Research Papers
Home

 

Principal Research Areas and Contributions of Michael J. Radzicki

 

Post Keynesian-Institutional Dynamics

    These papers represent my efforts to combine Post Keynesian economics and institutional economics with system dynamics computer simulation modeling to form a more powerful form of heterodox economics.

1.    “Institutional Dynamics: An Extension of the Institutionalist Approach to Socioeconomic Analysis.” Journal of Economic Issues 22(3): 633-666. 1988. This is the first paper ever to make the argument that system dynamics and institutional economics are strikingly similar and that the two fields should be combined to form a new field called “institutional dynamics.” This paper is written for institutional economists. Download the paper.

2.    “Methodologia Oeconomiae et Systematis Dynamis.” System Dynamics Review 6(2): 123-147. 1990. This paper is listed in “Desert Island Dynamics: An Annotated Survey of the Essential System Dynamics Literature.” M. Anjali Sastry and John D. Sterman, eds., System Dynamics Group, Massachusetts Institute of Technology. http://web.mit.edu/jsterman/www/DID.html. It is the second paper to make the argument that system dynamics and institutional economics are strikingly similar and that the two fields should be combined to create a superior form of heterodox economic analysis called “institutional dynamics.” This paper also shows precisely where the system dynamics method fits in the philosophy of science literature. The paper is written for system dynamicists. Download the paper.

3.    “An Institutional Dynamics Model of Sterling, Massachusetts: Indicative Planning at the Local Level.” Journal of Economic Issues 27(2): 481-492. 1993. With Donald A. Seville. [Also available as Memo D-4358. System Dynamics Group. Massachusetts Institute of Technology] The first two papers above outline the theoretical argument that system dynamics and institutional economics should be combined into a new area called “institutional dynamics.” This paper describes the first actual application of the proposed “institutional dynamics” method to an economics problem -- zoning and tax policy in Sterling, Massachusetts. This paper also extends the urban dynamics literature in the field of system dynamics because it identifies and models factors that attract in-migrants to Sterling that had previously not been included in the urban dynamics literature. Download the paper.

4.    “Evolutionary Economics and System Dynamics.” pp. 61-89 in Richard W. England, ed. Evolutionary Concepts in Contemporary Economics. Ann Arbor, MI: University of Michigan Press. 1994. With John D. Sterman, Massachusetts Institute of Technology. [Also available as Memo D-4352. System Dynamics Group. Massachusetts Institute of Technology and in The Role of Strategic Modelling in International Competitiveness: Proceedings of the 1993 International Conference of the System Dynamics Society. Cancún, Mexico. July 19-25. Enrique Zepeda and José A. D. Machuca, eds., pp. 388-1 to 388-26] This paper extends the notion that system dynamics should be combined with institutional economics and argues that system dynamics can be used by all “evolutionary economists” (institutional economics is a subset of the broader field of “evolutionary economics”). This paper includes an original system dynamics model that illustrates evolutionary economic behavior. Download the paper.

5.    "Mr. Hamilton, Mr. Forrester and a Foundation for Evolutionary Economics." 2003. Journal of Economic Issues. 37(1): 133-173. This is an invited paper for a special issue of the Journal of Economic Issues that was published to celebrate the fiftieth anniversary of the publication of David Hamilton’s book Evolutionary Economics. Hamilton’s primary insight was that neoclassical economic models are based on Newtonian concepts of change while institutional economic models are based on Darwinian concepts of change. The paper extends Hamilton’s insight to seven additional schools of economic thought and shows that many of them could profit from the use of system dynamics modeling and/or agent-based modeling. It also outlines how system dynamics modeling embodies the concept of Darwinian change and presented an extended version of the model originally presented in “Evolutionary Economics and System Dynamics”. Download the paper.

6.    "In Defense of System Dynamics: A Reply to Professor Hayden." with Linwood Tauheed. This paper will be submitted to the Journal of Economic Issues. It answers the egregiously inaccurate claims of Professor Greg Hayden that system dynamics modeling is inappropriate for institutional economic analysis. Download the paper.

7.    “A Post Keynesian Model of Macroeconomic Growth, Instability, and Income Distribution.” pp. 435-443 in Enrique Zepeda and José A. D. Machuca, eds. The Role of Strategic Modelling in International Competitiveness: Proceedings of the 1993 International Conference of the System Dynamics Society. Cancún, Mexico. July 19-25, 1993. With Khalid Saeed. Asian Institute of Technology. This paper outlines a Post Keynesian system dynamics model (to my knowledge, the world’s first). Post Keynesian economics complements institutional economics. In fact, many institutional economists refer to themselves as “Post Keynesian institutionalists.” Download the paper.

