Professor Yoo’s Flawed Study Flunks Test on Municipal Broadband

By Jim Baller and Joanne Hovis

On May 24, 2017, University of Pennsylvania Law School professor Christopher Yoo and  student Timothy Pfenninger released a paper entitled “Municipal Fiber in the United States: An Empirical Assessment of Financial Performance”. Applying a Net Present Value analysis to data on 20 municipal broadband projects for the years 2010 to 2014, the paper concluded that 11 projects generated negative cash flow and that only two projects generated sufficient cash flow to pay off the debt incurred during the networks’ estimated useful life.

As with many past industry-supported attacks on municipal broadband, it will take some time for interested readers to dig into the details of the Yoo study and fully understand its strengths and weaknesses.[1] That will occur in due course. There are, however, a number of serious problems with this study that leap out at once.

For one thing, almost immediately after releasing their report, the authors issued a press release acknowledging that they had “erroneously stated that the bonds used to finance the projects in Chattanooga, TN; Lafayette, LA; and Wilson, NC; call for balloon payments toward the end of their bond terms.”[2] They say that they made this error because they were reading summary tables on page 2 of the relevant sources rather than “[t]he more detailed discussions contained inside the Official Statements.”[3]

While the authors claim that this error did not affect their financial analysis, one wonders how many other serious errors exist in the study—and how many other times the authors took shortcuts instead of reviewing the full available data. Perhaps if they had contacted the cities at issue to verify the data, they could have caught this mistake in advance. Apparently, they skipped that step as well.

A particularly important shortcoming of the study is that the choice of 2010 through 2014 as the study period introduced significant selection bias. This occurred for at least two reasons. First, at least three of the key projects studied—Chattanooga, Lafayette, and Wilson—all began service in 2009. Because city-wide projects take years to build (as do all substantial communications networks), none of these projects was fully operational throughout the 2010 to 2014 study period.  This may also have been true of several of the other projects included in the study.[4]

Second, amplifying this problem, during the 2010-2014 study period, the U.S. was slowly recovering from the 2008–2009 collapse of the global economy. This undoubtedly dampened demand for communications services relative to what it would have been in more normal times. Yet, the authors chose to base their projections on these unrepresentative data.

That using 2010-2014 as the study period was inappropriate becomes all the more apparent when one examines Chattanooga’s, Lafayette’s, and Wilson’s actual post-2014 experience.  For example, the Yoo study predicted that it would take Chattanooga 412 years to “turn positive.” In fact, Chattanooga’s post-2014 sales have soared, its fiber utility has already paid off its debt, and its net revenues from sales of fiber-based communications services are helping to keep Chattanooga’s electric rates down.[5]  At the same time, according to studies by Dr. Bento Lobo, professor of finance at the University of Tennessee (Chattanooga), Chattanooga’s fiber network has contributed to the creation of between 2,800 and 5,200 jobs—and to economic and social benefits to the community ranging in value from $865 million to $1.321 billion.[6]

Lafayette and Wilson have also had strong post-2014 experiences. Wilson has reduced electric rates by 22 percent, a clear sign of the strength of its utility,  LUS Fiber has already paid $600,000 to the City’s General Fund, and by 2020, it will be paying about $1 million annually, which support Lafayette’s fire and police forces.  LUS Fiber has also helped to create nearly 1,000 high tech jobs in Lafayette.

Indeed, all three cities have merited strong bond ratings from Wall Street.[7] So, one must ask, who has more credibility in assessing the strength of these municipal fiber projects, a law professor and his student—or professional bond rating firms such as Moody’s Investment Service, Fitch Ratings, and Standard & Poor’s Ratings?

Another problem with the Yoo study is that the boldness of its conclusions is undermined by the many caveats and qualifications set forth at various points in the study. For example, at one point, the report states, “[t]he high-level analysis presented in this study may overlook key details that can help explain the results in particular cases. In addition, the financial performance of these projects may improve in the future.”[8]

At another point, the study notes that the authors did not have certain data for seven of the 20 projects being studied, so they made projections based on data that they had for the remaining 13 projects.[9] This effectively made a small sample of projects even smaller.

Likewise, in the course of discussing their linear model for calculating discounted cash flows from 2010 to 2014, the authors candidly caution, “[c]are should be placed on attributing too much weight to the results of these models. The relatively small number of observations and the simplicity of the model prevent them from achieving statistical significance.”[10]

Nevertheless, despite these and other similar qualifications, the authors did not hesitate to draw broad conclusion and opine that “the overall results provide a useful snapshot of the nature and size of the challenges that municipal fiber projects face.”[11] Unfortunately, in reporting on this study, several media outlets have focused on the authors’ sensational conclusions and have missed the significance of these caveats.

