Infrastructure and Firm Dynamics A micro-simulation approach Michiel de Bok Infrastructure and Firm Dynamics A micro-simulation approach Proefschrift ter verkrijging van de graad van doctor aan de Technische Universiteit Delft op gezag van de Rector Magnificus prof.dr.ir. J.T. Fokkema voorzitter van het College voor Promoties, in het openbaar te verdedigen op donderdag 3 mei 2007 om 15.00 uur door Michiel Adriaan DE BOK civiel ingenieur geboren te Leerdam Dit proefschrift is goedgekeurd door de promotoren: Prof. ir. F.M. Sanders Prof. dr. ir. P.H.L. Bovy Toegevoegd promotor: Dr. M.C.J. Bliemer Samenstelling Promotiecommissie: Rector Magnificus, voorzitter Prof. ir. F.M. Sanders, Technische Universiteit Delft, promotor Prof. dr. ir. P.H.L. Bovy, Technische Universiteit Delft, promotor Dr. M.C.J. Bliemer, Technische Universiteit Delft, toegevoegd promotor Prof. A. Daly, University of Leeds, UK Prof. dr. H.J.P. Timmermans, Technische Universiteit Eindhoven Prof. dr. F.G. van Oort, Universiteit Utrecht Prof. dr. G.P. van Wee, Technische Universiteit Delft TRAIL Thesis Series nr. T2007/5, The Netherlands TRAIL Research School TRAIL Research School P.O. Box 5017 2600 GA Delft The Netherlands Telephone : + 31 (0) 15 27 86046 Telefax : + 31 (0) 15 27 84333 E-mail : [email protected] ISBN 978-90-5584-086-1 Keywords: infrastructure, firm-demography, urban economics, micro-simulation Copyright © 2007 by Michiel de Bok All rights reserved. No part of the material protected by this copyright notice may be reproduced or utilised in any form or by any means, electronic or mechanical, including photocopying, recording or by any information storage and retrieval system, without written permission of the author. Printed in The Netherlands “When you really want something to happen, the whole universe conspires so that your wish comes true”. The alchemist (in the form of Melchizedek) says to Santiago, in “The Alchemist” by Paulo Coelho Preface Now, I guess it is common sense that transport infrastructure and urban plans have a strong and long lasting impact on the urban environment. The opening of a new highway, generates new traffic and attracts urban development. But why did I decide to try to make a simulation model that tries to predict such impacts? Answering this question only with a juvenile interest in Lego and the SimCity computer game would not do justice to the deeper motivation. During my education as a Civil engineer I discovered I was more interested in the “do we need an extra bridge?” question than the “what would be the dimensions of the pillars of this bridge?” question. Answering the first strategic question is far from straightforward and requires skills from many disciplines: transportation, econometrics, regional economics, policy management, and a healthy dose of common sense. During the past years I was lucky to be given the opportunity to apply the practical intuition of an engineer to a more strategic problem, that is spatial planning. I am convinced that the planning practice could benefit significantly from advanced simulation models to evaluate planning or policy alternatives. And before I discuss my ideas on such models, I will express my gratitude to a number of people that made a valuable contribution to this thesis. First, my promoter Frank Sanders, who provided me the opportunity to conduct a PhD study, in which I could develop my research skills. He made an crucial contribution by directing me towards an approach that has great potential for policy assessments: micro-simulation. Next, I owe much gratitude to Piet Bovy for reviewing many papers and draft model specifications in my research. I realize how important this supervision has been in the completion of this thesis. His ability to identify crucial elements and to suggest solutions show great experience and knowledge. Finally, I benefited much from the supervision of Michiel Bliemer who helped in particular on the mathematical formulations in research papers and my thesis. Next, the chosen micro-simulation approach would not have been possible without the co-operation of Frank van Oort from the Spatial Planning Agency. He enabled me to realize my research approach by providing an enormous amount of firm-data. Our discussions on spatial or locational externalities and how these should be implemented in urban simulation models have been very valuable. Also I would like to thank Leo van Wissen from the NIDI, who helped me with valuable feedback on specific firm- demographic components in my model. Apart from my affiliation as a PhD, I benefited a lot from working as an analyst in a number of research companies. With my colleagues at Significance, and previously RAND Europe, I am experiencing how rewarding it is to apply state-of-the-art techniques to actual planning questions. When I started my PhD I still worked as an analyst with ABF Research in Delft. I thank many of my former colleagues there for learning me to work with large datasets, to program simulation models and showing me that this can be fun as well. At the university I had much fun with a number of colleagues. I realize my list is not complete but I should at least thank Berry, Barry, Sander, Robert, Geertje and Olivier for all the coffee breaks and lunches. In particular Berry, thank you for all the additional cultural trips on international conferences: the Arctic Circle, Roque Noblo and New VII VIII Infrastructure and Firm Dynamics York are by far better experiences than any scientific session in the backroom of an uninspiring academic building. Ylva, thank you for all the tastefull and joyfull home- cooked diners you managed for the department from the first moment I joined our department until my last PhD-year. Unfortunately there is no opportunity to thank you in person but the warm corner you created at our department was inspiring. The last words of gratitude go out to my friends and family for being patient with me at the times that I was unavailable, in particular in mental terms. Your presence during these years, and the moments I could no other than surrender to happiness, joy and sometimes sadness, were crucial for keeping me going. And lastly I want to express my admiration for Leonie, and thank her for pulling me out of the shadow of this thesis once and a while. Michiel de Bok Haarlem, Maart 2007 Table of contents PREFACE...............................................................................................................................................VII NOTATION..........................................................................................................................................XIII 1 INTRODUCTION............................................................................................................................1 1.1 CONTEXT AND BACKGROUND.....................................................................................................1 1.2 RESEARCH OBJECTIVES..............................................................................................................3 1.3 