PROJECT APPRAISAL AND MACROECONOMIC POLICY Project Appraisal and Macroeconomic Policy TSJALLE VAN DER BURG Rotterdam School 01 Management, Erasmus University, Rotterdam, The Netherlands Kluwer Academic Publishers Dordrecht / Boston / London Library of Congress Cataloging-in-Publication Data Burg, Tja11e van der. Project appraisal and macroeconomic poliCy / by TSjal1e van der Burg. p. cm. Inc 1u des bl b 11 ograph i ca I references. ISBN-13: 978-94-010-6503-0 (hardcover alk. paper) 1. Economic pol lcy--Econometric models. 2. Equilibrium (Economics)--Mathematlcal models. I. Title. HD87.B87 1996 95-40137 ISBN-13: 978-94-010-6503-0 e-ISBN-13: 978-94-009-0033-2 DOI: 10.1007/978-94-009-0033-2 Published by Kluwer Academic Publishers, P.O. Box 17, 3300 AA Dordrecht, The Netherlands. Kluwer Academic Publishers incorporates the publishing programmes of D. Reidel, Martinus Nijhoff, Dr W. Junk and MTP Press. Sold and distributed in the U.S.A. and Canada by Kluwer Academic Publisbers, 101 Philip Drive, Norwell, MA 02061, U.S.A. In all other countries, sold and distributed by Kluwer Academic Publishers Group, P.O. Box 322, 3300 AH Dordrecht, The Netherlands. Printed on acid-free paper All Rights Reserved © 1996 Kluwer Academic Publishers Softcover reprint of the hardcover I st edition 1996 No part of the material protected by tbis copyright notice may be reproduced or utilized in any form or by any means, electronic or mecbanical, inc1uding photocopying, recording or by any information storage and retrieval system, witbout written permission from the copyright owner. Table of contents Foreword 1X Acknowledgements xi 1. Introduction 2. The macroeconomic effects of specific portfolios of projects 5 2.1. Introduction 5 2.2. The macroeconomic effects ofprojects which lead to savings on fuels and other raw materials 5 2.2.1. The essence of the analysis 5 2.2.2. Counterbalancing effects related to an increase in spending 10 2.2.3. Effects on financial markets 11 2.2.4. Savings on non-imported fuels 14 2.2.5. The results of quantitative studies 16 2.2.6. The effects of a policy of saving on fuels and other raw materials at the level ofthe European Union 20 2.2.7. Conclusions 23 2.3. The effects of some portfolios of projects on inflation 24 2.4. Conclusions 26 Appendix: A simple, quantitative analysis ofthe effects of a policy of saving on fuels at the level ofthe European Union 26 Notes 29 3. Existing methods for the social appraisal ofprojects in situations of unemployment 31 3.1. Introduction 31 3.2. Methods without a welfare-theoretical foundation 34 3.2.1. Primal methods based on an input-output model 34 3.2.2. Primal methods based on an econometric macro-model 36 3.2.3. A dual method based on an econometric macro-model 38 3.3. Methods with a welfare-theoretical foundation 40 3.3.1. Dual methods 40 3.3.2. Primal methods based on a theoretical model 48 3.3.3. A primal method based on an econometric macro-model 49 3.4. The consistency between project appraisal and macroeconomic analysis and policy 51 v vi Table 0/ contents 3.5. Implications for the development of an improved method 55 Notes 56 4. A method to calculate and evaluate the macroeconomic effects of projects, including small projects 57 4.1. Introduction 57 4.2. ~The calculation of the effects of a project 58 4.2.1. Introduction 58 4.2.2. The effects of financing the expenditures 59 4.2.3. The effects ofthe expenditures excluding the effects offinancing them 61 4.2.4. The effects ofthe output 64 4.2.5. Additional remarks 66 4.3. Potential eITors related to the estimates when large projects or port- folios of projects are analysed 67 4.3.1. Introduction 67 4.3.2. EITors related to project-specific characteristics 67 4.3.3. EITors related to the estimate of the effects of a project that leads to specific durable changes 68 4.3.4. EITors related to the estimates ofthe effects of other projects 69 4.3.5. Final remarks 72 4.4. Potential eITors related to the estimates when small projects are analysed 72 4.4.1. Introduction 72 4.4.2. The concept of risk-pooling 73 4.4.3. One ofthe problems related to the principle ofproportionality, in the context of a hypothetical setting 75 4.4.4. The principle ofproportionality, considered from a more general and realistic point ofview 78 4.4.5. The third step of the method 81 4.4.6. The justification ofthe method when it is used to analyse one project, or a few projects only 81 4.4.7. A common misconception 85 4.5. The advantages ofthe method ofthis study as compared with the conventional macroeconometric method 86 4.6. The evaluation ofthe effects 88 Appendix: The treatment of indirect taxes 92 List of symbols 93 Notes 95 5. The calculations done with the Freia model 97 5.1. Introduction 97 5.2. The Freia model 97 5.3. The base-run 102 5.4. The calculation ofthe H, and the V, 102 5.5. The extent to which the calculations fulfil the conditions of additivity 103 Table ofc ontents vii 5.6. Tbe differences between the effects ofincreases in government investment and development aid in different years 107 5.7. The period to which the analysis refers 112 5.8. Comparison of the calculations made for this study with calculations by the Central Planning Bureau 112 5.9. Evaluation ofthe errors 113 Notes 115 6. The macroeconomic effects ofwind energy in the Netherlands 117 6. 