The study also insultingly suggests that city officials are shortsighted in seeking to secure for their constituents affordable access to gigabit broadband: “[a]lthough some day people may need the download speeds that FTTH makes possible, the evidence suggests little need for such speeds today.”[12]

This breezy dismissal of the need for fast broadband demonstrates that the authors lack understanding of the goals of the local officials they seek to warn away from municipal fiber projects. Like electric utilities in the last century, fiber networks are multi-purpose platforms and drivers of simultaneous progress in just about everything that matters to communities. This includes economic development, education, public safety, healthcare, transportation, environmental protection, energy, government service, democratic discourse, digital equity, entertainment, and much more.

These critical community goals are why local government officials are eager to obtain the benefits of advanced communications networks—by working with willing incumbents, partnering with new entrants, building their own networks when necessary, or developing other innovative arrangements that work for them.

In any event, local governments are not alone in seeking broadband at gigabit speeds. Google Fiber, AT&T, Comcast, and many other entities have made the case for such robust networks—and have invested in building them in select areas, notwithstanding Mr. Yoo’s skepticism about the need for such networks.

Furthermore, throughout their report, the authors condescendingly treat municipal leaders as ignorant hayseeds.  Municipal leaders, the authors say, “should carefully consider all of the relevant costs and risks before moving forward with a municipal fiber program.”[13] Cities considering whether to initiate a municipal fiber project “should carefully evaluate the performance of prior efforts and assess whether differences exist that would likely lead to a better outcome.”[14]  In our experience, local government officials are well aware that capital-intensive infrastructure projects are costly and risky and actively seek  to learn from the experiences of other communities. They do not need a lecture on common sense from a law professor and a law student.

When a locality considers the possibility of developing a municipal fiber network, months—even years—of intensive discussion typically occurs at the local level, during which every fact, every assumption, and every technological, legal, financial, social, and other relevant issue is examined from multiple angles. Incumbent service providers invariably participate actively in this process, as do local entrepreneurs and competitive providers. In the end, this process ensures that only those projects that meet a significant unmet need and have a high probability of success will go forward.

Finally, as former Pennsylvania Governor Ed Rendell noted at the event where this problematic report was released, localities undertake broadband initiatives not for purposes of profit, but for public purposes such as education, economic development, healthcare, and civic engagement. These purposes are fundamental to the missions and obligations of local governments, and the value delivered by fiber networks in these areas goes far beyond income on a financial statement. As with so much else about municipal broadband, Mr. Yoo and his student fail to accurately understand that.

Jim Baller is the President, and Joanne Hovis is the Chief Executive Officer, of the Coalition for Local Internet Choice

[1]               The Yoo study was published under the auspices of Penn Law and its Center for Technology, Innovation and Competition. Among the Center’s supporters are AT&T, Comcast-NBC Universal, CTIA, NCTA, Time Warner Cable, and Verizon. https://goo.gl/oY1Ouv.

[2]               Steven Barnes, “Correction,” Press Release, University of Pennsylvania Law School, May 26, 2017, https://goo.gl/7XFWJ2.

[3]               Id.

[4]               Table V includes a column entitled “Age of Project as of 2010,” but the numbers listed in the column are unreliable.  For example, Chattanooga and Wilson are listed as two years old, and Lafayette is listed as three years old, even though the study elsewhere acknowledges that each began service in 2009.  According to Table V, at least five other projects were two or three years old as of 2010.

[5]               Colin Wood, “Report discredits municipal fiber financials — but experts balk,” Statescoop, May 30, 2017, https://goo.gl/cU6P1x.

[6]               Bento J. Lobo, The Realized Value of Fiber Infrastructure in Hamilton County, TN (May 2, 2017), https://goo.gl/g4vDDJ.

[7]               Dave Flessner, “EPB Readies $250 million bond issue with upgraded rating,” Times Free Press, https://goo.gl/Ait7w8, July 9, 2015 (Fitch and Standard & Poor’s upgrade EPB’s bond rating from AA to AA-plus); “Standard & Poor’s Upgrades LUS Fiber’s Bond Rating to A+”, https://goo.gl/YjASlL (April 2015); City of Wilson, NC, “Fitch Ratings upgrades City of Wilson bond rating,” https://goo.gl/CZbyxh December 30, 2016 (“The City of Wilson’s bond rating was recently upgraded by Fitch Ratings to AA+, a significant endorsement of the financial health of the city.”)

[8]               Yoo study, at 1. See also id., at 3.

[9]               Id., at 1.

[10]             Id., at 24.

[11]             Id., at 1.

[12]             Id.

[13]             Id., at 1.

[14]             Id.