📈 economics

Diversity in Schumpeterian games

This paper utilizes a population game model of Schumpeterian competition to demonstrate that changes in diversity driven by new products act as a primary force shaping the evolution of the economic system and the process of creative destruction.

Original authors: Fryderyk Falniowski, ElĊbieta Pliś

Published 2026-07-29

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Fryderyk Falniowski, ElĊbieta Pliś

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine the economy not as a cold spreadsheet of numbers, but as a bustling, chaotic ecosystem, much like a giant, ever-changing jungle. In this jungle, companies are the animals, constantly hunting for food (customers) and trying to survive. For decades, a famous economist named Joseph Schumpeter gave us a special way to look at this jungle. He called it "creative destruction." Think of it like a forest fire: sometimes, the old trees must burn down to make room for new, stronger saplings to grow. This isn't just about destruction; it's about how new ideas replace old ones to keep the whole system alive and growing.

But here's the tricky part: what happens when the new sapling isn't too different from the old tree, but also not exactly the same? In the world of evolutionary game theory—a field that uses math to predict how groups of animals (or companies) behave over time—there's a concept called an "Evolutionarily Stable State." Imagine a perfect balance in the jungle where some animals are brave explorers trying new foods, and others are cautious followers sticking to what they know. If this balance is stable, the jungle doesn't collapse into chaos; instead, the explorers and followers learn to live together. This paper asks a fascinating question: How does the "variety" or "diversity" of a new invention change this balance? Does a new product that is wildly different from the old one encourage more companies to take risks, or does it scare them away?

The Great Innovation Game

In this study, the authors, Fryderyk Falniowski and Elżbieta Pliś, set up a mathematical playground to watch how companies decide whether to invent something new or stick to what works. They call this a "Schumpeterian game." Imagine a massive crowd of companies, each holding a card. They have two choices: play the Innovator card (spend money to create a shiny new product) or play the Imitator card (keep selling the old, reliable product).

The authors use a clever way to measure "diversity." Instead of just saying "this is different," they look at the specific features or "attributes" of a product—like its color, weight, or special functions. They ask: How much do customers value these new features compared to the old ones? If a new product has features that customers love but the old one didn't have, it creates a "dissimilarity." The bigger the gap between the old and the new, the more "diversity" is added to the market.

The Magic Balance: The Schumpeterian State

The paper's main discovery is that there is a sweet spot where the economy settles down. They call this the Schumpeterian State. It's a unique, stable balance where both Innovators and Imitators coexist. It's not a world where everyone invents, and it's not a world where everyone copies. It's a mix.

Here is the magic formula they found: The more "diversity" a new product adds (meaning the more it differs from the old one in ways customers actually care about), the more companies decide to become Innovators.

Think of it like a video game. If a new level is just a slightly harder version of the old one, nobody bothers to try it. But if the new level introduces a totally new power-up that players have never seen before, suddenly, everyone wants to try it. The authors show that when an innovation brings a big increase in diversity (a big "dissimilarity"), it acts like a magnet. It encourages more firms to take the risk of inventing, even though inventing costs money.

The Tug-of-War

The authors also looked at the costs. Inventing is expensive (they call this cost $C_I$ ). If the cost is too high, fewer companies will try. But if the new product is "relevant"—meaning it solves a problem or adds value that customers really want—the market value goes up (represented by a parameter called $\alpha$ ).

The paper proves that if the increase in value from the new diversity is high enough, it can outweigh the high cost of invention. In fact, the more valuable the new diversity is, the more firms will jump in. The result is a stable market where a specific percentage of firms are always inventing, and the rest are copying. This percentage isn't random; it's determined by how much the new product differs from the old one and how much it costs to make that difference.

What If We Add a Third Player?

The authors also tested what happens if they add a third strategy to the game: a "Retaliator." This is a company that only invents if it sees someone else inventing, but copies if everyone else is copying. When they ran simulations with this extra player, the perfect, stable balance they found in the two-player game started to wobble.

In this three-player scenario, the stable mix of innovators and imitators still exists, but it's no longer the only place the system wants to go. Depending on how the game starts, the whole system could collapse into a state where nobody innovates, and everyone just copies or retaliates. This suggests that while the "Schumpeterian State" is a very strong attractor in a simple world, adding more complex behaviors can make the outcome less predictable.

The Bottom Line

So, what does this mean for the real world? The paper suggests that diversity isn't just a nice-to-have; it's a driving force. When a new product offers something truly different and valuable, it doesn't just replace the old product; it changes the entire ecosystem. It encourages more companies to take risks and innovate.

The authors don't claim this is a magic bullet that solves all economic problems. They show that in their specific mathematical model, diversity acts as a "spiritus movens" (a moving spirit) that fuels the engine of creative destruction. If you want a dynamic economy where new ideas keep flowing, you need innovations that are distinct and relevant enough to shift the balance of the game. The more distinct the new idea is, the more likely the whole system is to embrace it, creating a vibrant mix of pioneers and followers working side by side.

