101,99 €
Mehr erfahren.
- Herausgeber: John Wiley & Sons
- Kategorie: Fachliteratur
- Sprache: Englisch
From the Foreword by Dr Valmond Ghyoot, Emeritus Professor of Real Estate, University of South Africa:
‘The valuation profession, the legal profession, property industry participants in general and students will welcome publication of this book. Investors, environmental groups and affected property owners will find essential information for use in their decision-making, development objections and claims. My hope is that [it] will provide answers where required and that it will help to improve the professional standard of valuations and appraisals internationally. I trust that it will also raise the standard of testimony in damages cases. If so, the editors and contributors will have succeeded in documenting the state of the art in this relatively unexplored terrain.’
As a reference source, this book will help quantify the negative impacts on property values of high voltage overhead transmission lines, cell phone towers, and wind turbines. It gives a modern perspective of the concerns property owners have about the siting of industrial structures used to transmit or generate various forms of energy and how these concerns impact on property values.
Studies reveal concerns the public have about devices and structures that emit electromagnetic fields (EMFs) due to their potential health hazards. . Despite some research reports suggesting there are no potential adverse health hazards from high voltage overhead transmission lines (HVOTLs) and towers, there is still on-going concern about the siting of these structures due to fears of health risks from exposure to EMFs, changes in neighbourhood aesthetics and loss in property values. The siting of wind turbines is also receiving community opposition due to noise, light flicker, aesthetic concerns, and loss in property values. The extent to which such attitudes are reflected in lower property values is not well understood.
Towers, Turbines and Transmission Lines: Impacts on Property Value outlines results of studies conducted in the US, the UK, Australia and New Zealand and offers guidance to valuers as well as to property/real estate appraisal students and property owners around the world. The book provides defensible tools that are becoming widely accepted to assess the effect that these environmental detriments have on property prices.
Sie lesen das E-Book in den Legimi-Apps auf:
Seitenzahl: 560
Veröffentlichungsjahr: 2013
Ähnliche
Contents
About the Editors and Contributors
Foreword
1 Introduction
1.1 Valuation and Environmental Attributes
1.2 Risk and Stigma
1.3 Media Impact
1.4 Methodologies
1.5 Book Structure
2 Methods
2.1 Introduction
2.2 Sales Comparison Method
2.3 Regression Analysis
2.4 Hedonic Modelling
2.5 Spatial Hedonic Modelling
2.6 Qualitative Analysis
2.7 Triangulation
2.8 Conclusions
3 Risk Perception, Stigma and Behaviour
3.1 Introduction
3.2 Risk and its Perception
3.3 Risk Communication
3.4 Risk Behaviour
3.5 Perception and Risk Management
3.6 Property Advice
3.7 Property-related Stigma
3.8 Assessing Stigma
3.9 Property Behavioural Research
3.10 Conclusions
Part I: High-voltage Overhead Transmission Lines (HVOTLs) and House Prices
Introduction
1.1 Introduction
I.2 Residential Property Values near HVOTLs
4 HVOTLs in the UK
4.1 Introduction
4.2 Existing Research
4.3 Barriers to Research in the UK
4.4 Value Impacts in the UK
4.5 Conclusions
4.6 Additional Research
5 HVOTLs in New Zealand
5.1 Introduction: Electricity Distribution and Planning Guidelines
5.2 Health Concerns Relating to Proximity of HVOTLs Which Affect Value
5.3 Background to the NZ Research
5.4 Literature Review
5.5 Case Study Description
5.6 Market Analysis using a Hedonic Housing Model
5.7 An Attitudinal Study of Residents’ Perceptions
5.8 Summary and Conclusions
6 A Review of HVOTL Studies in North America
6.1 A Review of Existing Research
6.2 Hedonic Studies in the USA
6.3 Conclusion
Summary
Part II: Cell Phone Towers
Introduction
II.1 Introduction
II.2 Cellular Phone Systems
III.3 History
II.4 Siting Issues and Public Concerns
7 Cell Phone Towers in New Zealand
7.1 Introduction
7.2 Review of Existing Research
7.3 New Zealand Case Study
7.4 Research Procedure: Opinion Survey
7.5 Research Procedure: Market Study
7.6 Summary and Conclusions
8 Cell Phone Towers in North America
8.1 Introduction
8.2 Locating Cell Sites in the USA
8.3 Case Study Area and Data
8.4 Research
8.5 Summary and Conclusions
9 Cell Phone Towers in the UK
9.1 Introduction
9.2 Types of Cell Towers
9.3 Planning Considerations
9.4 Developing a Framework to Establish the Impact on Value
9.5 UK Research
9.6 European Study
9.7 Conclusions
Summary
Part III: Wind Farms
Introduction
III.1 Wind Energy
III.2 Cost of Building Turbines and Generating Energy from the Wind
III.3 Land Use
III.4 The Growth of Wind Energy
III.5 Planning and Development
III.6 Barriers to Development
III.7 The Need for Research
10 Wind Farms in the UK
10.1 Introduction
10.2 The Growth of Wind Energy in the UK
10.3 Existing Research
10.4 Proposal Objections: Case Studies
10.5 Valuation Research: Cornwall Case Studies
10.6 Conclusions
11 Wind Farms in North America
11.1 Introduction
11.2 Previous Research
11.3 Present Research
11.4 Robustness Tests
11.5 Conclusions
12 Wind Farms in Australia and New Zealand
12.1 Introduction
12.2 Existing Research
12.3 Methodology
12.4 Survey Results
12.5 Summary and Conclusions
Summary
13 Conclusion
13.1 Introductory Section (Chapters 1–3)
13.2 Part I (Chapters 4–6)
13.3 Part II (Chapters 7–9)
13.4 Part III (Chapters 10–12)
13.5 Concluding Remarks
Index
Other Books Available from Wiley-Blackwell
Valmond Ghyoot Ph.D.
The authors and contributors would like to dedicate this book to our friend and colleague Valmond Ghyoot, Ph.D., 61, who sadly passed away peacefully at home surrounded by family on the 29 December 2012. He is survived by his wife, two sons and a daughter.
Valmond is a well respected and admired academic internationally with an extensive career in the construction and real estate industry spanning more than 40 years.
He was the co-founder of the African Real Estate Society (AfRES), one of the regional real estate societies under the International Real Estate Society (IRES) umbrella. In recognition for his services rendered to the AfRES he was awarded the IRES Service Award in 2004. An engaged, thoughtful and supportive colleague and friend, this book was written through his encouragement and mentorship. His understanding of real estate valuation and feasibility as well as the complex nature of the real estate asset will be remembered by many, and are explored in this book.
Valmond will be sorely missed by family, students, and colleagues from around the world.
This edition first published 2013© 2013 Bond, Sims and Dent.
