Table Of Content

WORLD SCIENTIFIC SERIES ON NONLINEAR SCIENCE Editor: Leon O. Chua University of California, Berkeley Series A. MONOGRAPHS AND TREATISES* Volume 74: Physarum Machines: Computers from Slime Mould A. Adamatzky Volume 75: Discrete Systems with Memory R. Alonso-Sanz Volume 76: A Nonlinear Dynamics Perspective of Wolfram’s New Kind of Science (Volume IV) L. O. Chua Volume 77: Mathematical Mechanics: From Particle to Muscle E. D. Cooper Volume 78: Qualitative and Asymptotic Analysis of Differential Equations with Random Perturbations A. M. Samoilenko & O. Stanzhytskyi Volume 79: Robust Chaos and Its Applications Z. Elhadj & J. C. Sprott Volume 80: A Nonlinear Dynamics Perspective of Wolfram’s New Kind of Science (Volume V) L. O. Chua Volume 81: Chaos in Nature C. Letellier Volume 82: Development of Memristor Based Circuits H. H.-C. Iu & A. L. Fitch Volume 83: Advances in Wave Turbulence V. Shrira & S. Nazarenko Volume 84: Topology and Dynamics of Chaos: In Celebration of Robert Gilmore’s 70th Birthday C. Letellier & R. Gilmore Volume 85: A Nonlinear Dynamics Perspective of Wolfram’s New Kind of Science: (Volume VI) L. O. Chua Volume 86: Elements of Mathematical Theory of Evolutionary Equations in Banach Spaces A. M. Samoilenko & Y. V. Teplinsky Volume 87: Integral Dynamical Models: Singularities, Signals and Control D. Sidorov Volume 88: Wave Momentum and Quasi-Particles in Physical Acoustics G. A. Maugin & M. Rousseau Volume 89: Modeling Love Dynamics S. Rinaldi, F. Della Rossa, F. Dercole, A. Gragnani & P. Landi *To view the complete list of the published volumes in the series, please visit: http://wwwworldscientificcom/series/wssnsa Lakshmi - Modeling Love Dynamics.indd 1 Rational Econometric Man Rational Econometric Man Transforming Structural Econometrics Edward J. Nell Malcolm B. Smith Professor of Economics, The New School for Social Research, USA Karim Errouaki Special Advisor, The Foundation for the Culture of Peace, Spain With a Foreword by Lawrence R. Klein, Nobel Laureate in Economics Edward Elgar Cheltenham, UK • Northampton, MA, USA © Edward J. Nell and Karim Errouaki 2013 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical or photocopying, recording, or otherwise without the prior permission of the publisher. Published by Edward Elgar Publishing Limited The Lypiatts 15 Lansdown Road Cheltenham Glos GL50 2JA UK Edward Elgar Publishing, Inc. William Pratt House 9 Dewey Court Northampton Massachusetts 01060 USA A catalogue record for this book is available from the British Library Library of Congress Control Number: 2010930124 This book is available electronically in the ElgarOnline.com Economics Subject Collection, E-ISBN 978 1 84980 962 7 ISBN 978 1 84980 154 6 (cased) Typeset by Servis Filmsetting Ltd, Stockport, Cheshire Printed and bound in Great Britain by T.J. International Ltd, Padstow Contents Foreword by Lawrence R. Klein vi Acknowledgments viii Introduction xvii PART I FROM RATIONAL ECONOMIC MAN TO RATIONAL ECONOMETRIC MAN 1 Re- reading Hollis and Nell 3 2 Haavelmo reconsidered as rational econometric man 35 3 Induction and the empiricist account of general laws 61 4 Variables, laws and induction I: are there laws of nature? 79 5 Variables, laws and induction II: scientific variables and scientific laws in economics 111 6 The concept of the ‘model’ and the methodology of model building 151 PART II THE CRITIQUES AND THE FOUNDATIONS 7 Debating the foundations: a new perspective? 189 8 Scientific issues in structural econometrics 251 9 Haavelmo and beyond: probability, uncertainty, specification and stochasticism 291 PART III STRUCTURAL ECONOMETRICS IN ITS PLACE: MAPPING NEW DIRECTIONS 10 Conceptual analysis, fieldwork and the methodology of model building 353 11 Working with open models: lawlike relations and an uncertain future 401 Conclusion 484 References 489 Index 523 v Foreword The New School For Social Research has played an important role, contributing to many fields of advanced study in the United States – for example, a recent conference on the work of Franco Modigliani in rela- tion to the prevailing economic situation. This book starts from much earlier work at the New School, the early 1940s Seminar of Marschak and Haavelmo on econometrics, which laid the foundations for the work at Cowles. Nell and Errouaki have written a very welcome book, coming at a good time. Its message is one of support for the original Cowles approach, agreeing that our work then was on the right track. They correctly under- stand the intention of the founders, which was to bring mathematics together with economic theory, so as to develop precise hypotheses that could be confronted with data, using methods of statistical inference. The idea was to expand and develop economic theory, making it more realistic, so that it could be put to use to solve some of the world’s problems. We felt we had all the answers from a statistical point of view and from the point of view of econometric methodology and economic content; we could make it easy to have a well- organized, well- run economy after the war. It was generally expected that econometric investigations would build up a large body of agreed- upon findings, and that different investigators