A new book titled The Entrepreneurial State  that is now making the rounds gives credit to Uncle Sam for inventing many of the technologies we enjoy today, from the Internet to smartphones. It has given fuel to proponents of government grants, subsidies, and mandates for technologies like solar cells, windmills, electric cars, algae fuel, and cellulosic ethanol. Here, they claim , is justification for continued support of money-losing businesses that one day will be commonplace and profitable, yet might never exist at all without government support.
But getting past the sound-bites, the case for expanded technology subsidies is based on several propositions that require examining.
First, there are the facts. What role, precisely, did the government play? After all, success has many fathers while failure is an orphan.
And then, almost as important, there are the counterfactuals. Which of these technologies would likely have been developed without government involvement, when would they have been developed, and how might they be the same or different? What else could the nation have done with the money if the people who earned it, rather than government bureaucrats, decided how to invest it?
And finally, there are the opportunity costs. What other “investments” made by the government at the time resulted in massive destruction of wealth through the support of dead-end technologies that diverted talent and resources from the market?
A failure to look beyond industrial policy advocates’ cheery anecdotes and to ask the key “what if” questions can lead to very bad policies. As an antidote, I suggest reading French economist Frédéric Bastiat’s discussion in his classic essay, “What Is Seen and What Is Not Seen .”
How can we assess a gambler’s prowess at blackjack if we count only his wins and not his losses? And how can we know if mastering blackjack is the best use of his time, talent, and money if we don’t consider what else could be doing—such as starting a business? This is especially important to consider when a high roller like Uncle Sam tries to gamble with other people’s money.
So, let’s consider the leading example of a technology claimed to be invented, or at least enabled, by government, the Internet. There is no disagreement that the Internet began life as the Advanced Research Projects Agency Network (ARPANET ), a program funded by the Department of Defense to allow university researchers to more easily share data. The telecom connections between early ARPANET users—the actual physical network—was built by stitching together leased lines owned and operated by Ma Bell’s government-sanctioned telephone monopoly (more on that below).
The researchers contributed a set of conventions, or protocols, known as TCP/IP . While extremely long lived, these basic conventions also proved grossly inadequate for the commercial Internet. That’s because ARPANET was a closed private network designed for use by a small set of trusted insiders, not an open public network populated by spammers, Russian mobsters, bandwidth hogs, and a bewildering array of applications that require more than just “best effort” delivery. So, if this government “invention” were ever to be commercialized, it would be up to the market to do the actual design work.
But it was a market that almost didn’t come into existence, because government planners and self-styled “consumer advocates” like Ralph Nader, through his Taxpayers Assets Project , fought tooth and nail to keep commercial traffic off the fledging Internet. Instead, they envisioned the Internet as a great commons, like a public library, not to be sullied by commerce.
It took entrepreneurs like Bill Schrader, founder of PSINet , one of the first commercial internet service providers, to badger his way past the early Internet Acceptable Use Policy, which made the majority of what we do today on the Internet illegal. (How many of us remember that the only way to get an internet email account was to lie about having a university affiliation?)
Would another source of seed protocols more suited to open public networks have developed had ARPANET never been sponsored by government? Most likely, yes, once the telecom market was deregulated.
Which brings us to the actual physical wires upon which the Internet is built. Had Ma Bell’s telephone monopoly not been broken up in 1983, the rush to wire the country with broadband data pipes would never have happened. That is because government-sanctioned and -owned telecom monopolies were once the biggest examples of industrial policy outside the Soviet Union. AT&T and its Bell Operating companies got a 12 percent guaranteed rate of return on capital in exchange for delivering universal service—equality first, innovation maybe later. (Despite the hagiographies written about the brilliant innovations produced by Bell Labs, by the time I got there in 1978, it was a sclerotic bureaucracy plodding its way up from 2400 bps dial-up service.)
Creative destruction was poison to Ma Bell’s business model. The 20-year telecom boom, which led a ravenous crop of competitors to lay down the physical network upon which the Internet was built, happened only after the government got out of the way.
So by all means, read The Entrepreneurial State. But read it with a copy of Bastiat’s “What Is Seen and What Is Not Seen” by your side.