The Dynamic Theory of Economic Complexity

arXiv:2512.10672 · econ.GN, q-fin.EC · Submitted 2025-12-11 · Read on arXiv

Listen

Radio episode about this paper

Transcript

Introduction to the show: ident: Economics. The day in economics: theory, mechanisms and the evidence beneath them.

Ruth: We're looking at the paper: "The Dynamic Theory of Economic Complexity".

Warren: Here's the gist of the paper. Capability accumulation and conditional convergence: Towards a dynamic theory of economic complexity develops an endogenous model that generates a transition between unconditional and conditional convergence as capability intensity rises Model Foundation The paper develops a dynamic model of economic complexity that endogenously generates a transition between unconditional and conditional convergence <ref:2512.10672#pg3> This model assumes economies reinvest part of their output into the activities that use those capabilities, inducing a Riccati-type dynamical system with a boundary separating phases of unconditional and conditional convergence <ref:2512.10672#pg3> The model predicts Solow-style convergence when activities require few capabilities and conditional convergence when capability requirements increase <ref:2512.10672#pg3> Additionally, this model provides a theoretical foundation for the principle of relatedness, suggesting that economies are more likely to enter activities similar to those they are already specialized in <ref:2512.10672#pg3> Weak Links and Convergence Dynamics Economic complexity models relate to weak-link models of development by emphasizing the importance of complementarities and/or linkages between inputs <ref:2512.10672#pg4> The model's production function is constructed by multiplying terms representing the probability an economy has each of the required capabilities, which can be expressed in closed form as 1 minus the probability of lacking the capability (1 − qpb(1 − rcb)) <ref:2512.10672#pg6> Convergence is observed when less developed economies grow faster than more developed ones, eventually catching up to them <ref:2512.10672#pg4> The growth rate depends on the availability of capabilities as an inverted parabola with a tunable peak, allowing the model to explain the transition from a convergence to a conditional convergence regime <ref:2512.10672#pg5> Phase Transition and Critical Thresholds The position of the peak growth rate depends on the intensity which activities require inputs (the complexity of the activities), and when activities are “simple” (not intense in the requirement of many capabilities) the model predicts unconditional convergence <ref:2512.10672#pg5> As the intensity with which activities require capabilities increases, the peak moves towards the “right,” crossing the origin and leading to a second order phase transition that starts the gradual onset of conditional convergence <ref:2512.10672#pg5> The critical value for an interior peak of growth rate (and hence conditional convergence) arises when investment dominates depreciation, requiring hqi > γ/(2γ − δ) <ref:2512.10672#pg11> Principle of Relatedness The model provides a theoretical explanation for the principle of relatedness, showing that the rate of accumulation of a capability will increase more rapidly in economies where other complementary capabilities are available <ref:2512.10672#pg18> This effect grows with the complementarity of capabilities, providing a deductive theoretical foundation for estimating relatedness <ref:2512.10672#pg19> The derived expression shows that the rate of increase of a capability increases with the presence of others and grows faster when this is complementary to others <ref:2512.10672#pg19> Kinematics and Speed of Convergence The time scale of the model, representing the speed of convergence, depends on the complementarity of capabilities Ecpb and their intensity of use Qpb <ref:2512.10672#pg15> This means that when capabilities are intensely required, EcpbQpb accumulation is faster for economies that are better endowed with them <ref:2512.10672#pg15> The solution to the Riccati equation shows that the speed of convergence depends on the complementarity of capabilities and their intensity of use <ref:2512.10672#pg15> The model provides a simple and parsimonious way to recover historical behaviors, such as the divergence experienced by Rhode Island due to complex production processes <ref:2512.10672#pg20> Extension to Multiple Capabilities The general form still involves a Riccati equation even for multiple capabilities, which is solved by focusing on the case where depreciation is small compared to investment <ref:2512.10672#pg14> The solution shows that the time scale of the model depends on the complementarity of capabilities and their intensity of use <ref:2512.10672#pg15> The final expression for capability accumulation shows that growth depends on both activities that do not heavily rely on it and those that do depend on its availability <ref:2512.

Paper summary: Ruth: So we're looking at this paper, "The Dynamic Theory of Economic Complexity," which is trying to build a dynamic model for how economic complexity evolves. The main idea here is that it endogenously generates a transition between unconditional and conditional convergence as the intensity of the capabilities rises.