8.    "Institutional Economics, Post Keynesian Economics, and System Dynamics: Three Strands of a Heterodox Economics Braid." in John T. Harvey and Robert F. Garnett, Jr., eds. Future Directions for Heterodox Economics. Ann Arbor, MI: University of Michigan Press. Forthcoming. This paper outlines the ways in which Post Keynesian economics, institutional economics, and system dynamics complement one another and provides an overview of my Post Keynesian-Institutional-System Dynamics model. Some of the first uses of the model will be to examine the dynamics of an "employer of last resort program" and the “future of the euro.” Download the paper. Download the Table of Contents.

9.    "Was Alfred Eichner a System Dynamicist?" In Money and Macroeconomic Issues: Alfred Eichner and Post-Keynesian Economics. Marc Lavoie, Louis-Philippe Rochon, and Mario Seccareccia, eds. Armonk, New York: M. E. Sharpe, Inc. Forthcoming. This paper was written in honor of the twentieth anniversary of the publication of Alfred Eichner's The Macrodynamics of Advanced Market Economies. It makes the case that Eichner's Post Keynesian methodology is strikingly similar to the system dynamics paradigm and postulates that Eichner might very well have used system dynamics had he known of its existence. Download the paper.

10.    "The Circular and Cumulative Structure of Administered Pricing." Journal of Economics Issues 40(2): 517-526. 2006. With Mark W. Nichols and Oleg V. Pavlov. This paper reviews the literature on Post Keynesian pricing and inflation, and then presents the mark-up pricing sub-sector of my Post Keynesian-Institutionalist-System Dynamics Model. Some preliminary results from simulation of the sub-sector are also presented. Download the paper.

11.    “A Note on Kelsey's ‘The Economics of Chaos or the Chaos of Economics.’” Oxford Economic Papers 40(4): 692-693. 1988. This paper corrects some erroneous statements made by an economist who wrote a paper outlining how chaos theory could be useful in economics. Download the paper.

12.    “Institutional Dynamics, Deterministic Chaos, and Self-Organizing Systems.” Journal of Economic Issues 24(1): 57-102. 1990. This paper outlines how chaos theory, implemented via system dynamics modeling, is in harmony with much of institutional economics and can be used to produce more realistic evolutionary economic models. Download the paper.

13.    “Chaos Theory and Economics.” pp. 42-50 in Geoff Hodgson, Warren J. Samuels, and Marc R. Tool, eds. Institutional and Evolutionary Economics. Brookfield, VT: Edward Elgar Publishing Limited. 1994. This invited paper provides a “sweeping overview” of chaos theory for institutional and evolutionary economists. Download the paper.

14.    “An Institutionalist Perspective on Law and Economics (Chicago Style) in the Context of United States Labor Law.” Arizona Law Review 35(2): 397-443. 1993. With William A. Wines and James B. Zimarowski. This paper extends the institutional dynamics method to the sub-field of economics known as “law and economics.” It won the 1991 William O. Douglas Award for "best paper" in the Pacific Northwest Legal Studies in Business Association Conference. Download the paper.

15.    “An Institutional Dynamics Approach to the Study of Peace and World Order.” pp. 543-552 in Jac A. M. Vennix , Jan Faber, Wim J. Scheper, and Cees A. Th. Takkenberg, eds. Proceedings of the 1992 International Conference of the System Dynamics Society. Utrecht University. Utrecht, The Netherlands. July 14-17, 1992. With W. Scott Trees, Siena College. This paper extends the institutional dynamics approach to the area of peace studies. Download the paper.

16.    “MicroWorlds and Evolutionary Economics.” pp. 533-542 in Jac A. M. Vennix, Jan Faber, Wim J. Scheper, and Cees A. Th. Takkenberg, eds. Proceedings of the 1992 International Conference of the System Dynamics Society. Utrecht University. Utrecht, The Netherlands. July 14-17, 1992. This paper discusses how an institutional dynamics model can be turned into a “microworld” or “learning laboratory” so that it can be used in a gaming environment. Download the paper.

17.    “A System Dynamics Approach to Sustainable Cities.” pp. 191-210 in Toshiro Shimada and Khalid Saeed, eds. Proceedings of the 1995 International Conference of the System Dynamics Society. Gakushuin University. Tokyo, Japan. July 30-August 4, 1995. With W. Scott Trees, Siena College. This paper extends the urban dynamics literature by tying it to the (at the time) emerging area of “sustainable cities.” Traditional urban dynamics models generally omit environmental and quality of life variables that are thought to be crucial to the notion of a sustainable city. This paper presents a system dynamics model that incorporates many of these factors. It also outlines a sustainable cities research program. Download the paper.

