On November 2, the citizens of Kaysville, Utah, will decide if the municipality will build a new broadband network, often called a government-owned network (“GON”). Citizens and leaders face a tough choice. Putting an Internet service provider (ISP) under the auspices of a local government doesn’t sidestep the endemic issues of cost faced by network builders. Since the entire network must be built before any revenues can be collected, new networks owned by municipalities face the same kind of financing problems and risks as any other entrant.
Survey results from citizens of Kaysville conducted by the city reflect what many think will happen with these projects.1Design Nine. “Kaysville Fiber Planning Recommendations.” Accessed September 30, 2020. https://www.kaysvillecity.com/DocumentCenter/View/2413/Design-Nine—Kaysville-Broadband-Recommendations Eighty-seven percent of respondents said that they expect better Internet service in the form of lower prices and higher quality service. Moreover, citizens as well as boosters of the project think the project will invigorate economic development. How- ever, research into publicly owned networks presents a far more nuanced picture than one obtains from surveys of public opinion.
As Kaysville’s own internal study notes, “it is inefficient and wasteful to build full duplicated digital” infrastructure, as this will likely raise the cost of telecom services to all public and private users.2Ibid. Kaysville already has two wired ISPs and three wireless ISPs,3Ibid. so the introduction of a new municipal Internet provider doesn’t necessarily guarantee a better, more efficient market, or an invigoration of the local economy, but it is likely to cost taxpayers handsomely.
The effect of a new entrant for consumers and businesses depends largely on the price and quality of the new broadband system in conjunction with the reactions by other providers in their price-quality offerings.4Sharkey, William W. The Theory of Natural Monopoly. Cambridge: Cambridge University Press, 1982. The following report details some of the key insights from the academic literature on whether government-owned networks are effective at increasing quality, competition, and economic growth. Three lessons for policymakers shine through:
- Building a new network is costly and risky;
- Broadband competition is complicated but is largely a product of local conditions like population density; and
- Municipal broadband is often a popular investment politically, but the actual benefits to citizens are often mediocre.
When Does Building a Broadband Network Make Sense?
Broadband networks require a large initial investment, but the costs drop substantially for each additional subscriber.5 Sutton, John. Sunk Costs and Market Structure: Price Competition, Advertising, and the Evolution of Concentration. Cambridge, Mass. Eng.: The MIT Press, 2007. High fixed cost and low marginal cost structures tend to give rise to markets with only a few competitors. Economists call markets consisting of a small number of competitors oligopolies or oligopolistic markets.
Central to the cost of operating a network is the technology necessary to make it run. ISPs have a range of technological options when it comes to deploying service.67. Calzada, Joan, and Fernando Martínez-Santos. “Broadband Prices in the European Union: Competition and Commercial Strategies.” Information Economics and Policy 27 (2014): 24–38. https://doi.org/10.1016/j.infoecopol.2014.04.001. For example, CenturyLink and EarthLink utilize existing telephone networks, while Comcast depends on the cable network. An ISP could also rely on a completely new system that is designed from the ground up, which is what is being proposed in Kaysville.
Costs also vary widely depending on population density, local conditions, and importantly, regulatory compliance costs. A broadband project in rural Wisconsin to upgrade an old telephone system cost roughly $8,000 per household,78. Brommerich, David. “Cochrane Co-Op Telephone (CCT) and Town of Milton Weighing Proposed Plan for Broadband Online Services,” Winona Daily News. November 25, 2017. https://www.winonadailynews.com/news/local/cochrane-co-op-telephone-cct-and-town-of-milton-weighing/article_2e1547f4-c5be-54ae-9ffe-d09e083896c3.html. whereas in rural Tennessee the cost for a new fiber broadband network hovered around $5,000 per household.8 Cawley, Elena. “Second Grant Will Help Expand Broadband into Rural County,” February 3, 2018. https://www.tullahomanews.com/news/local/second-grant-will-help-expand-broadband-into-rural-county/article_308e0d30-d650-5905-b10a-9985826129e5.html. Expanding broadband to rural areas tends to be more costly, because the large fixed costs must be spread over fewer people. The last time the Federal Communications Commission ran the numbers, the agency estimated that the hardest to reach households would cost nearly $90,000 per premise to connect to broadband.91de Sa, Paul. “Improving the Nation’s Digital Infrastructure.” Federal Communications Commission, 2017. https://transition.fcc.gov/Daily_Releases/Daily_Business/2017/db0119/DOC-343135A1.pdf.