1. Introduction 117 6.2. Characteristics ofthe project from a microeconomic point ofview 118 6.2.1. Introduction 118 6.2.2. Costs and benefits from a private point ofview 119 6.2.3. Modifications needed for the calculation ofthe macroeconomic effects 121 6.2.4. The private profitability ofthe project 122 6.2.5. InpuH:mtput analysis 122 6.3. The macroeconomic effects ofthe project 124 6.3.1. The macroeconomic effects 124 6.3.2. Evaluation ofthe effects 127 6.3.3. Sensitivity analysis with regard to the microeconomic characteristics 130 6.3.4. The dependency ofthe outcomes on macroeconomic factors 132 6.4. Conclusions 134 Notes 135 7. Converting the method into a dual method 137 7.1. General outline ofthe method 137 7.2. Shadow prices based on the Freia model 139 7.3. Conclusions 145 Notes 145 8. International aspects and the debate about import substitution 147 8.1. Effects on the world economy 147 8.2. Some comments on the debate about the effects of import substitution on the domestic economy 150 Notes 152 9. Summary and conclusions 155 References 163 Index 167 Foreword Environmental economics addresses the issues that arise on the boundaries between economic systems and natural systems, such as pollution and natural resource de pletion and degradation. Like any other branch of applied economics, it has drawn its tools and techniques from the wide range already available in economics gener ally, selecting, adapting and extending these to meet its own particular requirements in its own particular context. Here, as elsewhere in economics, public policy analy sis requires quantitative assessments of the economic impact of different policy choices. Perhaps the most distinctive contribution of environmental economics has been the development of techniques for the economic valuation of environmental goods and services in the absence of markets for such goods and services, or in the presence of markets that are at best imperfect or incomplete. Nevertheless policy analysis still relies on one or another of three broad groups of methods used in eco nomics generally. One is project appraisal, which at the micro level provides an evaluation ofthe costs and benefits ofinvestment options to inform the choice among them, while at the macro level policy analysis rests either on computable general equilibrium (CGE) models or on economy-wide macroeconometric models. All models are simplifications, designed to focus attention on the important fea tures ofthe problem at hand, and neglecting other features that might for a different problem assume greater importance. Project appraisal typically takes the macro en vironment as given, and focuses on individual projects in a partial equilibrium setting. In contrast, when aggregate macroeconometric models are used to assess the consequences of an increase in public expenditure, no attention is typically paid to the nature ofthe projects that might be funded by the incremental expenditure. CGE models provide greater sectoral detail, having focused in early applications on trade and public finance questions. With a foundation in Walrasian general equilibrium theory, they are essentially comparative static models, in which policy exercises rest on a comparison of a 'base' solution and a 'new policy' solution, each comprising a set of relative prices that support flow equilibria in all product and factor markets. The nature of the adjustment process that might lead from one equilibrium to an other is not addressed. On the other hand, dynamic adjustment in disequilibrium situations has been a focus of attention in macroeconometric models, and issues of inflation. unemployment, and balance-of-payments disequilibrium are high on the agenda. ix x Foreword Each approach in this stereotype form has obvious limitations, and the practical methods have been adapted to overcome them. Project appraisal in a partial equilib rium context, in which the markets for the project's inputs and outputs are ana lyzed in isolation, gives misleading guidance if other markets are affected in important ways. Shadow prices that take account of general equilibrium interactions might then be generated by a CGE model. Such models in turn might require modi fication or extension if their neoclassical assumptions are at variance with the be havioural or institutional rigidities found in many economies. However, ad hoc changes designed to increase the realism ofthe model may be difficult to interpret in the received theoretical framework; on the other hand a theoretically pure model is unlikely to yield reliable policy guidance in a situation where its assumptions are not satisfied. There is always tension between theoretical purity and empirical accuracy, and macroeconometric models have tended to emphasize the latter, having been concerned to provide a good statistical description of the actual economy, market imperfections, non-optimizing behaviour, and all. In-this case recent developments are in the opposite direction to that of the CGE developments noted above, namely towards a greater theoretical coherency, particularly in respect ofthe model's steady state or long-run implications, as economic theories heavily based on equilibrium concepts are more relevant to these than to the process of short-run adjustment. Environmental policy studies typically require a longer horizon than the macro economic policy analysis of everyday political debate, hence such modifications make macroeconometric models more useful to the environmental economist. Exciting developments in quantitative methods