Technical Summary: Diversity in Schumpeterian Games

Problem Statement
The paper addresses a gap in Schumpeterian economic theory regarding the interplay between innovation, creative destruction, and economic diversity. While innovation is widely recognized as the driver of economic evolution, the conditions under which both old and new technologies coexist in a stable market structure remain under-explored. Specifically, the authors investigate how the change in diversity introduced by a new product influences the evolution of the economic system. The central question is how the dissimilarity between a new, relevant innovation and an existing technology affects the strategic choices of firms (to innovate or imitate) and the resulting stable population structure.

Methodology
The authors construct a theoretical framework combining evolutionary game theory with a specific measure of economic diversity.

Diversity Model: Building on the attribute-based approach of Nehring and Puppe (2002), the paper defines a product or technology as a set of attributes. A "diversity function" $v(D)$ is defined based on the weights of attributes relevant to a continuum of consumers. The "dissimilarity" between a new technology $y'$ and an old technology $y$ is quantified as the difference in their diversity values, specifically the weight of attributes possessed by $y'$ but not $y$ ( $d(y', y)$ ).
Population Game: The economic system is modeled as a population game with a continuum of firms (modeled as the unit interval $I=[0,1]$ $I = [0, 1]$ ). Firms choose between two Schumpeterian strategies:
- Innovation ( $I$ ): Adopting a new, relevant technology $y'$ , incurring a fixed R&D cost $C_I$ .
- Refraining ( $R$ ): Continuing to use the established technology $y$ .
Payoff Structure: The game is a symmetric $2 \times 2$ $2 \times 2$ anti-coordination game. Payoffs depend on the market value of the technology, the cost of innovation, and the parameters $\alpha$ $α$ and $\beta$ $β$ :
- $\alpha$ : Represents the increase in market value due to the new attributes of the innovation (driven by dissimilarity $d(y', y)$ ).
- $\beta$ : Represents the market value retained by non-innovators when the innovation is present (driven by dissimilarity $d(y, y')$ ).
- The payoff matrix accounts for scenarios where both firms innovate, both refrain, or one innovates while the other refrains.
Dynamics: The evolution of the firm population is analyzed using revision protocols (specifically pairwise proportional imitation), leading to replicator dynamics. The authors seek an Evolutionarily Stable State (ESS) that is globally attracting.

Key Contributions

Formalization of the Schumpeterian State: The paper defines a unique, globally attracting, interior Evolutionarily Stable State (ESS) called the "Schumpeterian state." In this state, innovators and imitators coexist in fixed proportions, representing a stable market structure where creative destruction does not eliminate the old technology entirely.
Linking Diversity to Strategic Choice: The authors mathematically demonstrate that the rate of innovators in the stable state is directly dependent on the parameters $\alpha$ and $\beta$ , which are functions of the product's dissimilarity (diversity change).
Extension of Diversity Metrics: The paper adapts the Nehring and Puppe diversity function for use in a population game with a continuum of agents, allowing for the measurement of diversity based on consumer valuation of product attributes.

Results
The analysis yields several specific findings regarding the relationship between diversity and market structure:

Existence of a Unique Stable State: Under specific conditions on the payoffs (Condition 23), the game possesses a unique interior ESS ( $x^*$ ). In this state, the fraction of innovators ( $x^*_I$ ) is given by:
$x^*_I = \frac{V + 2(\alpha - \beta - C_I)}{\alpha + \beta}$
where $V$ is the original market value and $C_I$ is the innovation cost.
Impact of Diversity Increase (Corollary 2): The rate of innovators increases as the increase in diversity caused by the innovation grows. Specifically, if the innovation provides a greater increase in market value ( $\alpha$ ) relative to the loss of old attributes ( $\beta$ ), more firms choose to innovate.
Population Payoff (Corollary 3): The total payoff for the population depends on the relationship between $\beta$ and $V/2$ . If the imitators' share of the market with innovation ( $\beta$ ) is sufficiently large ( $\beta > V/2$ ), an increase in the fraction of innovators leads to a higher overall population payoff.
Robustness and Extensions: The paper notes that while the 2-strategy game yields a globally attracting ESS, introducing a third strategy (a "retaliator") can destabilize global attraction, leading to convergence on states with no innovators depending on initial conditions.

Significance and Claims
The paper claims that diversity change is a fundamental "spiritus movens" (moving spirit) of innovation. The authors argue that:

Relevance of Innovation: The "relevance" of an innovation is not just about its novelty but its ability to increase economic diversity (dissimilarity) in a way that consumers value.
Coexistence Mechanism: Creative destruction does not necessarily lead to the total elimination of old technologies. Instead, the specific properties of the new product (its diversity relative to the old) determine a stable equilibrium where both innovators and imitators coexist.
Policy and Strategic Implications: The findings suggest that high costs of innovation ( $C_I$ ) can be offset if the innovation significantly increases diversity (high $\alpha$ ). This provides a theoretical basis for understanding why certain innovations (e.g., eco-innovations) succeed in creating new market niches without completely destroying existing ones.
Integration into Growth Models: The authors propose that product diversity should be explicitly incorporated into Schumpeterian growth models as a determinant of the rate of innovation and the structure of the economic system.

The paper concludes modestly, noting that while the model focuses on a specific 2-strategy interaction, it provides a foundational link between the relevance of innovation and economic diversity, suggesting avenues for future work on market entry, externalities, and endogenous attribute generation.