COVID-19 has, among other things, brought home the costs of the digital divide. Numerous op-eds have offered solutions, including increasing subsidies to schools, providing eligible low-income people with a $50 per month broadband credit, funding more digital literacy classes and putting WiFi on school buses. A House bill would allocate $80 billion to ideas meant to close the digital divide.
The key missing component of nearly every proposal to solve the connectivity problem is evidence — evidence suggesting the ideas are likely to work and ways to use evidence in the future to evaluate whether they did work. Otherwise, we are likely throwing money away. Understanding what works and what doesn’t requires data collection and research now and in the future.
Why is research so important?
Consider President Trump’s belief in hydroxychloroquine as a cure for the novel coronavirus based simply on his “gut.” That resulted in the government ordering the drug to be produced, distributed to hospitals, and 63 million doses put into a strategic national stockpile.
The well-meaning folks offering up multi-billion dollar broadband plans probably recognize the foolhardiness of the president’s gut-check approach to guiding virus treatment plans. But so far, policy makers and advocates are promoting their own gut beliefs that their proposals will treat the digital divide. An evidence-free approach is likely to cost billions of dollars more and connect fewer people than an evidence-based approach.
It doesn’t have to be this way. The pandemic did not only lay bare the implications of the digital divide, it also created a laboratory for studying how best to bridge the divide. The most immediate problem was how to help kids without home broadband attend distance learning classes. Schools had no time to formally study different options — it was a race to find anything that might help. As a result, schools incidentally ran thousands of concurrent experiments around the country.
We should be learning from those experiences.
Unfortunately, to our knowledge, nobody systematically collected that data. In other words, while people have busily promoted their pet projects and ideas, nobody stepped back to study the vast amount of information being created by organizations that were trying to solve that very problem. There still is time to look at these experiments and learn from them, but it’s running out.
Some might reply that the connectivity problem is simple to solve — broadband is too expensive; therefore, the solution is equally simple — give people more money to connect. But it turns out that’s not quite true. A series of experiments the FCC conducted in 2013 as well as programs offered by private ISPs revealed that although lower prices encourage adoption, real-world responses to lower prices are much smaller than one might expect from looking at survey responses.
The change in adoption we should expect at different price levels is just one of many questions that need to be answered to effectively address the digital divide.
Does the demand curve for broadband change with income? The answer to this question will make it possible to estimate how many additional subscribers we should expect at different price and performance levels.
To what extent do smartphones and the hotspots they can generate substitute for a fixed home connection? We know they are not perfect substitutes, but they are for certain activities. You can’t stream an HD movie on Netflix through your phone to your TV without hitting a data speed constraint. But to what extent is the presence of a mobile hotspot and lack of fixed connection a contributor to the “homework gap” relative to the lack of a computer? Might allowing unlimited data transfer at some minimum speed between schools and students be a more cost-effective and speedier fix?
What is the role of digital literacy? Countless digital literacy classes are offered around the country, but almost no independent evaluations of them exist. Some are likely fantastic, while others a waste of time. In one of the FCC’s 2013 experiments, subscribers were willing to pay an extra $10 a month to not take digital literacy classes. That result seems odd at first glance, but the marginal value of time to a low-income person can be extremely high. If a low-income person is going to add digital literacy classes to her already full week of managing possibly several jobs and a household, those classes better deliver a lot of value. Yet, so far, we know very little about the effectiveness of those programs.
Sometimes, studying a problem is a convenient way to avoid doing anything about it — but doing something without studying it is a way to feel good without actually doing any good.
In addition to the natural experiments taking place already because of the pandemic, new experiments can take months, not years, and cost a small fraction of the money they can save. They can be built into programs we start now so that the program can change course if initiatives prove to not work.
This research would not be expensive, particularly compared to the billions of dollars we are planning on spending on digital divide programs, and, if done well, could radically increase their effectiveness.
A growing consensus finally wants to address the digital divide. The first step toward making real progress is recognizing what we don’t know and doing something to fill that gap in our knowledge.
Gregory Rosston is the Gordon Cain Senior Fellow at the Stanford Institute for Economic Policy Research and Director of the Public Policy program at Stanford University. He previously served as Deputy Chief Economist at the Federal Communications Commission. Follow him on Twitter @grosston
Scott J. Wallsten is president and senior fellow of the Technology Policy Institute, and a senior fellow at the Georgetown Center for Business and Public Policy. Follow him on Twitter @scottwallsten