occur on the interface between exist ing methods, when complementary strengths overcome individual weaknesses. At the rnacro level an input-output model might be incorporated into a macroeconometric model to provide missing sectoral detail, or a macroeconometric model might be used to pro vide the macroeconomic closure ofa CGE model. The interface between rnicroeconornics and macroeconornics, with which this book is concerned, is a more difficult one. Al though the use of a CGE model to generate shadow prices for use in project appraisal, as mentioned above, provides an example, the author argues that the theoretical character istics of the CGE approach are at variance with the practical airns of project appraisal. Instead, he seeks to integrate project appraisal and rnacroeconornic policy analysis by using macroeconometric models for the evaluation of small projects in a situation of disequilibrium. The method is applied to a wind energy project, and it is estimated that such projects and a number of sirnilar projects have positive macroeconornic effects. I was honoured to serve as an external examiner of the dissertation on which this book is based and, coming to this work from a background in rnacroeconometric modelling, I found many challenging and provocative ideas. As a contribution to the discussion oft he macroeconornic effects of environrnental policy and to the development of quantitative methods more generally the book now deserves a wide readership. Kenneth F. Wallis ESRC Macroeconomic Modelling Bureau University of Warwick Acknowledgements This book is a revised version ofmy dissertation, which I wrote when I was working at the Faculty ofEconomics in Groningen. In writing the dissertation and the book I have received aid and encouragement from a number of people. I would like to express my thanks to the following people in particular: Professor A. N entjes for his supervision, his many stimulating comments, and bis patience in dealing with my intuition. Professor J .A. Kregel for bis useful comments and his confidence in the project in its early stages, wbich made it possible to get started. Professor J. Muysken and Professor K.F. Wallis for giving me some critical and helpful comments in the last stage of the research. Bemard ter Haar has done the calculations with the Freia-model, and wrote the main part of the computer program for calculating the effects of the wind energy project. I have been lucky in having an assistant with very high scientific standards. I would further like to express my thanks to Diana Galjaard-Vlijenhoek and Theresa Hekman-Huizinga for typing parts ofthe manuscript. Without the good library ofthe Faculty ofEconornics in Groningen it would have been rather difficult to do research, and I am grateful to Mr. H.A. Woensdregt and his colleagues for their friendly way of helping me. Finally, I would like to thank Caroline Cost Budde, Ellen Koen, and Maaike van der Rijst for improving my English grammar. xi CHAPTER I Introduction It is curious how common sense, wriggling for an es cape from absurd conclusiops, has been apt to reach a preference for wholly 'wasteful' forms ofloan expendi ture rather than for partly wasteful forms, which, be cause they are not wholly wasteful, tend to be judged on strict 'business principles'. (J.M. Keynes) In the present western world there is a lack of consistency between proj ect appraisal and macroeconomic policy. Project appraisal takes, of course, the specific charac teristics of projects into account. In most cases, it is based on the assumption of fuH employment. In a few cases, it takes account of the existence of unemployment. However, as will be explained later in this study, in these cases it is nearly always based on a partial analysis, or at least on an analysis which is not consistent with the macroeconomic analyses used for decisions about macroeconomic policy. Macroeconomic policy refers to policy decisions on aggregate variables (total demand, the money stock, the exchange rate). In many cases it is based on an inte gral analysis ofthe economy that takes account ofthe existence ofunemployment. A restriction of such an analysis is that specific factors at a lower level of aggregation tend to be neglected. F or instance, many studies about the rnacroeconomic effects ofKeynesian policies concerning increases in government expenditures are based on the implicit as sumption that there is only one type of government expenditure. A number of economists have combined macroeconomic analysis with analyses at a more disaggregated level in order to investigate the effects of specific portfolios of government projects, or of other mesoeconomic policies. Some ofthese analyses will be discussed in this study. At this point it can be said that despite these analyses there is still a lack of integration between project appraisal and macroeconomic policy. This study is concerned with situations ofunemployment. It is based on the idea that, ifunemployment exists, fuH employment cannot always be (quickly) reached by using conventional macroeconomic policies, or by supply-side policies aiming to improve the free functioning ofmarkets, or by a combination ofboth. Given this starting point, the lack of integration between project appraisal and macroeconomic policy could lead to a non-optimal total policy. The selection of individual public projects is generaHy not based on the criterion that they should I