Your internet is slow because of where you live, not what plan you buy
AILSA CHANG, HOST:
Is your internet painfully slow even though you are paying for high-speed service? Well, according to a new investigation, you are not alone. Speed - it actually depends on where you live, not what you pay. The Markup is a nonprofit news service that keeps an eye on big tech, and their research found that service was faster in higher-income white areas than in lower-income, less-white areas, even when everyone was paying pretty much the same price.
To discuss this story, I'm joined now by Leon Yin, who's an investigative data journalist with The Markup. Welcome.
LEON YIN: Hi, Ailsa. Thanks for having me.
CHANG: Well, thanks for being with us. OK. So you looked at internet service in major cities in the U.S. Can you just tell us about some of the findings that stood out most to you?
YIN: The thing that stood out to me the most was how close addresses were that were given the slowest and fastest speeds. So one example that comes to mind is in New Orleans. In Central City, if you looked at a household there - in case you don't know, Central City is a lower-income, majority non-white and historically redlined area. And a household there was offered 5 megabits per second for $55 by AT&T, which is a speed that's so slow the FCC does not consider it to be broadband internet. If you walk just nine minutes away to the 11th Ward, which is a predominantly white, upper income and not historically redlined area, you'd find addresses offered 300 megabits per second for the same cost by the same provider.
CHANG: Wow. OK. And then as you were looking at other major cities across the U.S., like, how common were these disparities between neighborhoods not that far apart?
YIN: They were widespread and rampant. So we looked at 38 major cities across the United States through four different service providers. We found that 92% of these cities had disparities based on income. Two-thirds had disparities based on race and ethnicity, between areas that had the highest concentration of people of color and those that had the highest concentration of white residents. And then 100% - 22 of the cities that we looked at that had historical redlining maps, they all had disparities.
CHANG: That is so fascinating. I am curious, like, what first prompted you all to look into this?
YIN: Well, I was actually assigned it from my editor. So my editor assigned me to reproduce an academic study by Princeton researchers. And they had found that the FCC's national broadband map was inaccurate because providers self-reported their own services, and they overreported often. I was tasked to find out if they are still doing this.
CHANG: Well, how have internet service providers been explaining why there are such stark inequalities between internet speeds when you're talking about neighborhoods pretty close to each other?
YIN: Yeah. So the FCC earlier this year had a proposal to create a digital discrimination policy. And AT&T left a comment for them and said that, we don't discriminate willingly, so it's not discrimination. Like, there's no intent to discriminate. In fact, a lot of these are associated with business factors. So, like, sound logic to explain why certain areas got developed. And we are actually able to account for some of this logic that both industry professionals and critics alike commonly cite. So that includes things like population density, the number of other competitors in the area and adoption rates of broadband, so who's already a subscriber. And so we actually were able to account for these statistically and see if it explained the observed disparities that we saw rampant in all of these major cities. And for the most part, what we found is that these disparities persist.
CHANG: Interesting. Well, then what is the role of the federal government in a situation like this? Like, how can they prevent disparities like the ones you found or impose consequences on internet service providers that don't provide equal service for equal price?
YIN: So there's no pricing transparency that exists. So there's no way that anyone would know that they're getting a bad deal. And partially, that's because internet is not treated as a utility. So telephones, cable TV, electricity, water - it's all regulated, and so pricing and availability is subject to regulation. Internet currently is not.
CHANG: So there is nothing the federal government can do to directly address these disparities?
YIN: Well, one thing that they could do, which has been proposed already, is change the definition of broadband. Currently, broadband is considered fast internet that is above 25 megabits per second download and three up. Most advocates and people who actually work on getting people connected know that this speed is actually too slow. For example, the advocacy group Common Sense Media says that 200 megabits per second are necessary to have concurrent connections in your household, necessary for remote learning and also working remotely.
CHANG: That is Leon Yin, an investigative data journalist at The Markup. Thank you very much.
YIN: Thank you so much for having me.
NPR transcripts are created on a rush deadline by an NPR contractor. This text may not be in its final form and may be updated or revised in the future. Accuracy and availability may vary. The authoritative record of NPR’s programming is the audio record.