Warren: That sounds like it's proposing some sort of boundary in economic development where things change from one pattern to another. What does this actually claim about how economies move over time?

Iris: It assumes that economies reinvest part of their output into the activities that use those capabilities, which triggers a Riccati-type dynamical system with a boundary separating phases of unconditional and conditional convergence. That's the core mechanism they are using to describe the process, right?

Ruth: Exactly. They predict Solow-style convergence when activities require only a few capabilities, but they see conditional convergence emerge when the capability requirements increase. So it’s not just one simple path for development anymore.

Warren: And this model also lays some groundwork for what we call the principle of relatedness, suggesting that economies are more likely to enter activities similar to what they're already specialized in. That's a big thing because it explains why certain economic links stick around.

Iris: The paper argues that this model provides a theoretical foundation for that relatedness idea, showing how the rate of accumulation for one capability gets faster when other complementary capabilities are available. It’s based on the complementarity of those capabilities eighteen nineteen.

Ruth: But they also get into how fast this convergence happens through kinematics. The speed depends on the complementarity of capabilities and their intensity of use. So if activities really demand many inputs, that changes how quickly things accumulate for economies with good endowments.

Warren: And if you look at the transition itself, it’s not a simple switch. They find that the peak growth rate depends on the intensity of what activities require as input.

Iris: Right, and when those activities are simple—not intense in their requirement for many capabilities—the model predicts unconditional convergence. But as the intensity with which they need capabilities increases, that peak moves to the right and crosses zero, which starts this gradual onset of conditional convergence through a second order phase transition five.

Paper summary: Ruth: That sounds like a way to frame the conditionality not just as an arbitrary threshold, but as a structural shift in how activity requirements are set. It’s about the complexity driving the change.

Warren: And to find that critical value for when you get that interior peak of growth rate—which is where conditional convergence kicks in—they say it happens when investment dominates depreciation, which requires some specific relationship between hqi and gamma and delta eleven.

Iris: That brings us to the principle of relatedness again. The paper shows that the rate of increase for a capability grows faster when other complementary capabilities are present, which gives them a deductive way to estimate that relatedness effect nineteen.

Ruth: So, if you look at how fast things accumulate, it's not just about having one capability; it's about the structure of all the other capabilities around it. That’s where the complexity comes in.

Warren: And when we look at the speed of convergence, that time scale is tied to those specific values—the complementarity and the intensity of use—so better endowed economies accumulate faster when activities are intensely required fifteen <ref:2512.10672#pg2>.

Iris: The paper also looks at extending this to multiple capabilities, where they still use a Riccati equation but focus on the case where depreciation is small compared to investment fourteen <ref:2512.10672#pg2>. This means the speed of accumulation depends on those same factors we just talked about.

Ruth: It's interesting how they link these different parts together: the dynamic system, the phase transition, and the relatedness effect all feeding into one unified structure describing economic complexity.

Warren: And to make it practical, they point out that this model can recover historical behaviors, like the divergence seen in places such as Rhode Island due to complex production processes twenty <ref:2512.10672#pg2>. It gives a way to see those real-world patterns through the lens of capability accumulation.

Iris: One thing I want to point out is their limitation. They state that they solve the Riccati equation by focusing on cases where depreciation is small compared to investment, which simplifies things but might not capture scenarios with high depreciation rates fourteen <ref:2512.10672#pg2>.

Ruth: That makes sense. So, while the model gives us a powerful framework for understanding how complexity drives the move between convergence types and relatedness, it relies on that simplification regarding depreciation.

Paper summary: Warren: What this means for the real world is that we need to look at activities not just in terms of what capability they use, but also in how intensely they require those capabilities as input. That's a crucial operational detail.

Iris: So, the paper suggests that understanding economic development isn't just about tracking whether poor countries catch up to rich ones; it’s about modeling the structural dependence and complementarity within economies two <ref:2512.10672#pg1>. It moves beyond simple capital accumulation models by endogenizing capability building.

Ruth: And if you think about this for a moment, it implies that policies or market structures that encourage specialization in certain areas might trigger a shift from one type of convergence to another based on how much those activities rely on each other.

