Romer Growth Model

The Romer growth model, developed by economist Paul Romer in 1990, explains sustained economic growth through endogenous innovation—the idea that a nation’s growth rate depends on its investment in research and development, not just factor accumulation.

Earlier growth models (Solow, 1956) treated technological progress as exogenous—falling from the sky, unexplained. Romer’s insight was that firms rationally invest in R&D to capture rents from new ideas, and those ideas spill over to the broader economy, lifting productivity and wages permanently. In Romer’s framework, growth is endogenous: a government policy that encourages R&D investment (tax credits, IP protection, university funding) directly raises a nation’s growth rate, not just its temporary income level.

The three-sector model

Romer’s canonical model separates the economy into three sectors:

1. Research (R&D): Invents new blueprints/knowledge. Investment of labor and capital yields new product designs, process innovations, or drug patents. These blueprints are non-rival (one firm using it doesn’t stop another), creating a public-good aspect but with partial IP protection (patents, trade secrets).

2. Manufacturing: Uses blueprints and capital to produce consumer goods. This is the traditional production function where labor, capital, and blueprints combine.

3. Human capital development: Education and training. Workers accumulate skills and knowledge, enabling higher productivity and R&D engagement.

Each sector grows at a steady state. If the research sector devotes 5% of the labor force to R&D, the stock of blueprints expands at a constant percentage rate. That growth in blueprints means manufacturing becomes more productive (the same inputs yield more output), raising wages and living standards permanently. Contrast this to Solow: once capital-deepening exhausts itself, Solow growth stalls. Romer growth continues as long as ideas keep flowing.

The knowledge spillover and increasing returns

Romer emphasizes that knowledge has increasing returns: the cost to discover the first antibiotic is $1 billion and a decade; the cost to produce it at scale is $10 per dose. Private R&D firms bear the discovery cost but capture only partial returns (through patent protection). The consumer surplus and broader productivity gains accrue to society—a positive externality.

This spillover is why growth is endogenous. Each new idea makes subsequent innovation easier: calculus unlocked physics, which unlocked electric motors, which unlocked every modern engineering field. A critical mass of researchers and accumulated knowledge accelerates the pace of discovery. A country with 10,000 researchers achieves more than ten countries with 1,000 each, because idea-merging and cross-pollination raise the marginal product of the 10,001st researcher.

Policy implications and the growth-enhancing agenda

Romer’s model justifies substantial government R&D spending and IP protection. Patent laws, university funding, and tax credits for private R&D all raise the return to innovation, incentivizing firms to invest more in blueprints. Countries that under-protect IP or under-fund education will under-invest in R&D and fall into slower, secular stagnation. Conversely, aggressive IP enforcement and human-capital investment unlock higher potential GDP growth.

This reasoning underpins 21st-century industrial policy: government subsidies for semiconductors, EV batteries, and quantum computing are pitched as “investments in the knowledge stock.” They are attempts to move the economy rightward along the Romer growth curve.

Empirical challenges and the scale effect puzzle

Romer’s model predicts a scale effect: larger populations should yield more innovation, higher growth. Yet empirically, the richest countries (large populations, massive R&D budgets) do not grow faster than smaller rich countries. The U.S., with 330M people, grows at ~2% annually; Switzerland, with 8M, also grows near 2%. This has motivated refinements: perhaps the scale of world innovation matters (globalizing R&D), or population size matters for sustained growth only if institutions are strong enough to convert ideas into productivity gains (a governance effect).

A second puzzle is that patents and IP protection have not clearly risen in step with growth, questioning the feedback loop. Some researchers argue Romer’s model works best in manufacturing-heavy economies and underweights the role of institutional quality, rule of law, and financial market development.

Distinction from other endogenous models

Romer’s model is one of several endogenous-growth frameworks. AK models (Rebelo, 1991) assume constant returns to reproducible capital, yielding faster growth. Endogenous-tech models with human capital (Lucas, 1988) emphasize skills accumulation. Romer is distinctive in modeling the research industry explicitly, with profit-driven innovation as the growth engine—making IP policy and R&D incentives the central policy variables.