.1...ch n!

.3 .j Taxi;
35.3.}. $1,:
1.! 3:29?

i I!
:lfsvlqlxif
19.. 3.133.. ..
nl lvllﬁlarie
'51-.)

53.635...
.: 53..) .

srniz... 2
sin "Wu.

1:.
.53.. .I «19
z. Av S. :1 u

.r'IIIsl-lrn

.33:

i.

.1. in t.
‘ 2?...
2.2::

..1,1.
1

er

.. wifc‘
8.....7 o‘.
n . 1:3...

\-
.
54

III. I. SIX.» :1

 

THESIS

W \\\\l\ll\l\l‘~l\\\\\\\\\\\\\\\\\\l

1932

This is to certify that the

dissertation entitled

Housing Investment in Developing Countries:
The Case of Korea

presented by

JeHoon Lee

has been accepted towards fulﬁllment
of the requirements for

Ph . D . degree in We e

 

ﬂ Major prcuessor

Date 2%,“ /9{ His”

MSU is an Afﬁrmatiw Action/Equal Opportunity Institution 0- 12771

4 ‘~‘-h——___ FVAI—ﬁ—mm

 

 

LIBRARY
Michigan State
University

 

 

 

PLACE ll RETURN BOX to romovo this checkout from your record.
TO AVOID FINES rotum on or boforo dot. duo.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

DATE DUE DATE DUE DATE DUE
. W
7;
2’ .9
usw-Anm A -1 "w A “,:...;..:.:...

 

 

 

 

 

 

 

   
       

HOt' a! . .
HOUSING INVESTMENT IN DEVELOPING COUNTRIES:

THE CASE OF KOREA

By

Je—Hoon Lee

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

treaties » ‘ ~. '
M to SAMWOFPHIWHY‘ ‘ I _-, ‘ ~‘I‘ J n '3‘.»

. H'nvcwjr .i‘fw 'i ~ A ; ' ' " '1'" .-‘=-‘ZT."i: as:

he
’ .mﬂim 0" :l'wk‘ggfismnftsqs In? {mtg - 31.1“.»4‘133‘3' conﬁde!
1995

 

ABSTRACT

HOUSING INVESTMENT IN DEVELOPING COUNTRIES:
THE CASE OF KOREA

By

Je-Hoon Lee

The main contribution by this dissertation is the addition of macroeconomic and
political context to empirical analysis and interpretation of housing investment in
developing countries. An additional contribution is systematic analysis and description
of housing policy in Korea as a case study. The example of Korea allows us to examine
the results of forty years of housing investment in a country in its middle stage of
development. An assessment of Korea’s present housing situation reveals housing
investment to be inadequate and this has caused distortions in the housing market,
thereby lowering the consumption of housing services in a market faced with growing
demand.

A general review of housing theory discloses that more and better housing is
used as a solution to societal stress, and that state intervention is justiﬁed in the
provision of housing. However, although existing models for housing investment are
related to overall allocation of resources in a country, they failed to adequately consider

macroeconomic factors.

Korea’s macroeconomic plans are linked to housing conditions, development,
and operation of housing plans based on policy, that has been organized through a
series of ﬁve year plans. Central organizing structures of development have a strong
impact on private sector housing investment, and also strong linkages with the
macroeconomy. Both of these circumstances indicate a signiﬁcant impact on the general
components of the Korean housing market.

New predictive models developed in this dissertation, which deﬁne the
determinants of residential construction investment, reveal a statistically signiﬁcant
effect on housing sector. These results are made more meaningful by adding
international, domestic, and institutional variables to the context in which housing
policy was formulated.

Analysis shows a statistically signiﬁcant relationship between historical patterns
of investment in housing sector and the structure of Korea’s total investment. Based on
theoretical framework, factor analysis, Pearson Correlation Analysis, and stepwise
regression analysis are used to select variables. Regression models using time-series
data from Korea covering the period of 1953-1993 are estimated to establish the
empirical relationships. Results show that housing investment in Korea is statistically
explained by levels of income, military spending, political stability, housing policy,

foreign affairs, and global context of ﬁnance.

© Copyright by
Je-Hoon Lee

1995

   

     
   
   
   
      

. i "I n' 1.3-. ' .
-~ '- i hit ass.‘
I ﬁlm: ',
CMITF1LLZ..

’ toe in m.- .

dedicawd to

mYPmntS. wifeandson

“I?“ IN iIIV'II ‘ v‘
MUM” it x'Ir‘e. n.“

l; -- in and an! 0‘ ,1... .,
\ .

_ M '. W 33119:]

 

- ‘1!
-n r'r
'--‘ .‘lhr-

79' .mﬁ, t (59“. . _-' v.-'.

 

ACKNOWLEDGMENTS

I owe a great debt of gratitude to many people for their encouragement and help

with this dissertation.

Many thanks and gratitude to my dissertation committee members, Professors
Roger Hamlin, Gill-Chin Lim, Mary Ann Kniseley, Kenneth Verburg, and Richard
Hula for their assistance and invaluable comments during the preparation of this
dissertation. I should express my appreciation to Professor Roger Hamlin, my
dissertation chairman, for his kindness, sincerity, and wit all the time. He has
supported me in many ways. I am deeply obliged to academic mentor Gill-Chin Lim,
Dean of International Studies and Programs, for his continuous enlightening ideas, and
scholarly guidance. This fact enabled me to complete this research. I am deeply
indebted to Professor Mary Ann Kniseley who freely gave of her time to guide me to
conceptualize ideas and their logical connections through the research process. In
addition to meticulously reviewing the thesis, Professor Kenneth Verburg taught me
valuable lessons in and out of the classroom. Professor Richard Hula made precious

criticism that enriched the study.

I take this opportunity also thank the faculty of Urban Planning. I cannot forget
their useful advice and challenges. I particularly appreciate Prof. Rene Hinojosa, and
Prof. Patrick McGovern for their useful advice and challenges in my work. I want to

thank Carl Goldschmidth for his comments on early drafts of this work. Several

vi

professors took time to discuss ideas with me. I especially want to thank Dr. Bruce
Pigozzi, Dr. Jon Burley, Dr. loo An Kwon and Yeong Sik Chae, who advised me in

the statistical ﬁeld, for their thoughtful suggestions.

I would like to extend my appreciation to my academic colleagues for their
encouragement and help. Particularly I am grateful to Dr. Hong Koo Yuh, Dr. Man-
Hyung Lee, Mr. Tae-Yul Lee, and Chung Won Suh for their enthusiastic help in
collecting data. My thanks also go to my colleagues and friends Mike Miller, Shelley
LeMahieu, Sandy Holy, and Grace Plaza de Meza at International Studies and

Programs.

My best thanks must go to my parents, parents-in—law and my wife, Mi-Young
for their invaluable support and love. Without their help, this work could never have

been completed.

vii

TABLE OF CONTENTS

LIST OF TABLES ................................................................................ VIII
LIST OF FIGURES ................................................................................ XII
LIST OF ABBREVIATION ..................................................................... XVI
CHAPTER ONE: INTRODUCTION ............................................................ 1
1. OVERVIEW OF THE STUDY ....................................................................... 1
2. STATEMENT OF THE PROBLEM .................................................................. 5
3. OBJECTIVES OF THE STUDY ...................................................................... 6
4. METHODOLOGY ................................................................................... 7
5. SCOPE OE THE STUDY ............................................................................. 8
6. RESEARCH CONTEXT ............................................................................. 9

CHAPTER TWO: GENERAL THEORETICAL FRAMEWORK AND LITERATURE

REVIEW OF HOUSING THEORIES .......................................................... l4

1. INTRODUCTION ................................................................................... 14

2. DEBATE ON THE ROLE OF GOVERNMENT AND ITS INTERVENTION IN THE HOUSING

SECTOR ................................................................................................ 14
3. RELEVANT THEORIES ON HOUSING INTERVENTION IN THE HOUSING MARKET ......... 18
3-1. State Intervention ........................................................................... 19

4. THE GENERAL THEORIES oN HOUSING INVESTMENT IN RELATION TO NATIONAL

DEVELOPMENT ....................................................................................... 25

viii

CHAPTER THREE: EXISTING HOUSING INVESTMENT MODELS ................ 30

1. INTRODUCTION ................................................................................... 3O
2. EXISTING MODELS IN HOUSING INVESTMENTS AND POINTS OF DEPARTURE ........... 31
2—1. Burns and Grebler’s Model: Theoretical Model ..................................... 31
2-2. Renaud’s Model: A Model for A Single Country ................................... 35
2-3. Annez and Wheaton’s Model: A Cross—National Context ......................... 38
2-4. Chang and Linneman’s Model: A Cross—National Context ........................ 40
2-5. Lim and Lee’s Model: A Model for A Socialist Country .......................... 41
3. PREVIOUS STUDIES OF URBAN HOUSING IN KOREA ......................................... 41
3—1. Housing Policy Studies in Korea ....................................................... 42
3-2. Housing Market Studies in Korea ...................................................... 43

CHAPTER FOUR: BACKGROUND OF HOUSING MARKET: DIMENSIONS AND

PATTERNS OF HOUSING DEVELOPMENT ............................................... 48
1. iNTRODUCTION ................................................................................... 48
2. OVERVIEW OF NATIONAL CONTEXT OF KOREAN HOUSING ................................ 48
2-1. Historical Background .................................................................... 49
2-2. Land ......................................................................................... 50
2-3. Economy .................................................................................... 51
2-4. Urbanization ............................................................................... 54
2-5. Income Distribution ....................................................................... 55
3. OVERALL HOUSING CONDITIONS .............................................................. 56

3-1. Housing Supply ............................................................................ 56

3-2. Physical Housing Conditions ............................................................ 59
3-3. Housing Occupancy and Tenure ........................................................ 60
3-4. Housing Prices and Affordability ...................................................... 62
4. HOUSING INSTITUTIONS ......................................................................... 65
5. HOUSING FINANCE ............................................................................... 67
6. HOUSING PROGRAM .............................................................................. 70
7. GENERAL HOUSING POLICY ISSUES AND EVOLUIION OF HOUSING POLICIES ........... 74
7-1. Summary of Korean Housing Policy .................................................. 75
8. HOUSING ISSUES AND PROBLEMS .............................................................. 96
CHAPTER FIVE: MODELS OF HOUSING INVESTMENT 1N KOREA .............. 99
1. INTRODUCTION ................................................................................... 99
2. RESEARCH DESIGN ............................................................................... 99
2-1. Theoretical Rationale ...................................................................... 99
2-2. Conceptual Model Structure ............................................................ 105
2—3. Formulation of Hypotheses, Variables and Models of Analysis ................. 110
3. PATTERNS OF HOUSING INVESTMENT AND FLUCTUATIONS .............................. 115
3-1. The Level of Housing Investment and Fluctuations ............................... 115
3-2. The Share of Housing Investment and Fluctuations ............................... 119

CHAPTER SIX: TESTING MODELS OF HOUSING INVESTMENT IN KOREA 127

1. INTRODUCTION ................................................................................. 127

2. DATA DESCRIPTION AND DATA COLLECTION ............................................. 128
2-1. Correlation Analysis ..................................................................... 130
3. FACTOR ANALYSIS AND FACTOR MATRIX .................................................. 139
3—1. Orthogonal Rotation with the Varimax Criterion .................................. 141
4. EMPIRICAL RESULTS AND INTERPRETATION ............................................... 151
4-1. Housing Investment Share Models ................................................... 152
4-2. Housing Construction Area Models ................................................. 172
CHAPTER SEVEN: CONCLUSION AND FUTURE DIRECTIONS ................. 178
1. INTRODUCTION ................................................................................. 178
2. SUMMARY OF FINDINGS ...................................................................... 178
3. LIMITATIONS OF THE STUDY ................................................................. 181
4. RECOMMENDATIONS FOR FURTHER RESEARCH ........................................... 183

5. THE APPLICATION OF THE THEORETICAL BASIS To THE KEY HOUSING PROBLEMS 184

6. POLICY REORIENTATION AND POLICY RECOMMENDATION ............................. 185
BIBLIOGRAPHY ................................................................................. 187
APPENDDI 1. LIST OF VARIABLES ........................................................... 205
APPENDIX 11. SITE OF NEW TOWNS AROUND SEOUL ..................................... 208
APPENDIX III. LIST OF SAMPLE NATIONS IN MODEL .................................... 209
APPENDIX IV. GLOSSARY .................................................................... 210
APPENDIX V. VARIABLES FOR TIMES-SERIES ANALYSIS ....................... 211

APPENDIX VI. EXCHANGE RATES, WEIGHT, AND MEASURES ......................... 219

xi

LIST OF TABLES
TABLE 1 REGRESSION ANALYSIS OF BURNS AND GREBLER'S MODEL ...................... 34
TABLE 2 REGRESSION ANALYSIS OF RENAUD'S MODEL ....................................... 36
TABLE 3 SHARE OF HOUSING IN TOTAL OUTPUT (SHTO): KOREA COMPARED To BURNS-
GREBLER SAMPLE ............................................................................. 37

TABLE 4 RENAUD'S ESTIMATED REGRESSION EQUATIONS (EFFECTS OF DEMOGRAPHIC

VARIABLES) .................................................................................... 38
TABLE 5 SUMMARY OF PREVIOUS HOUSING MARKET STUDIES IN KOREA ................ 46
TABLE 6 SUMMARY OF PREVIOUS HOUSING STUDIES ......................................... 47
TABLE 7 INDICATORS OF TRADE AND FOREIGN DEBT ........................................ 5 3
TABLE 8 URBANIZATION .......................................................................... 55
TABLE 9 A SUMMARY OF THE KOREAN HOUSING STOCK CONDITIONS ..................... 57
TABLE 10 HOUSING SUPPLY BY YEAR ........................................................... 58
TABLE 11 HOUSING QUALITY TREND AND STATUS QUO ..................................... 59
TABLE 12 DISTRIBUTION OF HOUSEHOLDS BY TENURE ....................................... 61
TABLE 13 TRENDS IN HOUSING-RELATED ECONOMIC VARIABLES: 1974-1991 ........... 63
TABLE 14 HOUSING SUPPLY OF PUBLIC / SEMI-PUBLIC / PRIVATE SECTORS ............. 65
TABLE 15 THE CONDITIONS OF MORTGAGE LOAN ............................................ 70
TABLE 16 AGGREGATE INDEX OF THE HOUSING ACTIVITIES ................................ 71
TABLE 17 DWELLING CONSTRUCTION: PLANNED AND REALIZED BY INITIATION ........ 72
TABLE 18 HOUSING CONSTRUCTION PLAN: PLANNED AND RBALIZED: 1962-1992 ..... 73
TABLE 19 RESIDENTIAL CONSTRUCTION ACCOMPLISHMENTS: 1951-1961 ................ 77

xii

TABLE 20 HOUSING POLICY GOAL IN HOUSING RELATED LAW .............................. 83

TABLE 21 HOUSING POLICY MEASURES ......................................................... 86
TABLE 22 THE Two—MILLION HOUSES PROGRAM ............................................. 89
TABLE 23 NEW TOWN DEVELOPMENT PLAN ................................................... 90
TABLE 24 HOUSING POLICIES IN ECONOMIC PLANS ............................................ 94
TABLE 25 MAJOR CHARACTERISTICS OF ECONOMIC DEVELOPMENT PLANS .............. 95
TABLE 26 VARIABLES FOR TIME-SERIES ANALYSIS ......................................... 113
TABLE 27 THE LEVEL OF HOUSING INVESTMENT ........................................... 1 16
TABLE 28 HOUSING INVESTMENT PER CAPITA .............................................. 118
TABLE 29 THE SHARE OF HOUSING INVESTMENT As A PERCENTAGE OF GNP ......... 120
TABLE 30 NEW HOUSING CONSTRUCTION .................................................... 122
TABLE 31 THE GROWTH RATE OF NEW HOUSING CONSTRUCTION ....................... 125
TABLE 32 DESCRIPTIVE STATISTICS ........................................................... 129
TABLE 33 CORRELATION MATRIX AMONG DEPENDENT VARIABLES ..................... 131

TABLE 34 CORRELATION MATRIX AMONG HOUSING INVESTMENT GROWTH PATTERNS135

TABLE 35 CORRELATION COEFFICIENT AMONG INDEPENDENT VARIABLES ............. 137
TABLE 36 SUMMARY OF MAJOR FACTORS AND VARIABLES ............................... 140
TABLE 37 FACTOR MATRIX: COMPONENT LOADINGS ...................................... 143
TABLE 38 FACTOR MATRIX: ROTATED LOADINGS .......................................... 144
TABLE 39 FACTOR 1,2 AND DETERMINANTS ................................................. 145
TABLE 40 FACTOR 3, 4 AND DETERMINANTS ................................................ 146
TABLE 41 FACTOR 5, 6 AND DETERMINANTS ................................................ 146

xiii

TABLE 42 THE LEVEL OF HOUSING INVESTMENT EXPLAINED BY PRINCIPAL COMPONENT

SCORES ........................................................................................ 147
TABLE 43 FACTOR SCORES BY TIME-SERIES ................................................. 149
TABLE 44 CLASSIFICATION OF HOUSING INVESTMENT MODELS .......................... 151
TABLE 45 THE HOUSING iNVESTMENT SHARE MODEL BY STEPWISE SELECTION ...... 153
TABLE 46 REGRESSION RESULTS: HOUSING INVESTMENT SHARE MODEL .............. 154

TABLE 47 TESTING: RELATIONSHIPS BETWEEN THE INDEPENDENT VARIABLES AND
SOCIOECONOMIC VARIABLES .............................................................. 155
TABLE 48 TESTING: RELATIONSHIPS BETWEEN THE DEPENDENT VARIABLES AND
DEMOGRAPHIC AND URBANIZATION VARIABLES ....................................... 156
TABLE 49 TESTING: RELATIONSHIPS BETWEEN THE DEPENDENT VARIABLES AND
POLICY EFFECT .............................................................................. 157
TABLE 50 TESTING: RELATIONSHIPS BETWEEN THE DEPENDENT VARIABLES AND
HOUSING CONDITIONS ...................................................................... 157
TABLE 51 TESTING: RELATIONSHIPS BETWEEN THE INDEPENDENT VARIABLES AND
INSTITUTIONAL SETTINGS VARIABLES .................................................... 158
TABLE 52 TESTING: RELATIONSHIPS BETWEEN THE DEPENDENT VARIABLES AND
GLOBAL AND FOREIGN AFFAIRS EFFECT ................................................ 159
TABLE 53 TESTING: RESULTS FROM STEPWISE REGRESSION ANALYSIS OF COMBINED

MAIN EFFECT VARIABLES, ILLUSTRATING THE BEST EQUATION ..................... 160

TABLE 54 SYNTHESIS OF THE HYPOTHESES TESTING RESULTS OF REGRESSION ANALYSIsl61

xiv

TABLE 55 NONLINEAR RELATIONSHIP BETWEEN THE SHARE OF HOUSING INVESTMENT
(CONNP) AND GNP PER CAPIIA ........................................................ 162
TABLE 56 RELATIONSHIP BETWEEN THE SHARE OF HOUSING INVESTMENT (CONNP)
AND DEMOGRAPHIC AND URBANIZATION VARIABLES ................................. 164
TABLE 57 RELATIONSHIP BETWEEN THE SHARE OF HOUSING INVESTMENT (CONNP)
AND HOUSING CONDITIONS VARIABLES .................................................. 165
TABLE 58 RELATIONSHIP BETWEEN THE SHARE OF HOUSING INVESTMENT (CONNP)
AND MILITARY SPENDING VARIABLES ................................................... 167
TABLE 59 RELATIONSHIP BETWEEN THE SHARE OF HOUSING INVESTMENT (CONNP)
AND POLICY EFFECT VARIABLES .......................................................... 168
TABLE 60 RELATIONSHIP BETWEEN THE SHARE OF HOUSING INVESTMENT (CONNP)
AND FINANCE VARIABLES .................................................................. 169
TABLE 61 SUMMARY OF COMPREHENSIVE HOUSING INVESTMENT SHARE MODELS 170
TABLE 62 ELASTICITIES AND BETA COEFFICIENTS, INVESTMENT ANALYSIS ............ 170
TABLE 63 THE HOUSING CONSTRUCTION AREA MODEL BY STEPWISE SELECTION 172
TABLE 64 BASIC REGRESSION EQUATION: THE RELATION BETWEEN PHAREA AND
OTHER VARIABLES .......................................................................... 173
TABLE 65 HOUSING CONSTRUCTION MnnFI .. ......................... 174
TABLE 66 THE HOUSING INVESTMENT HOUSEHOLD MODEL BY STEPWISE SELECTION175
TABLE 67 BASIC REGRESSION EQUATION: THE RELATION BETWEEN HIHSH AND OTHER