Wiley-Blackwell is an imprint of John Wiley & Sons, formed by the merger of Wiley’s global Scientific, Technical and Medical business with Blackwell Publishing.
Registered OfficeJohn Wiley & Sons, Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK
Editorial Offices9600 Garsington Road, Oxford, OX4 2DQ, UKThe Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, UK2121 State Avenue, Ames, Iowa 50014-8300, USA
For details of our global editorial offices, for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at www.wiley.com/wiley-blackwell.
The right of the author to be identified as the author of this work has been asserted in accordance with the UK Copyright, Designs and Patents Act 1988.
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, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher.
Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought.
Limit of Liability/Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this book, they make no representations or warranties with the respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. It is sold on the understanding that the publisher is not engaged in rendering professional services and neither the publisher nor the author shall be liable for damages arising herefrom. If professional advice or other expert assistance is required, the services of a competent professional should be sought.
Library of Congress Cataloging-in-Publication Data
Towers, turbines and transmission lines : impacts on property value / edited by Sandy Bond, Sally Sims, Peter Dent.p. cm.Includes bibliographical references and index.
ISBN 978-1-4443-3007-6 (cloth)1. Real property–Valuation. 2. Overhead electric lines. 3. Towers. 4. Wind turbines.I. Bond, Sandy. II. Sims, Sally. III. Dent, Peter.HD1387.T69 2013333.33′2–dc23
2012040172
A catalogue record for this book is available from the British Library.
Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books.
Cover design by Meaden CreativeCover image courtesy of iStockPhoto
About the Editors and Contributors
Dr Sandy Bond
Sandy Bond is the Professor of Property Studies in the Commerce Faculty at Lincoln University in Christchurch, New Zealand. She is a Registered Property Valuer and Senior Member of the Property Institute of New Zealand (SPINZ), is the President-elect (2013) of the International Real Estate Society and a past President of the Pacific Rim Real Estate Society (PRRES). She has lived and worked in NZ, the USA, UK and Australia, and her career has encompassed property valuation, valuation consultation, academic research and university teaching. She is the only non-American member of the real estate consulting firm, American Valuation Partners.
Sandy holds a PhD from Curtin University in Western Australia, a Commerce Degree in Valuation and Property Management, a Postgraduate Diploma in Business and an MBS in property from Massey University in New Zealand. She was awarded the Property Institute of NZ Academic Award in 2010 and the PRRES Achievement Award in 2002 for her significant contribution to property education and research.
Sandy’s research is of an applied, multi-disciplinary nature that aims to solve globally significant property issues such as the impact of detrimental conditions (including climate change) on property values. Her current focus is on sustainability of the built environment. However, as a result of the 2011 earthquakes in Christchurch, NZ her research now encompasses the impact of natural disasters on property values.
A frequent presenter at international real estate conferences and seminars, Professor Bond is also the author of numerous articles and monographs published in prominent real estate journals within New Zealand, Australia, Malaysia, the UK and USA. She was responsible for drafting the first PINZ Practice Standard on the Valuation of Contaminated Sites in NZ.
Mr Peter Dent
Peter Dent is a Fellow of the Royal Institution of Chartered Surveyors. He has held various posts at Oxford Brookes University, most recently as the Comerford Climate Change Fellow in the Department of Real Estate and Construction. He has had considerable experience of managing both academic development and research projects both in the UK and overseas.
His research interests range from property valuation to sustainable behaviour and literacy. His co-authored book Real Estate: Property Markets and Sustainable Behaviour was published in 2012. In addition to publishing his research, he has presented his research findings around the world to both academics and practitioners.
Mr Ben Hoen
Ben Hoen is a Principal Research Associate in the Electricity Markets and Policy Group at the Lawrence Berkeley National Laboratory. His primary responsibilities include investigating individual and community responses to a number of different renewable energy sources and assisting in the preparation of the Annual Wind Technologies Report. Ben has conducted a nationwide analysis of the impact of wind facilities on local property values and an analysis of the effect of solar energy systems on home transaction prices in California.
He is a graduate of the Bard Center for Environmental Policy at Bard College in New York with a Master of Science Degree in Environmental Policy. Previous to graduate school he worked in the paper recycling industry in Baltimore, and set up and ran his own business in Brooklyn, NY. He holds Bachelors degrees in Finance and in General Business from the University of Maryland.
Dr Sally Sims
Sally Sims is a senior lecturer in the Department of Real Estate and Construction at Oxford Brookes University. She was a Member of the Stakeholders Advisory Group on EMF and is a member of the Editorial Advisory Board for the International Journal of Housing Markets and Analysis.
Her research expertise is in the use of hedonic modelling and survey-based techniques to establish the impact of environmental features on property and land values. She was recently awarded the Outstanding Paper Award Winner Emerald Literati Network Award for Excellence 2011 for her paper on refurbishing Polish housing stock with Miroslaw Belej and was awarded the 2007 Appraisal Journal Prize for research on wind farms with Peter Dent.
Recently completed projects include: a RICS-funded project into wind farms; client-based research on rural exception sites; an analysis of the real versus perceived impact on house values from the presence of overhead electricity lines; and developers’ attitudes towards microgeneration, in particular, the inclusion of small wind and solar power in residential development.
Dr Elaine Worzala
Elaine Worzala is the Director of the Richard H. Pennell Center for Real Estate Development and a Professor in Real Estate within the College of Architecture, Arts and Humanities at Clemson University in South Carolina. Elaine holds a PhD in Real Estate and Urban Land Economics (1992) and an MS in Real Estate Appraisal and Investment Analysis (1984) from the University of Wisconsin-Madison.
An active member of several academic associations, Dr Worzala has served as the President of both the American Real Estate Society and the International Real Estate Society and is currently serving as a board member for both organisations. With an interest in applied research, Dr Worzala has completed research grants for the Royal Institution of Chartered Surveyors (RICS), the International Council of Shopping Centers (ICSC), Pension Real Estate Association (PREA) and the Real Estate Research Institute (RERI). She has published articles in many international real estate academic journals as well as industry-based publications. She sits on the editorial boards of many of the real estate academic- and industry-based journals. Primary teaching interests include real estate valuation, investments and feasibility analysis.
In 2000, Dr Worzala was awarded the IRES Achievement Award that recognised her outstanding achievement in real estate research education and practice at an international level and was also awarded the Bert Kruijt Service Award for outstanding service to IRES. Dr Worzala is a Fellow of many organisations including: the Weimer School that is part of a real estate think tank devoted to expanding the knowledge of real estate educators; the Royal Institute of Chartered Surveyors; the Institute of Green Professionals; and was the first woman Distinguished Fellow of the National Association of Industrial and Office Properties (NAIOP). In 2006, she was initiated into Lambda Alpha International. Elaine is also a member of the Counsellors of Real Estate, the Pension Real Estate Association and the Urban Land Institute. In 2011 she graduated from the ULI-SC Sustainability Leadership Institute and currently serves on its Advisory Board.