In the middle of a large, dense city, hooking up another household might cost only $800.10Yarow, Jay. “How Much It Would Cost Google To Become A National Cable Company Like Comcast.” Business Insider, December 7, 2012. https://www.businessinsider.com/how-much-it-would-cost-google-to-build-a-cable-network-2012-12. However, even in such cities, costs can climb. Networks typically string their wires on already existing poles, but if the poles are crowded or if the city requires utility wiring to be underground, the costs can quadruple.11 Columbia Telecommunications Corporation. “An Engineering Analysis of Public Rights-of-Way Processes in the Context of Wireline Network Design and Construction,” 2011. https://www.ctcnet.us/2011%20CTC_Study%20NationalLeagueCities%2011-59.pdf. Per-household cost drops with more density, but projects can still be slowed by the municipal permitting process, state regulations, or federal requirements like a National Environmental Policy Act (NEPA) review.
Obtaining permission from the municipality adds cost and time to a build, both of which can vary greatly depending on the area. Stephen Milton, who helped to design and build the Gigabit Now service in Sea Ranch, California explained that his company had to obtain permission from 23 separate local, county, and federal agencies to get the new project up and running.12Milton, Stephen. “Great Communities.” Broadband Communities Summit. Speech presented at the Broadband Communities Summit, May 2, 2017. Broadband provider Sacred Wind out of New Mexico wrote in a filing to the FCC that an application involving one landowner and one authorizing jurisdiction commonly takes 2–4 years to complete, while something more complex, that involves more than one piece of land spanning multiple authorizing jurisdictions, can take anywhere from 4 to 8 years to complete.13Canfield, Jill, and Jesse Ward. “Reply Comments of the National Telecommunications Cooperative Association.” Electronic Comment Filing System, 2011. https://web.archive.org/web/20191120202800/https://ecfsapi.fcc.gov/file/7021712146.pdf. Indeed, when San Francisco first considered building a GON, the feasibility study named the city’s knowledge of how to traverse its own red tape as a critical asset.14Columbia Telecommunications Corporation. “Fiber Optics for Government and Public Broadband: A Feasibility Study.” CTC Technology & Energy, January 2007. https://www.ctcnet.us/SFFiberFeasibilityReport.pdf. GONs have to go through the permitting process like anyone else, they just tend to know how to do it much better.
Users in a market vary greatly in what they demand. Gamers tend to like low latency so they can quickly react in an online setting. Other users might want high speeds to download documents and watch streaming video on a couple of devices. Meanwhile, some broadband subscribers might only need the service for a limited range of activities like email or surfing the web. The variety of service offerings from broadband providers reflect the variety in consumer demand. However, with Internet services, consumers tend to find little value is added beyond 100 Mbps.15 Liu, Yu-Hsin, Jeffrey Prince, and Scott Wallsten. “Distinguishing bandwidth and latency in households’ willingness-to-pay for broadband internet speed.” Information Economics and Policy 45 (2018): 1-15. As a result, firms tend to be limited in the services they can provide and the price points at which they can offer service.
The break-even point at which a new broadband project will be a good investment can vary massively.16A good investment is one that would suffice either a net present value or internal rate of return investment decision rule. Within the industry, take rates express the percentage of customers who eventually subscribe to the service as compared to all of the homes passed in a given broadband build. These take rates in turn influence the number of potential competitors that a market can maintain. As the organization Community Networks notes, “The [average] 30% take rate minimum means most markets would support, under ideal circumstances, no more than three competitors.”17“Competition.” Community Networks, 2020. https://muninetworks.org/ content/competition. In dense urban settings, a 20 percent take rate might be sufficient to cover costs, while upwards of 80 percent may be needed for a rural project to break even.18Treacy, Ann. “Lessons from Rural Minnesota Broadband Feasibility Studies.” Blandin Foundation, 2014. https://blandinfoundation.org/content/uploads/vy/Feasibility_Studies_-_01-31-14.pdf. Indeed, research into GONs finds that many are cash flow negative and thus fail to break even.19 Yoo, Christopher, and Timothy Pfenninger. “Municipal Fiber in the United States: An Empirical Assessment of Financial Performance.” University of Pennsylvania Law School’s Center for Technology, Innovation and Competition, 2017. https://www.law.upenn.edu/live/files/6611-report-municipal-fiber-in-the-united-states-an.