Warren: It means the speed at which an economy gets there isn't just determined by how much capital it has, but by the intricate network of what it can do with that capital. That’s where the quantitative side gets interesting for me, because we need to model that network structure.

Iris: So, to summarize what they are doing with "The Dynamic Theory of Economic Complexity," they are proposing a dynamic model where an economy's ability to accumulate capabilities drives a transition between unconditional and conditional convergence based on the intensity of activity requirements.

Ruth: And they provide a formal explanation for the principle of relatedness by showing how having complementary capabilities accelerates accumulation, which is tied to how fast things happen through the kinematics of capability use.

Warren: The big picture seems to be that economic complexity isn't just an outcome; it's a dynamic process with specific rules governing when and how development paths change.

Iris: And the model’s limitation is that by focusing on cases where depreciation is small relative to investment, it might miss dynamics where depreciation plays a more significant role in slowing down the accumulation process fourteen <ref:2512.10672#pg2>.

Ruth: It gives us a way to see historical economic divergence, like Rhode Island's experience, not as random events but as outcomes of these underlying structural dynamics.

Warren: So what this implies for how we think about growth, it’s that simply looking at averages might miss the critical structural shifts happening when activity requirements change their complexity level.

Conclusion: Ruth: So we've been looking at this paper, "The Dynamic Theory of Economic Complexity," which uses a dynamic model to show how economies move between different kinds of growth patterns based on how complex their activities are.

Warren: It’s about this endogenous model generating that switch from unconditional to conditional convergence as the capability intensity goes up. That sounds like it sets up a boundary in development.

Ruth: Exactly, it assumes economies reinvest output into the stuff that uses those capabilities, which creates this Riccati system with two phases of growth. It predicts simple convergence when activities need few things and conditional convergence when they need more.

Iris: And the model suggests that by looking at how many capabilities you have, you can actually explain why economies tend to move towards activities they are already good at because of relatedness.

Warren: That relatedness part is interesting because it gives a theoretical reason for why certain economic links stick around over time. It's not just random chance.

Ruth: Right, and the speed of this convergence depends on both the nature of those capabilities and how intensely they are used in activities, which is a bit more nuanced than just looking at total capital stock.

Iris: The transition point itself is where things get tricky; it’s a second order phase transition that starts conditional convergence when the intensity of capability requirements increases beyond a certain level.

Warren: So, if you're listening just to the show, this means development isn't one straight line anymore; it has these structural turning points based on how demanding the economic activities are.

Ruth: Precisely; and we need to look at those specific structural shifts when we try to understand long-term economic trajectories.

Cásar A. Hidalgo, Viktor Stojkoski

Center for Collective Learning, IAST · Center for Collective Learning, CIAS · Alliance Manchester Business School, University of Manchester · Ss. Cyril and Methodius University in Skopje

econ.GN, q-fin.EC

Submitted: 2025-12-11

Updated: 2026-11-03

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Importance score: 77/100

The gist: Capability accumulation and conditional convergence: Towards a dynamic theory of economic complexity develops an endogenous model that generates a transition between unconditional and conditional…

Key concepts

Unconditional Convergence
This occurs when an economy converges to a steady state regardless of its initial level of development. The model predicts this happens when the economic activities requiring capabilities are simple, meaning they do not depend heavily on many different types of skills or knowledge.
Conditional Convergence
This is a convergence that depends on the initial conditions and the specific requirements of the activities. It arises when activities demand an increasing number of capabilities, causing economies to converge only if they possess those necessary skills.
Principle of Relatedness
This principle suggests that economies are more likely to enter activities similar to those they already specialize in. The model shows that the rate at which a capability accumulates increases significantly when other complementary capabilities are also available, reinforcing specialization.
Capability Intensity
This refers to how much an economic activity requires specific inputs or capabilities. As this intensity increases, it shifts the system toward conditional convergence, meaning economies must acquire more complex skills to participate effectively.