VARIABLES .............. . . ..................................... 176

 

TABLE 68 HOUSING INVESTMENT HOUSEHOLD MODEL .................................... 177

LIST OF FIGURES

FIGURE 1 RESEARCH DESIGN ..................................................................... 13

FIGURE 2 THE NONLINEAR RELATIONSHIP BETWEEN SHARE OF HOUSING INVESTMENT

AND GDP PER CAPITA ........................................................................ 33
FIGURE 3 GNP PER CAPITA (1970 CONSTANT PRICES) ....................................... 52
FIGURE 4 THE GROWTH RATES OF GNP PER CAPITA ......................................... 53
FIGURE 5 TRENDS IN HOUSING-RELATED ECONOMIC VARIABLES: 1974-1991 ............ 64

FIGURE 6 PUBLIC/PRIVATE CLASSIFICATION BY PARTICIPATION, FUND, AND TARGET

GROUPS IN HOUSING SUPPLY ................................................................ 66
FIGURE 7 GRAPHICAL REPRESENTATION OF MODEL STRUCTURE ......................... 109
FIGURE 8 THE LEVEL OF REAL HOUSING INVESTMENT .................................... 1 17
FIGURE 9 HOUSING INVESTMENT PER CAPITA ............................................... 118
FIGURE 10 THE SHARE OF HOUSING INVESTMENT AS THE PERCENTAGE OF GNP ..... 121
FIGURE 11 NEW HOUSING CONSTRUCTION BUILDING PERMITS ........................... 123
FIGURE 12 THE GROWTH RATE OF NEW HOUSING PERMITS .............................. 126

FIGURE 13 THE RELATIONSHIPS AMONG HOUSING INVESTMENT PATTERNS BY
NORMALIZED PLOTS ........................................................................ 134

FIGURE 15 FACTORS SCORES OF HOUSING INVESTMENT BY TIME-SERIES ................ 150

BOK
CPI
EPB
GNP

KHB
KLDC
KNHC
M1

M2

LIST OF ABBREVIATIONS

Bank of Korea

Consumer price index

Economic Planning Board

gross national product

Korea Development Bank

Korea Housing Bank

KOIea Land Development Corporation

Korea National Housing Corporation

narrow money supply (currency plus demand deposits of the deposit

money banks)

broad money supply (M1 plus time and savings deposits at the deposit
money banks)

Ministry of Home Affairs

Ministry of Construction

Ministry of Trade and Industry

National Housing Fund

the Newly Industrializing Countries

CHAPTER ONE

INTRODUCTION

1. Overview of the Study

Urban housing problems are among the most severe issues confronting nations
world wide. Both developed and developing countries presently lack sufﬁcient and
affordable housing. In market based societies, the proﬁt motive directs builders toward
high end products, while low proﬁts derived from low cost housing discourage
investment. Government subsidy to housing faces problems of sustainability and social
stratiﬁcation. While such issues are gaining global attention, there has been little
systematic analysis and development of models to achieve answers. Those that have,
emerged tend to limit their perspective to selected criteria without considering broader
contextual issues affecting government from within and without.

This dissertation empirically examines housing investment using Korea as a case
study. Housing investment is one of the fundamental housing activity measurements at
the macro level. It represents the magnitude of aggregate housing ﬂuctuations and
“reﬂects housing policies in the public and private sector” (Chang, 1986z4). It can be

measured in terms of the total output of capital investment1 or dwelling unit production,

 

‘ Homing investmem includes “not only new construction but also capital spending on the alteration and
modernintion of existing dwellings."(Burns and Grebler, 1986: 138)

or that total as a percentage of gross domestic product2 (GDP) or gross national
product3 (GNP). This measure represents the overall performance4 of a housing
industry5 and housing production and is a useful tool of comparison between different
periods and between countries.

The research investigates a theoretical foundation by reviewing existing theories
and literature concerned with housing investment with respect to overall allocation of
resources in a country. This approach includes broader contextual variables such as
level of income, military spending, political stability, housing policy effect, foreign
affairs, and global context of ﬁnance.

The example of Korea6 allows us to examine the results of housing investment
in a country in its middle stage of development, a condition of many developing
countries. An assessment of Korea’s present housing situation reveals housing
investment to be inadequate and this has caused distortions in the housing market,
thereby lowering the consumption of housing services in a market faced with growing
demand.

To identify problems of overall housing investment, the question of policy and

related decision-making processes must be addressed. It is essential to take a

 

3 The gross domestic product is a “measure of total output in national income and product accounting”. It includes
exports bu excludes imports. The gross national product takes account of the excess of imports over exports or the
excess of exports over imports (Burns and Grebler, 1977: 45)

3 The gross national product is the aggregated value of all goods and services produced annually in the mtion.

‘ The yardstick for measuring the performance of the housing sector is shifted from the mimber of newly built units,
the conventional standard, to residential investment. Because investmem in monetary calculation reﬂects quality
changes and assessment the future of housing markets.

5 Housing irxlustry includes “all firms which share inthe receipts of expenditures for housing." (Shih, K. 1990:1)

‘ This dinettation deals exclusively with South Korea; I will use the term Korea in stead of South Korea. So, all
refereneesto Koreainthethesis areto South Korea

methodological approach that takes into account multidisciplinary and comprehensive
relation to national policy and economic development. This approach needs to
recognize a model for describing and predicting the cause-and-effect relations between
economic growth and housing investment. We consider the interdependency of various
components and aspects of the housing sector in the national economy as another
determining factor associated with the functioning of complex housing policy
backgrounds.

Analysis of the role of the housing sector in economic development has been a
controversial issue, because most housing studies in Korea to date are descriptions of
general housing policies, analyzing housing supply and demand or, case studies of
speciﬁc geographical areas. Few empirical analyses use rigorous analytical
frameworks. In the area of housing investment, there is only one systematic empirical
study7 which has the historical pattern of housing investment at the national context
been observed in Korea. However, this study has limited its analytical tool. This fact
led to the fact that current empirical knowledge far too inadequate to understand the
nature of housing investment.

To achieve the goal of understanding economic factors shaping the nature of
housing investment and problems in Korea, this study incorporates into the
investigation the determinants of housing investment at a large scale that previously
attempted. It is at the national policy level. Housing policy was considered not as a

separate issue but as one of many policies in national politics, and bureaucratic

 

' Renaud's study (1980), for more detail see Chapter Three

structure. The scale of this approach takes in the international pressures and context
applied by Korea’s economic integration in to a global market. By making explicit
comparisons among different sectoral policies, the effect of competition among sectors
and resulting impact on the housing sector is discernible.

To comprehend the historical development of housing policies, this study
follows the evolution of policies over time, identifying inﬂuences by political and
economic transformations and resulting investment shifts for military, social welfare
and environment. The effects of such inﬂuences are expected to emerge as
interruptions, in continuities or ﬂuctuations in housing policies.

An interdisciplinary approach permits the combination of a broad range of
qualitative and quantitative analyses associated with politics, economics, and planning.
A broad survey of literature reveals qualitative inquiries are based sociohistorical
interpretations, and documentation. For quantitative analysis, a new model based on
historical data of housing investments was developed. It includes international
comparisons where appropriate. The results gained by different disciplinary and
analytical approaches are compared for consistency of observation. The relative
signiﬁcance of housing situations is determined by comparing the results of the Korea
study with that of other countries to assess the reliability and usefulness of results.

Finally, the observations yielded by this analysis have been brought together for
policy recommendations. Through a comprehensive study the results are reinterpreted

policy implications are suggested and some generalizations made.

2. Statement of the Problem

Korea has shown substantial economic growth during the last three decades. In
this regard, much scholarly attention (Brown, 1973; Cole and Park, 1983; Mason,
1980; Lim: 1991; Kuznets, 1988:11-45; Balassa, 1988:273—290; Chan, 1991:79-103)
has been directed toward Korea as one of the Newly Industrializing Countries (NICs)
to explore the nature and extent of this rapid growth. A careful survey of the current
literature reveals that the development experience of Korea has been the foundation of
theory building for economic growth.8 Yet Korea’s housing market conditions have
remained unsatisfactory9 (Renaud, 1992). In contrast to the continued emphasis on
economic development in the current literature, the role of the government housing
sector is rarely discussed. This study is an effort to ﬁll this gap of at least partially.

In sum, this dissertation focuses on Korea's housing investment policy based on
the following issues and research questions:

0 What are the current housing investment policies and problems?

0 How can we explain the determinants of housing investment?

0 How have urbanization and political and economic policies inﬂuenced
housing conditions and policies?

0 What are the patterns of housing investment?

0 What housing policies should Korea consider in the future and what role
should the government play?

0 What lessons can we learn from Korean housing investment experiences?

 

' Most studies have focused onthe “strong" and “developmental” state. However, recently a few studies try to ﬁnd
how the state changed as a result of developmem. For more detail see Kim, Bun Mee, 1992.

’ Renaud criticized restriction of housing ﬁmnce by the government is the main reason for inefﬁciency of the
Korean housing market. For more details see Kim. K. Y. 1988 and Lee. K. and Sohn. 1989.

3. Objectives of the Study

The main objective of this study is to provide a comprehensive analytical
framework for determining the causes of housing investment, ﬂuctuations of housing
investments and housing policy in Korea. This study is important not only because it
clariﬁes trends and determinants of housing investment but also because it provides
some insights into the housing development in developing countries.

Political and economic factors are more difﬁcult to deﬁne. Yet, in developing
housing investment framework, they are the main features of housing in Korea over
several distinctive periods since 1953. They have molded the greater investment
policies, which have then affected housing. Therefore, it is essential to investigate the
larger context and its effect on housing investment and policy development in Korea to
if one is to create an effective analytical tool or predictive model.

For the purposes of theoretical discussion, this study primarily reviews the
theories on housing investment and their relationship to macroeconomy policy and
government intervention in the housing sector. Moreover, housing investment models
which represent empirical explanations can be derived from analysis of housing market
and the linkage between housing spending and macroeconomics policies. More
speciﬁcally, this study seeks to evaluate empirically the actual housing investment

effects on economic development in Korea.

This empirical housing investment analysis and its interpretation also describe
and evaluate other experiences of NICs and China in order to understand housing
policy and development in those countries. Few studies explain the housing problems in
a developing countries by looking at the government’s policies on a macro level. In
addition, this study attempts to investigate to inspect housing policies in a social-
historical context, and to provide empirical evidence for the conceptual model by using
economic and statistical analyses of Korean housing problem. The conceptual and

empirical models could be used for other developing countries to test for generalization.

4. Methodology

The key method of analysis is based on multiple regression analysis and time
series analysis. The Pearson correlation analysis and factor analysis are used in
conjunction with the Burns and Glebler model (BG), which is taken as starting point,
and hereafter is referred to as BG Model. The BG model associates levels of housing
investments with levels of economic development, levels of urbanization, and natural
population growth.

86 analysis has the advantage of simplicity but lacks of a comprehensive scope
that takes into account political and socioeconomical factors. Although its shortcomings
were acknowledged, it is still worthwhile as a regression model to acquire a base
analysis for this empirical work. From this point of departure, theories of government

intervention in housing markets and correlation to macroeconomic policies are used to

develop an improved model of housing investments. The new model considers the
impacts of political-economic factors by examing the role of international, domestic,
and institutional arrangements in housing policy formulation.

The improved model takes into account current conditions of the Korean
housing market and Korean housing policies and relates them to historical context.
With this model we may address the question, How do macroeconomic policies and
urbanization affect housing investment and to what extent may the pattern of housing
investment undergo change?

By developing theoretical and empirical models, the impacts of political-
economic factors on housing investment in Korea have been tested. In addition,
economic, political, and social variables affecting the level of total housing investment
are investigated with respect to internal social structures and governmental formations
that have inﬂuenced levels of housing investments by relying on the utility

maximization model approach as a theoretical guideline.

5. Scope of the Study

This study examines the housing policies, political, and economic development
of Korea as well as the decision process supporting its housing policy for a relatively
short but signiﬁcant period from 1953 to 1993. During this period Korea has engaged
in an aggressive program of national and international economic integration as an

independent state.

Korea is unique in many ways, not only in its historical background and in the
impact of external inﬂuences on recent development, but also because, in spite of such
peculiarities, Korea is a typical replication of Western-style economic and social
development, seemingly ﬁtting into the capitalistic development path of traditional
developmental ideals. Most of all, Korea's intensive economic growth within such a
short period provides a rare opportunity to study a full range of aspects of housing
investment and housing policy development.

Some scholars of past housing investment studies in developing countries (M.
H. Lee, 1990, Renaud, 1990) have been concerned about the lack of valid and constant
data for research. In the case of Korea, there are some difﬁculties ﬁnding data that
relate to the proportion of informal housing, such as squatters and illegal constructions,
which are not recorded in the housing investment accounts (Renaud, 1980). Such

omissions can bias a time series study and this should be considered carefully.

6. Research Context

This dissertation is divided into seven chapters. The Research Design Flow

Chart in Figure 1 shows general research directions.

 

Chapter Two is devoted to a general literature review of housing theories and

discussion of urban housing studies in Korea with a view to housing intervention in

l0

relation to economic development theories. Current housing investments theories are
explained with a critique in terms of resource allocation and with notes concerning the
main role of governmental in the housing market. The key of this review is to classify

housing theories in relation to national economic development.

2 ZZ'E"EL'I MI!

This chapter describes and veriﬁes the reliability of existing housing investment
models, including BG model and hypotheses. It focuses on the usefulness of these
models for evaluating political development, economic policy, and urbanization, as

they have inﬂuenced the historical development of housing policies in relation to

economic development in developing countries.

 

This chapter addresses three questions: What are the current housing investment
policies and problems?, What housing policies should Korea consider in the future? and
What lessons can we learn from the Korean experience?

This chapter also investigates how Korea’s macroeconomic plans have been
linked to housing conditions and to development and operation of housing plans.
Korean housing policy issues are explained in the context of social and economic
transformation, which provide a link between the macroeconomy and that of the
general components of housing markets. Major housing problems are identiﬁed in

relation to housing investment.

11

2 E' 'MII EEI 'L “I:

Chapter Five answers three questions: How can we explain the determinants of
housing investment? How have urbanization and political and economic policies inﬂuenced
housing conditions and policies? and What are the patterns of housing investment?

This chapter presents a quantitative analysis of housing investment. Economic
models are constructed to explain the historical pattern of the housing sector's behavior
in investment. The empirical results are used as a basis for commenting on major
housing policies, including housing construction plans, and the pattern and
determinants of housing investment in Korea. Relationships between housing
investments and key economic and demographic variables are described along with
elasticities regarding housing investments. Comparisons are made between results of the
Korean data and that of previous studies from other countries, including comparison of

income elasticities for housing investment.

1 3.1]. III! SEE 'L .5:

Chapter Six provides hypotheses and its interpretation by testing regression
analysis concerning housing investment models characteristic of Korean housing
policies during the period of investigation. This chapter identiﬁes how Korea's
macroeconomics policies in turn have inﬂuenced housing policies. Special attention is
given to the impact on housing policies of certain economic policies with respect to
stabilization policy and anti-speculation measures. In this framework, problems of

dominance and consistency among various policies are reviewed, and tested with

12

variables for urbanization and population as factors affecting Korean housing policies.

This chapter critically examines the reliability of the hypotheses.
1 S _ 2 I . I E D' .

This chapter provides a summary of the study’s major ﬁndings. The limitations
of explanatory models are addressed and offer a conclusion of this study. Policy

recommendations for Korea and some general observations are suggested with the

future goals.

Figure 1 Research Design

  
   
  
        
 
   
   
   
   

  

CH. 1
Introduction:

 
 

CH. 2
Literature Review

  

Framework

    

4
Housing Policy Analysis
Demographic

Models of Housing Investment

Housing Investment

Housing Policy Recommendations

CHAPTER TWO

GENERAL THEORETICAL FRAMEWORK AND LITERATURE
REVIEW OF HOUSING THEORIES

1. Introduction

This chapter reviews and evaluates the existing framework for housing theory.
It attempts to clarify the role of government in the housing sector while focusing on
housing investment in relation to national economic development. Furthermore, it
highlights the relevant theories concerning intervention in the housing market. This
chapter reviews these theories both from the perspective of housing advocates, who
view more and better housing as a solution to societal stress, and economists, who see
housing as only one set of options to social development. The second section of this
chapter discusses the relationships between housing investment and economic

development issues.

2. Debate on the Role of Government and Its Intervention in the Housing Sector

In developing countries, the role of government and its intervention in the
housing sector during various phases of economic growth is a subject plagued with
controversy concerning the position of housing in resource allocation. Yet housing

issues have acquired paramount importance. Massive rural-urban population shifts have

14

15

aggravated already serious dwelling shortages in developing countries since the 1950’s.
Persistent dwelling shortages have resulted in deteriorating living conditions for the

poor and have contributed to general increases in social stress.

2-1-1. Housing Advocates

Housing advocates and urban planners have fully recognized these development
problems. The have engaged in a lively debate with economists about the optimal
allocation of scare resources for residential construction to relieve rising problems. of
Housing advocates are convinced that better dwellings and neighborhoods are the most
effective and direct means of improving the human condition (Burns and Grebler,
1977: 100-103). They assign a high priority to housing for its own sake.

Housing subsidies for the poor have been proposed as a means to “correct
unequal income distribution” in developing countries (ibid.:101). Support for this
strategy comes from the urgent realities of rapid population growth and the
extraordinary pace of urbanization in most parts of the developing world. Housing
advocates argue that as the housing shortage worsens, the need for shelter becomes
more compelling. Levels of societal stress increase, affecting everyone, not just the
poor. In addition to its high social utility, housing advocates claim that improved
shelter will contribute to political stability by moderating frustration with the slow

tempo of betterment in their general living conditions (I-lowenstine, 1957: 26).

16

To support their premise, housing advocates argue that residential construction
absorbs large quantities of labor with minimal input of scarce capital. In addition to the
favorable labor-capital ratio, construction offers rural migrants a port of entry to the
urban labor market and provides opportunities for the acquisition of skills. It has been
pointed out that housing construction may employ the vastly underutilized pools of
human and national resources available in low—income countries, and at zero or near-
zero opportunity cost (Burns and Grebler, 1977).

The problem with this approach is that long term planning objectives can be
sacriﬁced to short term goals driven by numbers of households.10 This fact can lead to
improper standards11 and the imposition of consumption12 patterns that intended
beneﬁciaries might not choose if offered alternatives. Since households do not have a
housing problem as much as an income problem, a better recourse is more options.
Such an alternative depends on greater knowledge of the housing market structure and
respect for proﬁt related alternatives.

When approaching the problem of quickly increasing housing units, decisions
based on the aggregate level can lead to overestimates of the total volume of resources

which “should” be available to the housing sector. With rushing of large scale housing

 

'0 A ‘household’ is deﬁned as ‘two or more persons living together with common housekeeping or a person living
alone who is responsible for providing his or her own meals.’ The deﬁnition of households varies from census to
census inKorea. The basic deﬁnition is that a households is deﬁned as a group of persons sharing living quarters
and households expenditures. The censuses also distinguish between ordinary households and quasi-households.

n For example, in regard to squatter settlements, we have to consider Turner's (1967. 1970) notion that governmem
policies aggravated housing problems and disregarded the economic and social needs of the poor. Turner proposes
applying site and service programs instead of providing housing as an end product. In the Korean case, the
mjodty of squatter settlers move to the city seeking jobs. not housing.

