Broadband and Unemployment

Economists at the University of Tennessee at Chattanooga and Oklahoma State University conducted a study that correlates broadband speeds to unemployment. They concluded that unemployment rates are 0.26% lower in counties with faster broadband. They further concluded that broadband has a bigger impact on jobs in rural areas than in metropolitan ones.

The lead economist on the project, Bento J. Lobo, lives in Chattanooga and began the investigation because of the high-speed municipal fiber network in the City. He was curious if that network had contributed in a measurable way to jobs. The study also looked at FCC data from the National Broadband Map dataset and looked at 95 other counties in Tennessee.

The study looked at broadband availability and unemployment over the period from 2011 to 2016. The study measured broadband availability by considering places that have more than one landline broadband provider defined as served, with the rest either unserved or underserved. They concluded that Tennessee looks a lot like the rest of the country in that urban areas have decent broadband while broadband options in rural parts of the state are limited.

I have to wonder about the extent to which poor FCC broadband mapping data suppressed the findings of the study. I wrote a blog earlier this week that highlighted a Penn State study that showed the inadequacies of the FCC data in Pennsylvania, where the number of people that have broadband availability was overstated in every county in the state. For example, the Penn State study showed that there are counties in the state that the FCC considers as fully covered by broadband, but in which the average actual download speed in the county is at half of the FCC’s 25/3 Mbps definition of broadband. That kind of mapping error has to be affecting the results found in this unemployment study by overstating the rural areas that have good broadband.

The fact that the authors found a correlation is impressive after understanding the nature of the FCC dataset. The authors of this report say that the topic is worthy of more granular studies looking at specific counties that get broadband for the first time. At CCG we work with such counties and we’ve gathered a lot of anecdotal evidence over the years that broadband brings jobs to rural America.

Everywhere we go we see evidence that rural people are hungry for good-paying jobs. In one rural county we studied in Minnesota we saw that every single farm in the county had an incorporated home-based business that was separate from farming. Somebody at every farm was trying to supplement farming income. The rural folks in that county hoped that they could find better-paying jobs after getting broadband. In this particular county, the farms didn’t even have rudimentary DSL, and their broadband options were limited to satellite broadband or cellular data.

Parts of this county have gotten wireless broadband that is advertised at speeds between 25 Mbps – 50 Mbps. I agree with the researchers that more granular study ought to be done and it would be illuminating to have studied the rural households in this county before and after the introduction of broadband. My guess is that broadband has a bigger impact than calculated by this study.

Good broadband enables rural residents to find home-based online jobs – an exploding part of the new economy. In this particular county the unemployment rate might not change due to broadband – but household incomes are likely to increase as farm family members find better-paying jobs online to replace the ones they are tackling today without broadband. That kind of job upgrade would not be measured by looking at the unemployment rate, but would be discovered in more granular analysis.

The impacts in bigger cities like Chattanooga must be a lot harder to quantify. Again, we know anecdotally that programmers and other high-tech folks moved to Chattanooga due to the ubiquitous gigabit fiber network. It has to be very hard to somehow pinpoint those fiber-related new jobs out of a diverse big-city economy. This has to be particularly hard to pinpoint the impact of broadband in an economy where unemployment rates fell nationwide during the whole study period.

4 thoughts on “Broadband and Unemployment

  1. Try having this conversation to corporate right wing lobbyists such as Richard Bennett or misinformation darling Michi Iljazi whose job is to convince government officials and an uneducated public that our broadband connectivity is wonderful if you leave it all up to the private sector.

  2. Is there anything in here that can validate what follows what? Lower unemployment areas would likely have more money, so could we just be seeing that areas with more disposable income have better broadband?

    Anyone reading this blog is probably predisposed to believe in the power of the internet (me too!) so I’d worry about confirmation bias in these interpretations.

    Similarly there was just talk about studies “proving” broadband inhibits population loss. And I had the same question. What if broadband just tracks to wealth and people leave (relatively) impoverished areas at higher rates? How different does that look?

    I think we need more direct before-and -after studies.

    As a broad spectrum skeptic, I’m also interested in more data on both the impact of home work and estimates of the potential limits — I think we saw peak eBay in y2k and the gig economy’s big impact has been to lower wages, ask any graphic designer. As part of a farm family I know that small improvements help low margin businesses a lot, but I think that’s a farm policy and global competition problem not a technology shortfall (my Illinois cousin Jeff also doesn’t do precision ag, use drones, etc., for the thousands of acres he farms, just for the record, so the whole 5g-in-a-tractor thing isn’t a fully global need yet, either.)

    Last thought: Chattanooga is an awesome test case. I have as yet to see any big new things come out of there. Is there some strong evidence of, say, lots of medium new things coming up, general gdp growth better than equivalent towns, quantifiable impact of home work, etc.

    The Google fiber experience was that people really, really wanted triple play. Cord cutting or cheaper media != economic growth.

    Super happy if I’m misinterpreting, here, somehow…

    • Your observations point out the difficulty of undertaking economics studies at a global level. I agree that case studies that look at before and after would be a lot better, but those are incredibly labor intensive and would be difficult or impossible in larger cities.

  3. Pingback: Weekend Reads 062819 – rule 11 reader

Leave a Reply