The US space sector is widely seen as having been transformed over the past two decades by a wave of new commercial entrants – SpaceX, Blue Origin, Rocket Lab, and others – that displaced a stagnant incumbent industry built around cost-plus government contracting. This New Space narrative has become the standard account in policy discussion and popular writing, and it has shaped expectations about how innovation in the sector arose (Weinzierl 2018).

In a recent paper (Gaetani and Whalley 2026), we examine where and when US space-sector innovation surged, and which firms accounted for it. The descriptive evidence reframes the historical narrative and raises questions for current industrial policy debates, in which the space sector is frequently invoked as a model of how private enterprise drives innovation.

Building a measure of space sector innovation

A central obstacle to studying innovation in the space sector has been the lack of a comprehensive, consistent measure of space-related innovative activity over time. Standard patent classifications do not isolate space-related patents cleanly, and prior measures have either relied on patent office classifications or key words.

We address this by constructing several complementary measures of space-related patenting. These range from a narrow classification based on patent classes explicitly tied to space technology, to broader classifications that use large-language-model predictions and text matching to an expert-developed taxonomy of space-sector activities. We validate these measures against patenting by known space organisations and make the resulting data publicly available. With these measures in hand, we can document the timing, composition, and geography of US space-sector patenting at a level of detail that prior data have not permitted.

When the surge happened, and who drove it

The first descriptive fact concerns timing. Our new broader measure shows that the largest sustained surge in US space-sector patenting occurred in the 1990s, more than a decade before the entry of the commercial firms typically credited with transforming the sector. The surge was concentrated among incumbent aerospace firms – Lockheed Martin, Boeing, Hughes, and their peers – rather than among new entrants.

Figure 1 Space-sector patenting by incumbent and entrant firms, 1976–2017

The continuing record is also striking. Throughout the period covered by our data and through to the present, incumbent aerospace firms account for most broadly defined space-related patenting. Entrants contribute a growing but still minority share. The post-2005 emergence of SpaceX and its peers is real, but it has not displaced incumbents as the dominant source of US space-sector innovation.

The surge was not accompanied by a rise in NASA’s own patenting, which remained roughly flat across the period; private firms account for nearly all the increase. Nor was the response confined to incremental innovation. Applying a standard text-based measure of technological novelty (Kelly et al. 2021), we find that the novelty of space-sector patents rose sharply in the same window. Whatever the 1990s patenting reflected, it was not paper pushing.

Figure 2 Technological novelty of space-sector patents, 1976–2014

The geography of the surge is also informative. Space-sector patenting in the 1990s was concentrated in metropolitan areas that had been the centres of Apollo-era investment three decades earlier, the same regions identified in Kantor and Whalley (2025) as the loci of public R&D effects on local innovation, and these regions have remained dominant since.

What the patterns are consistent with

These are descriptive facts, not causal estimates. We cannot, from this evidence alone, conclude what produced the 1990s surge. But the four facts (timing, firm composition, novelty, and geography) are jointly informative about which interpretations of the historical record are plausible.

The standard New Space narrative, which attributes the transformation of the sector to post-2005 commercial entry, is hard to reconcile with the data. The patenting surge predates the entrants by more than a decade and is driven by the firms that the narrative casts as incumbents to be displaced, and the incumbents have continued to dominate space-related patenting since.

The interpretation we find most plausible, considering the historical record, is one of directed technical change (Acemoglu 2002). NASA’s R&D budget was flat to declining across the period, so a story in which more government research funding produced more private patenting does not fit. What changed in the policy environment was on the demand side: the Commercial Space Act of 1998 committed federal agencies to purchasing space-based services from commercial providers; the 1996 Telecommunications Act and parallel reforms of the Federal Communications Commission deregulated satellite communications and opened downstream markets; the civilianisation of GPS created new commercial possibilities; and NASA’s procurement strategy shifted from cost-plus contracting for bespoke hardware toward fixed-price purchasing of commercial services. These policy changes made the relevant downstream markets appropriable for private firms, that is, made it possible for private investors in R&D to capture returns from the resulting innovations.

In this reading, incumbent aerospace firms directed R&D toward the newly appropriable markets that were accessible from their existing capabilities. The post-2005 commercial entrants, SpaceX above all, represent a different kind of contribution: bringing science-based insights into domains requiring new technological paradigms, particularly reusable launch. Both incumbent and entrant innovation are real, but they are different phenomena. The 1990s surge is best understood as directed technical change by capability-rich incumbents in response to policy-created markets, not as a creative-destruction wave led by entrants.