'2 Consumption is deﬁned as ﬁnal consumption expenditure of residemial households for new durable and non-
dunble goods and services less the net sales of second-hand goods, scrap, and wastes.

17

projects there is a good chance that few “high quality” (i.e., “high standards”) or more
desirable forms of housing investment will be produced. The end result is often a
misallocation of resources within a context of no overall increase in the total volume of

investment.

2-1-2. General Economists

In contrast, economists traditionally view housing as but one of many
alternative uses of scarce resources, and not necessarily one deserving high priority.
From their perspective, residential building may have a favorable labor-capital ratio,
but its capital-output ratio is unfavorable in comparison to many other investments. The
basis of their argument is whether rent or rental value is considered an adequate
measure of housing output. In this situation, residential construction shows an
unusually high capital-output ratio and consequently a long capital-recapture
period(ibid.: 101).

Among economists, beneﬁt cost or capita—output ratio provide a basis for a
theoretical justiﬁcation of delaying government action on housing investment. In other
words, housing is considered welfare spending with an excessively high capital-output
ratio. It is seen as an option that less developed countries could not possibly afford
until enough productive industrial capital had been secured (Strassman, 1993).

Economists who deal with thirty year perspectives come into conﬂict with
planners whose views often extend to much longer periods. Developing nations are

receiving apparently good but conﬂicting advice based on length of perspective. The

18

question is which will make great progress: 1) by concentrating on capital outlays
yielding a more rapid ﬂow toward other purposes as well as for housing; or 2) a
greater social investment where true cost of residential improvements may be measured

in terms of the beneﬁts foregone in capital outlay alternatives.

3. Relevant Theories on Housing Intervention in the Housing Market

In the preceding section of this chapter we see that both planners and
economists view housing intervention as a justiﬁable economic strategy, but with
differing outcomes, depending on the long— or short- term of analysis. Another
consideration is the impact of market imperfections which produce and distribute
housing services. Such imperfections are found in the existence of a social demand
curve reﬂecting internal unperceived benefits to housing consumers, which it perceived
would remove reason for public programs. Government intervention would be
substantially necessary if the market mechanism assured optimal allocation to housing13
(Burns and Grebler: 102—125).

The general conditions for optimality have long since been established in
economic theory. On the demand side, consumers will allocate their budgets in a way
that equalizes the marginal utility of expenditures for all goods and services. They will
spend more on goods and services yielding greater satisfaction and less on those of

smaller utility, until the satisfaction gained is equalized over the whole spectrum of

 

‘3 The theory of housing intervention ottmnes that housing ﬁts the general criteria for justiﬁable intervention that
have been established by modern economists. This theory analyzes the imperfections of the markets which
produce and distribute housing services.

19

consumption possibilities. On the supply side, output will be allocated in a way that
equalizes proﬁt rates. Resources will be shifted from the production of goods and
services yielding lower proﬁts to those yielding higher proﬁts, until rates of return are
equalized. These simpliﬁed abstractions will hold in practice only if several conditions
are met.

The conditions for transforming the theory of allocation into practice apply to
housing as they do to all other goods and services. Housing markets must operate
efﬁciently. Consumers must know the utility of housing services relative to alternative
objects of spending, and investors must have accurate information on rates of return on
housing relative to other investments. The beneﬁts must fully accrue to the demanders
and suppliers of housing services. These conditions are unlikely to be met in the real
world because of the cost of information, which requires expenditure of time, energy,
and/or material resources.

Furthermore, the characteristics of housing such as its obvious indispensability,
immobility, durability, and externality deter optimal allocation of resources in
competitive markets. The above facts points to a strong presumption that the market

mechanism fails to assure optimal allocations to housing.

3-1. State Intervention

Several standards are commonly imposed to judge whether state intervention is

warranted in the provision of goods and services. The question is whether housing

20

qualiﬁes under any of the standards speciﬁed below. Burns and Grebler identify four
major characteristics of standards: (1) merit goods, or public goods (2) goods
distributed unequally (3) large projects and economies of scale, and (4) market

imperfection.

3-1-1. Public Goods

The economic theory of public goods supports the foundations of government
intervention in the housing provision. By deﬁnition, public goods have two aspects that
differ from those of private goods involving characteristics behind their production and
consumption. First of all, public goods can be characterized by their nonrival
consumption or use at a given level of production among the people, i.e., consumption
by one person need not diminish the quantity consumed by anyone else. In other
words, a number of people may simultaneously use the same goods without interfering
with others' use even though it may still be possible for one person to use the goods
while others do not. The second characteristic of public goods is nonexclusion. This
means that it is impossible, or prohibitively costly, to conﬁne the beneﬁts of the goods
(once purchased) to selected persons. A person will then beneﬁt from production of the
goods regardless of whether or not he pays for them. Consequently, public goods are
not marketable (Browning and Browning, 1979: 23-25).

Without any provision for coercion, each individual will not participate in cost

sharing for public goods because there is no incentive for him or her to do so. In other

 

21

words, each individual will be better off by choosing non—participation as his
"dominant strategy" regardless of the others' choices. "The rational economic response
to such a situation is to refuse to pay for public goods" (Hula, 1988:6).

Because of these characteristics of public goods, it is difﬁcult to expect that
public goods can be provided by the private market at an optimal level. If provided
voluntarily by a few contributions, they will nonetheless remain rare and insufﬁcient.

When we assume a society has utilized all its available resources, there will then
exist a boundary of the possible maximum level of the production of goods (production
possibility frontier). Accordingly, if we categorized all goods supplied in a society as
public and private goods, then there will be numerous possible combinations of the two
types of goods to achieve the Pareto Equilibrium. According to this above discussion,
housing is a private good, but sometimes it is necessary to be provided by the public

sector.

3-1-2 . Distributional Justice

The market has little to do with equity. The problems of the market associated
with public goods and externalities are essentially concerned with the question of
efﬁciency - how to allocate resources in a productive way. There is a public consensus
that "fairness in income distribution is a desirable social characteristic that can be
consumed jointly by all members of the society." (Lim, 1988: 102) In attempting this
goal, some policy objectives should be targeted for those who are unable to compete in

the market. In this case, government intervention in the lower income housing markets

 

22

would be justiﬁed as an aim to assist individuals and groups suffering economic
hardships and/or social inequalities because of the inequality of the housing stock
distribution. In this way the government could stand in affordable housing by

intervening in market processes (Merret, 1979).

3-1-3. Large Projects and Economies of Scale

Intervention at too small a scale does not alleviate the problem but still
consumes state resources without the beneﬁt. Therefore, public support may be
justiﬁed for projects requiring substantial investment in order to generate any return, or
for projects characterized by substantial internal scale of economies. State intervention
under this criterion appears justiﬁed for two important activities related to housing.
First, land assembly in quantities sufﬁcient to permit large-scale, comprehensive
redevelopment may be possible only by a government expanding its broad ﬁnancial
base and exercising its right of eminent domain. Public support of urban renewal has
been justiﬁed on this criterion. Second, public subsidy may be required to mobilize
adequate resources for experimentation in building technology or for demonstrations of

new and risky types of housing projects that otherwise may not gain public acceptance.

3-1-4. Market Failure

Market mechanisms do not inevitably work well and indeed often fail in the

absence of state support. The nonexclusive and nonrival characteristics of public goods

 

23

are responsible for market failure. Moore (1978: 391) deﬁnes market failure as the
inability of a market economy to allocate goods efﬁciently or to distribute them in a
manner society deems equitable. Head (1974) states, "The causes of market failures
may be classiﬁed in three broad categories: nonappropriability, large numbers, and
imperfect information.” Nonappropriability means that it is impossible for producers or
consumers to appropriate the full social costs or beneﬁts of their production or
consumption. To a large extent, the concept of nonappropriability is equivalent to the
nonexclusive property of public goods already mentioned and what economic literature
refers to as externalities, spillover, or neighborhood effects

The existence of externalities is another rationale for public intervention into the
housing market. The problem of large numbers constitutes a second cause of market
failure. The third cause of market inefﬁciencies is imperfect information. To make
decisions about the future, individuals and ﬁrms need accurate information. To the
extent that this information is uncertain, inaccurate, unavailable, or expensive, one
should expect decisions to be less efﬁcient. The nonexclusive and nonrivalrous nature
of public goods is likely to cause goods information to be undersupplied.

The existence of market failure is a necessary condition for government
intervention in housing markets. Therefore, by the transitive property, the existence of
public goods is a necessary but not a determining condition for government
intervention. The theory of public goods and their responsibility for market failure
provides the basis for a theoretical justiﬁcation of public intervention and planning in
the housing sectors. Furthermore, it expands to the issues of housing provision with the

notion of economic efﬁciency and social equity.

 

24

3-1-5. Prisoner‘s Dilemma

Without cost sharing, each rational individual stops purchasing additional
amounts of goods to maximize beneﬁts when the marginal cost is equal with his
marginal utility, regardless of the beneﬁts the group might receive from that purchase.
The result is suboptimal. Incidentally, this result is the same as the "prisoner's
dilemma" in game theory, where resources used to increase the production of goods in
the public sector will be available only at the price of contracting other goods in the
private sector.”

In other words, when we assume that society produces only two goods, public
and private, in addition to the previous assumptions, i.e., nonexclusiveness as well as
the efﬁciency of government, it is possible to have an exchange rate between public
goods and private goods. This is the social rate of substitution between the public
goods and private goods we assumed on the basis of the above.

Individuals will stop purchasing public goods when marginal costs equal
marginal beneﬁts, and they will enjoy extra beneﬁts from the public goods supplied by
others as "free riders”. They know they cannot be prevented from beneﬁting even if
they do not pay. Participation for individuals will require members to bear costs, even
though it is surely more desirable for the increase of beneﬁts to society as well as to

each individual by supply of optimal public goods.

 

" Inthis regard, Samuelson (1954:387-389) set a condition for the optimality of public goods' production. He
showed that a different pricing rule would be optimal for goods subject to nonrivalry comumption.

25

One sure way to solve the under-optimal production problem is through
coercion. Actually, government ﬁnancing of public goods with taxes can be seen as an
efﬁcient instrument to overcome the free-rider problem (Browning and Browning, 33).
To ﬁnance public goods, we have to compete by relying upon tax revenues as levied in
a political process (Heilbrun, 1981: 426-431).

However, the real problem is that coercive pricing on public goods through
taxes does not necessarily solve the under—optimal production problem stated above,
even though this intervention will generally be beneﬁcial to more individuals of
society.” In relation to the argument, policy makers can decide on the nature and

extent of intervention in the housing sectors.

4. The General Theories on Housing Investment in Relation to National
Development.

The stage of national development is one of major factors in housing
investment. There are several major housing theorists who have proposed stages of
output allocation for housing with respect to the general stages of economic
development (Howenstine, 1957, Dennison, 1967, Kuznets, 1960, Rostow, 1971 and
Sn'assman, 1970). Their ﬁndings point out that stage of housing investment is geared to

levels of general development (Chang, 1986).

 

'5 McKean states (1965:496505) that a democratic voting procedure somewhat succeeds in registering citizen
preferences for public goods in the same way the market does for private ones.

 

26

In relation to the stages of economic growth, Rostow (1971, 1990:420—21)
deﬁnes ﬁve stages: (1) the traditional society, where the major source of employment is
farming, (2) pre-conditions for takeoff, (3) economic takeoff, (4) stages of economic
maturity, and (5) the stage of mass consumption. When a country is in the stage of
economic takeoff, a primary city exists in a nationally deﬁned urban system.

According to Rostow's stages of economic growth, Korea is in the stage of
economic maturity. This stage is characterized by the fact that the economy
successfully passes the takeoff stage and begins to allocate 20 to 30 percent of the total
output to maintain the process of economic development, and allocates the rest for
welfare programs such as housing and medical services.

Strassman (1970) suggests a theory of general development and housing
investment. He proposes that levels of economic development deﬁne the resources
allocated for housing investment. The rate of residential construction is slow in both
poor and advanced economies due, in part, to the lack of capital in the former and the
decrease in population growth in the latter. Middle-income economies have the highest
rates of residential construction because building materials and technology are
domestically produced; also, entrepreneurs and labor are free to enter the housing
market through their tests of the share of housing construction cost compared to GDP.
Strassman’s ﬁndings also lead to the conclusion that housing investments potentially
can increase continually along with a growing national income.

Howenstine (1957) describes the relationship between housing investment and

housing consumption and begins with the proposition of three stages of economic

27

development (Burns and Grebler, 1976: 100). They are priority of resource allocation,
stewardship, and the necessity of housing. Generally resource allocation for
construction resources is the major target objective, intended to build factories and
other essential producers' goods. Housing investment could be allowed only to the
extent necessary for the success of this objective. Then, in the following step, housing
investment should be expanded to the meet the 'minimum standard of health and
decency'- given the problems of existing unemployment and underemployment can be
solved- and workers are provided with the necessary capital. In the ﬁnal stage, it is
necessary to invest in housing. The improvement of housing conditions should be a
major policy goal.

Donnison (1967:75-78) also suggests a three—stage scheme similar to
Howenstine's, but places greater emphasis on central planning. Donnison deﬁnes the
stages of economic growth as (1) early stages of industrial growth and urbanization, (2)
economic growth, and (3) economic stages of maturity.

In the ﬁrst stage, the investment priority should be placed on education,
industrial investment, health, and defense. In relation to planning, the ﬁrst priority can
enforce minimum standards for urban planning. In the second stage, there is a trend in
which the rate of population growth has been decreased, and individual incomes are
increasing. Social needs can be obtained through a properly organized building
industry. In addition, the resources can allow builders to contribute to the housing
sector. Governmental set-up of a subsidy system for better housing for workers could
be implemented from the beginning of the second stage. In the last stage, the pressure

to meet the needs of special groups, such as large families and the elderly, is applied to

28

government. Howenstine and Donnison have in common that their research has partly
theoretical characteristics, and that they are internal to state allocation and address
housing at a national scale. Donnison's stage scheme identiﬁes that Korea is in the last
stage, and the government should actively take responsibility for the solution of the
housing problems. The government should start to focus on the needs of special groups
such as large families, the elderly, and handicapped. Therefore, housing investment is
to be a major policy goal.

Kuznets (1960), a traditional economist, states his empirical estimates of the
correlation between housing investments and levels of economic growth. He also
recognizes the role of political institutions in macroeconomic development plans. He
argues that governments should have blueprints for economic growth to ensure
economic development. From his analysis of cross-section post—war data covering 34
countries, Kuznets proposes a positive correlation between per capita income and the
ratio of the total construction investment to gross domestic product. According to his
estimation, "for the lowest income groups in the sample, total construction averaged
8.5 percent of gross domestic product. In the next two income ranges, construction
increased to 11.0 and 11.9 percent"(Burns and Grebler, 1977:21).

For the share of housing in construction investment, he states that the lowest
income countries are 30.3 percent and the middle income countries are 42.5 percent.
Kuznets attributes the generally rising trend to supply considerations: dwelling costs
have risen relative to other construction costs. He later modiﬁes his argument by

reporting on a longitudinal analysis of data for 11 developed countries. This ﬁnding

29

suggests that on the supply-side the cost of housing is relatively high compared to that
of construction cost, and that on the demand-side the rates of population growth
decreases explain the diminishing need for housing. The cross section covers a broad

development spectrum, with the developed nations at the upper end of that distribution.

CHAPTER THREE

EXISTING HOUSING INVESTMENT MODELS

1 . Introduction

This chapter reviews the conventional model, a macro-level analysis of housing,
developed by Burns and Grebler as a base with which to explain housing investment in
a country. Since publication of their work, several studies have appeared which
examine determinants of housing investment and its political impact on macroeconomic
policies. The models reviewed are: Renaud’s Model (1980), Annez and Wheaton’s
Model (1984), Chang and Linneman’s Model (1990), and Lim and Lee’s Model
(1992). These studies have assumed that on the whole national housing investment is
strongly affected by the level of economic development and associated with population
and urbanization. In addition several urban housing studies conducted in Korea are also
evaluated with respect to the above models. Therefore, this chapter serves as the

theoretical foundation for the empirical analyses presented in Chapter Five and Six.

30

31

2. Existing Models in Housing Investments and Points of Departure

2-1. Burns and Grebler’s Model: Theoretical Model

Burns and Grebler (1976) conduct an empirical model applicable to national
housing investment and housing markets in developing and developed countries.16

They initially select per capita GDP, population growth, and urbanization levels
to predict the percentage of the GDP represented by housing investments. Then, they
associate levels of housing investments to the level of economic development. To
interpret the housing problem in a given country, they use three independent
determinants: (1) level of urbanization, (2) the status of the national economy, and (3)
the social structure by cross-section analysis. They assume government actions affect
housing investments and consumption.

Findings of BG model show that the share of housing construction in the total
output of an economy is associated with the different levels of economic growth. They
introduce a nonlinear relationship between housing construction and the production of
an economy. Figure 2 indicates that the functions are truly parabolic, ﬁrst rising, then

peaking, and declining thereafter.

\

16

The countries included in their studies are listed in Appendix III.

32

The BG model is a structurally stable, using economic and demographic
variables. It addresses three major theories that describe housing investments on a
macro level concerning the proper size of housing in total output.

First, theoretical structure tests whether good housing directly improves human

conditions.

Second, theoretical structure used to explain levels of housing investment is
called the productivity theory of housing. It proposes that better housing will
improve health and labor productivity.

Third, theoretical structure tests whether housing investment is closely related
to the opportunity cost of capital in the country and the expected return on

housing projects.

The ﬁrst theoretical structure tends to insist on high housing standards that the
poor do not afford and /or do not usually require. Such high standards as assumed by
Burns and Grebler lead to problems of affordability and resource availability. Housing
policies based on these theories produced fewer dwellings, thus exacerbating the
housing problem.

The second argument relates to the fact that better housing improvements could
be achieved with limited improvements in roads, water supply, drainage as well as
public services. All of these will be accounted for social overhead in national accounts
(Renaud 1980).

The third theoretical issue is the fact that the share of housing investment will

increase when the scarcity of capita becomes less tense. This fact leads to the

33

conclusion that non-residential construction could yield higher returns to capital than

housing.

Figure 2 The Nonlinear Relationship between Share of Housing Investment and
GDP Per Capita

 

 

 

 

 

 

GDP Per Capita

 

Source: Burns, L. and Grebler, L. “Resource Allocation to Housing Investment: A
Comparative International Study.” Economic Development and Cultural Changes, 25(1):95-
121. 1976.

 

34

Table 1 Regression Analysis of Burns and Grebler's Model

 

 

H =a +31 GDP +p2 GDP2+ [33 DPOP + [34 DPOP’ + as URB2
DependenLYariahle

H: H is the share of housing in total output measured as average residential
construction as a percentage of average annual gross domestic product.

Independentlariahle

Y: GDP is development level measured as the average annual gross domestic product
per capita
g: DPOP is the population growth measured as average annual rate of increase in

national population

u: URB is urbanization measured as average annual rate of population increase in
cities of 100,000 persons and over, divided by the average annual rate of increase in
national population.

 

The abstract market analyses of BG model presented above (Table 1) are based
on neo-classical economics whose model provides a very useful analysis for getting an
idea of the determinants of housing investment. Therefore this model provides a valid a
point of departure for a new approach, and establishes an overall picture of the role of
government and its relation between economic development and population growth.

However, Burns and Grebler do not consider the speciﬁc way in which the
government is related to social agents and institutional arrangements in the housing
supply and planning. Even though the BG model has its clarity, it lacks speciﬁc
institutional arrangements of the housing market, an omission which may lead to
misunderstanding of the real situation. Improvement of the BG model should include

the institutional arrangement for housing provisions.

