You are here: Home › Research › Transdisciplinary Concepts & Methods › Research Areas › A Model of the German Economy
 

A Model of the German Economy


A Model of the German Economy (Speaker: C. Jaeger)

 

Equilibrium as a transdisciplinary concept as well as the strengths and weaknesses of its use in economics have been explored in a rich literature. Since the discovery of the Sonnenschein-Mantel-Debreu theorem on the shape of excess demand functions, it is known that the equilibrium concept used in contemporary economics cannot deliver what it is expected to grant: a reference point for policy design. There simply are too many possible equilibria, and therefore reference points, without a sensible selection mechanism. The present activity develops new tools to address the problem. The present activity develops a multi-agent model of the German economy that explicitly addresses the issue of equilibrium selection in view of the linkages between climate policy and unemployment.

Over the past years, at PIK a series of ideas have been patiently developed that now provide the basis for a focussed modelling effort. Key concepts used for that purpose include coordination by conventions and non-linear dynamics on networks.

In his classic “The mythical man-month”, F.P. Brooks develops the concept of “surgical teams” to overcome the endemic delays in developing highly innovative software. Here, the problem is to develop a highly innovative socio-economic model. Therefore, the team is built not around a senior software engineer as with Brooks, but around an economist with a strong transdisciplinary background. But here, too, the team is built with a careful balance of complementary skills and based on close face-to-face interaction. The relevant skills draw on economics, sociology, mathematics, computer science, and management.

 

First versions of the model have been implemented and presented to leading researchers in the field at international workshops (Berlin, November 2008; Maryland, March 2009; Venice, April, 2009; Berlin, November 2009) and at a Dahlem conference in December 2008. The model was first presented in a peer rewiewed paper (Mandel et al. 2009) that also reports the results of simulations obtained in a three-sector representation of the German economy. The generic model can be applied to model other economies as well. Its basic structure can be summed up as follows:

Agents in Lagom generiC are many firms and households, as well as one government, one financial system and an import export sector. Firms produce the good of the sector they belong to and have private prices for selling it; households make a living by working for firms as well as owning them, and consume goods according to private parameters; the government raises taxes and pays unemployment assurance; the financial system sets the inflation rate and gives credit to firms; the import export sector represents trade with the rest of the world.

Upon initialization of the model, the agents' private characteristics are created from randomness within given bounds, so that firms and households are similar but not perfect clones of each other. In a simulation, steps representing some actions, such as trade between firms and households, production, consumption, accounting etc., are carried out more frequently than others, which represent learning of the agents. Learning, that is a change of characteristics of the agents, occurs first by imitation: firms that are not doing well copy the characteristics of firms that are doing well, households observe another household and adapt their characteristics towards those of the observed household if it is doing better. Secondly, learning occurs by evolution, that is, random mutations of an agent’s characteristics with small probability.

The model has provided the conceptual basis for two major studies commissioned by the German ministry for the Environment, the first assessing the economic implications of German climate policy, the second designing sustainable answers to the financial crisis. In these studies, the problem of equilibrium selection plays a crucial role because the German economy is seen to currently have at least two possible equilibria: one with low growth, high unemployment, and slowly decresing emissions; and another one with high growth, low unemployment, and quickly decreasing emissions. On this basis, policies that can lead to the selection of the latter equilibrium have been designed.

 

Presently, the model is being further developed in several respects:
- a more complete representation of the financial sector as well as governments
- making the model spatially explicit
- representing further aspects of the natural environment explicit

 

The team's division of labor in this research is as follows:

 

Transdisciplinary oriented economist:

Develop the theoretical structure implied by the model and embed the model development in on-going conversations with decision-makers.

Mathematical economist:

Develop the mathematical structures involved in the model in view of their economic content.

Two computer scientists:

Design algorithms, use existing libraries, program the code to implement the model.

Mathematician:

regionalize the model, investigate applicability of results from the theory of randomly perturbed stochastic dynamical systems.

Empirically oriented economist:

Provide access to relevant data and stylized facts, retrieve data needed to calibrate the model.

Management Assistant:

Document the research process, organize communication both within the team as with the scientific networks it is embedded in (Dahlem conference, GSD project, etc.).

 

Papers and reports about the model of the German economy:

 

Document Actions