Endogenous Technological Change
The final part of the text develops models where technological progress is the result of purposeful, profit-driven R&D investment, making the long-run growth rate endogenous.
Foundations
- Non-Rivalry of Ideas: Technology (as blueprints or ideas) is a non-rival good: its use by one firm does not prevent its use by another. This inherent property leads to increasing returns to scale at the economy-wide level.
- Ex-Post Monopoly Power: To incentivize private firms to incur the costs of R&D, they must be able to earn profits from their innovations. This is typically modeled by granting innovators perpetual patents, leading to monopolistic competition in the market for new goods (using the Dixit-Stiglitz framework).
- Schumpeterian Creative Destruction: In models of competitive innovation, growth is a process of “creative destruction,” where new, better technologies replace existing ones, and innovator firms displace incumbents. This creates a conflict of interest, as incumbent monopolists have an incentive to block new technologies that would destroy their rents.
Key Models of Endogenous Growth
| Model Type | Engine of Growth | Key Mechanism | Implications |
| Expanding Variety | Creation of new intermediate goods (machines). | R&D expands the number of specialized inputs (N), which increases total factor productivity. | Equilibrium growth is typically below the social optimum due to knowledge spillovers and business-stealing effects. Exhibits “scale effects” where growth depends on population size (L). |
| Competitive Innovation (Schumpeterian) | Quality improvements in existing goods (“creative destruction”). | Firms innovate to replace incumbents and capture monopoly rents. Growth is a “step-by-step” process up a quality ladder. | Market structure is endogenous. The relationship between competition and growth can be non-monotonic. Creates political-economy conflicts between incumbents and entrants. |
| Directed Technical Change | Innovation is directed towards specific factors of production. | R&D is allocated to developing technologies that complement either skilled labor (H) or unskilled labor (L), depending on relative profitability. | Explains skill-biased technical change as a response to an increased supply of skilled labor (market size effect dominates price effect). Provides a rationale for why long-run growth is labor-augmenting (Harrod-neutral), as capital accumulation makes labor the relatively scarce factor. |
These models represent a major step forward by opening the “black box” of technological progress. They demonstrate how policies affecting profit incentives (e.g., R&D subsidies, patent law, competition policy) and factor supplies can have a direct impact on the long-run growth rate of an economy.