18. "System Dynamics and Its Contribution to Economics and Economic Modeling." Encyclopedia of Complexity and System Science. Robert A. Meyers, ed. Heidelberg, Germany: Springer-Verlag. In addition to summarizing some of the controversial applications of system dynamics modeling to problems in economics, this paper provides a sweeping overview of how system dynamics can be used for economic analysis. Among the techniques discussed are translating existing written, static, difference equation, and differential equation economics models into their equivalent system dynamics representation. These techniques are offered in contrast to simply building economics models from scratch using proper system dynamics modeling techniques. Download the paper.

Agent-Based Modeling

    These papers present my efforts to apply agent-based modeling to problems in economics from a heterodox perspective. Properly created, agent-based models are behavioral, so they should be valued by behavioral economists and economic psychologists, non-ergodic, so they should be valued by Post Keynesian economists, and can be considered evolutionary "pattern models," so they should be valued by institutional economists.

1.    "Rebuilding New Orleans: An Agent-Based Modeling Approach." With Munaf Amir, Diana Damyanova, Mark W. Nichols, and Esra Unluaslan. Presented at the 2006 Meetings of the Midwest Economics Association. March 26, 2006. Chicago, Illinois. This paper describes an agent-based model that is used to analyze proposed policies for rebuilding the city of New Orleans. In particular, the effectiveness of an organization such as Habitat for Humanity going in to New Orleans to build houses is examined. One of the primary findings reported in the paper is that an organization such as Habitat must simultaneously run an "employer of last resort" program to effectively jump-start the rebuilding of the city. Download the paper. Download a PDFed version of the Power Point slides presented at the 2006 Midwest Economics Association meetings.

2.    "An Agent-Based Model of Behavior in "Beauty Contest" Games. With Mark W. Nichols, University of Nevada, Reno. Presented at the Ninth International Post Keynesian Conference. September 16, 2006. Kansas City, Missouri. This paper describes an agent-based computer simulation model of a p-beauty contest game. These games are designed to mimic the expectation formation process as described by Keynes in Chapter 12 of the General Theory. In a traditional p-beauty contest game, human subjects choose a number over a fixed interval [L, H], where L is the low number (e.g., zero) and H is the high number (e.g., 100). The choices are then made known to all subjects, as is the mean and p*mean, where p is some number greater than zero, but frequently less than one. The winner(s) is the subject(s) whose choice is closest to p*mean. The game is then repeated until (optimally) a Nash equilibrium is reached. Learning is generally measured by the degree of convergence of the subjects' (assumed rational) choices toward that equilibrium. Curiously, most of the experimental results in the literature suggest that humans play the game by following simple decision rules and adjusting their strategies based on feedback from their environment rather than by utilizing deductive, rational, iterative thinking.

For this paper we ran a p-beauty contest game with human subjects and then used protocol analysis to uncover their actual decision making strategies. We then created an agent-based computer model of the game and programmed the processes by which the subjects formulated and revised their strategies. We found that the model could accurately replicate the results of three separate plays of the game. Our preliminary conclusions are that (a) agent-based modeling combined with protocol analysis can be effectively used to mimic p-beauty contest games and, possibly, actual expectation formation and investment behavior; (b) simple, inductive, behavioral decision rules appear to explain the results of p-beauty contest games more effectively than merely positing players’ depths of reasoning; and (c) once the model can mimic the game it can be used to run "what-if" experiments that are not possible with human subjects to answer questions such as "can a small number of really 'smart' agents, both in terms of their depth of reasoning and ability to anticipate the actions of others, win most of the time and become 'rich'?" Download the paper. Download a PDFed version of the Power Point slide presented at the 2006 Association for Institutional Thought meetings.

History of Economic Thought via System Dynamics

    These papers present my efforts to examine and teach some of the history of economic thought via system dynamics. The models presented in these papers are part of a graduate course in “macroeconomic dynamics” and will also appear in a book by the same title.

1.    “Expectation Formation and Parameter Estimation in Uncertain Dynamical Systems: The System Dynamics Approach to Post Keynesian-Institutional Economics.” Proceedings of the Twenty Second International System Dynamics Conference. July 28, 2004, Keble College, University of Oxford, Oxford, UK. This paper presents a bounded rational expectations structure that nicely mimics actual human expectations data. The structure is then added to the Harrod growth model in a manner consistent with Harrod’s writings on how entrepreneurial expectations can cause macroeconomic instability. The modified Harrod model generates both a trend and a cycle (as Harrod originally argued) and nicely fits macroeconomic data from the U.S. economy. Parameter estimates provided by the fitted version of the model are both reasonable and consistent with other macroeconometric studies. The results of this paper also suggest that Harrod’s famous “knife-edge” theory (to which Harrod himself objected) is incorrect and that is should be removed from the textbooks. Download the paper.

