Mauro Gallegati

Investment confidence, corporate debt and income fluctuations☆

In this paper we present a macroeconomic model with New Keynesian features which endogenously generates dynamic paths of income and the stock of corporate debt of a cyclic and chaotic nature. From the market clearing condition on goods and money markets we derive the dynamic paths of income and corporate debt whose stability properties depend upon the propensity to invest out of the flow of internally generated funds, which in turn is a positive function of income. If it is relatively ‘low’, the dynamic paths on income and corporate debt converge to their steady-state long-run values. When the propensity to invest is neither too ‘low’ nor too ‘high’, the system can exhibit either bounded cycles or chaotic dynamics. Finally, if it exceeds a critical upper value, an explosive growth of debt occurs and a financial crisis is likely to ensue.

Economics with Heterogeneous Interacting Agents: A Practical Guide to Agent-Based Modeling

This book offers a practical guide to Agent Based economic modeling, adopting a learning by doing approach to help the reader master the fundamental tools needed to create and analyze Agent Based models. After providing them with a basic toolkit for Agent Based modeling, it present and discusses didactic models of real financial and economic systems in detail. While stressing the main features and advantages of the bottom-up perspective inherent to this approach, the book also highlights the logic and practical steps that characterize the model building procedure. A detailed description of the underlying codes, developed using RandC, is also provided. In addition, each didactic model is accompanied by exercises and applications designed to promote active learning on the part of the reader. Following the same approach, the book also presents several complementary tools required for the analysis and validation of the models, such as sensitivity experiments, calibration exercises, economic network and statistical distributions analysis. By the end of the book, the reader will have gained a deeper understanding of the Agent Based methodology and be prepared to use the fundamental techniques required to start developing their own economic models. Accordingly, Economics with Heterogeneous Interacting Agents will be of particular interest to graduate and postgraduate students, as well as to academic institutions and lecturers interested in including an overview of the AB approach to economic modeling in their courses.

Structure and Macroeconomic Performance: Heterogeneous Firms and Financial Fragility

In this paper we adopt a new macrodynamic tool, i.e. a system of non-linear difference equations describing the evolution over time of the first and second moments of the distribution of firms’ degrees of financial robustness captured by the ratio of the equity base to the capital stock - the equity ratio for short - which affects supply and capital accumulation decisions. For particular configurations of parameters the dynamic patterns of the average equity ratio and the variance generate irregular and asymmetric time series in which growth and fluctuations are jointly determined (fluctuating growth).

A New Keynesian Model of the Business Cycle and Financial Fragility

Two competing procedures are currently adopted in modelling the cyclical dynamics of output. According to the first one, output movements must be conceived of as responses of a linear system to stochastic disturbances (Frisch, 1933; Slutstky, 1937; Lucas, 1975). Alternatively, the second one interprets the oscillating path of income as a built-in feature of a non-linear deterministic system (Kaldor, 1940; Hicks, 1950; Goodwin, 1982; Grandmont, 1985). According to us, there is no reason why these two explanations should not co-exist: in this paper output dynamics is determined by a non-linear system of difference equations which takes into account also stochastic shocks (similar attempts in this direction have been recently carried out by Greenwald and Stiglitz, 1988a; Day and Lin, 1991; Delli Gatti et al., 1992). In our model the basic ingredients of both explanations are incorporated in the investment equation. Infact the propensity to invest is modelled as a non-linear function of income affected by a stochastic disturbance. Stochastic disturbances enter the process of income determination also through the capital assets price equation. Therefore the dynamics of the main macro-variables is jointly determined by the deterministic non-linear difference equations system and the random process which generates “clouds” around a deterministic cycle (when the shock is additive), and “jumps” from a cycle to another (when the shock affects the value of the parameters).