This impressive book contains contributions from some of the most inquisitive minds in economics. The essays are organized around five major themes.
The first theme concerns the current state of macroeconomic theory and modeling. This part of the book constitutes a great read for those unfamiliar with the technical details of modern macroeconomics. Axel Leijonhufvud provides an episodic account of the history of macroeconomics over the last century, emphasizing historical links to the Post Walrasian view. David Colander highlights the continuity between classical and Post Walrasian macroeconomics, and the importance of agent optimization in limited-information environments. According to Colander, the lack of sophisticated tools has kept economists from dealing with the questions raised by Post Walrasian economics and allowed the Walrasian approach, which focuses on manageable issues, to dominate macroeconomics. Perry Mehrling explores how time and uncertainty are treated in classical, Dynamic Stochastic General Equilibrium (DSGE), and Post Walrasian models. Peter Matthews discusses Post Walrasian man and what might be needed to replace the rational agents of the current paradigm. He argues that building a new model of man will require insights from experimental, behavioral, and evolutionary approaches to microeconomic behavior.
Section Two highlights some limitations of the DSGE framework, although its contributors are relatively optimistic that this framework can incorporate new methodologies and tools. William Brock and Steven Durlauf introduce sociological considerations into economic models and discuss the econometric challenges of modeling social interactions. They conclude that the empirical literature is still too young to make claims about the importance of social interactions. In the following contribution, the same authors explore model uncertainty. William Branch discusses Restricted Perceptions Equilibrium — an equilibrium between the economy's stochastic process and optimally misspecified beliefs. He argues that it constitutes a natural alternative to Rational Expectations equilibrium because it allows for bounded rationality and is consistent with Muth's original hypothesis. Masanao Aoki observes that in dealing with the tradeoff between analytical tractability and real-world relevance, existing macroeconomic models err on the side of tractability. He uses Einstein's "Not More So" criterion, which suggests that simplification beyond a point is self-defeating as it destroys the very insights offered by the model. Aoki then proposes using new tools that go beyond the assumptions of DSGE models.
Section Three challenges the Walrasian paradigm in macroeconomics and proposes using different tools from what the DSGE framework can easily accommodate. Leigh Tesfatsion explains the new opportunities for macroeconomic modelers due to the growing capabilities of computers. Agent-based computational economics, the computational study of economic processes as dynamic systems of interacting agents, is especially important. Robert Axtell argues that microbehavior can be studied using this technique without relying on strong and simplistic assumptions. Blake LeBaron considers the use of agent-based models in finance to resolve some puzzling aspects of the existing literature related to market heterogeneity and argues that our understanding of finance can be improved by adopting a behavioral approach.
Part Four has a hands-on flavor, focusing on data and econometric issues. Kevin Hoover questions the feasibility of a Post Walrasian program replacing the current orthodoxy because its multifarious agenda prevents it from producing conviction and fixing beliefs — a necessary condition for successfully challenging the Walrasian view. He favors a more Marshallian approach that focuses on confronting the data. Roger Farmer contrasts the quantitative macroeconomic approach that dominates the North American literature with the general-to-specific econometrics approach, focusing on the possibilities that emerge from combining the two approaches. Rational expectations constitute an integral part of the proposed synthesis. Robert Baumann rejects the widely held belief that the economic theories supply sufficient independent and consistent restrictions to allow identifiability in modern macroeconometrics. Soren Johansen alerts us to the mismatch between theory and evidence, and discusses the choice of appropriate models so that we can make valid inferences. However, problems remain. Asymptotic results derived from stationary processes cannot be used to make inference regarding highly persistent stationary processes. What should macroeconometricians do? Katarina Juselius and Johansen set forth an approach for analyzing data that can indicate inappropriate assumptions and guide model choice. In particular, they show how the (co-integrated) VAR approach can extract short-run and long-run information from the data.
The last section includes two papers on the challenges of economic policymaking. Russell Cooper focuses on the implications of multiple equilibria for policy and whether government policy interventions constitute a source or a solution to this multiplicity. Peter Howitt focuses on the implications of the Post Walrasian research program for monetary policy. He argues that it provides a number of insights regarding policy rules and learning, which are not visible to those using the Walrasian toolkit. An environment with informational imperfections and an adaptive view of monetary policy can offer additional insights.
Whether the current Walrasian consensus will remain dominant depends on, among other things, its ability to accommodate issues such as the ones raised in this volume. Most papers here recognize that the attempt to create an alternative to the Walrasian paradigm is only beginning and that Post Walrasian economics is really a set of diverse approaches. Post Walrasians are dissatisfied with some assumption of the Walrasian framework, but they do not have a coherent structure that can incorporate distinct contributions. This lack of unanimity is natural when a new paradigm develops, as Leijonhufvud points out. But to become a dominant paradigm, as Hoover notes, requires the ability to fix beliefs. This may be good news for supporters of the Post Walrasian and the Walrasian approaches alike. The former have a huge array of new ideas and techniques. At the same time, analysis has to guide policy; hence, this task will keep busy those wedded to the Walrasian toolkit. This book provides an excellent introduction to this titanic project as well as clear indications of the direction that macroeconomics is moving toward.