Psychology and System Dynamics

    These papers represent my efforts with Professor Jim Doyle, Professor Susan Ganter, and Professor W. Scott Trees to utilize tools and techniques from psychology to develop new ways of assessing the effectiveness of system dynamics models and interventions. They also represent our efforts to devise more “psychologically rich” system dynamics models and to devise improved ways of eliciting information from human decision makers.

1.    “Using Cognitive Styles Typology to Explain Individual Differences in Dynamic Decision Making: Much Ado About Nothing.” Center for Quality of Management Journal. 6(3): 5-15. 1997. With James K. Doyle, W. Scott Trees. and Andrew Rose. This paper describes a study Radzicki et al. did on WPI undergraduates in which they investigated how, if at all, a student’s individual cognitive style or preferred way of learning related to his/her actual learning from a system dynamics tool (an economics micro-world). The results were that, generally speaking, a student’s cognitive style was unrelated to how much he or she learned from the micro-world. Download the paper.

2.    "Measuring Change in Mental Models of Complex Systems." With James K. Doyle and W. Scott Trees. Currently under review for inclusion the book: Complex Decision Making: Theory and Practice. Hassan Qudrat-Ullah, J. Michael Spector, and Paal I. Davidsen, eds. Springer-Verlag. Measuring change in the mental models of the participants in systems thinking and system dynamics interventions is unavoidable if the relative effectiveness of different interventions in promoting learning is to be assessed. However, existing methods for organizing, representing, eliciting, and mapping mental models are designed primarily to facilitate change in mental models, rather than to measure change, and so new methodologies for measuring change in mental models are needed. This paper defines the necessary features of any methodology that aims to rigorously measure change in mental models of dynamic systems and describes the development and implementation of one specific new methodology designed to fulfill these criteria. An exploratory application of the method to a systems thinking intervention designed to change mental models of the causes of the economic long wave, or Kondratieff cycle, is also described. Results indicate that the intervention produced statistically reliable changes in the content and size of subjects’ mental models, as well as the degree of feedback thinking that they contained, but had no significant effect on the degree of detail complexity or dynamic complexity in subjects’ mental models. The method was able to capture even subtle changes in mental models due to the intervention, and it can be widely applied to answer important questions about the cognitive effects of alternate interventions. The limitations of the described work and suggestions for future research are discussed. Download the paper.

3.    “Measuring the Effect of Systems Thinking Interventions on Mental Models.” pp. 129-132 in George P. Richardson and John D. Sterman, eds. Proceedings of the 1996 International Conference of the System Dynamics Society. Cambridge, Massachusetts. July 22-25, 1996. With James K. Doyle and W. Scott Trees. Available as Memo D-4557-2, System Dynamics Group, Massachusetts Institute of Technology. This paper describes an original set of “mental model mapping” instruments and protocols Radzicki et al. designed, and their use on WPI undergraduates to measure how much they learned from a system dynamics tool (an economics micro-world). It is an earlier and abbreviated version of "measuring Change in Mental Models of Dynamic Systems: An Exploratory Study." Download the paper.

4.    "Mental Models of Dynamic Systems." UNESCO Encyclopedia of Life Support Systems. UK: EOLSS Publishers. 2001. With James K. Doyle, David N. Ford and W. Scott Trees. This article describes the history of the mental models concept in the fields of system dynamics and psychology, and offers a comprehensive definition of the term for use in system dynamics research. The characteristics of mental models of dynamic systems identified by the empirical literature are reviewed, with an emphasis on important flaws and limitations, as well as their underlying causes, which typically limit the utility of mental models for dynamic decision making. A mental model-based theory of dynamic decision making is presented that is consistent with this evidence, and the mechanisms by which system dynamics computer modeling can improve mental models within this theoretical framework are described. The implications of the theory for developing appropriate techniques for studying mental models, as well as specific priorities for future research, are discussed. Download the paper.

5.    “Assessing System Dynamics Curricula: Past, Present, and Future.” pp. 484-493 in Toshiro Shimada and Khalid Saeed, eds. Proceedings of the 1995 International Conference of the System Dynamics Society. Gakushuin University. Tokyo, Japan. July 30-August 4, 1995. With Susan L. Ganter and James K. Doyle. This paper critiques the curriculum assessment that has been done over the years in the field of system dynamics and presents a set of guidelines for a comprehensive system dynamics curriculum assessment program. Download the paper.