Studies on Broadband Competition
Official statistics from the Federal Communications Commission in June 2019 suggest that 70 percent of Americans have access to two or more broadband providers when satellite options are excluded, while only a quarter, mostly in dense areas, have access to three or more.2021. FCC Fixed Broadband Deployment. Accessed October 1, 2020. https://broadbandmap.fcc.gov/#/area-comparison?version=jun2019&tech=acfow&speed=25_3&searchtype=county Investments made in the past by telephone and cable companies, as well as more recent upgrades, tend to be the key determinants of the price and quality of broadband available in a region.2122. Wong, Siew Fan, Luke Younghoon Chang, and Myeong Choel Parl. “Analysis of the Dynamic Broadband Technology Competition: Implications for National Information Infrastructure Development.” Research Gate, 2016. https://www.researchgate.net/publication/296638105_Analysis_of_the_dynamic_broadband_technology_competition_Implications_for_national_information_infrastructure_development. As expected, regions with more broadband providers tend to have faster download speeds.22Reed, Markum, and Alison Watts. “Product Quality and Competition: Evidence from the Broadband Industry.” Applied Economics 50, no. 24 (2017): 2719–32. A longitudinal study found that most markets have shifted toward higher quality service tiers over time.23Flamm, Kenneth, and Pablo Varas. “The Evolution of Broadband Competition in Local US Markets: A Distributional Analysis.” SSRN, March 19, 2018. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3142329 Molnar and Savage find that markets with two wireline ISPs have faster download speeds than regions with only one.24Molnar, Gabor, and Scott J. Savage. “Market Structure and Broadband Internet Quality.” Wiley Online Library. John Wiley & Sons, Ltd, April 27, 2017. https://onlinelibrary.wiley.com/doi/abs/10.1111/joie.12106.
Research on new firm entry over time, rather than static snapshots of a single year, tend to show dynamic markets. Prieger et. al. examined broadband markets from 2011 to 2013 and discovered that DSL service gets better when a cable player enters the market, and also when cable operators start to offer faster speeds.25 Prieger, James, Gabor Molnar, and Scott Savage. “Quality Competition in the Broadband Service Provision Industry.” AEA Web, 2015. https://www.aeaweb.org/conference/2016/retrieve.php?pdfid=371. More recent work from Flamm and Varas (2019) found that entry or exit by wireline competitors in areas with existing legacy networks like cable or telephone “has essentially no impact on maximum download speeds offered by wireline ISPs.”26 Flamm, Kenneth, and Pablo Varas. “Quality Competition at the Competitive Margin in US Residential Broadband Markets.” SSRN, July 29, 2019. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3427582. When wireless competitors enter and exit the market, in contrast, there are “large changes in maximum download speeds offered by wireline ISPs.” This suggests competition in the near future might be increasingly influenced by wireless carriers.
Another line of inquiry in the entry literature follows from pioneering work by Bresnahan and Reiss that found the population required to support a second firm is often much larger than the population needed for a single firm.27Einav, Liran, and Jonathan Levin. “Empirical industrial organization: A progress report.” Journal of Economic Perspectives 24, no. 2 (2010): 145-62. In their example, a small town of 800 people could support one dentist, but if the entry of a second dentist intensifies competition, it will take more than 1,600 residents to support two dentists. Sometimes, it will take nearly double the population. In Bresnahan and Reiss’s example, a total of 2,400 residents would be needed to support a second dentist. The addition of a third or fourth firm often requires an incremental population increase similar to the second. Moreover, competition often kicks in quickly, but doesn’t increase much with each additional entrant. Using the Bresnahan and Reiss framework, Xiao and Orazem estimated that, “Once the market has one to three incumbent firms, the fourth entrant has little effect on competitive conduct.”28 Xiao, Mo, and Peter F. Orazem. “Does the Fourth Entrant Make Any Difference?: Entry and Competition in the Early U.S. Broadband Market.” International Journal of Industrial Organization. North-Holland, December 9, 2010. https://www.sciencedirect.com/science/article/abs/pii/S0167718710001384.
In one of the few studies focused on municipally-owned broadband networks, Steve Landgraf exploited variation in cities with and without state-based restrictions on municipal broadband to understand how GONs affect the market. Some states don’t allow local cities to build broadband networks, so by looking at similar cities that have municipal power companies but differ in their state restrictions, Landgraf was able to isolate the effect of municipal broadband after the network became operational. Landgraf discovered that GONs are “associated with lower maximum upload and download speeds offered by private cable and DSL providers.”2930. Landgraf, Steven W. “Entry Threats from Municipal Broadband Internet and Impacts on Private Provider Quality.” Information Economics and Policy. North- Holland, June 30, 2020. https://www.sciencedirect.com/science/article/pii/S0167624520301220?dgcid=rss_sd_all. In short, that research suggests government broadband projects like Kaysville’s can crowd out investment.
