Innovation And R And D

Innovation is the part of economic life where firms deliberately try to make their own markets obsolete.

7 frameworks, from 1942 to today · Industrial Organization · Economics

still a working view ended · each leads to its entry below

Overview

Innovation is the part of economic life where firms deliberately try to make their own markets obsolete. A new product can create a market that did not exist; a new process can let one firm undercut every rival; a new business model can rearrange who holds power in an industry. Industrial organization (IO)—the branch of economics that studies how firms and markets actually work—has therefore treated innovation as both a source of competition and a threat to it. The subfield of innovation and R&D in IO asks a set of linked questions: why do firms spend on research when the results are uncertain and easily copied? Does market power encourage innovation or smother it? How do patents, secrecy, and lead time shape who captures the gains? And how should competition authorities treat a merger or a dominant firm when the relevant market is being transformed from within?

What makes the field distinctive is that its central object—knowledge—behaves unlike ordinary goods. It is costly to produce but cheap to reuse, hard to keep inside a firm, and often valuable only in combination with other knowledge. Much of the history below is the story of economists learning to model that oddity, and then learning to measure it.

Schumpeter and the Creative Destruction of Market Structure

The field's founding provocation came from Joseph Schumpeter1. In Capitalism, Socialism and Democracy (1942) he argued that the relevant competition in modern capitalism is not price competition among many small firms but the competition from a new product, technique, or organization that threatens existing firms' very existence—"creative destruction."2 Schumpeter also suggested, in a famous and much-debated passage, that large firms with some market power may be the main engine of technical progress, since they can fund research and appropriate its rewards.

This posed a sharp question for IO: the standard structure–conduct–performance tradition held that concentrated markets produced poor outcomes, yet Schumpeter implied concentration might buy faster innovation. The empirical answer came in stages. In the 1960s and 1970s, economists such as Jacob Schmookler and later studies of patents and R&D spending found only weak or ambiguous correlations between firm size and inventive output—large firms did more R&D in absolute terms, but not clearly more per dollar or per employee. The Schumpeterian hypothesis survived as a set of distinct claims (about firm size, about market power, about the importance of new entrants) that empirical work has treated separately ever since.

Demand-Pull, Technology-Push, and the Turn to Empirical R&D Studies

A second strand asked where the demand for invention comes from. Schmookler's Invention and Economic Growth (1966) argued, using historical patent data across industries such as railroads and agriculture, that inventive activity responds to expected profitability—demand pulls invention—rather than invention arriving exogenously and pulling growth. This "demand-pull" view competed with a "technology-push" view in which scientific opportunity drives innovation independently of market conditions. The debate mattered because it changed what IO researchers looked for: not just the supply of scientists and labs but the market signals that direct them.

The same period produced the first systematic R&D statistics. The U.S. National Science Foundation began collecting industrial R&D data in the 1950s, and economists used these surveys to study how R&D spending varied with firm size, industry, and expected returns. These empirical studies established the basic regularities—R&D is concentrated in a minority of firms, is persistent within firms, and responds to appropriability conditions—that later theory would try to explain.

Appropriability: Why Knowledge Is Hard to Own

The pivotal theoretical move was to treat knowledge as a partially public good3. Kenneth Arrow's 1962 essay on economic welfare and the allocation of resources for invention made the case crisply: information is costly to produce but, once produced, can be reproduced at near-zero marginal cost, and it is difficult to exclude others from using it. Left to itself, a market would therefore underinvest in invention relative to the social optimum.

This created the field's central trade-off. Patents and other intellectual property rights solve the underinvestment problem by granting temporary exclusivity, but that exclusivity creates deadweight loss—prices above marginal cost—and can block follow-on innovation. Richard Nelson (1959) and Arrow framed the problem; later work by economists such as Edwin Mansfield measured how firms actually protect innovations, finding that for many industries patents were less important than lead time, learning-curve advantages, and secrecy. The "appropriability regime"—how easily an innovator can capture the returns—became a standard variable in the analysis, and it explained why innovation rates differ so much across industries even when R&D spending is similar.

Patent Races and Preemption: Innovation as a Game

From the late 1960s onward, IO theorists began modeling innovation as a race. The key insight, developed in work by William Fellner, Morton Kamien and Nancy Schwartz, and later Glenn Loury and others, is that the reward for invention is often "winner-take-all": the first firm to patent or to market gets the prize, while losers recover little. This turns R&D into a strategic game with preemption—firms invest not only for the expected return but to deny rivals the prize.

The game-theoretic literature produced sharp and sometimes counterintuitive results. Under certain conditions, competition in R&D can lead to overinvestment relative to the social optimum, because each firm's private race duplicates effort that a coordinated planner would avoid. The "patent race" model also showed how the structure of the prize (a patent of fixed length, a market lead, a standard) shapes investment. These models gave IO a language for analyzing rivalrous innovation that pure Schumpeterian reasoning lacked, and they connected directly to antitrust questions about whether incumbent firms can preempt entrants by patenting around them.

