This post, which considers the potential of TV White Space (TVWS) spectrum as a tool for expanding rural access, uses two recent and related events as jumping off points for the discussion.
1. Microsoft’s July 12 announcement of its Airband initiative, which aims to use unlicensed TV White Space (TVWS) spectrum to provide broadband Internet connectivity to 2 million rural Americans.
2. A July 26 Wired piece by Susan Crawford entitled Microsoft is Hustling Us With White Spaces in which Susan expresses skepticism about Microsoft’s claims regarding the value of Airband and TVWS in general as a connectivity solution for rural America.
My decision to focus on these two events and how they relate to the potential of TVWS to help connect rural communities is tied to multiple factors:
1. I’ve been a rural TVWS enthusiast since first researching the subject roughly a decade ago, which led me to write a policy paper entitled Spectrum Policy 2.0: White Space, the Internet and the Public Interest.
2. I’ve played a small role helping to catalyze the initial Microsoft-supported TVWS projects planned for Michigan, which are part of the broader 12-state Airband initiative. And I’m glad to see Microsoft launch Airband and continue to provide leadership and support for the still-nascent TVWS ecosystem.
3. I’m a longtime admirer of Susan and her work, agree with her on most telecom-related policy issues, had the pleasure of working with her around the time her Captive Audience book was published, and would welcome the chance to do so again in the future.
4. I think Susan’s piece is helpful in that it; a) reviews some TVWS history (as she notes, it’s been long, slow and messy); b) reminds readers it’s wise to: i) assume that Microsoft, like virtually all publicly traded corporations, is motivated primarily by self-interest and; ii) avoid blindly accepting heavily promoted claims regarding new technologies, products and initiatives; c) highlights the capacity advantages of fiber optics relative to TVWS and virtually any other wireless or wired forms of connectivity.
5. I’m concerned that the article’s headline and a few of its statements and implications cast an overly negative light on what I believe is an initiative that can be a key component of an effective rural connectivity strategy for the U.S. and other countries facing similar or even more severe rural connectivity challenges. More specifically, I see rapid development of a healthy TVWS ecosystem as a key component of a FAST PIIPS strategy that can expand digital connectivity and empowerment in rural areas, in the U.S. and around the world.
I hope that, by addressing some points asserted or suggested by Susan’s piece, this post will help put Microsoft’s Airband initiative in a realistic and constructive context that will: a) make sense to those concerned about our nation’s rural connectivity gaps, including Susan and those inclined to agree with her Airband/TVWS critique and/or: b) invite feedback about where I’m misguided in my analysis of Airband as part of a “fiber + wireless” public interest-focused strategy for bringing the Internet’s benefits to those who today remain insufficiently unconnected.
In addition to its headline, the section of Susan’s piece that strikes me as unhelpfully negative is the following paragraph (bolding is mine):
“And hustled is what we will be if we believe that Microsoft’s plans, by themselves, will fix America’s desperate internet access problem in rural areas. You see, while using white spaces will certainly be better than nothing in rural locations, those guard bands simply aren’t wide enough to allow for genuine, world-class internet data transmission to human beings in living rooms. Not possible. Not enough bandwidth. True, where commercial mobile radio (like AT&T and Verizon) isn’t available at all, white spaces will definitely help. You could use it for Internet of Things applications that are very very useful, as in advanced agriculture—don’t need to send much data to do that. But you would never use a white spaces transmission service alone if you didn’t have to. You’d end up with maybe a handful of Mbps or even less—hundreds of times less than what people with fiber would be getting. White spaces will definitely be another arrow in the quiver used by local fixed wireless operations, but they are no kind of substitute for actual great consumer internet access in rural areas.”
Though I’d replace the word “hustled” with “mistaken,” I don’t have a real problem with the first sentence in this paragraph since, by including the “by themselves” phrase, it can accurately be applied to virtually any connectivity technology. What I do take issue with is the rest of the paragraph’s characterization of the capabilities and potential benefits of TVWS in rural areas. For example, I think Susan’s reference to TVWS as “guard bands” delivering “a handful of Mbps or even less,” doesn’t do justice to the large amount of TVWS spectrum available in rural and small town America, and its potential impact on the provision of affordable access is those areas.
Among the policy-related virtues of TVWS spectrum is that it is most abundant in rural areas where connectivity tends to be most lacking and network economics are most challenging. This contrasts with the situation in major metro markets, which tend to have relatively few available TVWS channels but, compared to rural areas, enjoy relatively robust broadband speeds and competitive options, as well as more favorable economics for investments in fiber optic networks. So, more so than most other connectivity options, TVWS availability is well matched to today’s unmet needs.