Dr David Wyman
David Wyman is the Associate Director of the Spiro Institute of Entrepreneurial Leadership at Clemson University and a successful entrepreneur with over 50 inventions licensed to toy companies. He has a PhD from the University of Aberdeen in Scotland, an MBA from Cranfield University and a BA (Hons) in Economics from Queens College, Cambridge University.
A passionate teacher, Dave won the Excellence in Teaching award for undergraduate students at Clemson University in 2011 and was selected Professor of the Year by the University of San Diego business students in 2006. His academic publications include articles based on spatial hedonic modelling, valuation and complexity.
Foreword
There has long been a need for a reference source with data and guidelines on how to quantify the value impacts of negative externalities such as towers in close proximity to a parcel of real estate. The effect of such negative externalities on property values is a problem for the affected property owners, who fear that their property investment will suffer, and for valuers, who regularly have to determine their impact. No guidelines are usually available and the valuer is not aware of similarly affected properties. The valuer or appraiser simply has to rely on their own experience or extrapolate from other, dissimilar, cases. This is subjective and the quality of the resulting valuation depends to a large extent on the experience of the valuer. Legal practitioners also need guidelines and facts on which to base their arguments. This text addresses that need. The discussion focuses on towers: electrical, cell phone and wind turbine.
The concept for this book originated when I encountered Professor Sandy Bond at an international real estate conference, delivering a paper on the impact of external influences (externalities) on real estate value and suggested that she write a book on the topic, adding that students and practitioners desperately need such knowledge in a readily accessible form. She listened politely, but did not show any reaction. After similar suggestions at three consecutive annual conferences where she delivered further papers on the topic of quantifying externality impacts on real estate value, she finally took my suggestion seriously. A few months later, she e-mailed me with the news that this book was on its way!
Sandy’s continued interest in the topic stems from having received numerous requests for help over the years. Such requests are usually from affected landowners who wish to quote material from previous studies in their current development objections or legal proceedings, or who require expert testimony. Professor Bond has assembled an international team of scholars and professionals to prepare the text. This gives the reader access to a wide body of experience with examples from New Zealand, Australia, the UK and North America. Together, the authors and contributors have made the book an indispensable resource for industry participants, providing many guidelines and examples of the methodology to be followed in a scientific investigation. The practical emphasis of the text will make it useful for researchers who wish to replicate the studies, and for practitioners who need to analyse the impact of an externality on value or who need arguments to motivate a claim.
The valuation profession, the legal profession, property industry participants in general and students will welcome publication of this book. Investors, environmental groups and affected property owners will find essential information for use in their decision-making, development objections and claims.
My hope is that this book will provide answers where required and that it will help to improve the professional standard of valuations and appraisals internationally. I trust that it will also raise the standard of testimony in damages cases. If so, the editors and contributors will have succeeded in documenting the state of the art in this relatively unexplored terrain.
Dr Valmond GhyootEmeritus Professor of Real EstateUniversity of South Africa
1
Introduction
Peter Dent and Sally Sims
1.1 Valuation and Environmental Attributes
Over the years it has become apparent that valuers1 have experienced problems in determining the impact of high-voltage overhead power transmission lines (HVOTLs), cell towers and wind turbines on the value of proximate residential property. This book attempts to provide the best of current knowledge on the subject, and to lay out options for valuers based on the authors’ own personal research together with studies undertaken by others.
Valuation is a valuer’s opinion of the value of a particular interest in a property, on a specified date for a specified purpose. The process of valuation ‘requires the valuer to make impartial judgements as to the reliance to be given to different factual data or assumptions in arriving at a conclusion. For a valuation to be credible, it is important that those judgements can be seen to have been made in an environment that promotes transparency and minimises the influence of any subjective factors on the process’ (IVSC 2011; now also contained within RICS 2012, p.15).
A valuation is therefore a valuer’s written opinion of a property’s value, taking account of the purpose for which it is required as expressed in the terms of engagement. Investigations and enquiries undertaken to arrive at an appropriate figure will include an examination of the state of the market and analysis of any relevant comparable evidence. The valuer is therefore not simply concerned with the physical property when coming up with a value; the valuer is seeking to interpret a range of data about the property including its structure, size, location and market in order to advise a client with an opinion of value.
Property value is therefore not determined from a straightforward spreadsheet calculation, and it is not neatly incorporated into a rational model. There is a considerable amount of interpretation of facts from client, property, environment and market which has to be undertaken in order to enable the valuer to offer a professional opinion. In such circumstances, ‘a valuation is not a fact, it is an estimate. The degree of subjectivity involved will inevitably vary from case to case, as will the degree of certainty, or probability, that the valuer’s opinion of market value would exactly coincide with the price achieved were there an actual sale at the valuation date’ (RICS 2012, p. 87).
This book sets out to consider specific circumstances where a valuer’s opinion is sought. These circumstances include the impact on the value of homes that are proximate to HVOTLs, cell towers and wind turbine installations. While the impact from these structures on property values has been widely debated and researched, there is still uncertainty around their measurement and quantification. This book aims to provide guidance to valuers (and property/real estate appraisal students) around the world with defensible tools that are becoming widely accepted to assess these impacts whether on freehold owners or leaseholders. The authors have been undertaking research in this area for many years and believe that their experience and that of others who have contributed to this book will assist practitioners and their clients.
The book focuses on HVOTLs, cell towers and wind turbines partly because the siting of these environmental features has become controversial at various times in different countries and has consequently given rise to research. The process of researching these structures and their impacts has also helped to establish a methodology which can be universally applied across a much larger range of environmental factors.
When considering the impact of general locational factors on the value of any real estate development, there are certain overarching criteria which will influence the level of value impact of specific factors. These will range from the nature of the market at any one point in time, geographic location, physical structures, the prevailing sentiment towards these factors and, to some degree, the methodologies used to evaluate the impact of these factors. This introductory chapter therefore attempts to set a general context within which a valuer will have to evaluate the impact of perceived environmental detriments on specific properties.
1.2 Risk and Stigma
The valuation of properties impacted by disamenities is partly affected by the perceptions of risk and stigma of the various stakeholders in the buying/selling process, both of which can be influenced by media coverage. In any valuation exercise it is therefore important to have some understanding of what these two factors are (in the context of the valuation exercise) and how they may be perceived as influencing value.
