Valuing the Built Environment

GIS and House Price Analysis

Scott Orford

• Publisher: Taylor & Francis
• Year: 2017
• Print ISBN: 9781138263697
• eBook ISBN: 9781351876131
• Edition: 1
• Topic: Scienze biologiche
• Subtopic: Biologia

This book critically assesses the hedonic pricing technique as a method of imputing monetary values for the implicit attributes of housing. The hedonic technique is widely used, particularly in the US, but increasingly in Europe and Asia and has proved to yield important results and influence cost-benefit analysis. Scott Orford breaks new ground in this volume by exploring hedonic house price models within a geographical rather than purely economic context. He reevaluates the microeconomic theory of housing markets and concludes that only by treating housing market dynamics as inherently spatial can empirical results conform to the theory that underpins them. He also makes conclusions with respect to locational externalities, which have important implications as to how the built environment is valued.

1 Introduction

It is often said that the most important determinant of property prices is location, location, location. Hence the valuation of the built environment has long been a traditional concern of geographers, economists and planners. Over the years numerous conceptual, theoretical and empirical studies have attempted to formulate, model and quantify how the built environment is valued by its inhabitants. This book aims to complement and extend some of this work by investigating how the built environment is valued via the mechanisms of the urban housing market. More specifically, the aim of the book is to move towards a valuation of locational externalities through the modelling of housing market dynamics.

Locational externalities are becoming an increasingly significant component of local housing market economies, particularly as gentrification and urban regeneration continue to change the urban landscape. The pressures being placed upon the redevelopment of brownfield sites is resulting in town cramming and the loss of open space. Recent evidence is now suggesting that over-development and scavenging for land is encouraging a switch from NIMBY-ism (not in my backyard) to BIMBY-ism (bungalow in my back yard) (Evans, 1991), confusing the traditional debate of how local residents act to exclude unwanted (usually negative) externalities from their neighbourhood. What is emerging from these changes to the built environment is the need for a clearer understanding of how landuse impacts upon house prices. More importantly, there is a growing need to understand how these types of changes will ultimately be incorporated into the local housing market.

Empirical research relating to affects of externalities within a housing market is quite scarce. Work that does exist tends to be obscure and quite often either contradictory or counter-intuitive. To overcome this anomaly, this book intends to draw together existing work on locational externalities into a common structure based around the mechanisms of the urban housing market. In particular, the book aims to explore how housing market dynamics and locational externalities can be modelled using the method of hedonic pricing.

Hedonic pricing is an econometric technique used for estimating the monetary value of attributes of complex commodities. It is based upon the principles of pleasure seeking and consumer sovereignty; the belief that individuals are the best judges of their own needs and will maximise the set of attributes of a commodity that give them the most satisfaction. These attributes do not have directly observable market prices. Instead, the sum of their values is equivalent to the market price of the commodity. Within this context, the price of a house can be regarded as the sum of the implicit price of its specific bundle of attributes. Since location is an integral attribute of a house, its value within a bundle can be estimated using a hedonic house price function. This function relates house price to housing attributes, with the resulting parameter estimates corresponding to the implicit prices of these attributes. Hedonic house price research has a long, established history, with its origins within the location and landuse theories of the 1960s. Although it has since moved away from the central tenants of these theories, it still continues to be intimately bound up with the micro-economic literature of housing markets and residential location. Moreover, the treatment of location within hedonic house price research often reflects the naïve treatment of location within these theories. It is therefore hardly surprising that previous work on locational externalities is somewhat problematic. Nevertheless, an understanding of housing market economics and residential location is necessary to comprehend how hedonic pricing can be used to value the built environment. The remainder of this chapter is devoted to a review of this literature and the extent to which hedonic house price theory has since departed from its original formulations. The review is necessarily discursive since the economic framework of hedonic house price theory is explained in detail in subsequent chapters.

Housing and the Housing Market

Housing as a Commodity

‘It is fixed in geographic space, it changes hands infrequently, it is a commodity which we cannot do without, and it is a form of stored wealth which is subject to speculative activities in the market… In addition, [it] has various forms of value to the user and above all it is the point from which the user relates to every other aspect of the urban scene’ (Harvey, 1972; pp 16).

Housing is unlike most other commodities. It is a complex package of goods and services that extends well beyond the shelter provided by the dwelling itself. Housing is also a primary determinant of personal security, autonomy, comfort, well being and status, and the ownership of housing itself structures access to other scarce resources, such as educational, medical, financial and leisure facilities (Knox, 1995). As such, housing has been viewed as a ‘composite demand for a flow of services embodying a variable mix of characteristics’ (Maclennan, 1982. pp. 41) – a multidimensional commodity. It is so intimately bound up with the lives of individuals that only one is usually consumed by a household at any time. It is typical to talk of a household purchasing packages or bundles of housing services that vary between housing types and housing markets. However, defining what a particular bundle actually is can be complicated. In addition, housing has a number of relatively unique attributes. It has a fixed location, a long durability, and a limited adaptability in response to changing demands. Housing stock is complex and diverse and is sensitive to changes that are external to the local market. Housing is also subjected to a multitude of institutional regulations imposed by government.

