We already know how to affordably expand connectivity; government-run networks ain’t it.

Never-ending infrastructure week is back, baby, and it’s bigger than ever!

The Biden-Harris administration’s proposed infrastructure-jobs bill is both expensive—with a $2 trillion sticker price—and expansive. What isn’t infrastructure, these days? According to some senators, everything from personal care to parental leave could fit the infrastructure bill.

It’s a deliberate reframing of what most people think “infrastructure” is. President Biden says that infrastructure is about “putting Americans to work to get the job done.” (I, for one, require high quality organic coffee, and lots of it!) His press secretary Jen Psaki does admit that this is mostly intended to be a “job-creating package that creates good, union jobs, higher paying jobs.” Well, it is called the “American Jobs Plan,” not the “We Will Build the Roads Plan.” Oh, and somehow all of this is going to be green, too. What is the climate but a bit of natural infrastructure?

It’s obvious why the inner party would prefer a broad definition of infrastructure. The more that qualifies, the more that can be funded. How convenient that so many of the proposed neo-infrastructure programs just so happen to be largely operated by or for the benefit of Democrat constituencies.

It’s much more reasonable to include high-speed internet deployment as part of an infrastructure package. Beyond the fact that it involves installing literal utility equipment in the physical environment, having access to the internet is an essential part of contemporary life.

The White House is going for some New Deal retro appeal, harkening back to the 1936 Rural Electrification Act. “Broadband is the new electricity,” folks. As we’ll see, the administration is drawing from more than just Great Depression nostalgia in selling their plan, adding a good dose of real deal central planning on top.

Okay, okay, so broadband is infrastructure. Apparently, it is super-infrastructure, judging by the funds they want to spend. The package includes some $100 billion intended to close the digital gap in America, bring down costs, and promote competition. That’s a lot of fiber optic cables. For frame of reference, previous broadband infrastructure packages only included single digits of billions for expansion.

So how is America’s broadband infrastructure, anyway? It’s pretty good—not the best, but getting better. Although the U.S. digital infrastructure was able to handle the stresses from COVID-19 remote work and home isolation fairly well, there are still parts of the country where people lack access to fast, affordable, and reliable internet—this is the “digital divide,” which the Federal Communications Commission (FCC) had made a priority to close under the Trump administration. There’s definitely room for improvement.

You can look at connectivity from a few different angles: speeds, prices, and access. The FCC tracks internet speeds across OECD countries; its most recent report finds that the U.S. ranked 10th among developed nations in 2016. New America has some handy international data on average internet costs; the U.S. falls a bit higher in monthly costs than the international average, but it’s far from the most expensive. The FCC has tracked the digital divide in its annual “Broadband Deployment Report” for several years; its most recent publication finds that the number of Americans living without access to acceptable internet has dropped to around 14.5 million in 2019—a fall of about 4 million from the previous year.

So if we want to improve America’s digital infrastructure, closing the digital gap in terms of access should be a top priority. You would think you could easily do that with $100 billion on the table. Unfortunately, the broadband funds will almost certainly not be wisely spent. From the looks of it, the Biden-Harris administration’s plans to expand broadband access will be a big waste. The reason is that the massive spending plan mostly doubles down on policies that have already failed in the past.

One of the biggest problems is that the plan “prioritizes support for broadband networks owned, operated by, or affiliated with local governments, non-profits, and co-operatives”—a long way to say “funds government-run internet.” The plan fact sheet claims that these providers have “less pressure to turn profits” and therefore more of a “commitment to [serve] entire communities.”

That’s not what the experience with government-run internet has shown. Time and again, forays into such networks have resulted in higher costs, not lower costs. Municipal broadband hasn’t done much for access or jobs, either.

It is because government-run internet providers don’t feel pressure to turn a profit that costs and service suffer. One reason is implied in the fact sheet itself: the network is run for the benefit of employees and incumbents rather than the community it is supposed to serve. These government-run networks tend to end up as a hand-out for the people lucky enough to get the largesse, not the community as a whole.

This vintage throwback to FDR-style government-run services may presage yet more big government policies from the bad old days of central planning and austerity. Although the plan doesn’t explicitly state it, there is a risk that this broadband push may result in a move towards old-school rate regulation. Much of the language discusses “overpriced internet service” and “price transparency.” Keep an eye out for such rhetoric to morph into out-and-out calls for government price regulation of internet services.

There’s a lot of pointed language about “future-proofed” infrastructure too, which isn’t specifically defined, but has been interpreted to mean a bias towards fiber broadband. This is an odd choice at a time when one of multiple delivery mechanisms—fiber, wireless, and increasingly high-speed satellite—could be a better fit for a particular area or customer.

Any serious plan to expand broadband in America should not prioritize a particular channel, even if you really, really like the aesthetics of the Tennessee Valley Authority. You should set a goal and incentives for a variety of organizations to compete and best provide services. This is what we’ve largely done to close the digital gap so far, and it’s how we’ll keep finding success.

Actually, there is already a ton of federal and state money up for grabs explicitly intended to expand broadband access. The federal government has allocated some $100 billion (there’s that number again) to promote fixed, high-speed broadband in rural areas for the past two decades through the FCC’s Universal Service Fund’s Connect America Fund. Before literally doubling down with more money on this effort, why don’t we improve the programs that already exist?

My former Mercatus Center colleagues Brent Skorup and Michael Kotrous have thought a lot about how to reform USF spending to better effectuate the connectivity that everyone wants. Right now, the funds mostly subsidize telecom providers, and unequally at that: the bulk of the 2018 funds were doled out to providers in five measly states. Plus, these programs are mired in high overhead costs and inefficiencies.

Here’s a better idea: why not rearrange those funds as a consumer voucher payment that households can use to directly lower their broadband costs? This introduces more market choice, as households can make the best decisions for their own broadband needs, while smoothing out regional inequalities and getting away from the incentive problems involved with simply giving money to providers.

And what of the supply side? There is much that governments can do to encourage faster broadband infrastructure deployment. The goal should be to make it easier for companies to build out the structures that are needed. States should consider implementing policies like “dig once,” which makes it easier to install new fiber equipment whenever someone is digging in the ground anyway.

Consumer vouchers and smart fiber installation norms may not be as sexy as a 1930’s-inspired “Rural Broadbandification Act” spending bonanza on government-owned networks, but these ideas have proven track records of success. We already know how to expand internet connectivity. Biden’s municipal broadbandapalooza is not the way, but it should prove profitable to the party allies fortunate enough to grab some of the handouts.

ANDREA O’SULLIVAN is the Director of the Center for Technology and Innovation at the James Madison Institute in Tallahassee, Fla. Her work focuses on emerging technologies, cryptocurrency, surveillance, and the open internet.