Why Running the Government Like a Business Would Be a Disaster
Credit…Sam Drew
Mr. Fisman is a professor of economics at Boston University.
The fact that President-elect Donald Trump has tapped two businessmen — Elon Musk and Vivek Ramaswamy — to lead an advisory commission focused on streamlining government operations suggests that these efficiencies will come from making the government run more like a business. In 2017, when he was a White House adviser, Jared Kushner, who now runs a private equity firm, made a similar pitch that business thinking would help to “achieve successes and efficiencies for our customers, who are the citizens.”
It’s a popular idea; it’s also a terrible one. Businesses and government do fundamentally different jobs, and efforts at remaking government with an eye to cost-cutting can end in disaster. That’s because a lot of what the government does is hard to quantify and involves complicated tasks that inevitably require bureaucratic coordination and, yes, inefficiency.
Businesses often run more efficiently than governments do. So it’s natural to conclude that if only businesspeople were put in charge of public administration, everything would work better — shorter lines at the Department of Motor Vehicles, fewer cost overruns at the Pentagon, service with a smile at airport security.
The problem, though, is that business appears more efficient in large part because what it does is usually simpler than what the government does. Take auto glass installation, a classic example among economists because it was well studied by Edward Lazear in the 1990s. It’s a solitary and easy to evaluate activity: A single installer takes care of the job, and it quickly becomes obvious to the car’s owner if the installation is defective. It’s easy, then, to write a contract that compensates an installer based on how many windshields he takes care of each day, without worrying too much about the ill effects of the installer doing shoddy work.
Contrast this with, say, the job of preventing terrorist attacks on U.S. soil. Catching terrorists is a lot more complex a task than installing a windshield, of course, or even assembling a Tesla automobile. It involves cooperation among many agencies: F.B.I. field offices around the country, local law enforcement, the C.I.A. and other agencies across the Department of Justice and the intelligence community. They need to share information and deploy a range of skills, such as surveilling social media and tracking down and capturing criminals.
And if everything is done just right, nothing happens at all. It’s an elaborate job that’s measured by an absence of results, which could be because enforcement authorities are great at what they do or because there wasn’t much risk of attack in the first place.
Suppose that, to run the government like a business, the Department of Government Efficiency, which Mr. Musk and Mr. Ramaswamy will lead, decides to import some high-powered incentives from the auto glass installation business and, say, reward F.B.I. field offices based on how much intelligence they produce on potential threats. We’d surely get bloated, uninformative reports that nonetheless fill agents’ word count quotas. (If this sounds like parody, it’s not so different from the approach the military often took in evaluating success in Vietnam: What mattered were easy-to-measure enemy body counts, which turned out to be very different from the harder to quantify metric of winning the war, especially when civilian bodies were easily mislabeled combatants.)
Even worse, because there haven’t been any major terrorist attacks for a while, the masters of DOGE might be inclined to trim the F.B.I.’s budget. While disbanding the F.B.I. is unlikely to be on the agenda, Mr. Musk has already said the federal government “should be able to get away with 99 agencies” rather than the current figure, which he put at around 428. (The figure given by the Federal Register is 441.)
But as Mr. Musk and Mr. Ramaswamy wield their axes, it may be hard to tell whether they are aimed at red tape or more like D.I.Y. home renovators taking a swing at a load-bearing wall. The problem is that we won’t know what they’ve done until the house has already collapsed. And someone like Mr. Musk may be less attuned to such concerns precisely because when things go awry at his companies — a SpaceX rocket explodes or almost two million Teslas are recalled because of a software failure — the results are not as catastrophic as they can be for government misfires.
Certainly, the government can be run more efficiently than it is now. That’s something I wrote about in 2013, after a management consultant friend of mine raved about the fantastic experience he’d had at the Lower Manhattan passport office. The enlightened bureaucrat responsible for this wonder of government efficiency had put a lot of thought into and experimented with moving people as painlessly as possible through the application process. There was, for example, a separate line for people with complicated cases, so as to not hold up the people with simpler ones.
There should be ways that managers across government bureaucracy can learn from others’ successes. There are surely some ideas that can be imported from business — ideas that NASA and the Postal Service can pick up from SpaceX and FedEx. (Just as surely, there are ideas that SpaceX could learn from NASA.)
So yes, let’s have a commission devoted to making the government work more efficiently. But that’s different from cutting the government down to size, and it’s certainly not the same as simply running it like a business.