CONTRIBUTIONS.........................................................................................................................4 1.4 THESIS OUTLINE........................................................................................................................5 BACKGROUND 2 PROBLEM ANALYSIS...................................................................................................................7 2.1 INTRODUCTION..........................................................................................................................7 2.2 THE DOMAIN OF URBAN PLANNING............................................................................................8 2.3 INTEGRATED LAND USE AND TRANSPORT MODELS...................................................................12 2.4 FIRM LOCATION IN INTEGRATED MODELS.................................................................................16 2.5 CONCLUSIONS FOR RESEARCH..................................................................................................18 3 THEORIES ON FIRM LOCATION AND TRANSPORT.........................................................21 3.1 INTRODUCTION........................................................................................................................21 3.2 CLASSICAL THEORIES OF FIRM LOCATION AND TRANSPORT.....................................................22 3.3 AGGLOMERATION IN URBAN ECONOMICS.................................................................................27 3.4 MEASURES OF ACCESSIBILITY AND AGGLOMERATION..............................................................31 3.5 HYPOTHESES ON FIRM LOCATION, TRANSPORT AND AGGLOMERATION....................................35 3.6 CONCLUSIONS..........................................................................................................................37 4 FIRM-DEMOGRAPHIC MICRO-SIMULATION.....................................................................39 4.1 INTRODUCTION........................................................................................................................39 4.2 FIRM-DEMOGRAPHY.................................................................................................................40 4.3 MICRO-SIMULATION OF THE FIRM POPULATION.......................................................................43 4.4 CONCLUSIONS..........................................................................................................................50 IX X Infrastructure and Firm Dynamics THEORETICAL MODEL 5 DESIGN OF A SPATIAL FIRM-DEMOGRAPHIC MICRO-SIMULATION MODEL........53 5.1 INTRODUCTION........................................................................................................................53 5.2 GEOGRAPHICAL SCALE LEVELS................................................................................................54 5.3 FRAMEWORK OF THE SFM MODEL...........................................................................................55 5.4 FIRM-DEMOGRAPHIC COMPONENTS..........................................................................................57 5.5 SPATIAL ATTRIBUTES...............................................................................................................66 5.6 CONCLUSIONS..........................................................................................................................69 6 DESIGN OF THE LOCATION CHOICE MODEL...................................................................71 6.1 INTRODUCTION........................................................................................................................71 6.2 THE LOCATION DECISION OF FIRMS..........................................................................................72 6.3 DISCRETE CHOICE THEORY.......................................................................................................75 6.4 THE LOCATION CHOICE MODEL FOR FIRMS...............................................................................77 6.5 CONCLUSION............................................................................................................................83 CASE STUDY 7 DESCRIPTION OF THE CASE STUDY SOUTH HOLLAND.................................................85 7.1 INTRODUCTION........................................................................................................................85 7.2 GEOGRAPHY OF THE STUDY AREA............................................................................................86 7.3 DATA.......................................................................................................................................88 7.4 MODEL ARCHITECTURE............................................................................................................91 7.5 ACCESSIBILITY IN THE STUDY AREA.........................................................................................93 7.6 AGGLOMERATION IN THE STUDY AREA....................................................................................96 7.7 CORRELATION BETWEEN ATTRIBUTES....................................................................................100 7.8 CONCLUSIONS........................................................................................................................102 8 ESTIMATION OF THE FIRM-DEMOGRAPHIC MODELS................................................103 8.1 INTRODUCTION......................................................................................................................103 8.2 ESTIMATION RESULTS OF THE RELOCATION PROBABILITY MODEL..........................................104 8.3 ESTIMATION OF THE FIRM LOCATION CHOICE MODEL.............................................................106 8.4 ESTIMATION RESULTS FOR FIRM GROWTH BY INDUSTRY SECTOR...........................................117 8.5 ESTIMATION RESULTS FOR FIRM DISSOLUTION MODEL...........................................................124 8.6 FIRM FORMATION...................................................................................................................125 8.7 CONCLUSIONS........................................................................................................................127 9 VALIDATION OF THE SFM MODEL.....................................................................................131 9.1 INTRODUCTION......................................................................................................................131 9.2 DESCRIPTION OF INPUT DATA AND OBSERVATIONS................................................................132 9.3 VALIDATION OF POPULATION AGGREGATES...........................................................................135 9.4 EXAMPLES OF PREDICTION RESULTS AT MICRO LEVEL...........................................................136 9.5 VALIDATION OF SIMULATED EMPLOYMENT BY ZONE.............................................................138 9.6 CONCLUSIONS........................................................................................................................142