 

35

2-2. Renaud ’5 Model: A Model forA Single Country

Greater attention to economic variables has been found in studies published after
the BG model. Renaud (1980) tested the housing investment model in Korea by
focusing on variables of economic and ﬁnancial conditions. These add a new
dimension, which includes a contextual component addressing ﬁnancial conditions and
economic status.

Working with neoclassical economic assumptions, Renaud tried but failed to
examine the changes in share of housing investment over the total output in Korea. He
used a reduced form of the model and selected the per capita GNP as the economic
indicator, ﬁnding a nonlinear relationship between the share of housing in the total
output and GNP. Renaud tried to induce a more sophisticated model of housing ﬁnance
to obtain a more effective explanation of the behavior of the housing share in the total
output (Renaud, 1980: 397). Table 2 illustrates the overall summary of variables and
the results of model speciﬁcation.

In addition, Renaud’s time-series study of Korea investigated the effect of the
domestic savings ratio to the GNP and interest rates in an unregulated money market.
His ﬁndings show that the existence of a non-linear increase of share of housing
investments as the economy expands is supported.

However, there are apparent limits to the Renaud model. The variables used do

not explain the behavior of the variation of housing investment, which is a signiﬁcant

36

contextual factor. For example, domestic savings to GNP variable has related to the
effects of multicollinearity because the variable is a ratio of GNP.

Renaud also failed to consider the speciﬁc institutional context of the housing
ﬁnancial situation in Korea. Even he has agreed that restriction of housing ﬁnance by
the government is one of the main reasons for inefﬁciency in the housing market (ibid.:
397). His model simply explains the relation between two additional variables, such as
the domestic savings rate and interest rates. We, therefore need a different framework

of analysis to reach a better understanding of Korean housing.

Table 2 Regression Analysis of Renaud's Model

 

 

n =a +51 GNP +52 GNP2+B3 DPOP +134 DPOP: + as URB + [36 van2
+ 57 1/GNP + as DOMSAV + [39 l/DOMSAV + B10 UMM

Demudentlaﬁahle

H: H is the share of housing in total output measured as average residential
construction as a percentage of average annual gross national product.

Indemndentlariable

Y: GNP is development level measured as the average annual gross national product

per capita

g: DPOP is the population growth measured as average annual rate of increase in
national population

u: URB is urbanization measured as average annual rate of population increase in

cities of 100,000 persons and over, divided by the average annual rate of
increase in national population.

domsav: ratio of domestic savings to GNP

umm: unregulated money markets17

 

 

'7 Unregulated means that revolving aromid the private money market outside govermnera comrol. For more dcails
on the unregulated money market (UMM) in Korea see Yong-Chul Park.

 

37

Renaud analyzed four common nonlinear models for Korea. The four equations
were as follows; quadratic form, reciprocal transformation, logarithm-inverse
transformation, and a double log regression. After evaluating Korea’s situation, he
compared the results to BG’s, as demonstrated in Table 3. Among four equations,
Renaud showed that the reciprocal transformation and the logarithm-inverse transfer
produced results consistent with the behavior of housing investment in Korea and also

with the thirty-nine countries studied in BG’s analysis (shown Table 4).

Table 3 Share of Housing in Total Output (SHTO): Korea Compared to Bums-
Grebler Sample

 

 

Period Population Urbanization Per SHTO SHTO SHTO SHTO
Growth Indicator C apita actual projected” projected” projected
Rate“ GDP” Korea Korea Eq. 3

Eq.5

Post-War 2.98 1.88 144 1.51 2.85 1.46 1.4

(1957-61)

1S! Plan 2.59 2.17 165 1.73 2.95 1.79 1.89

(1962-66)

211‘] Plan 1.85 3.74 241 3.1 3.75 2.82 3

(1967-71)

3rd Plan 1.98 2.64 357 3.77 5.75 3.89 3.79

(1972-76)

4th Plan 1.59 2.2 865 3.6 4.55 6.15 4.9

(1977-81)

B~G 1.79 2.26 8.84 4.558 5.78 5.79 4.78

Sample

 

 

Note: 1) 5 year average
2) Constant 1970 US dollars
3) Burns-Grebler Model
4) Korea Equation 5: Y = a -b/x
5) Korea Equation 3: ln Y = a -bx
6) Sample means for Group of 39 countries

Source: Renaud, Bertland. “Resource Allocation To Housing Investment: Comments and
Further Results.” Economic Development and Cultural Change, 28(2): 393 January 1980.
Economic Planning Board, Korea Statistical Yearbook, 1982.

Table 4 Renaud's Estimated Regression Equations (Effects of Demographic

 

 

 

Variables)
cm Uni; 01:111 DROP Duo? 17sz a“ * * snail r-
at _ . ' ' _ Value _
(1)8HTO , -7.96 6.95” 4.062) -- .- -- 0.59 0.726 11.5
v (3.28) (2.49) (.45)
(2)8HTO 22.36 - -- -1536” 2.812) 0.80 0.51 16.2
7 . (5.56) (4.79) (.99)
_(3)SHTO, 22.14” 2.40 0.45 47.87” 3.35 -- 0.82 0.51 16.2
(9.87) (6.90) (1.08) (3.34) (1.46)
‘(4)SHTO -2.70 2.27 -0.28 3.45 -0.21 34.043) 0.93 0.34 33.1
, (8.71) (4.62) (3.44) (6.83) (1.27) (7.80)

 

Note: Numbers in parentheses are t-statistics.

1) SEE: standard error of estimate

2) Signiﬁcance at 95 % level

Source: Renaud, Bertland. ibid. p.396

2-3. Annez and Wheaton ’8 Model: A Cross-National Context

Annez and Wheaten (1984) propose a set of structural cross-sectional analyses

models for housing investment in a cross-national context. First, they estimate the total

growth in the housing stock and that portion of the growth that is recorded in GNP

accounts. Second, they predict the average level of housing services among the newly

constructed units and the price or cost of construction. They then assume that the share

of housing investment equals the product of change in stock, average size, and cost,

divided by

GNP.

As endogenous variables, Annez and Wheaten estimate the growth of housing

stock, the cost of new construction, and total volume of construction. In the case of

39

exogenous variables, they select demographic variables (population growth and
household formation), policy variables (the cost and availability of credit, public
housing production) and the level of income or economic development (GDP per capita
or per household).

Annez and Wheaton’s model18 is more comprehensive than the BG model; it is,
however, more difﬁcult to obtain some of the variables used in AW model, especially
since they draw upon over 30 cross-country comparison studies.19

The results of their research are probably more important for their
methodological implications than for their substantive interest. Aside from the problem
of data collection, various countries differ in the importance of the variables. For
example, the signiﬁcance of interest rates and savings rates in housing investment can
be interpreted differently in developed countries than in developing countries. In
addition, Annez and Wheaton use abstract models to explain the relation between
certain variables, whose applicability to the various countries' contexts, however, is not
proven. Even though there are high correlation between housing supply ratio and per
capita GNP, housing stock variable is not included in this conventional cross sectional
housing investment studies. Nonetheless, they conclude that government intervention in
the housing market has caused many problems that have contributed to a worsening

housing situation.

 

'8 Hereafter referred to as AW Model.
'9 See Appendix III for the list of countries which are included in the Annesz and Wheaton's model

40

2-4. Chang and Linneman ’5 Model: A Cross-National Context

Chang and Linneman (1990) estimate alternative models of the growth rate of
real housing investment. They analyze the different patterns of housing investment in
Taiwan, Korea, Japan, and the United States. They introduce homeowerhip, dwelling
size, and interest rates as independent variables. Using time series methods, they
investigate reliable forecasting models for housing investment trends in each country.
Their ﬁndings support the assumption by Buckley and Madhusudhan (1984) concerning
ﬁnancial deepening and inﬂation. In this respect, they demonstrate a clear relationship
between ofﬁcial interest rates and change in housing investment in the United States,
where a formal housing / ﬁnancial market exists. While in Korea and Taiwan, because
of the lack of a formal housing / ﬁnancial market, ofﬁcial interest rates are not
sensitive to the change of housing investment. Regardless of this difference, Chang and
Linneman cannot explain causation for instability and distortion in housing market
demonstrated in their ﬁndings, since they cannot analyze the speciﬁc institutional

elements for housing investment.

41

2-5. Lim and Lee ’5' Model: A Model for A Socialist Country

Lim and Lee (1993) conducted an empirical analysis of the historical trends of
housing investment in a socialist country. They postulate economic and political
variables to ﬁnd out whether politics play any role in behavior of housing investment.
In addition, they investigate the effect of social, political, and economic factors on
consumption of housing in China. They present income elasticities using time-series
and cross-section analysis. In order to overcome the problem of reliability of data, they
supplement the total product of society and national income data. However, their model
does not estimate the impacts of international and institutional elements in one country.
International and institutional elements exercise major impacts on decision-making. It

then affects housing investment and the provision of housing.

3. Previous Studies of Urban Housing in Korea

There are two major trends of urban housing studies in Korea. We can classify
housing policy studies and housing market studies by making the following

generalizations.

42

3-1. Housing Policy Studies in Korea

These policy-oriented housing studies were actively involved in the national
debate on housing problems and policy formations in the 1960s and 1970s. At that
time, research formats inclined toward a verbal description of housing policies or case
studies focused on a speciﬁc topic or area.

According to Lim’s classiﬁcation (Lim 1987, 1988), housing policies can be
divided into two major categories: explicit and implicit policies. Explicit housing
policies have focused on solving housing problems directly. Low-income public
housing20 projects and housing ﬁnance plans are examples of such a policy application.

Implicit housing policy refers to the indirect impact on housing issues and policy
by economic factors, including: taxes, military policy, and other investments, which
since the 19605 have become increasingly complicated by housing stock shortages and
affordability, two major issues needing to be incorporated into analysis of investment
behavior.

Related to this, is the question of scale where one of the main policy concerns
has been how many numbers of housing units should be built over a given period to
reduce the housing shortage. Attempts to answer this question address have paid the
most attention to changes in the housing supply ratio, while housing affordability has

also been a policy concern. Since the 1960s, as implicit policy concerns, housing stock

 

2° Public housing is deﬁned as the housing initiated by public agency and included publicly funded private rental
housing. (Ministry of Construction, 1989)

43

shortages and affordability could be identiﬁed as two major issues. Related to this, one
of the main policy concerns has been how many numbers of housing units that should
be built over a given period to reduce the housing shortage. Changes in the housing
supply rate have been paid the most attention. Housing affordability has also been a

policy concern.

3-2. Housing Market Studies in Korea

There are two major study topics on the housing market from the macro-level of
analysis: 1) the working of the housing market; and, 2) the efﬁciency of government
intervention in the market. The issues of housing market consist of demand, supply,
and housing prices; while capital gains taxes and price control of newly built
apartments are major discussions related to government intervention in the housing
market in Korea.

Since the late 1970s, some scholars have begun to pay attention to housing
markets as a topic Follain et al. 1980, and Malpezzi et al, 1985). These contributions,
like that of Chang and Linneman (1990) and Buckley and Madhusudhans (1984)
successfully or not deﬁned formal and informal housing market as factors against
which to compare government intervention. After that, the number of housing market
studies has increased, indicating a shift of concern from housing polices to the housing
market inﬂuences. This transition seems to coincide with the introduction to the

housing research of the neo-classical economics approach to analysis, a factor of

44

importance to the housing debate in Korea. Most neo-classical housing economists have
insisted that government intervention is one of the causes of housing problems rather
than a solution (Annesz and Wheaton, 1984).

In reviewing the urban housing studies in Korea (Table 5), we ﬁnd that most
studies have focused on the micro level of analysis. Though there are certain limitations
concerning these studies, which attempt to analyze urban housing market studies, they
provide depth and color to the more abstract macro-analyses.21 Most studies have
focused on the equilibrium of supply and demand and estimation of housing demands
by ex ante evaluation. Estimation of housing demandzzis the major study topic,
followed by estimates for price and income elasticities of housing demand among the
urban households in Korea.

Table 5 gives a selection of estimated income elasticies for housing demand and
consumption. Variations among these values are wide. Using Korean data for 1972,
Lluch et al. reported a fairly high value (2.48). In contrast, Mills and Song (1976),
using Korean time-series data for the years 1962 to 1975, present the lowest value
(0.027). Song and Struyk (1976), with Korean time-series data for the years 1960 to
1965 and 1966 to 1970 studying over 50,000 numbers of the urban area population
using census data through 18 measurements presents 0.91 income elasticies and -2.42

price elasticities.

 

2‘ As macroeconomics studies, time series studies relating income and housing expenditure were conducted in the
early 19608.

22 The demand for housing reﬂects the willingness to pay for a set of attributes or services which are provided by the
physical components of lot and housing structure. The most importam of these attributes are access, space, tenure,
on-site services, and shelter.

45

Sources of the variations are many. First of all, people living in different places
at different times may not have an identical demand function. Second, the estimates
may vary due to measurement errors. Third, speciﬁcation biases could widen the gap
between estimated elasticities among different studies. The third point is extremely
important in interpreting the result of demand studies and searching for reliable
estimates.

Follain, Lim, and Renaud (1980) analyze the Korean census data and obtain a
price elasticity and an income elasticity respectively. They also deﬁne four
determinants of home ownership in Korea. First, permanent income proves more
inﬂuential than current determinants in home ownership decisions. The researchers ﬁnd
a close association between mobility and tenure status. A family living in a rental
dwelling might be tempted to move. They compare Korean and U.S. data. They ﬁnd
that both income and price elasticities of housing demand are comparable in both
countries. Income elasticity (0.57) is less than one, and price elasticity (-O.2) is
negative or smaller than one in absolute value.

In addition, I. H. Kim (1983) analyzed a survey data and presents 0.09 ~O.24
income elasticities and -0.06~-0.16 in price elasticities. This result shows abnormal
inelasticity. Most of the housing market and demand studies strongly suggest that
overall, the consumer demand for housing is not elastic with respect to both price and
income. This implies that housing prices will depend upon not the quantities of

excessive demand through market mechanism but control of the government and its

46

effect. It seems to me, without national estimates, the order of magnitude could be used

for another country.

Table 6 lists estimated income elasticities for housing demand and consumption

in selected countries. The table shows that most income eleasticities fall between 0.5

and 1.00, suggesting that demand is inelastic. Compared to other countries, Follain et

al.‘study (1976) estimated that demand is inelastic in Korea.

Table 5 Summary of Previous Housing Market Studies in Korea

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Author Income Elasticity Price Elasticity Data Remark:
independent
variables
Song-Stryuk 0.91 —2.42 Time-Series and increasing rate of
(1977) Cross-sectional households
analysis: 1960
~1970
Mills- Song 0.027 —0. 134 Time-Series: total households
(1979) 1962—1975
Follain et a1. 0.57 -O.20 ~ 030 EPB special rate of housing
(1980) survey data in shortage, number
1976 of households,
the distance to the
CBD
Jeong Ho Kim 0.09 ~ 0.24 -0.06 ~ 016 housing market size of household
(1983) and need
analysis: 1982
JoonSooKim 1.14~1.18
(1984) a
Kilian-Young 1.726 Income Elasticity
Kim (1988) for Housing
Demand: 1.536

 

Source: loon Soo Kim. “Time Series-Analysis of Determinants of Housing Supply-
Demand in Korea. ” The Study of Korea Development, 5(4) 116, Korea Development

Institute, 1983.

 

47

Table 6 Summary of Previous Housing Studies

 

 

 

 

 

 

 

 

 

Author Place Survey Year Income
Elasticity

Pooled owners and

renters

Howe and Musgrove Guyaquil, Ecuador 1968 1.10
Howe and Musgrove Lima, Peru 1969 1.31
Howe and Musgrove Caracas, Venezuela 1966 1.09
Lluch et a1. Mexico 1968 0.93
Betancourt Central Chile 1964 0.79

Lluch et a1. Korea 1972 2.48
Lluch et a1. Urban Korea 1971 0.86

- Follain et a1. Korea 1971 0.54
Howe and Musgrove Bogota, Colombia 1968 0.98
Renters:

Follain et a1. Korea 1976 0.42
Mayo et al. Cairo, Egypt 1980 0.25
Mayo et a1. Beni Suef, Egypt 1980 0.50
Ingram Bogota, Colombia 1978 0.80
Ingram Bogota, Colombia 1978 0.72
Ingram Cali, Colombia 1978 0.16
Ingram Cali, Colombia 1978 0.47
Strassmann Cartagena, Colombia 1978 0.78
Jimenez and Keare Santa Ana, E1 1980 0.27

Salvador

Owners:

- Follain et a1. Korea 1976 0.62
Ingram Bogota, Colombia 1978 0.78
Ingram Cali, Colombia 1978 1.19
Strassmann Cartagena, Colombia 1978 1 .05

 

Source: Malpezzi, S. and Mayo, S. “The Demand for Housing in Developing Countries:

Empirical Estimates from Household Data. ” Economic Development and Cultural Change 35

(4): 687-721. 1987.

 

CHAPTER FOUR

BACKGROUND OF HOUSING MARKET: DIMENSIONS AND
PA'I'I'ERN S OF HOUSING DEVELOPMENT

1. Introduction

In this chapter, linkages are made between macroeconomic plan, housing policy
and housing conditions, operations, and markets. Korean housing policy issues are
explained in the context of the social and economic transformation Korea has
experienced in the course of its rapid economic development and urbanization since
19605. Accordingly, the legacy of past policies are evaluated with respect to their
effectiveness and efﬁciency. The housing operation system, which includes housing
institutions, housing ﬁnance, and housing programs is then described. Finally, an

overview of housing problems and issues is presented.

2. Overview of National Context of Korean Housing

In order to further discuss housing in Korea, it is necessary to summarize its

national context.

49

2- 1. Historical Background

From the 1300s to the twentieth century, Korea was a protectorate of China. It
was respected and valued for its scholars, and scientists and isolated because of its
geography which is mountainous with the Yalu River as a natural barrier in the North.
After 1854 Japan began modernizing and reaching for Korea as a doorway to the
continent and took Korea by force. Between 1905 and 1945 Korea was a colony of
Japan, which developed railroads and infrastructure, but focused development mainly
in the North, which was lost following World War Two. At that time the country was
divided at the 38th parallel, leaving the South with predominantly an agricultural based
economy at the end of the war.

In response to communist presence in North Korea an American military
government was set-up in Korea. In 1951 a democratic government was instituted but
shifted to a military government in 1961. Though elections were held, they inevitably
supported the military government. In addition, in 1987 for the ﬁrst time in Korea’s
history, a president was elected by the direct vote of the people without serious
illegitimate activities. Since then, Korea has experienced sound political development
and spawned a new hope for democratization (Lim, 1990, Kihl. Young W. 1990). The
shift of power and inﬂuence during the years from 1948 to 1995, have been dramatic
and yet persistently supportive of industrialization and development which extends from
the national to the local level. Furthermore, this trend accelerates local government

decentralization and local government reform as high visibility issues.

50

In terms of political stability, throughout the decades of the 60s, 708, and 80s,
South Korea has had ongoing conﬂict with North Korea. It has expended about 5
percent of its GNP (or 30-40 percent of the total government budget) on military
defense. This means Korea’s burden of military expenditure has been onerous. Yet in
spite of that commitment of resources Korea has managed to invest heavily in housing

and urbanization.

2-2. Land

Land taken by the Japanese during the ﬁrst half of the century was returned to
the land lord class in 1949 as part of a land reform and industrialization strategy. After
1961, large corporations (Clzaebolﬂ'3 have accumulated over seventy-ﬁve percent of the
land and industrial resources, while a wage labor class developed in textiles. Since the
1970s heavy industry in shipbuilding and automotive industries and information
technologies have joined textile as economic factors.

