Latest / The Productivity Podcast with Fexingo: Output, Efficiency, and Long-Term Economic Growth / The Productivity Cost of Low-Growth Expectations
Transcript
- Lucas: So here's a number I've been chewing on: the U.S. economy's long-run potential growth rate has dropped from about 3.2 percent in the 1990s to roughly 1.8 percent today, according to a 2025 Federal Reserve staff paper. Luna: That's a huge swing. And it's not just about population aging or slower productivity growth in the usual sense. Lucas: Exactly. The paper argues that a big driver is something economists call the 'natural rate of growth' — essentially the speed the economy can sustain without overheating. And when that expected speed drops, it changes behavior in ways that make the slowdown self-reinforcing. Luna: Meaning companies look at a lower growth horizon and decide it's not worth investing in new equipment or R&D at the same pace? Lucas: Precisely. If you're a manufacturer and you believe demand will only grow at 1.8 percent a year on average, you're less likely to spend on a new automated production line, even if the robotics are cheaper than ever. That lower investment itself drags down productivity growth, which further depresses the potential growth rate. Luna: It's like a feedback loop. And this isn't just theoretical — we can see it in the data on capital expenditure as a share of GDP. Lucas: Right. Nonresidential fixed investment — that's spending on factories, machines, software — has averaged about 13 percent of GDP since 2010, compared to over 14 percent in the late 1990s. Doesn't sound massive, but over time, that 1 percentage point gap compounds into trillions of dollars of lost capital stock. Luna: And that lost capital stock means workers have less and less productive equipment to work with, so their output per hour stalls. Lucas: Exactly. Labor productivity growth averaged about 2.2 percent per year from 1995 to 2005. Since 2005, it's been closer to 1.4 percent. That might sound like a small difference, but over a working lifetime of 40 years, it's the difference between doubling your living standards every 32 years versus every 50. Luna: So the Fed paper is basically saying part of that slowdown comes from a shift in expectations itself — not just technology or demographics. Lucas: Right. The authors use a model where firms form expectations about future demand based on recent history. If you experience a period of slow growth, you lower your forecast, and then you invest less, which makes slow growth a reality. They call it 'expectations-driven stagnation.' Luna: That's a pretty sobering frame. It suggests the economy can get stuck in a low-growth rut even if the underlying technology is there. Lucas: And there's historical precedent. The post-World War Two boom wasn't just about pent-up demand or new inventions — it was also about a broad belief that the future would be bigger and better. Companies invested aggressively because they assumed demand would keep growing at 4 or 5 percent a year. Luna: Which then made that growth happen. So the mindset becomes a self-fulfilling prophecy. Lucas: Exactly. But the flip side is also true. If we could shift expectations upward — say, through a sustained period of policy certainty or a visible technology shock — the same mechanism could work in reverse. Luna: Is there any evidence that something like that is happening now? I'm thinking about the AI investment wave. Lucas: It's interesting. ai related capital spending has surged — companies like Microsoft and Amazon are building data centers at a record pace. But the paper's point is that one sector's boom might not be enough to reset economy-wide expectations if the broader demand outlook stays muted. Luna: So you'd need something that lifts the whole baseline, not just a few high-profile bets. Lucas: Right. And the Fed paper is careful to note that expectations are sticky. They don't flip overnight just because of a few headlines. It takes years of consistent outperformance to convince firms that the trend has changed. Luna: Which brings us back to the productivity problem. If investment is low because expectations are low, and expectations are low because growth has been slow, how do you break the cycle? Lucas: The paper doesn't offer easy answers, but it does point to policy that directly targets investment — like investment tax credits or accelerated depreciation — as a way to jump-start the loop from the capital side, even before expectations shift. Luna: That makes sense. You give companies a reason to invest today, even if they're pessimistic about tomorrow. And if that investment actually boosts productivity, it could start to change the narrative. Lucas: Exactly. There's also a role for public investment. Infrastructure spending, for example, not only creates demand in the short run but raises the productive capacity of the economy, which can lift expected growth. Luna: So the core insight is that productivity isn't just a technology problem — it's also a belief problem. Lucas: I think that's a fair summary. And it's actually a somewhat encouraging message, because beliefs can change. It's not like we've run out of ideas or innovation. The potential is still there. Luna: You know, if walking through the economy with us has made something click — or if you just appreciate having a show that digs into these questions without ads — there's a way to support it. Lucas: Yeah. It's the smallest possible ask. If today was worth a coffee to you, that's the link — buy me a coffee dot com slash fexingo. Luna: Honestly, it helps us keep the show independent and focused on what we think matters. Lucas: And back to the loop idea — one concrete thing listeners could watch for is the upcoming revision to the Congressional Budget Office's long-term economic outlook, due later this year. If they raise their estimate of potential growth, that itself could start nudging expectations. Luna: Because the CBO projections are widely cited by businesses and policymakers. Lucas: Exactly. So a small change in a forecast can ripple through investment decisions. That's how powerful expectations can be. And it's why understanding this feedback loop is so important for anyone thinking about long-term productivity.