1.2.1 Risk
One important aspect in considering the impact of HVOTLs, cell towers and wind turbines on individual property values is the level of risk that an individual perceives as existing in a set of circumstances. This will vary according to the parties within prospective markets. For instance, the business sector might not be as concerned about some negative environmental externalities (e.g. HVOTLs) as home buyers. But even a price reduction may not be sufficient to persuade a lending institution to finance the purchase. It may be that the reluctance towards lending on residential units that are underneath or in close proximity to HVOTLs shown by some lending institutions in the UK, for example, is due to the still unknown potential health risks from exposure to residential Electrical Magnetic Fields (EMFs) produced by such structures.
In 1996, the Royal Institution of Chartered Surveyors (RICS) introduced Practice Statement (PS) 3.7 to its Appraisal and Valuation Manual (the Red Book2) instructing valuers that, while there was no clear evidence of a link between living near HVOTLs and a number of adverse health effects, ‘…public perception may, however, affect marketability and future value of the property’(RICS 1995). This PS provided professional guidance to RICS members when undertaking the valuation of property near a HVOTL; however, no additional advice was provided to aid valuers when they determine the likely impact of negative public perception on house price. This Practice Statement has been omitted from the Red Book since May 2003 and, instead, advice to valuers is now included in the RICS Practice Standard Contamination, the Environment and Sustainability (RICS 2010).3 This suggests that, ‘if, when a valuation is carried out, there is a cause for concern over the strength of a field, an appropriate specialist or chartered environmental surveyor should be consulted’ (p. 42). However, unlike PS 3.7, this Practice Standard does not impose a duty on valuers when surveying this type of property to instruct clients that HVOTLs may affect future value due to negative public perception.
1.2.2 Stigma
The effect on property values of stigma from the presence of structures that are the focus of this book – HVOTLs, cell towers and wind turbines – is related to numerous factors. These include:
the type of structure;
the proximity of the structure to the property;
visibility/audibility;
prevailing market sentiment;
ongoing media attention; and
the current state of the property market.
These aspects of both risk and stigma and their meaning and potential impact on the value of individual properties are examined in more detail throughout the book.
1.3 Media Impact
Media attention alone can play a major role in influencing the degree of stigma and risk (real or perceived) associated with property affected by proximity to the above-mentioned structures.
During the mid-1990s, for example, media attention highlighted a potential relationship between living near HVOTLs and childhood cancer. The resulting publicity stigmatised homes near power lines and subsequently had a negative impact on the desirability and value of this type of residential property (Sims and Dent 2005). Similarly, an example of the impact of media attention was evident in the UK in 2000 with regard to cell towers. In this case, the UK Government responded by commissioning an Independent Expert Group on Mobile Phones (IEGMP) to assess the possible health risks. This resulted in the Stewart Report (IEGMP 2000; principally about handsets) which was highly publicised in the media. Subsequently, the public began to object to the siting of cell towers near their homes. This led to a number of cases where phone providers were refused planning permission to erect cell towers due to public concerns. This type of negative media coverage and public response has been mirrored around the world, with the perceived risks of cell phone towers regularly highlighted. For example, public concern was expressed in New Zealand and clauses to address those concerns and ‘protect’ the public were inserted into the Resource Management Act (2008).
With the recent growth in renewable energy production to help combat climate change, public concern has focused on wind turbine location (particularly before and during the construction phase). However, public negativity towards environmental features viewed as ‘unpleasant’ is not a new phenomenon or limited to HVOTLs, cell towers and wind turbines. For example, other environmental features viewed as unpleasant include landfill sites (Clark 2004), household waste incinerators (BBC 2002), opencast mines (McLaughlin 2009), airports (Nelson 2004), power plants (Bobseine 2008) and nuclear facilities (Lean 2009).
The risks and stigma of perceived ‘unpleasant’ land uses near homes can be measured via impacts to property values, which serve as a somewhat unbiased proxy of the true effects on a surrounding community. That notwithstanding, accurately determining the impact on property values of these perceived detrimental conditions and environmental features remains one of the most challenging aspects of property valuation. Previous studies have established that public perception of non-physical contamination such as visual, noise and odour pollution can influence the value and marketability of residential property, especially when there is an association with a possible health risk (e.g. Fischhoff 1985; Slovic 1987; McClelland et al. 1990; Slovic et al. 1991; Krimsky and Golding 1992; Mundy 1992; Chalmers and Roehr 1993; Syms 1996, 1997; Bell 1999; Gallimore and Jayne 1999; Jayne 2000). This in turn has led to uncertainty, fear and, occasionally, a diminution of house prices (e.g. Bond and Beamish 2005; Sims and Dent 2005). These and other studies will be used to explore the issues in more detail in Chapters 4 and 7.
1.4 Methodologies
As market value is a valuer’s opinion as to the price that a buyer would be willing to pay and a seller would be prepared to accept for a property at a specific point in time, when quantifying the degree to which property values are likely to be affected by environmental features it is useful to start by surveying the attitudes of the stakeholders involved, i.e. buyers, sellers, agents and valuers. However, modelling the actual impact on property prices requires a far more complex methodology.
Such methodologies are covered in more detail in Chapter 2 of this book. However, as an introduction it might be useful here to provide an overview of the two accepted methods of assessing value diminution.
The first approach involves surveying market participants. The initial survey approach, known as ‘contingent valuation’, involves the use of interviews or questionnaires and attempts to predict willingness to pay. However, this method only produces a hypothetical valuation based on buyers’ stated preference rather than actual behaviour as evidenced by transaction data; the result is therefore not based on an analysis of real property transactions and as such it is not necessarily an accurate reflection of likely property value impacts. The dichotomy between public opinion and actual behaviour when faced with a real situation has, in the past, been one of the major criticisms of qualitative analysis as a reliable determinant of likely public response to environmental features (e.g. Kroll and Priestley 1992; Slovic 1987; Whitehead et al. 2008). This led to the use of psychometric testing (Slovic 1987) which identified a number of factors or ‘heuristics’ that could account for this dichotomy.4 This technique has been used to explore public and professional perceptions of a variety of environmental factors (Coy 1989; Slovic et al. 1991) including the siting of HVOTLs in proximity to residential property (Arens 1997).
Such surveys have generally been used to determine market sentiment, assess the monetary value impact of a detriment and estimate property values (Mitchell et al. 1993). These attitude studies generally involve surveying market participants about their perceptions or feelings towards particular environmental features and may also ask questions about the perceived impact of such a feature on property value. In the area of the impact of HVOTLs on adjoining residential property, such surveys have generally compared property professionals’ attitudes and opinions (in particular valuers and real estate agents) to those of homeowners in an attempt to determine the likely market resistance from buyers and the degree to which valuers perceived such market resistance would impact on value. While attitudes were generally negative, this approach often highlights differences in perception between residents and professionals (Bond 1995; Dent and Sims 1999; Bond and Hopkins 2000). More recently, discrete choice using conjoint analysis has been adopted to allow for simulation of a hedonic price dataset, but doing so with the use of a survey (e.g. Louviere et al. 2000).