Effect on Economic Growth
The availability of broadband tends to strongly correlate with economic growth and higher incomes. Correlation, however, doesn’t mean causation. While there is agreement on the broad positive link between the two features, there is less agreement on the mechanism that binds them.
Macro-level studies almost universally show a positive correlation of broadband deployment with growth.30 Czernich, Nina, Oliver Falck, Tobias Kretschmer, and Ludger Woessmann. “Broadband Infrastructure and Economic Growth.” Wiley Online Library. John Wiley & Sons, Ltd, May 12, 2011. https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1468-0297.2011.02420.x. One often cited report from the World Bank concluded that a 10 percentage point increase in fixed broadband penetration increases country-level GDP growth by 1.2 percent.31 Qiang, Christine Zhen-Wei, Carlo M. Rossotto, and Kaoru Kimura. “Economic impacts of broadband.” Information and communications for development 2009: Extending reach and increasing impact 3 (2009): 35-50.. Employment, as well, seems to be associated with broadband access.3233. Bai, Yang. “The Faster, the Better? The Impact of Internet Speed on Employment, by Bai, Yang.” Information Economics and Policy. Elsevier, 2017. https://ideas.repec.org/a/eee/iepoli/v40y2017icp21-25.html Rural areas with access to the technology in the early 2000s were more likely to have new firms locate there.33Kim, Younjun, and Peter F. Orazem. “Broadband internet and new firm location decisions in rural areas.” American Journal of Agricultural Economics (2016): aaw082. However, the effect was the most pronounced for “rural areas and those adjacent to a metropolitan area, suggesting that this effect increases with agglomeration economies.” Agglomeration economies is just a fancy term to describe the benefits from being located near a city. In related research, the National Telecommunications and Information Administration (NTIA) found that proximity to central cities may be more strongly associated with the availability of the highest speed levels of broadband service than population density.34Beede, David, and Anne Neville. “Broadband Availability Beyond the Rural/ Urban Divide.” National Telecommunications and Information Administration, 2013. https://www.ntia.doc.gov/files/ntia/publications/broadband_availability_rural_urban_june_2011_final.pdf. In other words, being close to a large population center might be driving broadband deployment.
The overall impact of broadband availability on firms, industries, and specific regions is more muddled than the relationship specifically with GDP growth. Broadband tends to have a positive impact on sales,35Canzian, Giulia, Samuele Poy, and Simone Schüller. “Broadband upgrade and firm performance in rural areas: Quasi-experimental evidence.” Regional Science and Urban Economics 77 (2019): 87-103. but little impact on firm level productivity.36Fabling, Richard, and Arthur Grimes. “Picking up Speed: Does Ultrafast Broadband Increase Firm Productivity?” SSRN, December 13, 2016. https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2884330. Economists investigating the impact of the technology in Ireland concluded there was “no statistically significant effect of broadband adoption on firms’ productivity (growth).”37 Haller, Stefanie, and Sean Lyons. “Broadband Adoption and Firm Productivity: Evidence from Irish Manufacturing Firms.” EconPapers, March 31, 2015. https://econpapers.repec.org/paper/pramprapa/42626.htm. Knowledge-intensive firms often locate their business based on broadband availability, leading to a clustering of these businesses in regions with widespread access.38Mack, Elizabeth A., and Tony H. Grubesic. “Broadband Provision and Firm Location In Ohio: An Exploratory Analysis.” Wiley Online Library. John Wiley & Sons, Ltd, August 28, 2008. https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1467-9663.2008.00487.x. But one study found digital connectivity in remote rural areas actually hurt local entrepreneurs because it opened these regions to e-retail.39Cumming, Douglas, and Sofia Johan. “The differential impact of the internet on spurring regional entrepreneurship.” Entrepreneurship Theory and Practice 34, no. 5 (2010): 857-884. Research focused on Swedish firms also uncovered a negative relationship between super-fast broadband and local retail sales.4041. Nordin, Martin, Erik Grenestam, and Joakim Gullstrand. “Is Super-Fast Broadband Negative? An IV-Estimation of the Broadband Effect on Firms’ Sales and Employment Level.” Lund University, 2019. https://project.nek.lu.se/publications/workpap/papers/wp19_8.pdf. For voters and leaders, these effects should be weighed against the potential consumer benefits of access to lower-cost products.