Supply and Demand of Housing

When discussing issues of supply and demand of housing, it is important to make the distinction between use value and exchange value. Use value generally refers to the net utility supplied by the bundle of housing services, whilst the exchange value is the capital value a property can realise in a competitive housing market. Although the use value of a property is a major determinant of its exchange value, this will also be influenced by the property’s potential for increasing capital gain, since the purchase of housing stock is often the largest and only source of a households accumulated savings (Muth & Goodman, 1989). Thus, two different housing markets can be identified. One deals with the supply and demand of bundles of housing services, whilst the other deals with an asset that can be thought of as housing stock. Although these two markets are conceptually different, they are integrated, since the majority of houses will offer similar services, such as a water supply.

Therefore, housing demand is a reflection of both its use value for consumption or occupancy purposes and its exchange value as an investment good. Housing demand tends to vary between income and racial groups and at different stages of the family life cycle. Other influential factors include migration, immigration and changes in tax rates and taxation policies, particularly mortgage interest rates. Demand for housing has been of central interest to economic theorists and has been contextualised in micro-economic models of land use (Alonso, 1964) and residential location (Evans, 1973). However, these demand led models have been severely criticised for ignoring the supply of housing.

The majority of properties supplied on the housing market come from the existing housing stock. Only a small proportion of the supply comes from newly constructed properties (Bourne, 1981). Substantial proportions of new supplies from existing stock arise through the subdivision of property and the conversion of non-residential buildings to dwelling uses. Even more occur through the death of a household, through the move of an existing household to shared accommodation or by a move outside the city. Supplies of housing stock may be ended by demolition or conversion to a non-residential use, or even merger by knocking together two or more dwellings (Knox, 1995). Therefore, housing supply is a complex phenomenon with new supply and existing supply requiring separate, but interdependent analysis (Maclennan, 1982). Furthermore, the supply of housing will experience time lags between the decision to supply housing services and these housing services coming onto the market. The rate at which these new supplies enter the market in the short term is sensitive to house price changes and fluctuations in interest rates. This is particularly so for new constructions in which the price and the availability of land, planning controls and the provision of infrastructure have important influences on decisions regarding the location and timing of a development (Muth & Goodman, 1989).

Factors Influencing the Decisions to Move

The household is initially assumed to be receiving a given utility in their present dwelling. For movement to be considered, a minimum threshold of housing dissatisfaction must be perceived. This may occur over a period of time as the household recognises its existing mismatch of housing attributes and household activities. The decision to enter the housing market and evaluate alternative housing opportunities may be triggered by a variety of factors. These may include increases in income, changes in family size, household formulation, or relative price changes across the market. Knox (1995) has made the distinction between voluntary and involuntary moves. Voluntary movements may be initiated by dissatisfaction with dwelling and garden space, housing repair costs and style obsolescence, as well as complaints about the neighbourhood. Reasons for forced moves include marriage, divorce, a death in the family, retirement, ill health, and employment changes. However, almost two thirds of household movement is due to changes in the family life-cycle, and their perceived space requirements (Short, 1982).

In recent years another factor has emerged that has had an important influence on the propensity to move in the UK; negative equity. This occurs when the market price of a house becomes less than the mortgage secured upon it. This became a widespread problem at the end of the 1980s and during the 1990s when house prices slumped dramatically in many parts of the UK (Dorling, 1995). Negative equity means that the household is liable to cover the additional money secured on the property when it is sold, and having to find this money prevented many households from being able to move in the early 1990s. Negative equity particularly affected first time buyers, young buyers and less affluent buyers, since these were more likely to take out relatively larger loans and then have less ability to pay them back via earnings, inheritance and other assets. Low levels of equity can also deter households from moving.

The Micro-Economic Theory of Housing Markets

Introduction

The 1960s and 1970s witnessed a proliferation of micro-economic theories and mathematical models of housing markets, residential location and landuse in both the UK and USA (e.g. Alonso, 1964; Muth, 1969; Batty, 1976). These formulated housing market dynamics in purely economic terms, based upon the theory of the firm and consumer behaviour. A key element to these formulations was the concept of a housing market in perfect equilibrium functioning under Pareto Optimum conditions. These micro-economic theories subsequently underpinned neo-classical approaches to residential location and in particular, the trade-off model of residential location. The trade-off model is perhaps the most influential economic model of residential location within hedonic house price theory. However, as will be discussed, hedonic house price theory has since abandoned much of the micro-economic theory concerning perfectly functioning housing markets and Pareto Optimum conditionality, in favour of a segmented housing market in disequilibrium.

