INET Meeting Room, Manor Road Building

We derive a tractable interactions-based macroeconomic model from microfoundations with the objective to improve our understanding of macroeconomic dynamics and, in particular, highlight sources of instability.

Investment decisions, like any decision, involve both independent thinking and exchange of opinions via interaction. The traditional approach is interactions-free because it is easier to describe mathematically systems with non-interacting agents. This approach, however, cannot explain certain observed behaviors, especially those related to critical phenomena such as regime transitions.

Models that take into account interactions among agents may do a better job at capturing these critical phenomena because with interaction the link between the micro and macro properties is no longer trivial (e.g. Lux, 2009). In particular, the collective effects emerging through interaction can generate instabilities, triggering a nonlinear large scale response to small perturbations. To the best of our knowledge, currently no interactions-based models can both fit empirical data and be tractable to highlight the mechanisms behind dynamic behaviors.

Here we derive an interactions-based macroeconomic model in analytic form. This derivation adapts the modeling framework, developed for the stock market in Gusev et al. (2015), to the general economy. This model has two types of interacting agents: investors who trade based on their expectations and analysts who interpret news, form expectations and channel them to investors. This micro-level formulation yields on a macro level a closed‐form dynamical system that interconnects information, expectation and price.

Further, Kroujiline et al. (2016) showed that this stock market model replicates past prices within reasonable tolerance and predicts future returns with a precision sufficient for designing a trading strategy. Kroujiline et al. (2018) linked it with a reduced-form economic model (Blanchard, 1981) to propose an endogenous mechanism of business cycles.


Research Themes

Research Programmes