In terms of land and population Korea is relatively a small country with an area
of 99,000 km2. Furthermore, approximately two thirds of the country is inhabitable
mountainous land. It has only a very small amount of arable land -— 30 percent -- with a
very high population density, 439 persons per square kilometer. Consequently, land

price is extremely high and the assembly of residential land is very difﬁcult?“4

 

23 For more details see Appendix IV. Glossary

24 For detailed accounts as to the way the residential land are linked together housing pricies, see Hannah, K.H. Kim
and Mills. 1993.

51

2-3. Economy

Korea’s economy has grown rapidly since the 1960s, showing GNP rising from
$2.4 billion in 1962 to $ 376.9 billion in 1994. Rapid economic growth has also been
represented by the considerable hikes per capita GNP, which jumped from $62 in 1962
to $84,483 in 1993 (Bank of Korea, 1994). The average annual economic growth rate
in 1953-93 was very high 7.56 percent. The growth of per capita GNP has been
reﬂected in real wage increases in Korea. Figures 3 and 4 show the postwar real GNP
per capita and its growth.

The growth of the Korean economy has been linked to the international
economy. Korea’s international trade has expanded signiﬁcantly (Lim, 1990). Its
exports expanded from $1.09 billion in 1971 to $57 billion in 1988, and its import
from $3.8 billion to $52 billion. It is particularly important to observe that the
expansion of exports ﬁnally created a trade surplus in 1986 for the ﬁrst time in Korean
history. In addition, Korea became a heavy debtor nation soon after it embarked on its
economic development plans.

The absolute amount of debt and its share as percentage of the GNP has
dwindled gradually since 1983. In 1989 the amount of total foreign debt was $ 30.3
billion and the net foreign debt was reduced to $1.6 billion (Table 7). In fact, Korea
became a middle-income nation with a large amount of transactions in international

markets.

52

Figure 3 GNP Per Capita (1970 Constant Prices)

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1000000 30000
900000
.25000
800000
700000
}.20000
600000
5 / c
3 500000 15000 £5
3 j 33
400000
I 10000
300000
5000
0

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

L—PerCapitaGhP..—GI\PI

 

 

 

 

 

 

53

Figure 4 The Growth Rates of GNP Per Capita

 

 

20

 

151

10.

 

 

 

 

 

 

88882

1962
1965
1968
1971
2 1974
1977

 

EGMmRateFUGIpKaG‘NP—ermRaeGhPl

 

 

 

Table 7 Indicators of Trade and Foreign Debt

 

 

1. I Account - - Total—Foreign Debt r Net-magma ‘ ~

 

1983
1989

 

1' 4.4 N 27.2 19.6

-0.02 46.8 35.6
4.2 44.5 32.5
7.7 35.6 22.4
11.4 31.2 7.3
4.6 30.3 1.6

 

'Nccc. Unitibillion uss

Source: Ministry of Trade and Industry, 1990

 

 

54

2-4. Urbanization

During the last three decades, Korea has undertaken the highest tempo of
urbanization in the world. Consequently, the growth rate of the urban population in
cities over 100,000 persons (a 4.461 percent per year average) was two and half times
that of the total population (1.768 percent). Korea has transformed from a rural to an
urban society. Table 8 demonstrates how Korea has become an urbanized nation. The
share of urban population rose from 50.1 percent in 1970 to 70.5 percent in 1982, 75.2
percent in 1986, and 83.7 percent by 1992. The rate of urban growth in Korea has
been quite high, although it has tended to decrease with the decline of rural-urban
migration. Urbanization has followed industrialization with considerable public sector
planning which has produced many new towns, resulting in a real estate boom. The

predominantly agricultural country found at the end of World War II is rapidly

disappearing or gone.

55

Table 8 Urbanization

 

 

Year Total Urban Rural Urban Population
Population Population Population (in percentage)
1970 31,435 15,750 15,685 50.1
1978 36,628 23,238 13,390 63.4
1982 39,114 25,577 11,537 70.5
1984 40,430 25,599 10,831 73.2
1986 41,161 30,936 10,225 75.2
1988 41,975 32,963 9090 78.4
1990 42869 35558 7832 81.9
1992 43663 37319 7249 83 .7

 

 

Note: Unit: 1000 persons
1. Nationwide population of 1982 -1988 is in accordance with residing population
census.
2. Urban area population includes that of Eup with more than 20 thousand residents.

Source: Ministry of Home Affairs (MHA), Municipal Yearbook of Korea. 1987 , 1993

2-5. Income Distribution

Income redistribution as well as land reform has also been very successful and
notable. A notable factor has been credit rationing. The long-term trend in income
distribution has been fairly stable. The Gini coqj'icient25 has changed slightly as
follows: 0.3439 in 1965, 0.3322 in 1970, 0.3908 in 1976, 0.3574 in 1982, 0.3567 in
1984, and 0.34 in 1988 (Lee, 1990). This means that Korea is relatively equitable
among the developing countries. Education and savings are particularly appreciated in

this society.

 

2’ This index is a measure of the degree of ineqality in the income distribution.

56

3. Overall Housing Conditions

This section presents an analysis of overall housing conditions along four
dimensions: housing supply, physical housing conditions, housing occupancy, and
housing prices. In general, current housing conditions in Korea are not good, since
serious problems of housing stock shortage, squatter settlements, small dwelling size,
lack of facilities, and overcrowding exist because of high population density and rapid
urbanization. Korea has witnessed some relatively poor conditions in housing stock
(72.4 percent of dwellings-to-households ratio), in physical housing conditions (small
dwelling size and lack of some dwelling facilities), and in housing prices and

consumption (housing speculation, high housing prices and expenditures).

3-1. Housing Supply

Housing quality has improved considerably with the sustained growth of
Korea’s economy since the Korean war, yet major stock shortages remain a problem.
Although housing investment has increased, the housing supply rate declined from 0.83
in 1960 to 0.791 in 1993 as shown in Tables 9 and 10. Table 9 shows the total number
of households (10.2 million in 1990) and the total housing stock (7.3 million in 1990)
in Korea. The imbalance between them is often referred to the great housing shortage

crisis in Korea.26

 

2‘ Some housing analysts in Korea take a skeptical view of the accuracy of the total housing stock (Kim, J. H.1988,
You, I. H. 1988). They argue that housing stock data does not represem the actual number of housing units. In
fact, in the multi-household—housing, they are considered as one household in census data.

57

The problem of housing shortage has been more severe in large cities. In Seoul,
the capital, the housing shortage rate was 44.7 percent in 1985 (Ministry of
Construction (MOC), 1989 and Chung, 1990).

The reasons for the decline of the housing supply rate27 are the high growth of
household formation and urbanization, as well as the demolition of old housing stock.
In any event, the Korean housing stock does not yet appear to be mature, and hence
housing experts have suggested that housing investment will continue to grow along

with the per capita GNP.

Table 9 A Summary of the Korean Housing Stock Conditions”

 

 

Year ' Per Capita ~ Housing Housing ; Population Dwelling Household” 7 Housing Supply
7 , ,gosz’.» Investment 3’ Investment(%)gGrowth(%) Units" , , amuse)
1955 x 30,916 11.97 1.78 2.96 .— 79.5

- 1960,; 34,517 18.67 2.18 3.00 3,464 4,198 82.5

’ 1965 48,074 32.16 1.69 2.28 3,867 5,133 75.3
19707 81,500 876 3.41 1.81 4,360 5,575 78.2
1975‘ 140,701 1,808 4.40 1.89 4,734 6,367 74.4
1980 216,672 2,507 5.78 1.81 5,463 7,331 74.5

’ 1985 308,897 4,864 4.33 1.95 6,104 8,751 69.8
1990} 503,182 145,773 8.22 0.92 7,357 10,167 72.5

 

 

Note: 1) 1970 constant prices
2) Unit: won
3) Unit: Billion won
4) Unit: 1000 dwelling units
5) Unit: 1000 households

Source: Korea National Housing Corporation, Housing Handbook, 1983, 1989, 1994.
Ministry of Construction, Yearbook of Construction Statistics, 1993.

 

27 The rate of the housing stock shortage equals (1- housing supply rate)

Table 10 Housing Supply by Year

58

 

 

 

 

Source: Korea Housing Bank. Housing Economics Statistical Yearbook. 183-184. 1994.
Ministry of Construction. Yearbook of Construction Statistics. 1993.

2) Unit: 1,000 households”
3) Unit: 1,000 units

 

Year Dlvmon - - 1 Total Persons per Ordinary Number Of ‘ Housing '
.-   ’* v f , I Households" each Households” Housing Units” Supply Rate
* * , Household f (95)
1970 Whole Countr 5,857 5.2 5,576 4,360 78.2
._,i;:_:,f;,;j; All Cities 2,525 5.0 2,404 1,398 58.2
21975 ‘ Whole Countr 6,754 5.1 6,367 4,734 74.4
x ‘ All Cities 3,412 4.9 3,216 1,809 56.3
1980 Whole Countr 7,968 4.6 7,470 5,319 71.2
g-;;;;je All Cities 4,668 4.5 4,294 2,542 59.2
' 1981 Whole Countr 8,147 4.7 7,712 5,435 70.5
All Cities 5,053 4.7 4,697 2,823 60.1
.1982 I Whole Countr 8,418 4.6 7,962 5,584 70.1
' 1 All Cities 5,309 4.5 4,973 2,951 59.3
.1983 Whole Countr 8,762 5.5 8,220 5,759 70.1
. 2 , All Cities 5,636 4.4 5,262 3,227 61.3
.7 1984 , Whole Countr 9,143 4.4 8,486 5,931 69.9
i _ All Cities 6,003 4.3 5,583 3,320 59.5
{11985 ; Whole Countr 9,575 4.2 8,751 6,104 69.8
All Cities 6,334 4.2 5,797 3,351 57.8
1986 '3 Whole Cormtr 9,859 4.2 9,037 6,303 69.7
; ; .. All Cities 6,337 4.2 6,137 3,593 58.5
1987 Whole Countr 10,175 4.2 9,320 6,449 69.2
: ‘ ,2 All Cities 6,532 4.2 6,328 3,747 59.2
1988 Whole Countr 10,513 4.0 9,612 6,670 69.4
All Cities 6,749 4.2 6,536 3,935 60.2
gfi1989 : Whole Countr 10,419 3.7 9,920 7,438 71.4
41990 A Whole Countr 11,354 3.71 10,168 7,357 72.5
Note: 1) Excluding nonblood relationship and one-person households

7’ Ordinary households are based on family groups related by birth or marriage including some persons living with
family group such as servams, housemaids, boarders or employee related the household business.

59

3-2. Physical Housing Conditions

In Table 11, one can observe some of the quality conditions. The average ﬂoor
area in the 1970s, rose from 55.11 m2 to 78.2 m2 in 1990. The number of persons per
room fell from 2.34 to 1.7. The level of sanitary facilities steadily rose. The share of
new dwellings equipped with modern facilities such as ﬂush toilets and running hot
water has been rapidly augmented. Clearly the quality of Korean housing made some

net gains.

Table 11 Housing Quality Trend and Status Quo
m’ff‘Tf'ififi‘fiff T 1970 , “1975 ' ' ‘ 1980 1985 ‘ ‘ , 1990 i '
Floor/unit” 55.11 57.7 66.26 73.4 778.2
rm’m “ 3-1 -- -- 3'6
person/room 2.34 2.3 2.115 1.934 1.7
area/person” -- 7.9 9.64 11.19 13.9
anemone-(75): -- 15 18.4 33.1 40
cc. -- -- 77.9 65.3 --

 

 

-- -- 10.0 20.0 -

 

 

 

Source: Economic Planning Board, Housing Census, 1970, 1975, 1980, 1990

60

3-3. Housing Occupancy and Tenure

Homeownership in Korea is not widespread because of the limited availability
and terms of mortgage money29 and absence of tax beneﬁts. In Korea, there are four
kinds of tenancy. First, is a substantial amount of deposit with monthly rent (Sak-Wol-
Sei). Second, a tenant who pays only a monthly rent and no deposit. Monthly renters
who do not have access to accumulated savings from relatives or from good incomes
face severe problems. The pure rent, western style, is used only by the poorest
households. In third form, called Chonsei, tenants occupy a whole house30 and pay a
lump sum without paying monthly rent but do not gain ownership. The Chonsei is
based on this capacity scarcity.:

Fourth, partial Chonsei, households occupy a part of the house and pay a lump
sum (Ministry of Construction, 1985, 80).

For a partial Chonsei the imputed rent is estimated as

R = Z[(A—D,t)i +D]/t
t=l
where A = amount of original Chonsei
D= monthly deduction from Chonsei
B = period
i = curb market interest rate

For security deposits with monthly rent, R is equal to the monthly rent to which

is added r the deposit multiplied by the monthly curb rate. With the real estate boom,

 

29 Borrowing from the extremely active unregulated money markets is not possible because the rates are very high
(between 3 and 5 percent momly) and the terms of maturity very short (typically a maximum of twemy momhs.)

3° The deﬁnition of a whole house include single family dwelling unit, multiple family dwelling units, duplex and
apartments.

61

Chonsei deposits have risen and therefore, implicitly, so have implicit rents. Table 12
reveals that the largest proportion of tenants do not have their own independent unit but
have to share their unit with others. One of the factors of the Chonsei system is that it
makes access to rental units dependent on accumulated wealth.

Both Chonsei system and credit rationing have distorted housing demand for the
last three decades as have underinvestment in the housing sector and increasing
concentration of real estate wealth in the hands of middle and upper income groups,

and increasing prices, while over crowding also continues to be a problem.

 

 

 

 

 

Table 12 Distribution of Households by Tenure
1970 1975 1980 1985 1990

Korea 5,857 (100.0) 6,754.3 (100.0) 7,969.2 (100.0) 9,571.4 (100.0) 11.3545 (100.0)
Owner 3,719 (63.5) -4,259.9 (63.1) 4,671.8 (58.6) 5,127.2 (53.6) 5,667.3 (49.9)
Chonsei 1,019 (17.4) 1,171.3 (17.3) 1,904.5 (15.3) 2,201.9 (23.0) 3,157.1 (27.8)
Rental “ 91405.6) 1,049.1 (15.5) 1,231.3 (15.5) 1,892.5 (19.8) 2,172.6 (19.1)
Others 205 (3.5) 273.9 (4.1) 161.6 (2.0) 349.8 (3.6) 357.6 (3.1)

Cities N/A 3,412.6 (100.0) 4,670.0 (100.0) 6,330.8 (100.0) 8,462.4 (100.0)
Owner 1,509.9 (44.2) 2,007.3 (43.0) 2,617.2 (41.3) 3,429.9 (40.5)
Chonsei 1,048.5 (30.7) 1,657.1 (35.5) 1,970.2 (31.1) 2,927.5 (34.6)
Rental " 738.8 (21.6) 931.9(199) 1,554.5 (24.6) 1,900.1 (22.5)
Others 115.4 (3.4) 73.7 (1.6) 188.9 (3.0) 205.0 (2.4)

Note: Unit: thousand households, numbers in parentheses are percentage

1) inchides monthly rentals with deposit

Source: Korea National Housing Corporation, Housing Handbook, 1983, p. 14, 1989, p. 12,
1994. p. 18.
National Statistical Ofﬁce, Population and Housing Census, 1970, 1975, 1980, 1990.

62

3-4. Housing Prices and Aﬁordability

High housing prices relative to household income have hurt the ability of urban
families to afford adequate housing, whether rental or owner-occupied. Large capital
gains from real estate may also have worsened the distribution of wealth and income.
Table 13 and Figure 5 provide indices (1974:100) of the prices of housing, land,
construction materials and all other goods as well as real GDP, urban household
consumption expenditure, money supply (M2), and the rates of return on savings
deposits and on curb market loans during the last 18 years. Housing prices increased by
960% , registering an average annual rate of 17.1 percentage, over a period of 16
years. This pace of increase was much faster than the rate of growth of the consumer
price index (CPI) at 10.8 percentage per rate of 5.7 percentage. After a dramatic
increase during a period of high inﬂation in the late seventies, housing prices stabilized
between 1984 and 1987 and then rose sharply in 1988 and 1989 (Kim, Kyung-Hwan,

1991:1-5).

63

Table 13 Trends in Housing-related Economic Variables: 1974-1991

 

 

vent Real UC" M," W’ LPK LPs” LP6 W—PI PC Cﬁ RYD RC
GDP " °‘ 7’ M8) 9’ ‘0’ ML

 

1974 700 100 100 100 100 100 100 100 100 100 100 10‘0-
1975 107 134 128 139 127 132 122 126 123 125 115 141
1976 121 167 164 176 161 153 148 142 132 144 134 199
1977 133 193 239 220 215 201 216 154 151 159 153 280
1978 146 259 323 287 320 474 387 173 175 182 181 406
1979 157 343 402 396 373 505 472 205 216 215 215 589
1980 154 415 510 502 417 572 553 284 281 276 257 853
1981 164 498 638 609 448 593 592 342 311 336 299 1154
1982 176 578 810 623 472 644 625 358 325 360 322 1452
1983 197 631 934 764 559 1016 824 359 338 372 348 1826
1984 215 688 1006 787 633 1253 1002 361 334 380 383 2264
1985 230 741 1 163 787 678 1354 1080 365 327 390 421 2808
1986 259 813 1377 766 727 1404 l 149 359 325 400 421 3456
1987 289 935 1640 820 834 1493 1309 361 328 413 510 4255
1988 322 1095 1992 929 1063 1912 1694 371 355 442 S60 5106
1989 342 1239 2387 1064 1402 2570 2235 375 379 465 -- --
I990 477 -- -- -- 1691 3348 2838 -- -- 508 -- --
1991 525 -- -- -- 1907 3722 3220 -- -- 556 -- ~-
Note: 1) UC: Urban Consumption
2) M2: Money Supply
3) HP: Housing Price
4) LPK: Average Land Prices in Korea
5) LPS: Average Land Prices in Seoul
6) LP6: Average Land Prices in 6 Major Cities
7) WPI: Whole Sale Price Index
8) PCM: Price of Construction Material
9) CPI: Consumer Price Index
10)RYD: Return on 1 Year Deposit
11) RCML: Return on Curb Market Loan
Source: Son. J. Y., An Economic Analysis of the Land Problem and Land Policy, Korea
Development Institute, p. 34, 1990.
Korea Housing Bank. A Demand for Housing Loan Model. 64-66. 1989.
Bank of Korea, Economic Statistics Yearbook, 1988, 1989.

 

Figure 5 Trends in Housing-related Economic Variables: 1974-1991

 

 

Index 1974-100

 

 

5m.1..____.-_____..._.._____.._._...._.__________.___.

p—-—______-___—————-____

 

———____-_-___-___——-—__—___

————-—-——-1

 

 

 

/ 4.
4,. s,

0 icii“i%i‘riii+
:4 :2 8 ° N g s g a
at at at 8 8 at on
v- v- 1— v- 1- v- 1- v- ‘-

 

 

+ Real GDP

—X — Urban Consunption

+ Miner 89999

+110th Pnces

+Average Land Prices (Korea)

.....__ Average Land Prices (Seoul)

.3. Average Land Prices (6 nnjor
clles)

_.._ Wiole Sale Prtoo hdex

—Prlce of Construction Nhterlals

*Consumer Fl'lce hdex

.g- Return on 1 year deposit

+ Return on curb narlet loan

 

 

 

 

 

 

65

4. Housing Institutions

In the 1960s and 708, Korea reorganized its housing institutions. There are three
types of organizations: public, semi-public (non-proﬁt or publicly supported), and
private sectors. The public sector is represented by the local Construction
Bureau/Ministry of Construction and the semi-public sector by the Korea National
Housing Corporation (KN HC).

Although the implicit importance of these institutions is extended into the
private sector, which is not easily quantiﬁable, we can still see how important is each
institution by the proportion of completed housing units during the last three decades.
Table 14 shows that the private sector plays the major role in housing supply in Korea.
Figure 6 shows the relationship of housing institutions and residents’ income levels.

There is also a public housing program for low-income families.

