area where methodology makes a meaningful difference. Standard approaches typically rely on Census geography and involve assigning population at the county level and, perhaps, even the census tract, or census block level. The problem is that EBS license polygons do not respect those boundaries. A census block may be partially inside and partially outside the license area. Simply including or excluding the entire block introduces error that compounds across dozens of boundary segments.
To address this, our valuation models incorporate high-resolution geospatial population data that distributes Census counts across the landscape based on actual land area, rather than assuming uniform distribution within administrative boundaries. This approach is considerably more accurate for licenses with irregular boundaries (i.e. virtually all of them).
This level of precision results in real tangible outcomes. MHz POPs, which is the the standard unit of measurement in spectrum valuation, is the product of bandwidth and population. It would be akin to square footage or acreage in real estate. An error in population estimation flows directly and proportionally into the appraised value. For a license covering a quarter-million people, even a modest percentage error in population can shift the valuation by tens of thousands of dollars. Accurately accounting for only these minute differences can alone justify the investment in a proper valuation.
Why Experience in the Band Is Not Optional
I would also emphasize a point that may be underappreciated by institutions evaluating potential advisors: direct experience transacting in the 2.5 GHz band is not a “nice to have”, it is foundational to producing a credible valuation.
The 2.5 GHz market has its own dynamics that do not neatly translate from other spectrum bands, and certainly not from telecommunications industry experience more broadly. The overlay auction structure of Auction 108, the encumbrance framework for incumbent versus overlay licenses, the competitive dynamics between major license holders and other potential acquirers, the practical implications of channel positioning within the band plan, the transition from legacy lease structures to outright sales. These factors all require specific knowledge that can only come from sustained engagement in this particular market.
Select Spectrum has been immersed in the 2.5 GHz band since our founding in 2010. Our team brings a combined 50 years of experience in the telecommunications industry and approximately 40 years of combined experience in 2.5 GHz, specifically. Our President, Robert Finch, led successful acquisition efforts for 2.5 GHz licenses covering approximately one-third of the United States while at MCI WorldCom, and subsequently led the 2.5 GHz spectrum team at Sprint/Nextel, which is the group that originated the majority of the lease structures T-Mobile relies on today as its primary mid-band spectrum solution. Our Managing Director, Broadband, Andreas Bitzarakis, has led more than 100 EBS transactions during his tenure leading our firm’s 2.5 GHz practice. I myself head up our analytics department and lead our valuation practice. I can safely state that this is the core of what we do, and we have facilitated more transactions in this band than any other spectrum advisor in the country.
Looking Ahead
The market for 2.5 GHz EBS licenses is entering an important period. Many of the long-term leases originally negotiated by Sprint and Nextel in the early 2000s will expire over the coming decade. Major license holders are already actively pursuing acquisitions for economic and strategic reasons. As of early 2026, they have filed FCC applications for EBS license purchases from at least ten educational entities. Other potential acquirers and intermediaries are beginning to approach licensees with offers.
Educational institutions are right to seek professional guidance. But we would encourage licensees to evaluate both the credentials of the firm and examine the specific tools and data that firm will bring to the engagement. The best way to do this is to ask a lot of questions:
- How many 2.5 GHz transactions have you/your firm completed?
- Will your valuation be benchmarked against proprietary transaction data or only publicly available information?
- How will the license geography be analyzed?
- How will population within the license boundary be estimated, and with what data source?
The answers to those questions will tell you a great deal about the quality of the valuation you are likely to receive, and ultimately, the outcome you are likely to achieve.
About the Author
Zachary Thompson is Director, Critical Infrastructure at Select Spectrum LLC. Zack is responsible for leading and supporting transactions with the Utility and Critical Infrastructure Industry (UCII) verticals but also leads the firm’s analytics department and valuation practice. During his time at Select Spectrum, Zack has led or supported numerous appraisals, with the majority focused on the 2.5 GHz band. Zack may be reached at zthompson@selectspectrum.com.
[1] It should be noted that for most licenses this is actually not the case, as the FCC “split the football” for co-channel licensees with overlapping radii; however, let’s assume a full 35-mile radius for this example.