The Perfectly Competitive Housing Market

The owner-occupied housing market is primarily an economic market set within a political framework for the purpose of exchanging housing services. In economic theory, the role of the market is to allocate scarce resources in an efficient manner so as to maximise output while minimising cost, using price as the allocation mechanism. The most precise interpretations of this conceptualisation of the housing market derive primarily from the micro-economics literature. Maclennan (1982) identifies nine assumptions that define a set of conditions sufficient for the existence of a perfectly competitive housing market. These focus on the behaviour of individual producers and consumers, and regard the matching of households to housing units as essentially an assignment problem. The allocation proceeds as to achieve a market clearing solution; one in which all housing units are allocated and all households are accommodated in the most efficient way. The assignment is also optimal in the sense that no household could be made better off with a different assignment without making another household worse off. This is known as Pareto Optimum conditions, and is a source of contention in the hedonic house price literature that argues against Pareto Optimum equilibrium conditions in favour of a segmented housing market in disequilibrium. However, before this can be explored in more detail, it is necessary to set out the conditions under which a perfectly functioning housing market is said to operate. Following Maclennan (1982, pp. 36), Pareto Optimum conditions can be achieved under the following nine assumptions:

1. There are many buyers and sellers.

2. In relation to the aggregate volume of transactions the sales or purchases of each house are insignificant.

3. There is no collusion amongst or between buyers and sellers.

4. There is free entry into and exit from the market for both consumers and producers.

5. Consumers have continuous, transitive and established preferences over a wide range of alternative choices of housing and non-housing goods.

6. Consumers and producers possess both perfect knowledge with respect to prevailing prices and current bids and perfect foresight with respect to future prices and future bids.

7. Consumers maximise total utilityl whilst producers maximise total profits.

8. There are no artificial restrictions placed on the demands for supplies and prices of housing services and the resources used to produce housing service. For instance, house purchases are not constrained by finance rationing or the non-availability of preferred housing choices.

9. The market is assumed to be in equilibrium.

It can be seen that these assumptions are extremely idealised, and as such, easily critiqued. For instance, the abstracted assumptions of perfect competition and rational buyers and sellers are frequently cited (Ball, 1985). However, a major source of criticism of the micro-economic theories of housing markets has been the disregard of the supply of housing. Compared to demand, very little micro-economic work has been done on the supply of housing, especially in the short-run (Muth and Goodman, 1989). For instance, in the micro-economic supply model developed by Muth (1969), a supplier has perfect information regarding present and future house price changes, and is assumed to be a price-taking profit maximiser. This allows the precise output level of housing to be identified deductively. This model has been used to formulate theoretical specifications for the estimation of price elasticity of supply of housing both in the short-run and long-run and for the elasticity of substitution between land and non-land inputs to housing supply. However, the durability of housing, and the difficulty of adapting existing stock to changes in demand has been ignored, even though these will effect long-and short-run housing market equilibriums.

The Neo-Classical Approach to Residential Location

Introduction Concurrent with the formulation of the micro-economic theories of housing markets was the development of new approaches to residential location. Under the auspicious title of new urban economics, these neo-classical approaches were underpinned by similar micro-economic assumptions, and used comparative-static utility maximisation to deduced urban rent gradients and individual household demand functions for housing space and city centre access. By the mid-1970s these models had coalesced into a general theory of residential location known as the trade-off model.

The trade-off model The trade-off model (Basset & Short, 1980), or ‘access-space’ trade-off model (Maclennan, 1982), describes how households trade-off travel costs to the city centre, against housing costs in an attempt to maximise utility subject to an overall budget constraint. The theoretical background of the trade-off model was developed in two stages (Anas & Dendrinos, 1976). The basic models were developed during the 1960s, principally by Alonso (1964), Beckmann (1968) and Muth (1969), and were based upon the micro-economic theory of housing markets operating under Pareto Optimum conditions. These initial models were elaborated during the 1970s by economists such as Evans (1973), Mills (1972) and MacDonald (1979). The principal contribution...

Table of contents

• Cover
• Half Title
• Dedication
• Title Page
• Copyright Page
• Table of Contents
• List of Figures
• List of Tables
• Preface
• Acknowledgements
• List of Abbreviations
• 1 Introduction
• 2 The Hedonic House Price Function
• 3 Housing Attributes and Spatial Data
• 4 Constructing a Context-Sensitive Urban GIS
• 5 The Spatial Dynamics of an Urban Housing Market
• 6 Towards a Valuation of Locational Externalities
• 7 Conclusion
• Bibliography
• Index