Table 14 Housing Supply of Public / Semi-Public / Private Sectors

 

 

 

 

 

- Housing Supply. ' i * Period. (1962281).
' Units Percentage

Plibli'éfswm 7f 7f; 7f

Local Government 534,049 20
semi-subtle{sectoral:11:

* KNHC 222,451 9
PrlvateSector

2,608,620 71

 

 

 

Source: Korea Housing Bank, Housing Economic Statistical Yearbook, 1994.
Korea National Housing Corporation, Housing Handbook, 1994

66

Figure 6 Public/Private Classiﬁcation by Participation, Fund, and Target Groups
in Housing Supply

 

 

 

| Public g— ﬁentral Government J————— Housing Policy Planning

Development of Housing Investment Plan
Minimum Living Standards

 

 

 

 

 

———§EOC&1 Government r—ﬁ Implementation of Housing Supply Scheme
Supply of Local Government Housing

 

 

 

 

 

 

 

 

ﬁ . , . Housing Quality Control
14%0‘71'855651351113: income Control of Private Sector’s Activity

I l Irn lementation of Public Hous' Su 1

MC P mg PP Y

t J Scheme

 

 

 

 

 

Low and middleoincomc
families (SALE)

we

 

 

 

 

 

Mobilization and Management of
Housing Fund

 

 

 

 

 

 

, - - 4 Supply of Private Housing
Private Build“ & Deve10pers 1 Implementation of Private Housing

Middle and high income New
I a .1. s |

——{Eonstruction Firms 1———1Construction of Public and Private Housing

 

 

 

 

 

 

 

 

 

 

 

 

——1 Private Money-LendaE}——— Supply of Private Housing Fund

 

 

 

 

 

 

 

:Institution Systems

f ] :‘ll‘argetGroup

 

 

 

 

 

Source: revised from KNHC, Housing Policy Development Research. pp. 307. 1983.

("D

(I

67

5. Housing Finance

Housing is not only a valuable and durable good but also expensive. The lack of
access to mortgage ﬁnancing by middle or lower income group causes a concentration
of housing demand in that economic class. Finance plays a major role in the housing
operation system and can be utilized to facilitate effective demand for housing. Effective
ﬁnancial strategies selected by government can promote the housing construction industry
by the efﬁcient allocation of funds in the housing market. In addition, housing ﬁnance is an
effective policy instrument, providing an anticyclical measure through promotion of
housing activities when the national economy is in a sluggish phase.

However, Korea's housing ﬁnance system has several problems. Lack of funds
to be mobilized into housing is a chief concern. Renaud criticizes such major problems
of a system as attributable to ﬁnancial repression (1987, 1988), which refers to a set of
regulations that forcing “domestic savings into pre-determined sectors of the economy
and which set deposit interest rates below the market rate"(Renaud, 1989z4).

In terms of mobilization and supply of housing ﬁnance, almost all of the housing
ﬁnance in Korea is covered by the Korea Housing Bank (KHB) and the National Housing
Fund (NHF), two of the major housing ﬁnance institutions.

KHB is the main organization responsible for raising and disbursing funds for
housing. Initially, it provides housing ﬁnance loans and raises its funds through

collecting deposits, sale of housing debentures, and borrowing from the government.

68

Among two different kinds of housing ﬁnance, KHB has taken care of both permanent
ﬁnancing and construction ﬁnancing. Permanent ﬁnancing gives a loan to the household in
lien of the house to be purchased. This will increase the household’s housing affordability
and facilitates transactions in the housing market. Construction ﬁnancing provides a short-
term loan to the house builders. This ﬁnancing promotes housing construction activities
(Lee, K and Sohn, K., 1989).

It is important to understand the source and use of funds by KHB in terms of
Korean housing ﬁnance. Deposits received show wide ﬂuctuation, which affects
mobilization of availability of loans. On the other hand, Worker’s Property Formation
Savings Deposits and housing installment savings deposits show a steady increase, as
expected. These two long-term deposits are the major sources of stability in KHB’s
fund mobilization, since the deposits are utilized solely for loans.

The share of loan collections by KHB shows a very low level compared with
other ﬁnancial institutions. In 1987, the ratio of loan collections to the total fund was
7.4 percent. This was mainly due to the high ratio of permanent ﬁnancing in asset
management. Therefore, the KHB is under a heavy pressure to mobilize sustained
banking activities.

NHF has, however, maintained a strong position in main housing loans. The
NHF system was established in 1981 in order to implement the Comprehensive
Housing Construction Plan. The fund is used to provide loans for site development and

national housing construction. The sizes of housing units are limited to below sixty m2

69

for housing construction loans. The amount of the loan31 per unit is between ﬁve
million won and six million won, depending on the size of housing units and the area
where the site is located.

A total loan amount of 731,202 million won was provided from the NHF to
build 118,953 dwelling32 units between 1982 and 1987. Among the total amount of
loans, about ﬁfty-eight percent was provided to the public rental housing projects of
local governments and KNHC.

Meanwhile, domestic saving rates and the propensity to save have an important
relationship to housing ﬁnance. The propensity to save is an important factor for
housing ﬁnance in terms of sources for homebuyers or housing loan lenders. Domestic
savings rates are high, for example 38 percent in 1988. In comparison, the savings
rates of many other developing countries are around 5-10 percent (Chang, 1986). One
of the important reasons for the high savings rate in Korea is the need to accumulate
funds to purchase a house. This unusual propensity to save partially accounts for
Korea’s active housing development. Table 15 exhibits mortgage loan conditions,

expressed in terms of amount, period, and interest rates.

 

3’ The maturity term is twemy years with a one—year grace period. The interest rate is 5 percent per annum for rental
housing and 10 percent per anntnn for owner housing.

32 A ‘dwelling’ is a “building or part of a building which provides structually separate living accommodation’.

70

Table 15 The Conditions of Mortgage Loan

 

 

 

Sector ’ ' ~ 1 Amount Period (years) Interest Rate - 7
; ; (% of Value) ; (1%) ; ‘

Public 4 50 20 11 - 13

Private. 7 h 70 or less 3 - 20 14

 

 

 

 

Source: Korean Housing Bank, 1982

There has been a continuous increase in the supply of housing loans by KHB
and in residential investment (Table 16). However, a comparison between 1986 and
1987 shows an unusual relationship where a larger number of houses were constructed
with a smaller supply of housing ﬁnance. This implies that the housing ﬁnance system

in Korea covers a small portion of the housing construction activities.

6. Housing Program

A housing program is a plan for implementing a housing policy. In general,
after a housing policy and priorities have been established, the housing authority
designs a housing program which consists of speciﬁc targets, such as the number of
housing units to be constructed within a given time. The targets are justiﬁed on the
basis of expected housing needs and demands among target groups and spatial
allocations by the public and private sectors. However, there is usually a gap between

targets and performance, which should be examined.

71

Table 16 Aggregate Index of the Housing Activities"

 

 

 

GNP Residential No. of Annual Ratio of
Constant Housing Housing Supply of Loans over
Investment Units Housing Housing
Constructed Loans Investment
(%)
1970 17,013.0 855.8 115,000 5.95 0.695
1975 26,1135 1490.8 179,951 35.94 2.41
1980 36,6723 2179.1 211,537 281.32 12.9
1985 52,7054 2742.3 227,362 889.82 32.44
1986 59187,8 3395.4 288,252 864.63 25.46
1987 66,3196 3658.1 244,301 1096.86 29.98
Note: 1) Unit: Billion won

 

Sources: Bank of Korea, National Accounts, 1987.
Korean Housing Bank, The Statistical Yearbook of Banking Services, 1987.

programs. This approach considers the public and private as a whole housing sector
rather than as speciﬁc targeted groups (Fable 17). Targets, ﬁnance, and land are the
three main components of housing programs. Since housing is not only used as a

service for social welfare but also as a tool for an economic development plan in

Korea has adopted a comprehensive approach toward the basic types of housing

Korea, housing programs have been designed as ﬁve-year short-run plans.

72

Table 17 Dwelling Construction: Planned and Realized by Initiation”

 

 

 

 

 

 

 

 

 

.. ; Public. I Private Public - .Privatc
8.4 91.6 12.2 87.8
6.0 94.0 12.8 87.2
30.0 70.0 30.0 70.0
38.4 61.6 44.5 55.5
43.2 56.8 47.6 52.4
45.0 55.0 33.3 66.7
15.0 85.0 8.0 92.0
5.2 94.8 13.8 86.2
.;:ie:i:i 19.5 80.5 18.9 81.1
{521977-81 22.7 79.3 31.9 68.1
21982-86" " 26.8 73.2 30.5 69.5
1988-92 2’ 20.4 79.6 - —
Note: 1) Unit.%

2) Two-Million Houses Program

Source: Korea National Housing Corporation, Housing Statistical Yearbook, 1990
Korea Housing Bank, Housing Economic Statistical Yearbook, 1994
Ministry of Construction, Yearbook of Construction Statistics. 1993.

Table 18 shows the results of housing programs which have a good record in
terms of the percentages of housing programs completed. From the viewpoint of
progress made in public housing programs, Korea has a high growth rate from a 12
percent total completed housing units in the ﬁrst housing program (1962-66) to 41

percent of total completed units in the fourth housing program (1977-81).

Table 18 Housing Construction Plan: Planned and Realized: 1962-1992

73

 

 

 

 

 

 

 

 

 

SYear Housing Units Eunds
Plan Period Planned “ Realized ” R/P (%) Planned 2’ Realized 2’ R/P (%)
1962-66

Public 40,266 39,915 99.1 5,864 4,602 78.5
Private 435,074 286,020 65.7 33,093 47,619 143.9
Total 475,340 325,935 68.6 38,957 52,221 134.0
1967-71

Public 30,000 69,613 232.0 4,793 38,344 800.0
Private 470,000 470,725 100.2 86,584 238,837 275.8
Total 500,000 540,338 108.1 91,377 277,181 303.3
1972-76

Public 250,400 228,766 91.4 254,000 369,473 145.5
Private 582,600 531,825 91.3 1,045,000 1,125,805 107.7
Total 833,000 760,591 91.3 1,299,000 1,495,278 115.1
1977-81

Public 512,000 497,792 97.2 600,000 2,973,317 495.6
Private 818,000 618,234 75.6 2,040,000 6,310,900 310.8
Total 1330.000 1,116,026 83.9 2,640,000 9,314,217 352.8
1982—86

Public 618,000 549,344 88.9 5,957,000 5,272,600 88.5
Private 813,000 605,727 74.5 16,531,000 12,039,800 72.8
Total 1,431,000 1,155,071 80.7 22,188,000 17,312,400 78.0
1988—92”

Public 900,000 905,000 101 13,136,000 - -
Private 1,100,000 1,812,000 136 51,147,000 — -
Total 2,000,000 2,717,000 165 64,283,000 — -

 

Note: 1) Unit: dwelTing unit
2) Unit: Million won

3) Two Million Houses Program

Source: Korea National Housing Corporation, Housing Statistical Yearbook, 1990

Korea Housing Bank, Housing Economic Statistical Yearbook, 1994
Ministry of Construction, A Study on Housing Problems and Policy Development in

Korea, 129,1985.
Korea National Housing Corporation, Housing Handbook, 5263,1987.

74

7. General Housing Policy Issues and Evolution of Housing Policies

Housing policy refers to the range of activities that public and private sectors
jointly undertake to provide housing services for a society. It may be used as an
instrument to inﬂuence the settlement patterns of a nation or a city. Korea recognizes
the overlap and importance of both sectors and integrates their involvement in its
housing policy. However, in such an inclusive approach, government always plays the
dominant role and the private sector lacks details of policy and documentations to
enhance its position.

Housing policy receives low priority in national development policies in many
developing countries because housing is an expensive investment involving large capital
outlays. However, this is not true in recent situations in Korea, where housing has been
regarded not only as a means to achieve social welfare and income redistribution, but
also as a tool for economic development. Thus, housing programs are connected to
economic development programs, which are examined below as aspects of housing
supply and demand policy.

Housing supply focuses on expanding housing units, and housing demand
relates to enhancing housing purchasing power. Over time both of these aspects of
housing policy have been greatly inﬂuenced by a general land policy, which has
evolved through historical land reform movements, referenced above, and has more

recently evolved as a new town policy where a decentralized urban population is

75

particularly encouraged.33 Such a diffusion strategy has introduced the question of
housing distribution into the supply question. However, supply, rather than
distribution, is more emphasized.

As a consequence of both supply and distribution issues, public housing remains
too expensive for low-income families, even though less stringent requirements for
eligibility to public housing have been established. Therefore, distribution as a
component of supply (spatial, scale, target groups, equity, etc.) should receive
heightened attention in Korea. Such factors become evident by their absence in the ﬁve

year plans made by Korea during the last half century.

7-1. Summary of Korean Housing Policy

The evolution of housing policy can be examined by dividing the postwar period
into 1945-61 and subsequent ﬁve-year planning periods (Chung, 1985). The following
segments review the evolution of Korean housing policy. Each economic development

plan includes goals and objectives of general policy, accomplishments and results.

7-1-1. Postwar Transition (1945-61)

There are two subtle divisions of housing policy and program in the Postwar
Transition Period, 1945-61. The ﬁrst part of the period from 1945 to 1954, was

marked by a huge immigration of about four million Koreans returning to their

 

33 See the site of new town near Seoul at the Appendix II.

76

homeland after the country’s liberation from Japan in 1945. This sudden increase in
population resulted in acute housing shortages which were worsened even further by
the Korean War of 1950-53. Matters were made more difﬁcult as new economic
structure was being introduced, as for example, institution of a capital gains tax in
1950, which reduced private sector incentive to invest in housing.

During the initial postwar period housing issues were addressed by the Ministry
of Construction. Government gave itself a target of building one million houses for the
period of 1950-56. To assist this goal, the Korea Development Bank (KDB) was
created in 1953, and in 1954 was given the job of providing housing funds. In 1958,
the Construction Industry Act was passed in order to promote a sound development
industry and two years later the capital gains tax was abolished. The effect of these
actions was felt immediately an increase in number of houses built between 1960 and
1961 over that of the previous years.

In summary, during the period of 1951-61, housing was considered as part of a
plan to address social needs. A total of 916,486 housing units were constructed,
including 260,000 fabricated housing units which were provided in the form of
governmental relief under the guidance of ofﬁce of Veterans Administration. Data are
shown in Table 19. It was, however, not enough. Although the national housing supply
rate was increased, the housing shortage remained at 17.3 percent, while the shortage
in Seoul was signiﬁcantly higher at 36 percent.

During the latter postwar period housing policy had evolved as new options for

housing ﬁnance strengthened demand and regulation of private and public sector

77

housing alone with relief in supply-related taxes expanded housing supply. In this

process, housing policy became useful as an overall price stabilization mechanism.

Table 19 Residential Construction Accomplishments: 1951-1961

 

 

 

 

Year Public Construction Private Construction Total
‘ ‘5l-‘56 * * 197,330 323,457 520,787
‘57 -,‘ 61 * 59,150 336,549 395,699
Total ‘   256,480 660,006 916,486
Ratio (96) g 28.0 72.0 100.0

 

 

Note: Unit: Dwelling Unit
Source: Special Task Force of Planning and Coordination Dept. of the Prime Minister Ofﬁce,
The Evaluation Study of 1 st 5 yr. Economic Development, 708. 1967 .

7-1-2. First Five-Year Economic Development Plan (1962-66)

A systematic approach to housing policy began with the ﬁrst ﬁve-year economic
development plan in 1962. At that time housing policy shifted from being an element of
social planning to an element in general economic policy (Rho and Ha, 1987). This
transition was based on the principal objective to establish a self-reliant economy. In
this case, housing policy was regarded as one of the additional components of a major
economic plan, and as result, priority status suffered.

The housing stock shortage and deterioration of investment rates were perceived
as major housing supply problems. They were addressed during this period to alleviate
the rate of housing shortage. In 1962, builders were given access to construction loans
for land and housing development. Manufacturers of construction materials were also

given assistance, with the result that about 326,000 new dwelling units were built. Such

78

numbers were considered as progress, though they were thirty-two percent short of the
target goal of 475,000. Attempts to address the demand side included a self-housing
program, which targeted the homeless. It was adopted in 1962, with the objective
enabling a household to borrow from the KDB an amount of ﬁfteen to forty percent of
that household’s own funds.

However, the most important development was the creation of the KN HC in
1962. This institution maintained broad responsibilities including land acquisition and
development, housing construction, management of public housing, and installation of
infrastructure for large scale housing development. The initial capital investment was
500 million won.

With such a national initiative, local administration structures were needed to
articulate policy effectively. This was accomplished with the 1963 amendment to the
Public Housing Act, which led to a distinction between the ﬁrst category (KNHC) and
a second category (local government) of public housing initiatives.

During the period of 1962-66, more than 52 billion won was actually invested,
thirty four percent more than planned. According to the plan, the private sector was to
build as many as 435,000 units but ended up with only 286,000, or sixty-six percent of
the target. The public sector was to produce 40,000 units and are this target (Ministry
of Construction, 1985: 22-26). The overall performance however, was quite
disappointing.

While the number of dwellings built fell far short of the target, money invested

rose above what was planned because the government had to keep up with rapid

79

increase in dwelling prices. The result was public sector share of housing supply at 85
percent and of ﬁnance funds at 15.1 percent, which were only 12.2 and 8.8 percentage
of what was to be realized. The private sector thus played a major role of housing
policy. However, it was still less than that of developed countries where level of
housing investment over the GDP averages 5.5 percent. In Korea during the ﬁve year
plan, the level of housing investment over the GDP was only 1.7 percent.

The ratio of housing supply also declined from 84.2 percent in 1960 to 74.6
percent in 1966. These facts suggest that at that time, the government’s direct
investment into the housing sector was limited because of greater demand for
investments in industry, production equipment and infrastructure facilities. Housing
investment had low priority. To sum up, the major development in this period was the

access of the homeless to long-term housing loans and the founding of the KNHC.

7-1-3. Second Five-Year Plan (1967-71)

The housing stock shortage was still a major housing problem during this plan
period. In order to alleviate the shortage, expansion of the housing supply became a
main objective of housing policy. In terms of achieving the goal of expansion, new
construction in this period was a phenomenal success (Chung, 1985). The initial target
of 500,000 units was exceeded by 8.1 percent. Sustained price hikes led to an
investment of 277 billion won, which was expanded by 800 percent. This surpassed the
planned amount of 91 billion because of bikes in the cost of housing construction

components.

80

Both public and private sectors did well. The public sector exceeded the target
by 132 percent while the private sector exactly met the plan’s goal. Such a success can
be explained by a series of measures. In 1967, Korea Housing Funds (KHF) was
created as a ﬁnancial institution with an initial capital investment of one billion won.
The fund was allowed to issue housing bonds. In the same year (May 19th), the Korea
Housing Funds Act was amended so that 15-25 year loans could be made for small
dwellings of 66 m2 ﬂoor area or less. On January 4th, 1969, the Korea Housing Funds
was renamed by KHB. These changes certainly strengthened the demand for housing.
However, faced with rapid urbanization, creeping inﬂation, and land speculation, the
government adopted, in 1967, anti-speculation measures including a heavy capital gains
tax and other price stabilization measures in 1971.

On the supply side, pre-fab techniques for high-rise apartments were
introduced. A series of amendments of the Building Industry Act during the 1967-71
period led to an upgrading of the industry’s technical and ﬁnancial capabilities. In the
meantime, the capital of KNHC was raised to 10 billion won, and initiators of public
sector housing (KN HC and local government) were given tax allowances. Further tax
allowances were made in the income tax and a 50 percent capital gains tax was limited
to cases where the lot size was less than ten times the building area. A few other
minor tax law changes were made.

The most important development of the period was the creation of KHB. This
period was marked by a huge expansion in the housing ﬁnance system and its funds

and relatively mild restrictive measures, with the result that the construction target was

81

exceeded. However, at the same time, the public and private sectors had to invest much
more money. The actual investment amount was 203 percent more than planned, and a
general decline began after 1969.