This kind of mechanism – government as constructor of appropriable markets rather than as funder of research – has been examined in other settings with cleaner identification. Kremer et al. (2020) review the advance market commitment literature, which finds that credible commitments to purchase can elicit substantial private innovation investment when the underlying capability exists. Clemens and Rogers (2020) document procurement design shaping the direction of medical innovation during the US Civil War and First World War, with fixed-price procurement producing greater cost-reducing innovation than cost-plus arrangements. The space-sector patterns are consistent with the same family of mechanisms operating on a sectoral scale across a decade-long window.

Alternative interpretations remain on the table. The 1990s also saw broader macroeconomic and technological changes, the end of the Cold War, the early commercial internet, the maturation of microelectronics. These could have contributed through channels other than the space-specific policy shifts. Disentangling these is a task for future work.

Implications for current debates

The findings are descriptive, and we are cautious about strong policy claims. But two implications seem worth taking seriously in current industrial policy debates, in which space is frequently invoked as a model (Juhász et al. 2023, Bloom et al. 2019, Mazzucato 2018).

First, government’s role in constructing appropriable markets may deserve more attention in the industrial policy toolkit than it currently receives. The empirical literature on R&D grants and subsidies is well developed (Howell 2017); the equivalent literature on procurement and demand-side market construction as innovation tools is considerably thinner. The space-sector patterns are consistent with appropriable-market construction having mattered substantially in at least one major sector. Given how widely such instruments are being deployed in current policy – advance market commitments for vaccines, procurement-led decarbonisation, defence-industrial base initiatives – the evidence base deserves to catch up with the practice.

Second, the assumption that policy-driven innovation responses come primarily from new entrants may not generalise. Much of the recent industrial policy revival has framed the goal as fostering entrants and disruption, in the spirit of the creative-destruction tradition celebrated in the 2025 Nobel Prize (Klenow 2025). The empirical literature on innovation by firm type (Akcigit and Kerr 2018) shows incumbents and entrants making distinct contributions. The space-sector evidence suggests that in mature sectors with concentrated capability stocks, established firms may direct R&D toward policy-created markets and account for much of the resulting innovation. Both incumbents and entrants matter, and they do different things. Policy designs that focus exclusively on supporting entrants may under-utilise the larger latent capacity of established firms.

The broader point is that the Old Space versus New Space framing, in which post-2005 commercial entrants transformed a stagnant sector, sits uneasily with the evidence on when, who, and how. A fuller picture of US space-sector innovation will need to engage with the 1990s, with incumbent firms responding to policy-created markets, and with government’s role in constructing the appropriability conditions under which private innovation can flourish.

References

Acemoglu, D (2002), “Directed technical change”, Review of Economic Studies 69(4): 781–809.

Akcigit, U, and W R Kerr (2018), “Growth through heterogeneous innovations”, Journal of Political Economy 126(4): 1374–443.

Bloom, N, J Van Reenen, and H Williams (2019), “A toolkit of policies to promote innovation”, Journal of Economic Perspectives 33(3): 163–84.

Clemens, J, and P Rogers (2020), “Procurement regimes and medical innovation”, VoxEU.org, 10 March.

Gaetani, R, and A T Whalley (2026), “Old Space, New Space: A commercial revolution in innovation?”, NBER Working Paper 35212.

Howell, S T (2017), “Financing innovation: Evidence from R&D grants”, American Economic Review 107(4): 1136–64.

Juhász, R, N Lane, and D Rodrik (2023), “The new economics of industrial policy”, VoxEU.org, 4 December.

Kantor, S, and A Whalley (2025), “Moonshot: Public R&D and growth”, American Economic Review 115(9): 2891–925.

Kelly, B, D Papanikolaou, A Seru, and M Taddy (2021), “Measuring technological innovation over the long run”, American Economic Review: Insights 3(3): 303–20.

Klenow, P (2025), “Sustained growth through creative destruction: Nobel laureates Philippe Aghion and Peter Howitt”, VoxEU.org, 22 October.

Kremer, M, J Levin, and C M Snyder (2020), “Advance market commitments: Insights from theory and experience”, AEA Papers and Proceedings 110: 269–73.

Mazzucato, M (2018), “Mission-oriented innovation policies: challenges and opportunities”, Industrial and Corporate Change 27(5): 803–15.

Weinzierl, M (2018), “Space, the final economic frontier”, Journal of Economic Perspectives 32(2): 173–92.