Research Highlights Podcast
November 5, 2025
Technological spillovers
Shawn Kantor and Alexander Whalley discuss the long-term effects of public R&D investment made during the Cold War–era space race.
NASA engineers discuss the Apollo project.
Source: NASA Glenn Research Center, Public domain
òòò½Íø
The launch of Sputnik by the Soviet Union in October 1957 led to a geopolitical crisis that reshaped American science policy. Within months, Congress established NASA, and by 1961, President Kennedy committed the nation to landing a man on the moon before the decade's end. The resulting investment was massive, and the program still serves as a model of government spending for advocates of public R&D.
In a paper in the òòò½Íø Review, authors and question whether the space race program succeeded as an economic policy that boosted economic growth and productivity.
To estimate the space program's effects on economic growth from 1947 to 1992, the authors used data on NASA contractor spending and a novel identification strategy based on declassified CIA documents that allowed them to determine which US industries in which counties specialized in space-relevant technologies before the space race began. Their findings complicate the conventional narrative about public R&D and provide important context for current proposals to replicate so-called “moonshot” models in other domains.
Kantor and Whalley recently spoke with Tyler Smith about the local effects of space race spending and why they didn’t translate into long-term productivity gains.
The edited highlights of that conversation are below, and the full interview can be heard using the podcast player.
Tyler Smith: In this paper, you're studying the knock-on effects of the Cold War–era space race program. Why did you want to explore this period of history?
Shawn Kantor: The space race is an iconic moment, so politicians and economists often use it. They think about that particular era as a time when public investment in R&D had amazing growth potential, stimulated economic activity, and led to all these great spillover effects that presumably we should be able to measure and know about. So, in terms of our overall research, we figured the space race seemed to be a nice place to look. Nobody's looked at it. Just measuring the returns to R&D seemed to be a vexing problem that economists are still facing. So we thought we would just take a stab at it.
Smith: How did you all figure out whether NASA spending actually caused economic growth versus just going to places where there was already capacity for growth?
Alexander Whalley: One of the concerns is that NASA could harvest effective technologies that are already in the United States and utilize those. The moon mission and R&D spending is not random. NASA was trying to win the race. They're looking for where they're going to have the biggest impact. We really tried to go back and measure which areas of the United States already had some kind of comparative advantage in space technology. It was challenging because there really wasn't space technology in 1958. How do you think about what that actually is? We used these historical files of the research the CIA was doing on Soviet space technologies. We used that to define what space technology was, the kind of things the Soviets were actually doing. We looked at that in the patent record in the United States and tried to figure out which areas and which locations were doing space technology before this started. That was really the essence of our research design.
Smith: Can you elaborate on that a little bit? Why would Soviet space technology help you estimate the impact of NASA spending?
Kantor: In terms of doing the causal identification, the nice feature of the paper is that it's looking at post-Sputnik Soviet technology as researched by the CIA. So we got the declassified CIA documents regarding Soviet space technology past 1958. And then we used that information to try to understand which American companies were doing that sort of work before Sputnik. And so we didn't want to use what NASA was doing because that could be correlated with NASA's harvesting of technologies, putting money where there was already potentially some opportunities in the American economy. We didn't want that. What we wanted is a situation where we had a true picture of what the demands were for getting into space and then looking at who was doing that research before Sputnik.
Smith: Once you had this estimation strategy in place, what did you find happened when NASA spending started flowing into large urban counties with space-related industries? What happened to the local economy?
Kantor: Not surprisingly, the companies that were getting this largesse grew substantially, like on the order of 35 to 50 percent in terms of employment and value added capital stock. They're just expanding rapidly relative to other counties and industries that didn't have that technology in place. So the companies that you would expect to be the beneficiaries of NASA spending were certainly expanding. But the critical thing that we found was that those benefits that accrued to those counties and space industries didn't spill over to other industries in the same county or to neighboring counties. So it was really heavily localized to those beneficiaries.
Smith: You calculate something called a fiscal multiplier. What is the fiscal multiplier in the context of your study and how big was it?
Whalley: We have this increase in output. Firms are getting bigger; they're hiring more workers. But how big is that effect and what would you compare it to? One thing that's really nice is that there are lots of estimates in the economic literature about different government spending programs. We have lots of estimates of how big the bang for the buck is there. Basically if you spend a dollar on an industry and location, how much does output increase? Does it increase by more than a dollar, which kind of indicates there's these positive spillover effects? Or does it increase by less than a dollar, which indicates it really isn't stimulating the economy tremendously. What we find is impacts that are less than one. We find a fiscal multiplier of about 0.3. So if you spend a dollar you get about $0.30 extra in output which really doesn't fit with the story of public R&D generating massive economic growth. We have this idea that there's these externalities from ideas. Ideas kind of make an economy more productive, so you can grow much faster. We don't really find that. We also don't find in the data much evidence for productivity increasing. Size gets bigger, firms hire more workers, they produce more output, but they don't really become more productive. We're not really seeing these kinds of big technological spillovers from the space race.
The space race was a very specific historical episode where this engineering feat cobbled together ideas that were already existing. It just sped up the development of those ideas, but it didn't necessarily produce new things, new advances in knowledge.
Shawn Kantor
Smith: You also looked at the long-run productivity gains. Did you find an absence of gains in that area as well?
Whalley: The long-run effects are not especially larger. Sometimes people talk about a general purpose technology, such as AI right now, or, historically, computers. It takes time for firms to learn how to use that. There's complementary innovations that have to happen. So you would expect maybe the longer-term effects would be larger than those short-term effects. But we don't really find that. The space race ends obviously with going to the moon and returning in 1969. But we studied the data up until 1992. We don't really see much bigger effects in the late 80s and early 90s than we do in the 60s and 70s.
Smith: There are still politicians and public figures who call for new moonshot programs like those coming out of the space race. What are reasonable expectations for what these sorts of investments can deliver economically?
Kantor: I think the lesson from the space race is that the devil's in the details. What kind of R&D program policy are we implementing? Is it one that's going to advance basic science and our fundamental knowledge about something, whatever it happens to be? Or is it effectively an industrial policy where we're just shifting rents to some particular favored industry or company? We might not expect the spillover benefits in that case to be great, as opposed to funding fundamental science and fundamental knowledge. I think the space race is more in that applied industrial policy framework. I think that's why we're not really seeing big spillovers. It was a very specific historical episode where this engineering feat cobbled together ideas that were already existing. It just sped up the development of those ideas, but it didn't necessarily produce new things, new advances in knowledge.
Whalley: For the general public thinking about mission-oriented policy, you want to think about the value of completing the mission. How much do you value sending someone to the moon and returning them? That should be how you think about the space program. I think that should be how you think about these other specific mission-oriented policies, rather than just general spillovers that go everywhere.
♦
“Moonshot: Public R&D and Growth” appears in the September 2025 issue of the òòò½Íø Review. Music in the audio is by .