The expansion of housing investment came following moderate economic
growth. The share of housing investment had increased from 1.7 to 3.4 percent.
Nonetheless, the level of housing investment was still almost half that of developed

countries.

7-1-4. Third Five-Year Plan (1972-76)

With the new KHB and KHNC in place, a systematic approach to housing
policy was easier, though it led to an overall housing policy emphasizing the role of
public housing. Evidence of this is found in an 18.2 percent increase of the share of
public sectors compared to the previous planning period. This period’s target was to
build 833,000 dwellings units, but ended up with 760,591 units, or 91 percent of the
target. About 1.3 trillion won was planned for investment, but the actual investment
was 1.5 trillion won. The creation of 760 thousand housing units may be attributed to
the 30.1 percent showing of the public sector. The overspending of housing investment
was mainly caused by hikes in land prices. For example, in Seoul, the share of land
cost amounted to 52.5 percent of total housing prices at that time (Planning and The
Coordination, Ofﬁce of the Prime Minister, 1977: 529), while in earlier periods the
percentage was low. The increase represented an inﬂation rate. Thus, the construction

target was almost met with relatively moderate price inﬂation.

82

Both public and private sectors met 91 percent of the target, in spite of a global
economic shock in 1972 in response to the oil shock induced by an aggressive OPEC
position. In response, Korea relaxed some anti-speculation measures, which deepened
the real estate market suddenly. In addition, to strengthen the economy, slackening
since 1969, the money supply was increased and interest rates were reduced.

In 1972, the government set up for the ﬁrst time a ten year housing construction
plan, which symbolized an intention for serious involvement in housing issues. The
MOC initiated the main activities of this plan and its independent control of
implementation. Furthermore, the government was now actively involved in
maintaining several newly enacted comprehensive housing laws. This plan included,
among other things, the use of foreign loans.

In 1973 the public sector, which includes KHNC, the central government, and
local governments, set a maximum housing size at 85 m2. The 1973 amendment of the
Housing Construction Promotion Act led to a precise deﬁnition of public sector
dwelling sizes: 60 m2 - 85 m2 (single-detached) dwellings, 40 m2 - 85 m2 (row-houses),
and 40 tn2 - 85 in2 (apartment). The capital of the KHB was raised to 20 billion won
(Ministry of Construction, 1986). However, as the economy became overheated by the
real estate boom in 1974, a series of restrictive measures were taken once again.

This period was marked by several changes in housing laws (Table 20) and
related-taxes, a greater distinction between public and private sectors, and international
factors, which were signiﬁcant, though without direct involvement as that in the initial
postwar period. The share of housing investment over the GNP increased a little from

3.0 to 3.6 percent; however, this portion was absolutely short compared to that of

83

developed countries (6.8 percent) during this period. Furthermore, the housing supply

ration also decreased from 71.2 percent in 1970 to 74.4 percent in 1975.

Table 20 Housing Policy Goal in Housing Related Law

 

 

 

{Laws i 3 f . . -  1 Content V. .i 7     -, ’

Public Housing Law (enacted in 1963, To improve housing situation and public

abolished in 1972) welfare by constructing public housing for
low-income households.

Housing Construction Promotion Law To improve overall housing service and

(enacted in 1972) public welfare by planned supply of

housing. For this, the law deals with
housing ﬁnance, production and supply of
housing materials.

 

Housing Construction Promotion Law To promote living security of tenant
(amended in 1977) households and to provide better housing
service for all

 

 

The law to promote living security of the prescription to promote living security of
working class and to support saving the working class and to support saving
(enacted in 1987)

 

 

Source: Lee, K., and Sohn, K. A Study on the Optimum Allocation of the Housing
Finance, Korea Research Institute for Human Settlement, February, 1989. p. 9.

7-1-5. Fourth F ive-Year Plan (1977-81)

This period was characterized by housing stock shortage, an imbalance of
housing supply and demand, and an economic recession beginning in 1978. There was
also a sudden rise of housing prices and poor quality of housing construction. To cope
with these difﬁculties, price stabilization became an important housing policy measure
(Table 21). Several major laws were enacted to stabilize dwelling prices. In addition
the government worked out a system of housing sales and strengthened a transaction

tax.

 

84

The target for new housing during this ﬁve-year period was 1,330,000 units
with 2.6 trillion won. However, no more than 1,116,000 units were built (83.9 percent
of the target) and about 9.3 trillion won was spent, or 253 percent more than planned.
This result reﬂected an unprecedented price increase since 1976, attributable to real-
estate land speculation. The speculation was partly due to the large inﬂow of funds
from the Middle-East as a result of construction exports (Chung, 1985). The reaction
of the government was immediate. To combat speculation, in 197 8 the government not
only sharply increased capital gains taxes but also imposed a system of land sales
permits, making speculative land transactions traceable and more difﬁcult.

Measures intended to curb rising housing costs were in conﬂict with the overall
goals of expanding housing supply. As a result, housing production, which attained in
1978 a record high of 300,107 units, fell sharply to 251,000 units in 1979 and 211,537
in 1980. The doubling of the price of oil in 1980, international economic recession,
and domestic recession, a drop of 5 percent in the GNP, were also major causes of
weakening housing construction at that time. The government therefore relaxed the
1978 measures by cutting the capital gains tax.34

In June 1981, broad measures were taken to recover from the economic
recession following 1978. In addition, there was the separation of the National Housing
Funds (NHF) from KHB Funds. The idea was to allow the MOC to control the NHF

better.

 

3" The action was applied differently. For real estate held less than two years, capital gains taxes were cut from 80 to
75 percent. For real-estate held for more than two years, captial gains taxes were cut from 70 to 50 percent.

85

On the supply side, on December 31, 1977, the Housing Construction
Promotion Act was substantially amended, and builders were classiﬁed into
“designated” and “registered” categories according to their qualiﬁcations. There were
also changes in regulations concerning building codes, infrastructure installment, and
house sales. However, the most important development was the creation in 1979 of the
Korea Land Development Corporation (KLDC), with the mission of providing low-cost
residential land. The KLDC’s capital was 200 billion won, raised to 500 billion won in
1980. The capital of KNHC was also raised to 50 billion won in 1978 (Ministry of

Construction, 1985).

7-1-6. Fifth F ive-Year Plan (1982-86)

This period was characterized by controlling the excess demand for housing and
sale prices of housing. The target for the entire period was 1,431,000 units, and the
target for the sub-period 1982-84 was 660,000 units, of which 579,000 (87.7
percentage of the target) were built. The swing of dwelling construction which started
in 1979 continued until 1981 when annual production was no more than 150,000 units.
In the meantime, the housing shortage rate which was 22 percent in 1970 reached 30
percent in 1980. This was enough to alarm the concerned authorities.

The government reacted by providing a series of incentive measures. All new
dwellings built within a prescribed period were given reductions in capital gains taxes.
Lands sold for new dwellings was exempted from a capital gains tax. Acquisition and

registration taxes on public sector dwellings (85 m2 or less) were cut by 30 percent.

86

All these measures were taken in 1982. In fact, dwelling construction increased from

150,000 in 1981 to 191,920 in 1982, 226,000 in 1983, and 222,000 in 1984. Thus,

there was a recovery, but it was slow.

Table 21 Housing Policy Measures

 

 

 

 

 

 

 

To Promote Construction 0 Pr1vately Operated o Natlonal Housmg Fund
Housing Funds 0 Worker’s Saving
0 Land Development 0 Public Development
Corporation
- Designated Firms
0 Advanced Sale
Equity 0 Transfer Income Tax 0 Housing Bond Bidding
- Control of Housing System
Prices Tenant Protection Law
0 Housing installment Rental Housing Supply
Deposit
National Housing Funds
Funds Subscriber’s
Deposit

 

Source: Lee, K., and Sohn, K. A Study on the Optimum Allocation of the Housing Finance,
Korea Research Institute for Human Settlement, February, 1989. p. 12

 

87

7-1-7. The Sixth Five-Year Economic Social Development Plan35 (1987 -1991)

This period was characterized by an increased scale of planning in which both
economic and social policies were combined. Improving housing supplies for low-
income groups was a speciﬁc agenda which emphasized the expansion of public land
development for housing sites. In addition, the plan aimed restrain and speculation by
raising property taxes for owners of multiple residences, and by stabilizing housing
rental rates with institutional support for the housing rental industry.

The sixth ﬁve-year plan was interrupted by the Two-Million Houses Program,
which led to creation of a new and more ambitious program. This shift of emphasis
coincided with an important political transformation. In 1987 the ﬁrst presidential
election took place without illegitimate activities. This introduced a new era of

democracy and social initiative.

7-1-8. The Two-Million Houses Program (1988-1992)

The Korean government initiated the ﬁve-year Two-Million Houses Program of
1988-1992 in order to meet ever-increasing housing stock demand. The plan was
developed in 1987, and has been implemented since 1988. It made a signiﬁcant

contribution to easing the housing stock shortage problem. It also stabilized housing

 

3‘5 From this plan. the governmem changed name of ﬁve year plan i110 the economic social development plan. This
fact represent: that they imend to focus on balancing economic and social development.

88

prices, provided new housing units of various sizes, and established higher quality
housing on a massive scale.

The plan called for dramatic expansion in the current housing supply capacity
by constructing a total of two million housing units. In addition, the rate of housing
investment over the GNP was to increase from 4.6 percent (1980-1987) to an average
of 6.5 percent. The level of construction capacity extended to 400 thousand housing
units per year from a level of 200 thousand in previous years, reﬂecting improved
housing productivity capacities, including human resources and materials (Yu, 1992).
As shown in Table 22, the housing program was successful in promoting new housing
construction on a massive scale. The ﬁrst year saw 317,000 new building permits. The
number of residential building permits issued accelerated to a maximum level of
750,000 units in 1990. In 1992, the four-year aggregate amounted to over 2.17 million.
That year saw 575,000 building permits issued, implying that over 2.77 million units
were supplied for the entire planning period, approximately 35 percent more than the

initially-targeted two million units.

89

Table 22 The Two-Million Houses Program

 

 

 

 

 

‘88- ‘88-’92 ‘88-’92 ‘88 ‘89 ‘90 ‘91 ‘92 ‘93
’92 results Ratio result result result result results results
Plan 8 s s s
Total 2,000 2,717 136 317 462 750 613 575 695
Public 900 905 101 1 15 161 270 172 195 227
Sector
permanent 190 190 100 - 43 60 50 37 -
rental
public 50 21 42 - - - - 21 189
housing
- rental 20 15 75 15 31
— sale 30 6 20 6 158
long term 150 171 1 14 52 39 65 15 - -
rental
worker 150 144 96 - - 61 36 47 38
- rental 50 43 86 20 12 1 l 10
-sale 100 101 101 41 24 36 28
Small-size 360 379 105 63 79 84 63 90 -
Sale
Private 1,100 1 ,812 165 202 301 480 449 380 468
Sector

 

 

 

 

Note: Unit: 1000 dwelling units and %

Source: Ministry of Construction, '94 Comprehensive Housing Construction Plan, 1994.2.

The plan was driven by the creation of a system of ﬁve new towns near Seoul.

These were Bundang, llsan, Pyongchon, Sanbon, and Jungdong. Table 23 provides the

scale and target population of new towns.

Table 23 New Town Development Plan

 

 

E‘EEClasSIﬁcanon Total Bundang” Ilsan Pyongchon‘ -: Sanbon .Jungdong
Areal) g 15,153 5 ,985 4,757 1:496 19267 19648
.;;‘:}Con3truction 294 97.5 69 42.5 42.5 42.5
housmg units 29’

T Population 1,176 390 276 170 170 170

 

 

 

 

Note: 1) Unit: 1000 pyung, 1 pyung equals 3.3 In1
2) Unit: 1000 dwelling units
3) Unit: 1000 persons
4) for the map of new town location, see appendix IV

Source: Ministry of Construction, Housing Construction Plan in the 6th ﬁve-year Economic
Social Development Plan, 1988.8.

It is clear that the Two Million Houses Program improved the short-term supply
of housing considerably. In addition to other factors, it must certainly have contributed
to the fall in housing prices in Seoul (Renaud, 1992). These achievements
notwithstanding, the plan has recently drawn much criticism. The primary one is that
the rising housing construction boom, which started in 1989, touched off excessive
investment in construction, resulting in a shortage of construction material and a
stampede of manpower into the construction industry, virtually jeopardizing other
segments of the national economy. The sale price of these housing units may not
reﬂect fully the cost of the resources involved. It also had a negative impact on
inﬂation and wages. Furthermore, this program achieved some of its results by
overriding the institutional problem of the housing sector in general. These measures
did not resolve the apparent problems. Nor did they provide a program of institutional,

regulatory, ﬁnancial, and ﬁscal reform.

91

7-1-9. The Seventh Five-Year Economic Social Development Plan (1992-1996)

The Seventh Five-Year Economic Social Development Plan set the average
economic growth target at 7.5 percent per year during the period. This downward
readjustment of the GNP could be pursued by reducing consumption and investment.
The average annual rate of growth in construction investment was lowered from 17.2
percent in the sixth plan period to 7.5 percent.

Where the housing sector is concerned, the plan implies reduction in housing
investment, because construction investment is considered to be less productive and
more inﬂationary. A large volume of housing investment is required to meet them,
suggesting that housing investment decisions must be somehow integrated with those
for the national economy. With respect to housing markets, demand will accelerate
with rapid increases in household and in per capita income, and the demand for better

quality housing will rise as middle-class households expand (Kim, J. 1993).

7-1-10. Summary of Evolution of Housing Policies

The evolution of the postwar Korean housing policy can be summarized in the
following terms. First, the primary objective of the housing policy so far appears to
have been the production of new dwellings. Second, housing policy in Korea has

evolved in the three following areas: (1) the expansion of housing ﬁnance through the

92

KHB, designed to increase demand, (2) the production of low income housing through
KNHC, and (3) local government and the promotion of private sector housing.

As far as housing ﬁnance and public sector housing production are concerned,
the policy has achieved what it set out to achieve. However, the most apparent failure
of the policy has been the difﬁculty of promoting private sector dwelling construction.
This could be explained by overregulation of the building industry on the one hand
and, to some extent, the industry’s lack of access to interim ﬁnancing. The third
shortcoming of housing policy in Korea is that it has been too often subject to an
overall economic stabilization policy. In each of the ﬁve-year plan periods examined,
an expansionary housing policy was accompanied either by anti-speculation or an
overall economic stabilization policy in such a way that expansion in dwelling
construction was short-lived (Table 24). This has perhaps been necessary for stable
economic growth (Table 25). However, the result is a mounting housing shortage.

The ﬁnal comment which can be made about the postwar evolution of Korean
housing policy is that taxes have not been used to effect a cyclical stability of dwelling
construction, with the exception of using capital gains taxes for ﬁghting speculation.
There is no reason why housing-related taxes should not be used for recovering the
slack in, or extending the expansion of dwelling construction. One conclusion that
emerges from a review of the evolution of housing policy in Korea is that housing
policy should somehow be a little more separated from the overall economic

stabilization policy without, of course, provoking real-estate speculation. In addition, it

93

appears imperative that the private sector play a more active role in housing ﬁnance

and construction.

IE!‘ 2..

The ﬁrst housing policy issue is the priority ordering of housing investment.
There is no doubt that the highest priority is the housing shortage. It must be pointed
out that the solution to the shortage problem will also attenuate the problems of
affordability and the equity in housing welfare distribution. In order to develop a
comprehensive set of alternatives for a ten year housing investment plan, the top
priorities are:

(1) New housing investment plans should be designed with attention to the
segmented housing markets. Given the distinct needs, affordability, and consumption
behavior of families, each sub market characteristic should be accounted for in
designing appropriate policies. (Lim, 1987: 183).

(2) To be effective, a housing program should deﬁne the target group and
formulate targeted minimum standards, taking into account the physical character of the
dwelling, dwelling control, environmental locus, and relative locus.

(3) The role of the public sector should be emphasized in housing production of
low-income housing. State-developed rental housing for the low-income groups should
be expanded.

(4) The government's contribution to public mortgage funds should be increased

and the long-term mortgage ﬁnancing system of the KHB should be further developed.

94

(5) Government intervention in the housing market should be based on social

need, and should limit private monopolies.

Table 24 Housing Policies in Economic Plam

 

 

 

 

 

 

 

 

Plans Objectives Planned Housing Supply Rate
The. lst To promote self-help housing 77.7%
(‘62-’66) construction
The’2ndn e To promote housing construction by the 76.8 %
(‘67-’7 1) 4 private sector
The 3rd To promote housing construction both 78.9%
(‘72-’76) by the private and the public sector
The 4th . To promote the supply of housing sites 80.0%
(‘77-’81) ~ and smaller housing units
The5th Stabilization of housing price 78.4% 3.24 pyung Person
(‘82-’86)
The 6th To promote the supply of housing for 78.4% 208,000 Rental Units
(‘87-’91), the low income households

 

 

 

Source: Lee, K., and Sohn, K. A Study on the Optimum Allocation of the Housing Finance,
Korea Research Institute for Human Settlement, February, 1989. p. 10

 

Table 25 Major Characteristics of Economic Development Plans

95

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2) Figures in parentheses denote the actual growth rate
3) PO represents Principal Objectives
4) P means Period
5) EG means Economic Growth Rate
6) MP meam Major Policies

Source:

543% MED? ’ . Second Third EDP PM EDP * Fifth EDP. , , Shah EDP: i“ "
0'31: ' EDP ' .7 _ ‘ . , j 3‘ ,, _ . 7 , , , 7 ‘ ‘ , , . I V ; T , 7 ‘7; _:
*5 Correction Mod-ermzat Coordination of Achievemem of Stabilization of Enlnncing growth
{0. of ‘socio- ion of growth, self-generation prices comrol

33.53}: economic inchrstrial stabilization growth Raising of Strengthening

C f vicious structure and equity Promotion of equity productivity Internationalization
- circle’ Promotion Achievement of through social Promotion of Improving the

{333.5 Establishing of self- self- developmem distortions in the living environment
the a self- reliam development Technical economy and

if; reliam economy Maximization innovation and promotion of rational

gigs. economy of land efﬁciency developmcm

ff ,4 cultivation and improvemem

.; ﬂ; development

,5 1967-71 1972—76 1977-81 1982-86 1987-91

, 7.0 (9.7) 8.6 (10.2) 9.2 (5.5) 7.6 7.0

I: ", Self- Self-sufﬁciency Scarring domestic Curving inﬂation Fiscal reform

“ V " suﬂiciency in food staples investmem Recovery of Reform of

1 of grains Improvement resources competitive power in monetary cortrol
lif’fzé and of living Achievemen of heavy industry and ﬁnancial

7 ,7 developme environmem balance in B.O.P. Consolidation of supervision

nt of industrial Raising agricultural policy Deregulation of
25$}; forestry structure iraernational Improvement of admiristrative

1T :. and Development of competitiveness ﬁnancial system comrol

ﬁsheries technology and Expansion of Overcoming energy Forming a new

3255'}: and social Formation human employment constraints pattern of economic
ff' ; 1? overhead of resources opportunities and Adjusting behavior

if f capital foundation Expansion of manpower government function Promotion of

322,}??? Effective use for social overhead developmera and rationalization of industrial structural
ff {3' of idle industrializ capital Expansion of fiscal managemem adjustment

gﬂi-f resources ation Improvement Saemaul campaigns Shift to competitive Development of the
32353;? Improvemen Improveme of welfare Improvemert of system and informative

Q t of balance at of living conditions promotion of open industry

{31; of paymeas balance of Expansion of economy Expansion of SOC
e: ”f position payments investmem in Developmem of and reform of the
ff Techm'cal position science and education and distribution systctn.
development Raising technology manpower and

{iii}; farmer-8’ lmprovemun of promotion of science

f: ,"a incomes economic and technology

lmproveme managemem and Establishing new

fjéifzifii nt of system relationship between

technology labor and

.6 and managemem

productivit Expansion of social
‘ u y developmem
Notes: 1) Economic Development Plan

Economic Planning Board. Korea ’s ﬁve-year Plan, 1961, 1966,1971, 1976, 1981, 1986

Economic Planning Board, Korea 's ﬁve-year Plan for the New Economy, July, 1993

 

8. Housing Issues and Problems

In spite of sustained increases in resources allocated to housing, Korea has
serious problems with housing shortages, worsening affordability problems, and the
lack of an adequate mortgage ﬁnancing system. Housing conditions in Korea have
regressed in terms of rates of housing supply and homeownership during the last
decade. The overall physical housing conditions, considering dwelling size and
facilities or shared dwellings and density, are relatively poor. Public housing cannot
meet the needs of low-income families in terms of quantity, quality, and price. The
most serious problem is land speculation by middle and upper income groups, which
causes high housing prices in the market. It will be a difﬁcult task for the Korean
government to solve theses issues as it seeks to improve housing quality and increase
housing quantity.