The second approach to assessing value diminution from environmental factors is to analyse actual house sales transaction data using a hedonic pricing model. In outline, this model assumes that the price of a property is determined by a number of key physical characteristics of the house (e.g. house size, number of bedrooms, layout, construction materials, age and condition, etc.) as well as significant characteristics present in the local environment (e.g. accessibility to schools, shops and local amenities; presence and level of any soil, water or air pollution; crime levels, etc). The model is used to estimate the extent to which each factor affects the price. Since each transaction reflects the value placed on the particular set of locational and physical attributes associated with that property, the implicit price placed on each attribute (characteristic) is not observed. Breaking down a property into its main characteristics allows the influence of each attribute on the total price to be determined.
This approach has been used to determine the impacts on property values of many different environmental features including: cell towers (Bond 2007; Filoppova and Rehm 2011); HVOTLs (Priestley and Ignelzi 1990; Bond and Hopkins 2000; Des Rosiers 2002; Sims and Dent 2005); and wind farms (Hoen 2006; Sims et al. 2008; Hoen et al. 2009). It is the accepted method of conducting a robust analysis of the impact of environmental features on house prices where sufficient property-specific data are available (Sims et al. 2009; Kauko 2002; Rossini et al. 2002). These and other studies will be used to examine value impacts throughout this book (e.g. Chapters 4, 7 and 10).
The data required for this type of analysis can be broken down into three categories as follows:
This enables a regression analysis to be undertaken for houses sold ‘near’ and ‘not near’ or with a ‘view’ or ‘no view’ of an environmental feature to calculate statistically the impact of a particular feature on the transaction price. Geographic Information Systems (GIS) is a modern tool that enables actual distance from each house to these features to be calculated and factored into the regression model.
These two approaches – surveying stakeholders and analysis of sales transaction data using a hedonic pricing model – are quite different from the traditional approaches of sales comparison, cost and income approaches used by valuers. The second of the two approaches, hedonic modelling, uses econometrics that provide a sophisticated statistical tool for practical use. This is easily understood in practice and is robust enough to withstand detailed scrutiny under examination.
1.5 Book Structure
Following an overview of the methods used in the book to assess property value impacts (Chapter 2) and an introductory chapter on risk and behaviour to set the context for all of the subsequent studies (Chapter 3), the remainder of the book is divided into three main parts.
Part I focuses on HVOTLs using case studies in the United Kingdom (Chapter 4), New Zealand (Chapter 5) and North America (Chapter 6). The introduction to Part I considers the nature of electromagnetic fields and background information on power distribution and legislative control. Chapters 4–6 provide an overview of how HVOTLs are legally sited on land in different countries. Relevant literature and research studies are used to highlight current thinking on the impact of HVOTLs and towers on adjacent property values.
A similar approach is adopted for Part II, which examines potential impacts on property values of cell phone towers. In this part, case study material from New Zealand (Chapter 7), the United States (Chapter 8) and the UK (Chapter 9) is used to illustrate the issues.
Part III of the book considers wind turbines that are being developed as part of the renewable energy solution to climate change. Case studies in the same three countries (Chapters 10–12) are used to examine the concerns raised by nearby property owners regarding the impacts of wind farm proximity on neighbourhood aesthetics, noise, loss of bird life and reduction in property values. In addition to the main case study countries, other examples from around the world are used to illustrate how such issues are being considered under differing circumstances and legal and cultural systems.
The final chapter attempts to bring together the findings from the range of research studies discussed in this book. It provides both a reflection on the current state of knowledge and emphasises the need to continue to analyse and interpret activities in the market.
A crucial aspect of the book is its practical nature in supporting practitioners and property students to recognise the issues identified and to respond in an appropriate way to provide a professional service to their clients. As such, the authors have provided guidance through models and templates on the ‘how’ to deal with these more complex valuation issues as well as the ‘why’.
Notes
1 The term valuer is used throughout this work to denote those who professionally value property. In different countries different terms are used to describe this activity. The term here is therefore used as synonymous with ‘real estate appraiser’, ‘valuation surveyor’, etc.
2 The majority of residential valuations within the UK are undertaken by Chartered Surveyors (also known as valuers or appraisers) who are also members of the RICS and, as such, have a mandatory requirement to follow the Valuation Standards (VS) within the Red Book. Failure to comply would constitute a breach of the RICS byelaws and regulations. Valuers are further advised to follow the Guidance Notes (GN) as a matter of good professional practice.
3 Similar standards operate in New Zealand: API and PINZ (2009), Australia and New Zealand Property Standards (specifically, the NZ Real Property Guidance Note 1 or NZRPGN 1), Valuation of contaminated land, and, in the US, Appraisal Institute (2012), Guide Note 6 to the Standards of Professional Appraisal Practice, ‘Consideration of Hazardous Substances in the Appraisal Process’. The relevant International Standards Committee Guidance Note is IVGN 7, ‘Consideration of Hazardous and Toxic Substances in Valuation’.
4 Paul Slovic was largely responsible for the use of psychometric testing to determine public perceptions of risk. This has evolved into two distinct paradigms: one based on the behaviour of individuals (psychometric theory) and the other on the behaviour of groups (cultural theory).
References
API & PINZ (2009) NZ Real Property Guidance Note 1 (NZRPGN 1), Valuation of contaminated land. In Australian and New Zealand Valuation and Property Standards. Australian Property Institute, p. 437.
Appraisal Institute (2012) Guide Notes to the Standards of Professional Appraisal Practice of the Appraisal Institute. Available at http://www.appraisalinstitute.org/membership/downloads/cpe/AI_Guide_Notes.pdf. Accessed September 2012.
Arens, S. (1997) The valuation of defective properties: A common sense approach. The Appraisal Journal, April 1997, 143–148.
BBC (2002) Protest over incinerator plans. Available at http://news.bbc.co.uk/2/hi/uk_news/wales/2050101.stm. Accessed September 2012.
Bell, R. (1999) Real Estate Damages: An Analysis of Detrimental Conditions. The Appraisal Institute, Chicago, USA.
Bobseine, B. (2008) Will power plant affect Walpole values. Viewed at http://www.dailynewstranscript.com/news/x478930342 on 2/3/2010.
Bond, S. (1995) The impact of transmission lines on property values. New Zealand Valuers Journal, September, 26–28.
Bond, S. (2007) Cell phone tower proximity impacts on house prices: A New Zealand Case Study. Pacific Rim Property Research Journal, 13(1), 63–91.
Bond, S. & Hopkins, J. (2000) The impact of transmission lines of residential property values: Results of a case study in a suburb of Wellington, NZ. Pacific Rim Property Research Journal, 6(2), 52–60.
Bond, S. & Beamish, K. (2005) Cellular phone towers: perceived impact on residents and property values. Pacific Rim Property Research Journal, 11(2), 158–177.
