by Christopher S. Yoo and Timothy Pfenninger
Today, we have published a response to comments about our study “Municipal Fiber in the United States: An Empirical Assessment of Financial Performance,” originally released on May 24. We are grateful for the time and thought so many have put into reviewing our work.
This debate has provided us with ideas of ways to improve our analysis, and we look forward to incorporating those insights as we continue our research in this area. Unfortunately, it has also yielded a degree of inaccuracy and noise, resulting in increased risk of confusion and misunderstanding.
The Best Data Available Selected Without Bias and Analyzed Using Standard Accounting Tools
One overarching point. In different ways, critics have questioned the relatively small sample size (20 networks in the original study) or the limited financial data points covered or the lack of longer term performance information or other data limitations.
These limitations are real. We acknowledge and address them in the paper.
At the same time, our study is based on the best data available. We base our analysis on audited financial statements and bond materials submitted to the SEC – the falsification of which would be a crime. Our study includes all projects that report their data in a usable way. We employ basic and widely understood analytical methods taught in every introductory business program.
Our belief is that it is better to work with the available data and acknowledge its limits than to leave policymakers to decide based on anecdote and conjecture.
Others are free to disagree and argue that the limitations of the data undermine the value of this report. They are not free to claim that the data are skewed or biased the data through selection devices, however. We have shown our work on this point, and it reveals an unbiased method applied systematically and transparently.
Two Modest Corrections
Our study is limited to municipal fiber systems. Thus, we should not have included the Fayetteville system which employs a hybrid fiber coaxial architecture, which reduces the number of projects that properly belong in the study to 19.
We have also acknowledged that the capital repayment schedules for the networks in Chattanooga, Lafayette, and Wilson do not require large balloon payments, a change that makes the narrative of the paper more precise but does not affect our findings regarding the solvency of these networks.
Responding to Flawed Arguments
Our response also walks through a series of flawed and inaccurate criticisms of the study.
Erroneous Claims About the Scope of the Study. Some critics assert the study period is non-representative because it covers only the first five years of any networks’ operations, because these networks performance was skewed by the 2008-09 recession, or because the covered years were outliers for networks in Wilson, Chattanooga, and Lafayette.
All these claims are incorrect. The study includes data from networks ranging from 0 to 12 years in business. The data shows the 2008-09 economic crash did not skew performance of these networks during 2010-14. And we tested whether these years were somehow unrepresentative for Wilson, Chattanooga, and Lafayette by including 2015-16 for those two networks, and the results are largely the same.
Erroneous Claims About Our Economic Methodology. Some critics question our treatment of depreciation, our focus on economic solvency, and our failure to account for benefits that accrued to the electric power business of some municipal fiber operators.
As the paper makes clear, we properly addressed depreciation by adding it back into operating income to ensure no depreciation charges were counted. We rightly focused on economic solvency because we whether a project can cover its actual interest payments is a core issue local leaders must assess. And we appropriately considered only broadband benefits of these networks reflected in the financial statements, because taking into account benefits that were not included in the financial statements would involve relying on non-authoritative sources and necessarily introduce a degree of conjecture.
Erroneous Claims About Our Ethics. Some critics claim the study was supported by corporate interests. It was not.
Extending Our Analysis
We have already identified ways to improve and extend our analysis, which we plan to incorporate in later updates to our research. We also plan to extend our analysis to cover all of the available years of data in the weeks to come. We hope that those interested in these issues will find our continued efforts to be helpful.