This chapter provides some reflections on the new macroeconomics of model uncertainty. Research on model uncertainty represents a broad effort to understand how the ignorance of individuals and policymakers about the true structure of the economy should and does affect their respective behaviors. As such, this approach challenges certain aspects of modern macroeconomics, most notably the rational expectations assumptions that are generally made in modeling aggregate outcomes. Our goal is to explore some of the most interesting implications of model uncertainty for positive and normative macroeconomic analysis. Our discussion will suggest directions along which work on model uncertainty may be fruitfully developed; as such it is necessarily somewhat speculative as we cannot say to what extent these directions are feasible.

Our interest in model uncertainty is certainly not unique; in fact, within macroeconomics, studies of model uncertainty have become one of the most active areas of research. Much of this new work was stimulated by the seminal contributions of Hansen and Sargent of which (2001, 2003a, b) are only a subset of their contributions. Hansen and Sargent's work explores the question of robustness in decision making. In their approach, agents are not assumed to know the true model of the economy, in contrast to standard rational expectations formulations.

Within economics, social interactions research constitutes a growing area of study. This research represents a good faith attempt to introduce substantive sociological factors into economic modeling. As such, this work represents a significant departure from the sorts of market-mediated interdependences between individuals that one finds in general equilibrium theory. While the substantive ideas underlying this work may be found in now-classic papers such as Loury (1977), the modern social interactions literature is quite young. Despite this, there are now an impressive range of applications of social interactions models in microeconomic contexts. Examples of phenomena where empirical evidence of social interactions has been found include (1) crime (Glaeser, Sacerdote, and Scheinkman [1996], Sirakaya [2003]), (2) welfare and public assistance use (Aizer and Currie [2004], Bertrand, Luttmer, and Mullainathan [2000]), (3) fertility (Brooks-Gunn et al. [1993], Rivkin [2001]), (4) housing demand and urban development (Irwin and Bockstaed [2002], Ioannides and Zabel [2003a, b]), (5) contract determination (Young and Burke [2001, 2003]), (6) employment (Oomes [2003], Topa [2001], Weinberg, Reagan, and Yankow [2004]), (7) cigarette smoking (Krauth [2003], Nakajima [2003]), (8) school performance (Boozer and Cacciola [2001], Graham [2005]) and even (9) medical techniques (Burke, Fournier, and Prasad [2004]).

INTRODUCTION

For the last few years, the Naval Research laboratory has been attempting to build robots that are similar to humans in a variety of ways. The goal has been to build systems that think and act like a person rather than look like a person, because the state of the art is not sufficient for a robot to look (even superficially) like a human person. There are at least two reasons to build robots that think and act like a human. First, how an artificial system acts has a profound effect on how people act towards the system. Second, how an artificial system thinks has a profound effect on how people interact with the system.

HOW PEOPLE ACT TOWARDS ARTIFICIAL SYSTEMS

“Everyone” knows that computers have no feelings, attitudes, or desires. Most people do not worry about hurting a toaster's feelings or cursing at a VCR. However, in a surprising series of studies, Cliff Nass has shown that people in some situations do, in fact, treat computer systems as social entities. Nass has shown that it takes very little “social-ness” for a person to treat computers (including robots, AI programs, etc.) as social creatures. For example, Nass and Moon (2000) examined people's application of social categories to computers. Nass and Moon (2000) compared users' interactions with two computer systems – a tutor and an evaluator – using different combinations of male and female voices. Even though the participants indicated that they knew they were interacting with a computer, and explicitly reported that the voice did not relate to the “gender” of the computer, or even the computer programmer, there were distinct genderrelated biases in the experiment data.

In Dynamic Macroeconomic Analysis: Theory and Policy in General Equilibrium, Altug, Chadha and Nolan, hereafter ‘ACN’, have undertaken the extremely difficult task of bringing the reader up to date on the vast literature that has developed in this key area of economics since such seminal books as Cooley's Frontiers Of Business Cycle Research (1995).

ACN tackle this formidable assignment by recruiting top scholars to write individual chapters. Each individual chapter is of exceptionally high quality because each scholar is a first-rate expert in the chapter's area.

The approach of the book is ‘quantitative theorising’, in the sense that each chapter presents not only recent developments in theory but also recent developments in compilation of the facts that confront the theory. Discrepancies between facts and theories are carefully discussed. Many of the chapters offer modifications to the baseline theory that make it do a better job of matching the facts.

I give the reader a quick overview of the contents of this book in a brief introduction to each of the chapters. Three chapters are discussed in more detail than the others in order to keep my foreword within standard space limits. But all of the chapters are equally exciting and have equal command on the reader's attention. I hope this somewhat unusual approach to writing a foreword will entice more readers to add this important book to their libraries. In addition to saying a few words about the chapters, I shall use the discussion of some of them to give some reflections about the field they cover as well as give some speculations and opinions about potentially fruitful future research.

Jim Pemberton (chapter 1) reviews dynamic life cycle consumption/savings models, the facts they are designed to explain, the struggles in attempting to modify the basic core model to fit these facts and other uses to which they are put. Particularly interesting is discussion of how minor modifications can lead to big and rather counterfactual predictions by some of the models. For example, adding a small safety net in one model caused the model's consumer to borrow too large an amount of money early on to be consistent with fact.