Chalmers, J.A. & Roehr, S.A. (1993) Issues in the valuation of contamination property. The Appraisal Journal, 7(1), 28–41.
Clarke, R. (2004) Beware of the landfill. The Telegraph, 14th January, available at http://www.telegraph.co.uk/property/newhomes/3320958/Beware-of-the-landfill.html. Accessed September 2012.
Coy, P. (1989) Radiation from VDTs, Power Lines Raises Fears. Register – Guard (Eugene, OR), December 3, 11C.
Dent, P. & Sims, S. (1999) Electricity distribution and residential property values. International Association of Housing Science (IAHS) World Housing Congress, San Francisco, USA.
Des Rosiers, F. (2002) Power lines, visual encumbrance and house values: A microspatial approach to impact measurement. Journal of Real Estate Research, 23(3), 275–301.
Filoppova, O. & Rehm, M. (2011) The impact of proximity to cell phone towers on residential property values. International Journal of Housing Markets and Analysis, 4(3), 244–267.
Fischhoff, B. (1985) Managing risk perceptions. Issues in Science and Technology, Fall, 83–96.
Gallimore, P. & Jayne, M.R. (1999) Public and professional perceptions of HVOTL risks: the problem of circularity. Journal of Property Research, 16(3), 243–255.
Hoen, B. (2006) Impacts of windmill visibility on property values in Madison County, New York. Project report submitted to the Faculty of the Bard Center for Environmental Policy, Annandale on Hudson, New York.
Hoen, B., Wiser, R., Cappers, P., Thayer, M. & Sethi, G. (2009) The impact of wind power projects on residential property values in the United States: A multi-site hedonic analysis. Ernest Orlando Lawrence Berkeley National Laboratory LBNL-2829E, Berkeley, CA.
IEGMP (2000) Mobile phones and health: The Stewart Report. Independent Expert Group on Mobile Phones, Didcot, UK.
IVSC (2011) International Valuation Standards 2011, IVSC, London.
Jayne, M.R. (2000) High voltage overhead lines and public perceptions of risk. RICS Research Foundation, London.
Kauko, T. (2002) Modelling the locational determinants of house prices. Neural network and value tree approaches. PhD thesis, University of Utrecht.
Krimsky, S. & Golding, D. (1992) Social Theories of Risk. Praegar Publishers. Westport, Connecticut, USA.
Kroll, C. & Priestley, T. (1992) The effects of overhead transmission lines on property values: A review and analysis of the literature. Unpublished report prepared for the Edison Electric Institute Siting and Environmental Planning Task Force, Washington, DC.
Lean, G. (2009) New nuclear plants will produce far more radiation. The Independent, 8 February 2009. Available at http://www.independent.co.uk/environment/green-living/new-nuclear-plants-will-produce-far-more-radiation-1604051.html. Accessed September 2012.
Louviere, J.J., Hensher, D.A. & Swait, J.D. (2000) Stated Choice Methods: Analysis and Application. Cambridge University Press, Cambridge, UK.
McClelland, G.H., Schulze, W.D. & Hurd, B. (1990) The effect of risk beliefs on property values: A case study of a hazardous waste site. Risk Analysis, 10(4), 485–497.
McLaughlin, M. (2009) New opencast mining plans are the pits, say protesters. Edinburgh Evening News, 11 August 2009.
Mitchell, R. & Carson, R. (1993) Using Surveys to Value Public Goods: The Contingent Valuation Method. Resources for the Future, Washington DC.
Mundy, W. (1992) Stigma and value. The Appraisal Journal, January 1992, 7–13.
Nelson, J. (2004) Meta-analysis of airport noise and hedonic property values: Problems and prospects. Journal of Transport Economics and Policy, 38(1), 1–27.
Priestley, T.J. & Ignelzi, P.C. (1990) A Guide to Assessing Transmission Line Impacts in Residential Communities. Edison Electric Institute, Washington, DC.
Resource Manage Act (2008) Resource Management (National Environmental Standards for Telecommunication Facilities) Regulations. http://www.legislation.govt.nz/regulation/public/2008/0299/latest/DLM1576701.html. Accessed October 2012.
RICS (1995) RICS Appraisal and Valuation Manual. RICS Business Services Limited, London, UK.
RICS (2010) Contamination, the Environment and Sustainability, 3rd Edition. RICS Coventry, UK.
RICS (2012) RICS Valuation- Professional Standards (Global and UK edition). RICS London, UK.
Rossini, P., Marano, W., Kupke, V. & Burns, M. (2002) A comparison of models measuring the implicit price effect of aircraft noise. In Proceedings of PRRES Conference, Brisbane, Australia, January 2003. http://www.prres.net/Papers/Rossini_Models_measuring_implicit_price_effect_of_aircraft_noise.pdf. Accessed October 2012.
Sims, S. & Dent, P. (2005) High voltage overhead power lines and property values: A residential study in the UK. Urban Studies, 24(4), 665–694.
Sims, S. & Dent, P. (2007) Property stigma: wind farms are just the latest fashion. Journal of Property Investment & Finance, 25(6), 626–651.
Sims, S., Dent, P. & Oskrochi, G. (2008) Modelling the impact of windfarms on house prices in the UK. International Journal of Strategic Property Management, 12(4), 251–269.
Sims, S., Dent, P. & Ennis-Reynolds, G. (2009) Calculating the cost of overheads: the real impact of HVOTLs on house price. Property Management, 27(5), 319–347.
Slovic, P. (1987) Perception of risk. Science, 236(4799), 280–285.
Slovic, P., Layman, M., Kraus, N., Flynn, J. & Chalmers, J. (1991) Perceived risk, stigma, and potential economic impacts of high level nuclear waste repository in Nevada. Risk Analysis, 1(4), 683–696.
Syms, P. (1996) The valuation of contaminated land: The problems caused by lack of accessible data in the United Kingdom. Journal of Property Development, 1(1), 17–25.
Syms, P. (1997) Contaminated Land: The Practice and Economics of Redevelopment. Blackwell Science, Oxford.
Whitehead, J., Pattanayak, S.K., Van Houtven, G.L. & Gelso, B.R. (2008) Combining revealed and stated preference data to estimate the non-market value of ecological services: An assessment of the state of the science. Journal of Economic Surveys, 22(5), 872–908.
2
Methods
David Wyman, Peter Dent and Sally Sims
2.1 Introduction
As discussed in Chapter 1, in arriving at an estimate of value for a property a valuer ideally quantifies benefits and disadvantages of one property in relation to other properties on the market or properties which have recently passed through the market. The way in which this process is carried out will differ from country to country and is largely based on culture and the experience of the valuer undertaking the valuation; this will unfortunately introduce varying degrees of subjectivity into the valuation process. However, it is possible to distinguish between ‘traditional’ approaches and ‘more advanced’ approaches to the valuing exercise. These have been categorised by Pagourtzi et al. (2003) as follows:
This book concentrates principally on the sales comparison method of valuation, as it is the most commonly used in the residential market. It is also the method that has provided the most evidence for the research that underpins the case studies in this book.
