Misinformation in the Internet Tax Freedom Act Debate

In 1998, Congress recognized that taxation could slow the growth of the Internet adoption and suppress the enormous potential of the digital economy and so it passed the Internet Tax Freedom Act (ITFA) to prohibit states from imposing new taxes on Internet access. After being renewed in 2001, 2004, and 2007, ITFA is once again set to expire and there is a lively debate over whether it should reauthorized and made permanent.

Unfortunately, not all participants in the debate are presenting the facts accurately. Michael Mazerov, a Senior Fellow with the Center on Budget and Policy Priorities’ State Fiscal Project, recently wrote a blog post opposing the extension, and one of his key assertions was that there are no differences in broadband subscription rates between states with and without taxes placed on Internet access. To back this dubious claim, Mazerov cites a 2006 report from the Government Accountability Office (GAO).

But there’s only one problem: the GAO didn’t say this.  In fact, when presenting its finding, the GAO states very clearly that their ability to properly analyze the problem was compromised by a lack of broadband pricing data. The only counterfactual the report was able to use was the nine states with broadband taxes grandfathered in by the original ITFA. This limited sample size decreases the ability to imply statistical causation. Despite these limitations, the report did actually find, at the statistically significant level of 10 percent, that Internet access taxes do indeed limit broadband adoption. While a five percent statistical confidence is the common threshold for concluding significance (which means that in 95 out of 100 samples the result will be true), concluding outright that the report demonstrates that there is no impact of taxes on broadband availability contorts the GAO study to Mazerov’s unfounded argument.  In fact, the GAO report finds that in at least 90 of 100 population samples the finding that broadband taxes do indeed have a negative impact on the expansion of broadband adoption would be true.

But Mazerov does not stop there. To back his next assertion, he cites two papers which he claims say that price is not a deciding factor for Internet subscription. Unfortunately, these sources also disagree with him. The first states very clearly that 41 percent of former users discontinued their home Internet service because of its expense. The second categorizes 18 percent of the general population as either ‘Digital Hopefuls’ or ‘Near Converts’ to broadband adoption, individuals without broadband who would be interested and are constrained only by high costs. The report also specifies that those who are constrained only by cost are the most likely to adopt broadband.

Why cite these sources? Most likely because many other studies find the impact of prices (and by extension taxes) on both wireless and wireline Internet connectivity to be quite high. A paper by Austan Goolsbee, former head of the Council of Economic Advisors in the Obama Administration, finds the elasticity for broadband to be between 2.15 and 3.50, with an average of 2.75. In other words, increasing taxes on wireless data and Internet services by $1.00 reduces expenditures on these services by an average of $2.75.  Likewise, Rappoport, Alleman, and Taylor found that for the average monthly U.S. consumer expenditure on cell phone service ($52 per month), every dollar of additional tax reduces expenditures by more than $1.60. Because wireless data services—including broadband Internet access—are an even more discretionary purchase for most consumers, the impact of taxes on wireless data and broadband are likely even higher.

In summary, not only would state Internet taxes have a proportionally large impact on the economy for the amount of revenue raised, it reduces consumption of a good that produces a positive externality, a cardinal sin for any tax policy. In short, making the ITFA permanent would increase broadband usage in the United States, which in turn would improve educational outcomes, spur productivity and have many other positive economic and social impacts.