Hillary Clinton’s technology and innovation agenda (0) promises to “finish the job of connecting every household in America to high-speed broadband.” How? Largely by taking money from taxpayers and funneling it to people who promise to expand broadband. In some universe, it might be possible to make the world a better place by taking money from businesses and consumers who were using it to produce wealth and value and putting that money into something that people are otherwise unwilling to pay for. But in the world in which we actually live, Clinton’s plan is likely to waste resources and make our economy worse for the experience.
Underlying the Clinton agenda is a belief that we have too little broadband and that we find ourselves in this situation because: (1) There are barriers to competition; (2) Customers don’t know what they are missing; and (3) We are not taking account of the positive economic spillovers of broadband.
Are there barriers to broadband competition?
Clinton actually addresses the first issue. Her agenda states that state and local governments have erected barriers to broadband competition and that these should be removed. If there are such barriers, she is likely to be right. However, she also wants to unleash the federal bureaucracy to search for anticompetitive practices. This is fine in principle, but there should be clear boundaries so that politically favored companies cannot use this to actually weaken competition.
Are people simply unaware of what they’re missing?
Some pro-subsidy activists implicitly assume that customers who choose not to buy broadband do so because they don’t understand how great it is. However, the real reasons are likely more complicated. Some low-income families (1) move often or otherwise have mobile lifestyles, making home broadband impractical. Demographics and primary uses of telecommunications also affect demand (2). Perhaps some people simply are not interested in spending their time watching Netflix and shopping online.
Ignoring these possibilities reveals that groups supporting a blanket subsidy approach (3) are probably suffering from an illusion of knowledge – an overconfidence that one possesses all the information that is needed to make informed decisions. As Friedrich Hayek noted (4) many years ago, social planners and those who aspire to be them rarely know as much as the individuals who are on the front lines in their work and lives.
Are we missing out on positive spillovers?
Citing Pew’s finding (5) that home broadband adoption in the US stands at 67% (and ignoring that residential mobile broadband jumped over 60% in 2015), economist Hal Singer argues (6) that increased residential broadband would have an economic spillover effect, meaning economic benefits would accrue to parties other than the customer and the broadband provider. Because broadband providers and customers do not internalize these spillovers, society ends up with too little broadband.
Placing so much importance on spillovers is problematic. Economist Christiaan Hogendorn outlines the basic arguments (7) in favor of believing in extensive broadband spillovers: (1) Broadband is used for nearly all business and broadband providers do not capture these downstream profits and (2) Broadband has network externalities.
Mr. Hogendorn’s first argument applies to all suppliers of business inputs, so it is unclear why broadband should be subsidized and not the others. Furthermore, it is a good thing that broadband does not capture all downstream profits: these profits are needed to encourage downstream innovation and competition. The second argument assumes that all network effects are externalities (they are not). Absent regulatory constraints, broadband providers can often develop prices and business arrangements (such as vertical integration and partnerships) that internalize network effects, resulting in appropriate supplies of broadband. Given this and all of the above, it is far from certain that America is actually suffering from too little broadband.
Clinton’s subsidies would waste resources
Clinton’s broadband plan is largely a subsidy program. The agenda says it would grow broadband by expanding and extending to new government institutions the subsidies currently provided under the Department of Commerce’s Broadband Technology Opportunities Program (BTOP (8)) and the FCC’s E-rate (9) program, growing the FCC’s Lifeline (10) program, continuing the Rural Utilities Service (11) program for broadband, and providing broadband grants to governments from a $25 billion Infrastructure Bank that she intends to create. All of these elements of the plan allow significant political discretion regarding who receives money. If experience is any guide, these initiatives will be replete with failed and uncompleted projects (12), political favoritism (13), and little if any positive impact (14).
Hal Singer (15) (not Clinton) suggests that a targeted subsidy of $4.3 billion annually could be used to induce 20 million customers, who are currently unwilling or unable to pay the full price for broadband, to buy it. This might be true if the government would and could dependably identify exactly those customers. But this never happens in practice.
The prevalence of flawed targeting is illustrated by the FCC’s current Lifeline program. Lifeline provides price discounts on telecommunications services for low-income households in an effort to make service affordable. Originally the program applied only to wireline voice services. Later the FCC expanded it to mobile services and now allows the subsidies to be used for broadband. Professor Olga Ukhaneva of Georgetown University studied the impacts (16) of the expansion to mobile services. She found that the expansion helped low-income customers, but not in the way the FCC probably intended. She finds that even if the FCC perfectly enforced its Lifeline policies (which is not true today — some people receive the subsidy even though they are ineligible), 19 out of 20 low-income households subsidized would buy service even without the subsidy. Even if Mr. Singer’s new broadband subsidy were only half as bad as the FCC’s Lifeline program, the cost would be about 10 times his estimate.
The bottom line is this: while everyone I know is interested in solving legitimate issues of connecting consumers who need it to the Internet, we need to avoid jumping to policies that are appealing in their simplicity and political popularity, but that create waste rather than solutions.