2.2 Sales Comparison Method
The most common approach in the UK, NZ and North America has been the sales comparison method when trying to advise on a likely value for residential property. This is principally because the volume of sales and the nature of the market often provide sufficient market comparables to indicate an expected selling price range. While it may produce a useful guide as to eventual transaction price, using this method alone does not always disaggregate the relative impact to value of key variables (such as location, vista, size, etc.). As such, it is therefore not often helpful in estimating marginal impacts to properties from exposure to the disamenities which are the focus of this book. By this we mean that, in a significant number of cases (but still a minority of cases), the sales comparison approach is still not helpful. Moreover, this method only gives an estimated value which is invariably based on the subjective opinions of the valuer, as opposed to what the market indicates. Finally, it can be helpful to use the comparable method to determine the key variables that create value; these variables are however fairly well explored in the literature (e.g. Sirmans et al. 2005).
2.3 Regression Analysis
If the focus of the research exercise is to identify the marginal impact of specific variables as opposed to estimating the final selling price, researchers can undertake some form of regression analysis. Such analysis helps to determine whether a causal relationship exists between price (Y), and potentially, a set of property or locational characteristics (X). Regression analysis calculates the effect of movement in an independent variable (X) on the dependent variable (Y), such as the relationship between price and floor size. Specifically, a regression model estimates the marginal contribution of a unit of change of the independent variables on the dependent variable, or in other words, the correlation between these two types of variables. When the relationship is between two variables, one dependent and one independent, a simple linear regression is used. Multiple regression analysis is used when there are more than two independent variables. In addition to estimating the marginal contribution to value (or correlation) between each independent variable and the dependent variable, a regression model estimates the strength of the relationship between each independent variable and the dependent variable using the underlying variance of the variables. A regression model also determines whether the relationship is statistically significant at a desired level of confidence.
The use of multiple regression analysis is particularly useful for the investigation of estimated impacts on home selling prices of neighbouring disamenities because those impacts can be tested while holding the influence of a variety of other characteristics of the home (e.g. square feet of living space, neighbourhood characteristics) constant. Moreover, the test is relatively transparent and flexible, allowing a variety of impacts to be tested simultaneously. For this reason, multiple regression models are often relied upon to determine whether or not a causal relationship exists between the presence of either (1) HVOTLs (e.g. Colwell 1990; Kroll and Priestley 1992; Bond and Hopkins 2000; Des Rosiers 2002); (2) wind turbines (e.g. Hoen et al. 2009); or (3) cell phone towers (e.g. Bond and Wang 2005) and the value of proximate residential property. A number of useful reviews of the various amenity and disamenity literature that use multiple regression models exist (e.g. Kroll and Priestley 1992; Boyle and Kiel 2001; Simons and Saginor 2006).
The level of sophistication of the model can enable forward or backward stepwise regression to include only those variables that show significance. Regression variables can therefore help to provide the valuer with a greater understanding of the impact that certain individual features have on the overall value of the property, or will go some way in helping to explain the price paid in the market.
When it is applied specifically to goods that have a variety of components that make up their overall price (i.e. composite goods), multiple regression analysis is generally called hedonic pricing. Homes, which are the focus of this book, are examples of composite goods in that their value can be broken down into a ‘composite’ of its characteristics (such as the size of the home, the number of bathrooms, the number of acres on which the home is located, and the neighbourhood in which the home is located).
2.4 Hedonic Modelling
The hedonic pricing (HP) method is a revealed preference model that theorises that a heterogeneous good consists of a bundle of attributes; each attribute has its own implicit price and the product’s price is the sum of these implicit prices (Rosen 1974). One major advantage of the revealed preferences model is its concreteness; the respondent indicates their preferences by actual market transactions (purchases) as opposed to an intended action (Dunse et al. 2007). The hedonic framework therefore allows for the implicit pricing of individual property variables for real estate that do not have an observable market price. In other words, the hedonic method measures the attributes that cause a product ‘to be different – it is virtually useless’ for truly homogenous products (Studenmund 2001, p. 404).
One of the problems with the hedonic method is that it does not necessarily reflect the total value of an attribute (Owusu-Edusei and Espey 2003). Typically, it does not capture the full benefits from the provision of public goods such as open space, wildlife habitat corridors or cleaner water. Furthermore, it does not capture passive-use values derived from individuals who may use the resource but do not live proximate to it (Dunse et al. 2007). For example, the provision of a park may create positive externalities so that its benefits extend well beyond the increase in property values accruing to proximate property. Nevertheless, since its use in pioneering housing studies (Grether and Mieszkowski 1974; Linneman 1980) the hedonic pricing method has been the primary choice of property researchers; this is particularly the case for pricing housing, due to its reliance on actual market transactions.
Wilhelmsson (2000) identifies four broad types of property factors that households normally take into account in the purchasing decision: structural characteristics of the property (number of bedrooms, square feet, attached garage, etc.); its location relative to urban services (such as school districts, jobs, etc.); its environmental attributes (such as the view or slope of the yard); and the impact of macroeconomic attributes (such as the prevailing interest rate for mortgages).
The implicit prices for a model can be estimated as follows (adapted from Plummer 2010):
where PRICEi is a vector of property prices observed at a particular location i; Xi is a matrix of observations of exogenous independent variables; β is a vector of the regression parameters; and εi is a random error term that is normally distributed.
In formal terms, the selling price of a property is termed the dependent variable while the independent variables comprise the bundle of attributes that are also described as predictors or explanatory variables.
Typically, the dependent variable (the vector of housing prices observed at a particular location i) is transformed into the natural log of property prices observed at a particular location i (Colwell 1990; Des Rosiers 2002; Wolverton and Bottemiller 2003; Chalmers and Voorvardt 2009). This is termed a semi-log or log-linear model. Malpezzi (2003, p. 79) observes that there is no strong theoretical basis for choosing the correct functional form. However, the log-linear (or semi-log) model is the overwhelming favourite because no other model has proved to be as consistently robust and parsimonious, given the often skewed distribution of the untransformed sales prices (Sirmans et al. 2005). The semi-log model is also convenient as the coefficient can be interpreted as the percentage change in price of the property given a unit change in the independent variable. For example, if the coefficient for acreage (or per hectare) is 0.025, then adding an acre or hectare of land would increase the value of the land by approximately 2.5%. Caution should always be applied in interpreting such results, as this approximation only operates over a limited range with all other independent variables being held constant.