Endogenous Growth and Knowledge Spillovers

In the 1980s and 1990s, the innovation question migrated into macroeconomics. Paul Romer's endogenous growth models (beginning in the mid-1980s) treated knowledge as a nonrival input that could generate increasing returns, so that investment in R&D could sustain long-run growth rather than merely producing a one-time level effect. This was a major shift: innovation became central to growth theory, not a residual.

For IO, the crucial concept was the spillover—the uncompensated transmission of knowledge from one firm or region to another4. Empirical work by economists such as Zvi Griliches and later Adam Jaffe, Manuel Trajtenberg, and Rebecca Henderson used patent citations to trace spillovers, finding that firms' R&D productivity depends on the research done by others, especially those geographically or technologically close. The spillover concept explained why firms cluster (Silicon Valley, biotech hubs) and why public policy might justify subsidizing R&D even in competitive markets. It also complicated the appropriability story: spillovers are a leakage that weakens private incentives but a benefit that raises social returns.

Escape Competition and Distance to the Frontier

A major synthesis came from Philippe Aghion and Peter Howitt's Endogenous Growth Theory (1998), which built Schumpeterian growth models in which innovation is driven by the prospect of temporary monopoly profits. Their "escape competition" result—that firms facing more competition may innovate more, because they need to escape rivals' pressure—challenged the simple Schumpeterian intuition that market power is good for innovation.

The resolution came through the "distance to the frontier" idea, developed by Aghion, Howitt, and later coauthors including Rachel Griffith and Fabrizio Zilibotti. For firms already near the technological frontier, competition can spur innovation because they can only survive by innovating; for firms far behind the frontier, competition may discourage innovation because they earn enough from imitation or from protected niches. This framework helped reconcile conflicting empirical findings across countries and industries, and it became a standard lens for thinking about how competition policy affects innovation in different settings—a question with direct antitrust consequences.

Structural Empirical IO of R&D

The most recent turn is methodological. Following the broader "empirical revolution" in IO, researchers have moved from reduced-form correlations to structural estimation: building explicit models of firm behavior—dynamic investment in R&D, entry and exit, patenting—and estimating their parameters from data, then using the estimated model to simulate counterfactuals such as a merger, a change in patent length, or an R&D subsidy.

Work in this vein, associated with economists such as Ariel Pakes and later many others, treats R&D as a dynamic investment problem with uncertainty and sunk costs. It allows researchers to ask questions that reduced-form data cannot answer: how much would a merger reduce innovation? What is the social return to a subsidy? How do firms' expectations about future competition shape current research? The approach is demanding—it requires rich data and strong modeling assumptions—but it has become the frontier for policy-relevant questions, especially in merger review and the evaluation of innovation policy.

The Present: Open Questions

Today the field is defined by several live tensions. The first is measurement: R&D spending is an input, not an output, and economists still struggle to measure innovation itself—patents are noisy, and much innovation (process improvements, organizational change, software) leaves little trace. The second is the effect of market structure: the Schumpeterian trade-off between static competition and dynamic innovation remains unresolved, and the rise of dominant digital platforms has made it urgent again. The third is appropriability in a world of complex, cumulative technologies, where a single product may infringe dozens of patents and where "patent thickets" can slow rather than spur innovation. The fourth is the geography of innovation: spillovers are local, but research is increasingly global, and the policy implications are unsettled.

What unites the field is a commitment to treating innovation as an economic activity with identifiable costs, rewards, and strategic interactions—not as an exogenous gift. The history above is the record of IO learning to model that activity, then to measure it, and now to estimate its consequences with enough precision to inform competition policy and innovation policy alike.

References

  1. Innovation (Economics and innovation, 20th and 21st centuries), Wikipedia
  2. Creative destruction, Wikipedia
  3. Internalization theory, Wikipedia
  4. Knowledge spillover, Wikipedia

The frameworks of Innovation And R And D

Each framework is a way of working the field has used: when it held, and the question it answers.

  1. 1942-Present

    Creative destruction

    How does innovation transform market structure?

    Fly to its tower in Nousfarer

  2. 1959-Present
  3. 1966-1985
  4. 1968-Present
  5. 1986-Present

    Endogenous growth theory

    What sustains long-run economic growth?

    Fly to its tower in Nousfarer

  6. 1990-Present
  7. 1998-Present

How the ideas connect

Each arrow runs from an earlier idea to the later one it shaped. The list below says what passed between them.

Creative destructionAppropriability and the Public-Good T…Demand-Pull versus Technology-PushPatent Race and Preemption ModelsEndogenous growth theoryStructural Empirical IO of R&DEscape Competition and Distance to th…

Beyond the field

Where Innovation And R And D's ideas came from in other fields, and what they led to there: the strongest 12 of 14 links.

A city in Nousfarer: a red-carpeted stair climbing terraces to a crowned tower on an acropolis, more towers round it, forest and snowy mountains beyond.
A city in Nousfarer, the game played in the browser.

Innovation And R And D is a planet in Nousfarer

Its city has a tower for each of these 7 frameworks. Land, walk in, and climb one concept at a time.

Land on Innovation And R And D

Related fields

Fields its ideas are linked to

More in Industrial Organization

See where Innovation And R And D sits on the Map of Human Knowledge