would normally replicate – or improve on – each other’s results. In fact agreement has been hard to come by; what Jacob Marschak very early on called the ‘model selection’ problem has stood in the way. Statistical inference alone will not do the job; but it is not necessarily a step forward to try to solve the difficulties by introducing hard-t o- justify assumptions – normality in probability distributions, ergodicity in time series. What is needed is greater realism, closer and more systematic atten- tion to what economic agents are actually thinking, planning and doing. I have advocated the use of survey data; the authors here call for fieldwork and drawing on vernacular knowledge. When the big models, along with every other form of economic inves- tigation, ran into trouble in the 1970s, many investigators turned against the approach. Nell and Errouaki rightly deplore this; structural econometrics got a lot of things right, and presented a reasonable picture of vi Foreword vii the macroeconomy. People have said that the models failed to predict the effects of supply shocks on the inflation of the 1970s, and that they didn’t predict the changes in structure. I believe the economy didn’t change in structure; instead exogenous inputs changed a great deal within a largely unchanged structure. And the large- scale models did a good job of predict- ing recession and inflation. This book is massive; it covers a great deal of ground, starting from philosophy of science, extending to methodology, and foundations of probability and statistical inference. It then goes on to the basics of structural econometrics and the Cowles approach, especially Keynesian econometric models, and finally covering the critiques of the Cowles approach and Keynesian econometrics, including the critiques of those critiques. The book also presents a number of the authors’ own contributions. These include their proposal to overcome the problem of induction and establish the existence of lawlike regularities in economics, justifying the assumption of a ‘data generating mechanism’; this leads them to their methodological triangle- circle (MTC) diagram, which summarizes their methodology. In addition to methodology they propose some specific modelling – for example, in regard to wage-p rice spirals, the analysis of money supply and demand, Keynesian uncertainty, and Minskyian financial instabil- ity. These ideas may seem unorthodox in today’s context; but they would not have seemed out of place to many of the early econometricians, for example at the Oxford Institute of Statistics. In developing econometric models some people became slaves of the neo-C lassical behavioural for- mulations; in their fear of being ‘ad hoc’ they chose theoretical lines which were not always well conceived. Many have forgotten, if they ever knew, the lessons of Keynes. Our authors propose to correct this, drawing on their program of fieldwork and conceptual analysis, and suggest some concrete steps along the path to reconceptualizing difficult and controver- sial areas of macro theory. The authors have succeeded in orchestrating a lively debate over the scientific foundations of structural econometrics. Their book deserves a broad readership. Lawrence R. Klein Gladwgne, PA, USA Acknowledgments HOW IT ALL BEGAN IN NELL’S OFFICE AT THE NEW SCHOOL A little over 25 years ago a graduate student, who had previously studied mathematics, econometrics and philosophy of science in Paris and had worked at INSEE, turned up in my office. He had just finished his studies with us, with perfect grades, and he wanted to work on a doctoral thesis. He had a straightforward project in mind: to rewrite Hollis and Nell’s Rational Economic Man (1975). It was a great book, he said, but focused on the wrong target. It was not so much economic theory that was distorted and undermined by the assumption of rational economic agents as it was econometrics. That was where the real problem lay. The arguments should be adapted and redirected before econometrics got lost any further in a morass of misspecifications and unrealistic assumptions. Quite a project! But Karim was persistent and I decided he was right. So we began work, and now Rational Economic Man (it was always ‘man’; feminist economics has generally been free of ‘rational’ fundamentalism) has become Rational Econometric Man. It took a long time, but we hope it’s worth it! INFLUENCES AND PATHS THAT LED TO THIS BOOK Professor Edward Nell was the principal PhD thesis advisor and long- standing mentor of Karim Errouaki, along with his two supervisors and mentors, the late Nobel Laureate Wassily Leontief and the late Professor Camilo Dagum at the New School (NY) in the late 1980s. Errouaki owes a great debt to his learned professor and humane friend Professor Nell for his constant help, critical guidance and generous encouragement; and is greatly indebted to all his mentors during the writing of his doctoral thesis and he thanks them for sharing some of their reflections on the scientific standing of econometrics. They were always open to his questions and ideas about econometric methodology, were a constant stimulus for his own thinking in economics, and gave him confidence that viii