A final problem confronted in hedonic modelling is the specification of the explanatory (independent) variables. Malpezzi (2003, 78) observes that theory is not necessarily a guide as there are ‘literally hundreds of potential housing characteristics’ that could be included as explanatory variables. He suggests that the choice of explanatory variables is an art as well a science. However, the vast number of possible attributes that explain the pricing of a property creates a problem of building a parsimonious model. This problem is exacerbated if locational characteristics that are ‘substantially more difficult to observe and quantify’ (Dubin et al. 1999, p. 79) are included. In addition, if the hedonic pricing equation does not provide for interaction between aspatial and spatial characteristics, the effects of the explanatory variables on the dependent variable are likely to be underestimated, mis-specified, undervalued or, worse, overvalued. The next section therefore explores in more detail the development of hedonic modelling with spatial analysis (i.e. spatial hedonic methods) and the way in which these can be applied to the main concerns of this book.
In summary, the hedonic approach is particularly useful when determining any potential impact on value ranging from a contaminant or detrimental condition to an enhanced location or positive physical attribute. It enables the combination of property-specific variables and external or condition-specific variables for every unit under consideration to be analysed by establishing a model, determining the parameters and then evaluating the result using multiple regression analysis (Kauko 2002). The hedonic methodology can therefore be applied to actual situations to help identify the impact of specific external features on value and, with the additional use of qualitative methods, help to explain these impacts in terms not only of facts but also of perceptions. Accordingly, the chapter ends with a section on some of the qualitative aspects of research that could complement the quantitative analysis. While the emphasis in the following sections is on spatial hedonic modelling, this does not minimise the need to conduct corroboratory forms of research utilising tools such as survey-based studies, case studies and paired sales analyses.
2.5 Spatial Hedonic Modelling
Tobler’s first law of geography states that: ‘…everything is related to everything else, but near things are related more than distant things’ (Tobler 1970, p. 236). Tobler’s first law of geography is at the heart of quantitative analysis using spatial statistics. A simplified method of understanding this concept is to consider that, if it is raining at one geographic spot, then it is more likely to be raining at a location which is geographically close as opposed to a location further away. Similarly, this analogy can be applied to trying to predict the price of property.
Dubin et al. (1999, p. 90) observe that real estate analysis has tended to employ ‘econometric tools designed for a spaceless world’. There has been considerable growth of spatial regressions in real estate literature since the mid-1990s (Conway et al. 2010). This has been facilitated by the increasing availability of Geographic Information Systems (GIS) and geographically GIS-based spatial data to analyse property. The advent of GIS with its ability to spatially link property addresses with geographic coordinates has revolutionised hedonic modelling. In a review of hedonic modelling, Malpezzi (2003, p. 86) observes: ‘Perhaps one of the most exciting areas for extending hedonic models is making use of the spatial structure of the data, using the emerging technology of geographic information systems and spatial autocorrelation.’
There are two main types of spatial problems that can arise in hedonic modelling: spatial heterogeneity and spatial dependence (LeSage 1998; Wilhelmsson 2002).1 Spatial heterogeneity refers to the systematic variation of underlying relationships over space (LeSage 1998). For example, if the mean price of housing differs from location to location, then the houses are spatially heterogeneous (Bowen et al. 2001). Holland (2008) points out that such spatial heterogeneity commonly occurs simply due to the tendency of cities to grow outwards over time and space. For example, the older neighbourhoods closer to the city centre tend to have smaller homes lacking some of the modern attributes such as air conditioning or triple garages. The relative scarcity of triple garages in the inner city compared to larger newer homes on the outskirts of the city means that the price per square foot for a garage may command a higher marginal price in the inner city. As a result, the marginal prices of housing attributes will be non-uniform and vary across space (Holland 2008, p. 48), leading to biased estimation of coefficients (i.e. the estimated coefficient for garages will be inaccurate in the above model) if geography is not included in the model. One method of dealing with spatial heterogeneity is the construction of spatial models employing dummy variables to represent disaggregated sub-markets (Bourassa et al. 2010).
LeSage (1998) defines spatial dependence as being where one observation at location i is dependent on other observations at location j. Thus, if the sale price of a house at location i is dependent upon the sale price of a house at location j, then these observations are said to be spatially dependent (Samarasinghe and Sharp 2008). Spatial dependency violates the standard hedonic model assumption that observations of variables are independent (Plummer 2010) leading to spatial autocorrelation. Spatial autocorrelation is also known as spatial homogeneity (Holland 2008).
The issue of spatial analysis is often complex in application. First, there is the fundamental problem of defining spatial dimensions. For example, different approaches can define the location of specific neighbourhood boundaries and spatial dependence may arise when artificial data boundaries, such as zip codes, county boundaries and census tracts, leads to measurement error (LeSage 1998). Secondly, traditional hedonic models may fail to include accurate descriptors that measure accessibility or locational amenities such as public schools, shopping centres or parks (Wilhelmsson 2002; Suriatini 2007). Plummer (2010, p. 148) concludes that in practice ‘it is often difficult to distinguish between spatial heterogeneity and spatial dependence in the cross-sections of data’.
In application, the most common method of dealing with spatial dependency is to include spatial descriptors (neighbourhoods, accessibility, etc.) as independent variables in the regression equation. Nevertheless, due to faulty specification or missing variables, such hedonic models may fail to uncover all spatial effects on prices. In this case, there are two standard methods of dealing with spatial dependency: spatial lag models and spatial error models.
2.5.1 Spatial Lag Model (SAR)
A spatial autoregressive lag model (SAR) or spatial lag dependency occurs when the weighted average of property prices in a neighbourhood affects the price of each property in the neighbourhood (Suriatini 2007). In a SAR, the correlated errors are related to the dependent variables as opposed to the independent variables. This has been described as substantive spatial dependence (Anselin 2002). For example, the price of a property inhabited by a drug gang may spill over and impact the price of a neighbouring property. In other words, there is an indirect spatial effect in addition to the direct structural and neighbourhoods effects that have already been included in a standard hedonic model. Omitting this spatial effect leads to a misspecification of the model and therefore the marginal price of specific characteristics may be biased. A SAR is less frequently utilised in real estate studies, but its impact can be substantive with more severe consequences (Suriatini 2007).
The implicit prices for a SAR model can be estimated as follows (Plummer 2010):
where Pi is a scalar of lot prices observed at a particular location i; p is a spatial autocorrelation parameter; WPj is the spatial lag operator; Xi is a matrix of observations of exogenous independent variables; β is a vector of the regression parameters; and εi is a random error term that is normally distributed.
The spatial lag operator WPj is the spatially lagged value of the variable Pj. Its value depends on the specification of the spatial weights matrix invoked. In the case of row-standardised continuity weights, WPj is the average value of its neighbours.