Contributions to System Dynamics Theory and Teaching

    Here are a few papers related to teaching system dynamics and the underlying theory of system dynamics.

1.    “Dyadic Processes, Tempestuous Relationships, and System Dynamics.” System Dynamics Review 9(1): 79-94. 1993. This paper presents a system dynamics exercise (“Romeo and Juliet”) that teaches about modeling oscillating systems. Today this exercise is used at a number of major universities including Cornell University, West Point, the University of Palermo, Italy, and the University of Bergen, Norway. It has also inspired several authors to extend the exercise. Download the paper.

2.    “Reference Modes and the Optimal Shape Parameter.” System Dynamics Review 5(2): 192-198. 1989. This paper describes an improved (mathematical) way of presenting system dynamics reference mode data. Download the paper.

3.    “Reflections on: ‘On the Very Idea of a System Dynamics Model of Kuhnian Science.’” System Dynamics Review 8(1): 49-53. 1992. This paper is a response to a published article that was critical of a well-known system dynamics model. It uses philosophy of science arguments and examples from economics to defend the system dynamics model. Download the paper.

4.    "The Early Years - A Voluntary Effort." This paper became part of a consensus story of how the System Dynamics Society came into being, commissioned for the 25th annual conference of the System Dynamics Society. It is integrated into: Andersen, David, John Morecroft, and Roberta Spencer (with Jay Forrester, Michel Karsky, Bernard Paulre, Jack Pugh, Michael Radzicki, Jorgen Randers, George Richardson, Khalid Saeed, and Eric Wolstenholme). 2007. "How the System Dynamics Society Came to Be: A Collective Memoir." System Dynamics Review 23(2-3): 219-227. Click here to see all of the papers commissioned for the consensus story. Download Radzicki's paper. Download the consensus story paper.

Theoretical Econometrics

    Here are two theoretical econometrics papers I did with my dissertation director Larry Marsh after I had received my Ph.D. I had worked for several years as Larry's statistics and econometrics TA and this work was a nice extension of our collaboration. Larry is an excellent econometrician and the driving force behind these papers.

1.    “Minimum Mean Square Error of a Generalized Stein Estimator.” Economics Letters 21(4): 333-335. 1986. With Lawrence C. Marsh, University of Notre Dame. This paper presents a contribution to the theory of econometrics. It presents a mathematical proof which shows, for the first time, that three well-known “biased regression estimators” (principal components regression, ridge regression, and Stein regression) share the same theoretical minimum mean square error. Download the paper.

2.    "A Generalized Family of Principal Components Regression Estimators." With Lawrence C. Marsh, University of Notre Dame. This paper develops a theoretical foundation for a broader family of principal components regression estimators. The key to this framework is a generalization of the transformation of the design matrix that is traditionally performed in principal components regression. In the second section of the paper this transformation is used to derive the theoretically minimum level of quadratic risk that is possible for this family of estimators. It is also used to demonstrate that this level of risk is smaller than the minimum level possible for Hoerl and Kennard's generalized ridge regression family of estimators. In the third section of the paper the boundary of the region over which the generalized principal components family dominates ordinary least squares is presented. In the fourth section of the paper the generalized principal components family is reinterpreted in terms of Bayesian regression, ordinary least squares regression, restricted least squares regression, Theil's best linear estimator, and the generalized Stein regression estimator. The fifth section of the paper reports the primary conclusion that the generalized principal components family provides both a theoretical framework for the comparison of regression estimators, and a lower bound goal for quadratic risk that all regression estimators could strive to attain. Download the paper. Download a long "referee's" version of the paper with all of the proofs completely specified.

Soft Variables in System Dynamics Models

    Here are two papers that address the issue of incorporating soft variables into system dynamics models.

1.    "Stability in a Superpower-Dominated Global Economic System." 2005. Journal of Economics Issues 39(2):  491-500. With Oleg Pavlov and Khalid Saeed. This paper illustrates how soft variables can be incorporated into macroeconomic system dynamics models. Download the paper.

2.    “Incorporating Christian Values Into Business Simulations: An Institutional Dynamics Approach.” pp. 157-172 in Norlin Rueschhoff and Konrad Schaum, eds. Christian Business Values in an Intercultural Environment. Berlin: Duncker and Humbolt. 1989. This paper illustrates how “values” or “soft variables” can be incorporated into corporate system dynamics models.

Book Reviews

1.    “A Review of Chaos: Making A New Science.” System Dynamics Review 5(1): 90-91. 1989.

2.    “Review of Frontiers in Ecological Economic Theory and Application.” Eastern Economic Journal. Forthcoming. Download the paper.