This matchup of availability and need can be appreciated by checking TVWS spectrum availability for small rural vs. large metro communities via the location-based search function available on Google’s (and other companies’) online spectrum database. It is also visually apparent in the color-coded map below (each channel is 6 MHz wide), as well as zoomable maps available on some of these spectrum database web sites.
For example, according to Google’s spectrum database there are 216 MHz of TVWS spectrum available in Alpena, a town of 10,000 located in a rural region of Michigan targeted by one of Microsoft’s first wave of Airband projects. In contrast, under current FCC rules, there is no TVWS spectrum available in New York City and Los Angeles, and only 24 MHz in Chicago. Given this, I’d suggest that using the term “guard bands,” while it may be accurate when considering metro markets, is unhelpful as a way to describe the very large swaths of contiguous TVWS spectrum available in many rural areas (e.g., Alpena’s TVWS spectrum includes two large contiguous chunks of spectrum, one 78 MHz wide, the other 60 MHz). As TVWS technology matures, this abundance of contiguous spectrum will enable substantially faster speeds than are being delivered today with what are still-early generations of TVWS technology.
Though I lack the expertise to estimate the average per-user data rates and pricing structures that Alpena’s 216 MHz of TVWS spectrum could support as TVWS technology matures, I’m pretty confident achievable speeds will be far beyond “a handful of megabits.” And, I suspect that, if TVWS equipment volumes are even a fraction of today’s Wi-Fi or LTE volumes, service and equipment prices will be at least reasonably affordable as long as they are not subject to profit-focused monopoly dynamics. And if–as Microsoft and other TVWS supporters are recommending–the FCC adopts policies ensuring that at least a handful of TVWS channels are available even in large urban markets like NYC and LA, it seems feasible that TVWS could reach volume and price levels roughly comparable to Wi-Fi, LTE and other mass-market technologies. And even if, as Susan suggests, Airband is driven more by self-interested corporate priorities than by the public interest goal of closing the rural connectivity gap, I see significant overlap between these two agendas and believe policymakers and local stakeholders would be wise to seriously consider how to leverage that overlap to more effectively connect currently underserved communities.
Understanding where TVWS (& other solutions) can help
All that being said, I share some of Susan’s skepticism with regard to Microsoft’s assertion that TVWS is “the best approach to reach the 80 percent of th[e] underserved rural population that live in communities with a population density between two and 200 people per square mile…[and that] fixed wireless and limited fiber to the home should be used for communities with a density greater than 200 people per square mile.”
This conclusion, derived from a directional study by Microsoft and Boston Consulting Group, presumably defines “the best approach” based on criteria and assumptions that, as far as I know, have not made made public, and may not apply very well to many real-world local situations (e.g., the specific mix of local needs, priorities, infrastructure, institutions and other resources, and the availability and cost of funding for a particular project). As one Michigan-based example that helps makes this point, I’d cite the case of Midwest Energy, a rural electric co-op that’s deploying a FTTH network in an area averaging eight premises per mile using mainly co-op owner equity and an RUS loan. If I remember my conversion ratios correctly, this translates into a population per square mile of roughly 60, well below the 200 cited by Microsoft as the threshold that can justify an investment in an all-fiber network. The implication here is that determining “the best approach” to expanding broadband connectivity in any given rural area will be influenced by the mix of local factors including, as in this case, the presence of a rural electric cooperative, its strategic priorities, and its financial health and access to low-cost capital.
Given this, I’d encourage Microsoft and Boston Consulting Group to share the findings and methods of their directional study as part of a broader collaborative effort aimed at developing analytical tools that can better inform broadband-related policy and investment decisions at the local, state and national level.
I’ll have more to say about the nature of such tools in the final post in this series, the next installment of which will consider Educational Broadband Service (EBS) spectrum. As that post will explain, EBS is another underutilized connectivity resource that is abundantly available in some rural areas, and is already being deployed in a handful of pioneering projects that will generate valuable lessons from which others can learn. And, as I’ll also discuss in a later post, there are some areas of rural America with large amounts of unused spectrum in both the TVWS and EBS bands, along with increasing amounts of open-access high capacity fiber backhaul linking community anchor institutions (CAIs) in these rural communities. As I’ll explain in that post, this combination of unlicensed and/or education-focused spectrum, coupled with open-access non-profit fiber and the active involvement of CAIs and other community stakeholders, is the essence of the FAST PIIPS model.
Below are links to the other posts in this series. Feedback from readers, especially constructive criticism, is welcome.
Evolving Human Systems 