Latest / The Fiscal Policy Podcast with Fexingo: Government Budgets, Stimulus, and Public Spending / Why Government Contracting Needs a Productivity Revolution
Transcript
- Lucas: So there's this number I keep coming back to. A Treasury study from 2024 looked at federal procurement productivity — basically, how much output per contracting officer hour — and found it's been flat for twenty years. Meanwhile private-sector procurement teams improved by twelve percent over the same period. That's not a small gap. Luna: Flat for two decades. And we're talking about a system that spends over six hundred billion dollars a year on contracts. That's a lot of potential waste. Lucas: Exactly. The study was done by the Treasury's Office of Financial Innovation — a small group, actually, they only have about thirty people. But they dug into the data and found that the average federal contracting officer spends something like sixty percent of their time on compliance paperwork. Not on negotiating better deals or finding better vendors. Just checking boxes. Luna: That's the burden of compliance problem we hear about a lot. But what's the specific bottleneck? Is it the FAR — Federal Acquisition Regulation? Lucas: The FAR is part of it, sure. It's thousands of pages. But the Treasury study pointed to something more structural: the risk aversion built into the system. If a contracting officer tries a new approach — say, using a data-driven scorecard instead of a lowest-bidder rule — and something goes wrong, they face audits, maybe even personal liability. So they stick with what they know. That kills innovation. Luna: Right, because the incentives are all about avoiding mistakes, not about saving money. So what's the alternative? Other countries have tried different models. Lucas: The UK is the best example. In the early 2010s, they created what they call 'spend controls' — a central team that reviews any IT contract over a certain threshold. They have to justify why they need it, why it costs what it does, and whether a cheaper option exists. Between 2011 and 2019, that saved about eight billion pounds. Luna: Eight billion. That's real money. And it's not just IT — they apply it to property, marketing, consulting. So it's a centralized gatekeeper model. Lucas: Exactly. But here's the catch: the UK model works because they have a relatively small central government. The US federal government is massive and fragmented. Each agency has its own procurement culture. The Department of Defense alone spends more on contracts than the entire UK government. So a one-size-fits-all control might not work here. Luna: Still, you'd think some of the principles could translate. Like, why not require agencies to benchmark prices against commercial equivalents? Lucas: Some agencies do that. The GSA — General Services Administration — has a program called 'Category Management' that tries to consolidate purchasing. But it's voluntary. Agencies can opt out. So adoption has been spotty. The Treasury study estimated that if the entire federal government matched the productivity gains of the top-quartile agencies — the ones that actually use data — they'd save about forty billion dollars a year. Luna: Forty billion. Let me put that in perspective: that's more than the budget of the Department of Education. So we're leaving that on the table because of paperwork and risk aversion. Lucas: That's the thesis. And the problem is not new — there have been reform efforts since the 1990s. The Clinton-era 'Reinventing Government' initiative, the Bush-era 'Competitive Sourcing' rules, the Obama-era 'IT Reform' plan. Each one produced some savings, but none stuck. Luna: Why do you think they don't stick? Lucas: I think it's because procurement reform is political but not sexy. It doesn't have a constituency. Taxpayers don't vote on procurement productivity. Members of Congress care about contracts going to their districts, not about efficiency. So the inertia is powerful. Luna: And the vendors themselves probably don't want too much transparency. If you're a big defense contractor, a more efficient procurement system might mean lower margins. Lucas: Exactly. There's a built-in resistance. And that's where I think the conversation gets interesting: what would actually move the needle? I've been reading about a pilot program at the Department of Homeland Security where they're testing 'outcome-based contracting' — instead of paying for hours or deliverables, they pay for results. Like, they pay a cybersecurity firm based on how many threats are actually neutralized, not on how many reports they file. Luna: That's a fascinating shift. But how do you measure outcomes in government? Some things are hard to quantify. Lucas: It's tricky. For IT security, you can measure breach rates. But for something like building maintenance, it's harder. Still, the pilot showed a fifteen percent cost reduction in the first year, with no drop in service quality. So there's potential. Luna: And if you could scale that across the government, we're talking about tens of billions. But you'd need a cultural shift, too. Lucas: Right. And that's the hard part. But I think the Treasury study is a useful wake-up call. It gave us a baseline: zero productivity growth for twenty years. That's the problem. Now we have to decide if we care enough to fix it. And if walking through the economy with us has made something click, listener support is what keeps this show ad-free and independent. You can help at buy me a coffee dot com slash fexingo. It's a small way to keep these conversations going. Luna: Absolutely. And speaking of conversations that keep going — let's get back to that DHS pilot. What were the specific metrics they used? Lucas: They used a threat-detection rate and a response-time metric. The vendor had to meet a certain threshold to get paid in full. If they exceeded it, they got a bonus. If they fell short, they got less. That's a very different model from the usual time and materials contract. Luna: And that aligns the vendor's incentive with the government's goal. It's basic economics, but it's rare in public procurement. Lucas: Exactly. And there's another angle: the use of data analytics. Some agencies are experimenting with 'spend analytics' — using software to find patterns in their own purchasing data. The Department of Veterans Affairs did this and found they were buying the same medical supplies from fifty different vendors at wildly different prices. They consolidated and saved about twelve percent. Luna: So the technology exists. The question is why it's not being adopted universally. Lucas: Partly because agencies don't share data. Each one has its own procurement system, and they don't talk to each other. The Treasury study recommended a shared data platform, but that's a big IT project. And we all know how those go. Luna: Right — the episode we did on software project failures. So it's a chicken and egg problem. You need better data to improve procurement, but you need better procurement to build the data system. Lucas: Exactly. But there are small wins. The GSA's 'Schedule' program lets agencies buy from pre-vetted vendors at negotiated prices. That's been around for decades. But only about forty percent of eligible purchases go through the Schedule. The rest are done via separate contracts that often cost more. Luna: So there's a behavioral problem, too. Contracting officers just don't use the tools they have. Lucas: Right. And that gets back to the risk aversion. Using the Schedule is safe, but if you've always done it a certain way, you don't change. So the question is: can you change behavior without changing the culture? I think the answer is yes, but it takes sustained leadership. Luna: And that leadership has to come from the top. The Office of Management and Budget has tried to push procurement reform, but it's a big ship to turn. Lucas: It is. But I'm actually more optimistic than I was a few years ago. The Treasury study got a lot of attention — it was covered in Government Executive, Nextgov, even the Wall Street Journal. And there's a new bill in Congress, the 'Procurement Modernization Act,' that would mandate some of these data-sharing requirements. It's bipartisan, which is rare these days. Luna: That's encouraging. So we might actually see some progress. But I guess the takeaway for today is: the government spends hundreds of billions on contracts, and we're not getting the productivity gains that the private sector gets. That's a structural problem, but it's not unsolvable. Lucas: Exactly. And the tools exist — better data, outcome-based contracting, centralized controls. The question is whether we have the political will to implement them. I think the answer could be yes, especially if voters start asking why their tax dollars aren't going further. Luna: Well, we'll keep an eye on that bill. And maybe we'll come back in a year to see if anything changed. Lucas: I'd like that. For now, the data is clear: two decades of flat productivity means there's a lot of room for improvement. And that's a story worth following.