Terminology

Summary

Capability accumulation and conditional convergence: Towards a dynamic theory of economic complexity develops an endogenous model that generates a transition between unconditional and conditional convergence as capability intensity rises

Model Foundation

The paper develops a dynamic model of economic complexity that endogenously generates a transition between unconditional and conditional convergence <ref:2512.10672#pg3> This model assumes economies reinvest part of their output into the activities that use those capabilities, inducing a Riccati-type dynamical system with a boundary separating phases of unconditional and conditional convergence <ref:2512. p. 110672#pg3> The model predicts Solow-style convergence when activities require few capabilities and conditional convergence when capability requirements increase <ref:2512. p. 110672#pg3> Additionally, this model provides a theoretical foundation for the principle of relatedness, suggesting that economies are more likely to enter activities similar to those they are already specialized in <ref:2512. p. 110672#pg3>

Weak Links and Convergence Dynamics

Economic complexity models relate to weak-link models of development by emphasizing the importance of complementarities and/or linkages between inputs <ref:2512.10672#pg4> The model's production function is constructed by multiplying terms representing the probability an economy has each of the required capabilities, which can be expressed in closed form as 1 minus the probability of lacking the capability (1 − qpb(1 − rcb)) <ref:2512.10672#pg6> Convergence is observed when less developed economies grow faster than more developed ones, eventually catching up to them <ref:2512. p. 210672#pg4> The growth rate depends on the availability of capabilities as an inverted parabola with a tunable peak, allowing the model to explain the transition from a convergence to a conditional convergence regime <ref:2512. p. 310672#pg5>

Phase Transition and Critical Thresholds

The position of the peak growth rate depends on the intensity which activities require inputs (the complexity of the activities), and when activities are “simple” (not intense in the requirement of many capabilities) the model predicts unconditional convergence <ref:2512.10672#pg5> As the intensity with which activities require capabilities increases, the peak moves towards the “right,” crossing the origin and leading to a second order phase transition that starts the gradual onset of conditional convergence <ref:2512. p. 310672#pg5> The critical value for an interior peak of growth rate (and hence conditional convergence) arises when investment dominates depreciation, requiring hqi > γ/(2γ − δ) <ref:2512. p. 910672#pg11>

Principle of Relatedness

The model provides a theoretical explanation for the principle of relatedness, showing that the rate of accumulation of a capability will increase more rapidly in economies where other complementary capabilities are available <ref:2512.10672#pg18> This effect grows with the complementarity of capabilities, providing a deductive theoretical foundation for estimating relatedness <ref:2512. p. 310672#pg19> The derived expression shows that the rate of increase of a capability increases with the presence of others and grows faster when this is complementary to others <ref:2512.10672#pg19>

Kinematics and Speed of Convergence

The time scale of the model, representing the speed of convergence, depends on the complementarity of capabilities Ecpb and their intensity of use Qpb <ref:2512.10672#pg15> This means that when capabilities are intensely required, EcpbQpb accumulation is faster for economies that are better endowed with them <ref:2512. p. 110672#pg15> The solution to the Riccati equation shows that the speed of convergence depends on the complementarity of capabilities and their intensity of use <ref:2512. p. 110672#pg15> The model provides a simple and parsimonious way to recover historical behaviors, such as the divergence experienced by Rhode Island due to complex production processes <ref:2512. p. 210672#pg20>

Extension to Multiple Capabilities

The general form still involves a Riccati equation even for multiple capabilities, which is solved by focusing on the case where depreciation is small compared to investment <ref:2512.10672#pg14> The solution shows that the time scale of the model depends on the complementarity of capabilities and their intensity of use <ref:2512. p. 610672#pg15> The final expression for capability accumulation shows that growth depends on both activities that do not heavily rely on it and those that do depend on its availability <ref:2512. p. 8

Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.

Improvements for AI systems

1

Improve capability accumulation models by incorporating a Riccati-type dynamical system that endogenously generates a transition between unconditional and conditional convergence based on activity intensity. This allows systems to model growth rates that rise and fall with an economy's level of development, predicting when convergence becomes conditional as capability requirements increase.

2

Implement path-dependent diversification by formally modeling the principle of relatedness within the capability accumulation framework. The model predicts that the accumulation of capabilities accelerates when an economy is endowed with complementary capabilities, providing a theoretical basis for how economies are more likely to enter activities that are similar to the ones they are already specialized in.

3

Develop predictive tools for growth regimes by using the derived quadratic differential equation to determine when the maximum capability growth rate occurs at a boundary or in the interior of feasible capability levels. This provides a mechanism where the onset of conditional divergence can emerge spontaneously as the production of activities crosses a threshold in the intensity of their capability requirements.

Sources

Related papers