Social studies students learn how a bill becomes a law, but Jennifer Pahlka would argue that you can’t stop the story at the president’s signature. The administrative infrastructure and vendor contracts that determine how a federal, state or local law is put into practice often act as an inadvertent, unconstitutional veto. Often, as the first people who tried to register for Obamacare discovered, the web developer has as much power as the congressional committee in determining what government does.
Pahlka is the founder of Code for America and the author of “Recoding America: Why Government Is Failing in the Digital Age and How We Can Do Better.” Her nonprofit aimed to improve the user interface of government with volunteers. When Pahlka was hired by President Barack Obama to found the United States Digital Service, she got to do the same work from the inside.
Her new book is a chronicle of the secret gridlock that can cripple even the most enthusiastically bipartisan policy. The government, culturally and legally, has a preference for contracting out the software that underpins our public policy. As Pahlka shows, this means our government often winds up serving its vendors, not its citizens.
Contracts and program constraints are hyper-detailed, going far beyond the text or the intent of the laws that set up the program or administrative agency. The government writes contracts adversarially, trying to bind vendors to every detail of the needed implementation, but leaving no room to course-correct as the team learns more or the needs of citizens change.
In one particularly ridiculous example, Pahlka recounts assisting the Veterans Administration with the Veterans Benefit Management System, which was so laggy as to be borderline unusable. But one of the VA leaders maintained they had no latency problem — latency is formally defined as taking two minutes to load a webpage — and that technically, they were under that time frame.
His obtuseness was a kind of learned helplessness. She pressed him about the way he applied the rules past the point of common sense, and he explained, “I’ve spent my entire career training my team not to have an opinion on business requirements. If they ask us to build a concrete boat, we’ll build a concrete boat.”
It didn’t matter to him if he was given constraints that lock veterans from the benefits they were promised. He was thinking about just one big-picture question, being able to prove that “when it goes wrong, it’s not our fault.”
A big part of the problem is that agency employees are accountable to everyone except the citizens they’re intended to serve. The vendors hold them to the letter of the byzantine contract, and Congress and the Government Accountability Office, lacking the context to supervise them well, hold them to detailed, barely relevant administrative requirements. Long-serving members of the civil service slowly have both their curiosity and their zeal to genuinely solve problems beaten out of them. It’s not worth gathering data on a problem if no one will let you try to fix it. As Pahlka summarizes, “Data is not a tool in their hands. It’s something other people use as a stick to beat them with.”
Pahlka discovered while working with the VA that the website wasn’t just laggy — key application processes were completely unusable for anyone not running a specific (and outdated) version of Internet Explorer and Adobe Acrobat. She and her team got the website functional, and applications increased tenfold. That’s when some VA employees started trying to force the team to shut down the fix. They weren’t prepared to handle the influx of applications. “The inability to apply had been largely invisible, both in the press and inside the building,” she writes, “but the backlog of unprocessed applications starting to build up was very visible.”
Pahlka and her team prevailed, letting veterans access in practice the benefits they were promised in theory. But the fight goes on at every level of government. In one particularly stark case, outside Pahlka’s stories, Derek Wu and Bruce D. Meyer, writing for the Becker Friedman Institute for Economics at the University of Chicago, described a serious benefits infrastructure breakdown at the state level. Indiana hired IBM to automate and manage enrollment for SNAP, TANF and Medicaid, and had to break off the contract because successful enrollments fell so sharply. SNAP food assistance, which meets an urgent, ongoing need, fell 15% in counties using IBM’s system in the first year after rollout.
The decline in all three services was so notable that Wu and Meyer noted that future researchers should look to this data as a gift for gauging the impact of these assistance programs. No one would authorize barring people from their benefits in randomly chosen counties in order to have a perfect control group to measure the impact of assistance, but that’s the result the government got when IBM’s program worked so poorly.
Pahlka’s book belongs on the desk of every lobbyist and citizen activist. Pressing for humane policy means also fighting for a humane implementation of the law. The most vulnerable are too often asked to get past burdensome, arbitrary hurdles, rather than being assisted in their applications and included in the design process. The people who get into civil service in order to help others suffer from moral injury and are painfully aware of how much they are shortchanging the people they want to serve.
One of the key solutions is to give more agency to government agencies. Rather than contracting out the job of building the key interfaces that citizens use to communicate with the government, agencies need to be willing to do more work in-house. It means accepting a higher explicit upfront cost in order to be able to draw on the agency’s ingrained expertise and be flexible enough to overhaul a project without lawyers having to renegotiate a contract. The cost of delegating out core government services is much higher than the sticker price on the vendor bid. And the cost is punishingly regressive, paid by the poor in forgone services and support.
Leah Libresco Sargeant is the author of “Arriving at Amen” and “Building the Benedict Option.” She runs the substack Other Feminisms, focused on the dignity of interdependence.