Terminology is key to unraveling the relationship between broadband and economic development. Broadband access or deployment isn’t the same as broadband adoption. Thus, while the technology has to be deployed before it is adopted, adoption seems to be a much better predictor of economic growth.41Whitacre, Brian, Roberto Gallardo, and Sharon Strover. “Does rural broadband impact jobs and income? Evidence from spatial and first-differenced regressions.” The Annals of Regional Science 53, no. 3 (2014): 649-670. After all, broadband only exists because there is demand for it by consumers and producers. Still, broadband rollout is tightly connected to economic growth. The few studies that try to understand this relationship by employing causal identification strategies muddle the picture even more. De Stefano et. al. found no effect from broadband on the performance of firms in En- gland,4243. De Stefano, Timothy, Richard Kneller, and Jonathan Timmis. “The (Fuzzy) Digital Divide: The Effect of Broadband Internet Use on UK Firm Performance.” EconPapers, November 8, 2014. https://econpapers.repec.org/paper/notnotecp/14_2f06.htm while Kolko found a positive effect on economic growth but not on employment rate or wages.43 Kolko, Jed. “Broadband and Local Growth.” Journal of Urban Economics 71, no. 1 (2012): 100–113. Work from Akerman et. al. discovered that broadband rollouts tend to skew skill demand, improving the labor outcomes and productivity of skilled workers while worsening the outcomes and productivity of low skilled workers.44 Akerman, Anders, Ingvil Gaarder, and Magne Mogstad. “The skill complementarity of broadband internet.” The Quarterly Journal of Economics 130, no. 4 (2015): 1781-1824. Other research tends to confirm this skill polarization effect.45Michaels, Guy, Ashwini Natraj, and John Van Reenen. “Has ICT polarized skill demand? Evidence from eleven countries over twenty-five years.” Review of Economics and Statistics 96, no. 1 (2014): 60-77.
Properly designed research on municipal networks is rare, but there are two standouts. An econometric study of municipal broadband conducted by economist Brian Deignan found that business establishments grew by about 3 percent after a GON entered.4647. Deignan, Brian. “Community Broadband, Community Benefits? An Economic Analysis of Local Government Broadband Initiatives.” Mercatus Center, 2014. https://asp.mercatus.org/system/files/MGPE_Deignan_0.pdf. However, worker income saw a drop of 1.3 percent, while private sector employment saw no growth. At the same time, local government employment expanded by around 6 per- cent. The effect may have been to shuffle the deck, not increase the size of the pie.
Using a similar method, Sarah Oh disentangled the various factors affecting growth to examine whether municipal broadband tends to stimulate economies. Oh detected no change in the unemployment rate, the household broadband subscription rate, or labor force participation due to GON entry.47Oh, Sarah. “What Are the Economic Effects of Municipal Broadband?” Technology Policy Institute, 2019. https://techpolicyinstitute.org/wp-content/uploads/2019/11/OhTPRC2019.pdf. After controlling for the nonrandom nature of these projects, Oh again confirmed that there was no change in these three key factors.
The Key Takeaways for Policymakers
Policymakers, especially those in Kaysville, need to attend to the specifics of each proposal and what they want to accomplish with a GON. Constructing a new ISP from the ground up is difficult and costly work. As one review of experiences in Europe framed it, “Municipalities have to dive into detailed operational details in order to achieve successful project results.” Importantly, the roles and responsibilities of the operator, especially as it concerns operation and maintenance, need to be clearly defined.48 Januschke, Oskar, Jakob Egg, Michael Mairhofer, Thomas Wimmer, Wolfgang Reichl, and Ernst-Olav Ruhle. “Operation, Administration and Maintenance of Municipal Fiber Networks.” EconStor. Calgary: International Telecommunications Society (ITS), 2017. https://www.econstor.eu/handle/10419/169468. In the United States, many local governments fail to conduct proper due diligence, which is reflected in the fact that so many broadband projects are cash flow negative.4950. Yoo, Christopher, and Timothy Pfenninger. “Municipal Fiber in the United States: An Empirical Assessment of Financial Performance.” University of Pennsylvania Law School’s Center for Technology, Innovation and Competition, 2017. https://www.law.upenn.edu/live/files/6611-report-municipal-fiber-in-the-united-states-an.
Kaysville’s broadband project remains difficult to assess. Proponents say a detailed market analysis complemented by clear survey data on customer demand has been conducted, but it has not been released to the public. Nor has the city produced a detailed sensitivity analysis that could help educated audiences understand the risks. Because the outcomes of broadband projects depend greatly on local market conditions, it’s difficult to know whether Kaysville’s project will be worthwhile for residents.
The research findings on economic impacts from broadband are not comforting. While there is a general positive relationship between broadband and the economy, building a new government-owned network does not automatically cause economic growth or other benefits. Indeed, many government broadband projects fail to break even. Municipal broadband often sounds good in theory and may therefore be popular politically, but the actual benefits to citizens can be outweighed by the costs.