One of the crucial problems of housing is the absolute housing stock shortage.
(Chung, 1990) Since the 1960s, there has been a growing shortage of housing, felt
particularly in Seoul and other large cities. The ever-increasing number of people and
industries has created a serious housing problem. The acute housing shortage can no
longer be ignored, and the government has built new towns within the Seoul Capital
Region.

Mayo, Malpezzi, Gross (1985) argue that in developing countries governments

have worsened housing shortages by trying to solve the problem through government

97

housing. Governments attempt to improve housing quality through imposing high
standards and building codes, which they argue have improved nothing.

In spite of a reasonable amount of resources allocated to housing, the low level
of housing production can be partially attributable to a skewed distribution of housing
demand and a low housing supply elasticity (Ha, 1989). The income elasticity of
housing demand in Korea is positive and tends to increase with income level.

While efforts to provide sufﬁcient housing has its own associated problems,
they are often complicated by the question of affordability. Traditional approaches” to
the problem of urban settlement have failed to provide affordable public housing for
low-income urban households because government policies have been ineffective in
providing adequate housing. The main reason is that these policies have been based on
a poor understanding of the nature of the problem. What they are missing is insight
into aspects of the reality of squatter settlements. The government had followed
unrealistic assumptions which regard squatter settlements as “cancerous growth” (Ha,
1990:297) and focusing on the physical conditions only not on socioeconomic needs.

Clearly, a better understanding of target groups should make it possible to
devise more effective housing policies. One approach by public institutions has to
produce houses for both sale and rent to low-income groups. However, the housing
authorities have concentrated on the policy of state-developed housing for sale to

encourage urban households' home ownership, which excludes the low income groups,

 

3‘ Traditioml approaches refer to the several urban redevelopment strategies such as urban renewal project,
Hapdong redevelopment project, clearance program fro squatter settlements, etc. For a detailed discussion of
urban redevelopmem policy, see Ha 1990.

98

to which are attached elderly and handicap individuals. This reference should cover
builders are reluctant to provide smaller units, because they are less marketable, less
proﬁtable, and therefore more risky. Under such circumstances, low-income and tenant
households ﬁnd themselves in a situation where suitable housing is more difﬁcult to

acquire.

CHAPTER FIVE

MODELS OF HOUSING INVESTMENT IN KOREA

1 . Introduction

As noted in the previous literature review and existing models of housing
investment, new housing investment models are now examined. A theoretical rationale
and a conceptual model structure are presented by this study. Then, hypotheses and
variables are constructed. Finally, models are developed to deﬁne the determinants of
housing investment. The potential factors of housing investment in general are
investigated ﬁrst. Then, the relationship between the behavior of housing investment
and the conditions of housing stock is analyzed. This relationship provides insights

regarding housing investment patterns.

2. ResearchDesign

2-1. Theoretical Rationale

The basic theoretical rationale for the study of housing investment determinants
lies in the assumption of traditional consumer utility maximization. Under the free
market economy, an individual decides how much to spend on housing following the

maximization of his/her utility subject to budgetary constraints. The aggregation of all

100

individual demands for housing results in demands for housing for the entire nation. In
other words, this assumption raises the question of how one country actually reveals
preferences for housing provisions in a context of changing circumstances. Most
housing investment determinant studies focus on the level of housing investments,
which are used as dependent variables. These studies use various independent
variables. Within the context of the consumer utility maximization assumption, housing
analysts have attempted to explain statistical variations in housing investment, which
are dependent variables, by using a regression equation. The independent variables
used in the equation have been socio-economic. An example of a conventional housing
investment model is

Y =fi (xi. x2. x3) Equation (1)

 

Y = a + bix, + bzxz + b3x3 ---- Equation (1) 1
where,

Y represents the share of housing in total output measured as average residential
construction as a percentage of the average annual gross domestic product;

x 1 represents the explanatory variable of development level measured by per
capita income;

x2 represents explanatory variables of population growth, measured by the
average annual rate of increase in national population; and

x3 represents the explanatory variable of urbanization, measured by the average
annual rate of population increase in cities of 100,000 persons or more divided by

the average annual rate of increase in the national population.

101

We have to pay attention to the fact that x 1 can be considered as a variable
measuring demand and x2 and x3 as variables measuring need. Not only for economists,
but for policy analysts as well, this conventional housing investment determination
model has become a standard; with only slight modiﬁcations along the basic line, they
have employed similar models.

The general housing investment determination model they have used is

 

Y = f2 (x1, x2, x3, , xi) Equation (2)
Y = a + bpc, +1)2 x2 + b3x3, ...... , +b,,x,, ----Equation (2)1
where,

x, represents each dependent variable which is mostly related to socio-economic

variables.

The current study also utilizes this general housing investment model, which is
the traditional model in the theory of housing investment determinants. In addition, the
linkage among the independent variables is measured in terms of the correlation
coefﬁcient in the study.

The proposed model is based on equation 2 above. We assume that potential
factors inﬂuencing housing investment, in general, can be categorized into six
dimensions. The model selects various independent variables: socio—economic status,
urbanization and demographic change, policy effects, housing conditions, institutional
setting, and global and foreign affairs effects. Speciﬁcally, the proposed model is

In SH = a +bIIn SEO + I); In DEM +b3 In POL + b4InH0C +b5In INS +

b6 In FOA + u ----------Equation (3)

102

where,

SH represents the dependent variable of housing investment;

In stands for the natural logarithm of the variable, 0 denotes the regression
constant, b,- depicts the coefﬁcients of each independent variable, SEO
represents the independent variables of socio-economic status, DEM represents
the independent variables of urbanization and demographic change, POL
represents the independent variables of policy effect, HOC represents the
independent variables of housing conditions, INS represents the independent
variables of institutional setting, FOA represents the independent variables of

global and foreign affairs effect, and u is residual, error term.

Equation (3) estimates housing investment using different measurements
and combinations of variables. The variables for socio-economic status
characteristics are per capita GNP, total national revenue, total national
expenditures, money supply, deﬂator index, savings rates, the consumer price
index, and producer price index. Urbanization and demographic change
variables include the growth rate of national population, rate of urbanization,
and growth rate of the urban population. Policy effect variables are two million
house program dummy variable,37 political stability38 dummy variable, and
election dummy variable.39 Housing condition variables are the housing supply

rate and the total number of households. Institutional setting characteristic

 

37 The two million houses program is a dummy variable that identiﬁes the actual housing plan period. For more
detail, see CH. Six

38 The political stability is one of a measuremem that identiﬁes the status of the political system. We assume that less
political stability is a country. the less housing investmem will occur. For more details, see CH. Six.

39 Election variable is a dummy variable that idemiﬁes the actual election and previous year. We assume that as one
of political factor, election conditions relate to housing investment.

103

variables are composed of defense expenditures and social welfare expenditures.
Finally, the global and foreign affairs variable is the trade balance.
Consequently, the new proposed model in detail becomes as follows:

In SH = a +b, In PERGNPO + b2 In PERGNPOZ +b3 In IRO + b,, In IEO +
b5 In MSO +b6 In DOMSAV + b7 In AEXPEN + b8 In CPI + b9 In PPI
+ c, In DPOP + c; In DPOP2 + Q; In URBRTE + d, TWOMIL + d;
DPLSTABL + d3 ELECTION + d,, REALPLYI+ d5 EALPLYZ + d6
REALPLY3 + e, In NHO+ ezlnNSIZE + e3 In NIHH +f, In RAIDEF
+f2 In SW6 + g] In IRABALO + 82 In INTR + g3 In EXCHANG +
u ---Equation (4)

where,

SH represents the dependent variable of the level of housing investment
at constant prices; In stands for the natural logarithm of the variable, 0
denotes the regression constant. For independent variables,
SE 0:
- PERGNPO and PERGNPOZ (square term of per capita GNP)
represent the independent variables of per capita income
- TRO represents the independent variables of total revenue
0 IEO represents the total expenditure variables
- MSO represents the money supply variables
0 DOMSAV represents the ratio of domestic savings to the GNP
(we can also use SAVR- saving rate, the rate of private
savings to disposable income)
0 AEXPEN represents the average expenditure per person in the
national budget
0 CPI represents the consumer price index
- PPI represents the producer price index

DEM:

POL:

HOC:

INS:

F 0A .°

104

DPOP and DPOPZ (square term of national population
growth) represent the independent variables of the growth
rate of the national population,

URBRTE represents the independent variable of growth rate

of urbanization.

TWOMIL represents the two million house program plan
dummy variable

DPLSTABL represents the political stability dummy variable
ELECTION represents the election dummy variable
REALPLYI, 2, and 3 represent dummy variable of real estate

regulation

NHO represents the number of constructed housing units,
NSIZE represents the number of persons per household,

NTHH represents the total number of households,

RAIDEF represents share of national defense spending over
total expenditure

SWPG represents social welfare spending as a percentage of
the GNP

TRBALO represents the trade balance

INTR represents the interest rates-the proxy of long-term
mortgage rate

EXCHANG represents the exchange rate of won to the U.S.
Dollar

and u is the residual, error term.

105

2-2. Conceptual Model Structure

We may make the following inferences for the modiﬁcation of the conventional
housing investment theory. Unlike conventional housing investment models, the new

proposed models address several issues. With this in mind, we set up sub-models.

2-2-1. Global Open Model (Domestic Housing Investment and Foreign Affairs
Approach)

Housing investment is affected not only by domestic affairs, but also by global
and foreign affairs. We assume that the global context of ﬁnance and foreign affair are
related to the determinants of housing investment. Government policies are responsive
to economic changes. For example, a modiﬁcation in the exchange rate affects the total
output of the Korean economy and inﬂuences the availability of the money supply,
which in turn affects rates of inﬂation and balance of payments.

We speculate that there is a global and foreign affairs effect on housing
investment. We designed this model as a global open model, in contrast to conventional
housing investment models which are closed domestic models. Korea’s major economic
strategy, which is distinguished by dependence on exports and foreign ﬁnance, has
affected the production and distribution of housing services. Foreign trade and exports
make up a portion of aggregate spending on the government just as do domestic

consumption, investment, and general government spending.

106

Foreign affairs and domestic policies are two essential factors that determine the
allocation of investment in developing nations. In this dissertation, foreign affairs and
global effects are measured by the exchange rate, foreign aid, and trade balance. The
exchange rate, EXCHANGE, is an indicator of foreign affairs because it is a
measurement of economic strength and competitiveness in the global economy. The
exchange rate is a key element inﬂuencing housing investments. We assume that the trade

balance variable, TRABALO, is also a major variable that affects housing investments.

2-2—2. National Allocation Model

Military expenditures affect Korea’s housing investment. This cannot be
understood by cross-national comparative studies. In the mid-19803, military spending
exceeded spending on health and education combined. For the analysis of housing
investment, we can no longer neglect the components of ﬁscal problems especially military
spending. It is important to put military spending decisions on the same footing as other
ﬁscal decisions, to examine possible trade-offs systematically, and to explore ways to bring
military spending into a better balance with development priorities such as housing
investment. We call this the national allocation model by focusing on this interaction
between housing investment, military spending, and social welfare spending subject to
budget constraints. Furthermore, this model accounts for an environment of decision-
making for allocating resources under special circumstances. Korea spends a considerable

amount of its national income on defense. This has a negative impact on the developmental

107

process and divert money from the housing market. RATDEF is an indicator of military

expenditures.

2-2-3. Policy and Institutional Impact Model

There is a policy and institutional effect. This model uses dummy variables to
account for the different phases of government intervention and operation. By
examining the role of international, domestic, and institutional arrangements in policy
formulation, the variation in housing investments in Korea can be understood. The
relationship between political determination and institutional settings can establish a
political understanding of housing problems in developing countries.

Furthermore, political factors are important to the behavior of housing
investment, particularly in developing countries. For example, the housing investment
decreased dramatically between 1979-81 in Korea when coups broke out after President
Park Chung-Flee was assassinated in 1979. We can thus recognize the importance of
political factors in housing investment. It is, however, difﬁcult to measure and predict

the inﬂuence of these factors.

108

2-2-4. Financial Model

Because the level and growth rate of savings are relatively high in Korea, we
assume that high savings rates will contribute to housing investment. Because of the
high levels and growth rates of homeownership and dwelling size, we believe that these
two variables are important in Korean’s housing investment. Some ﬁnancial variables,
mortgage interest rates and inﬂation are also expected to signiﬁcantly impact housing
investment.

Figure 7 is a graphical representation of the model structure. It shows the
relationship among the main independent variables. The ﬁgure also demonstrates the

variables computed by each factor category.

109

Figure 7 Graphical Representation of Model Structure

 

 

  
  

 

 

 

 

 

Wustruction

J

[residential building J ‘ """"
Jin-reeideztttiel buildiri] ‘ ........

...............

‘7

 

J Disposable Income J

 

 

   

 

 

 

 

 

 

 

 

 

Determinants of Housing Invesnnent

 

 

 

 

 

 

 

 

Jmoncy supply J i
rats; J /

J interestlrate (-) J

 

 

 

 

 

J total revenue ]

 

 

{total expenditure J

t

 

 

 

 

 

 

 

 

 

 

[GNP deﬂator

 

 

 

 

J producer price index

 

 

l
J consumer price index J
|
J

J housing costs

 

 

[land price and cost J

 

 

 

If:

%

 

 

 

 

 

[ goverment intervention J

I semi-public 4]

 

 

 

 

 

 

 

 

 

 

 

 

J tax burden rate (-) J——I

 

 

 

 

 

 

 

[private sector J
J

J military expenditutLJ

J teem welfare J

 

 

 

 

(+) : positive expected relation with housing investment
(-) : negative expected relation with housing investment

\\

 

 

Bernographic

I

 

population change
I +)

J

 

 

[urbanization (+)

]

 

 

 

 

Housing conditions J

 

 

[ﬁning supply ratiT]

 

J homeow nership J

 

‘e
\\ \\ [dwelling size M

 

O

]

 

0

household size
m M

 

 

 

 

1

 

 

 

Foreign Affairs J

 

 

exchange rate J

trade of balance

 

 

 

 

 

 

 

r
i

 

+_ SJ [political stability J

 

 

Wmﬁonal Settings J

 

 

 

 

J election J

 

 

 

J policy factor J

 

 

 

 

 

 

 

 

 

110

2-3. Formulation of Hypotheses, Variables and Models of Analysis

2-3-1. Hypotheses

What factors account for differences in the share of resources invested in
residential construction? One important factor is likely to be the stage of development,
an index of a nation’s ability to mobilize resources for new housing. Is the size of the
share systematically related to the economic development stage? There has been some
theorizing about such a relationship and its nature, supported by a modicum of
statistical veriﬁcation. How important are demographic variables such as urbanization?
The share of housing in total output is related to the economic development level,
population growth, political settings, institutional arrangements, foreign affairs and
urbanization.

Based on the purpose of this study and theoretical discussion, the study
postulates the main hypotheses developed for statistical investigation as follows:

First Ha,‘10 one of the most important determinants of the level of housing
investment is soda-economic factors in Korea.
The growth of the economy, population, and urbanization is positively
associated with housing investment. If we compare cases in other
developed countries, we can assume that when housing stock is not
“mature,” the factor of per capita GNP is highly important to housing
investment, and the linear or the quadratic form of per capita GNP is not

much different. Furthermore, other variables, particularly intereSt rates

 

‘0 Alternative Hopothesis

111

and inﬂation, become insensitive in housing investment unless the

housing stock reaches a mature level.

Second Ha, housing investment as a share of total output are explained by the
demographic and urbanization factors in Korea.

Third Ha, the level of housing investment is mostly determined by the policy
effects in Korea.
We assume that political factors play as critical a role in government

resource allocation.

Fourth Ha, the level of housing investment is mostly determined by housing
conditions in Korea.

If the postulate holds, then what is the magnitude of these forces?

Fifth Ha, among the most important detemtinants of the level of housing

investment are government and institutional factors in Korea. The budget

constraint factors have affected the level of housing investment in Korea.
Housing is part of a country’s politics and economy. Domestic politics
with its varying political stability and institutional settings affect housing
in total output in Korea. The excessive military expenditures leave little
in the national treasury for the other sectors of the economy. This means
that Korea spends large portions of its gross national product for
defense. If the postulate holds, then what is the magnitude of these
forces?

Sixth Ha, housing investment as a share of total output is previously determined

by global and foreign affairs affect in Korea.

112

Seventh H,, some combination of main effect variables from six different
domains will generate a predictive model of housing investment that is a better
predictor than the main effect variables selected from any one domain, based
upon predictor variables containing a p-values s 0. 05 and an overall

regression p-value $0. 05.

2-3-2. Deﬁnitions of Variables

Models are formulated to test the hypotheses and to evaluate the economic
development impact on housing investment. We investigate which factors affect
housing investment and to what extent the Korean economy can allocate its resources to
construct more housing. For dependent variables, we use three different measurements:
housing investment as a share of total output, residential housing construction area per
person, and housing investment per household.

The appropriate method for investing and explaining resource allocation in
shares committed to housing is multiple regression analysis. This analysis allows each
potential determinant to be examined individually while holding others constant, and
permits the derivation of the appropriate weight of each with respect to the other.
Variables are deﬁned and estimated for (1) testing the relative power of economic and
demographic factors in contributing to an explanation of the level of housing
investment and (2) identifying the relationship between the level of housing investment

and economic development.

113

2-3 -3. Dependent and Independent Variables for Time-Series Analysis41

Dependent variables representing housing investment are divided into three
different measurement categories: national housing construction investment as a
percentage of the GNP, housing construction area per person, and housing construction
investment per household.

Independent variables are classiﬁed into six categories: socio-economic status,
institutional setting, urbanization and demographic change, housing conditions, policy

effect, and global and foreign affairs effect.

Table 26 Variables for Time-Series Analysis

 

A. Dependent Variable

 

(a) CONNP: national housing construction investment as a percentage of the GNP
(b) PHAREA: housing construction area per person
(c) HIHSH: housing consn'uction investment per person

B. Independent Variables

S . -E . 51 ll . I l g
(a) PERGNPO: per capita constant gross national product
(b) TRO: constant total revenue

(0) TEO: constant total expenditure

(d) MSO: constant money supply

(e) Deﬂator: deﬂator index based on 1970

(f) CPI: consumer price index based on 1970

(g) DOMSAV: ratio of domestic savings to GNP

(h) UMM: unregulated interest on money markets

(i) DCPI: growth rate at consumer price index

(j) SAVR: ratio of private savings to disposable income

 

 

 

‘1 For more information - source, units, formular see Appendix V.

‘2 In order to investigate the relationship between housing investments and key variables in the national economy,we
focus on estimating the elasticity of housing investmems with respect to income. In addition to income and price
variables, we can include several other independera variables which may affect housing investment.

114

(k) AEXPEN: average income per person in the national budget
(1) HSGEXPIN: average expenditure per household in the national budget
(m) PPI: producer price index based on 1970

u l' ”H . . H .Hﬂ

(a) DPOP: annual rate of increase in the total population

(b) IRUPOP: rate of increase in the urban population

(c) URBRTE: rate of urbanization

(d) RURPOP: rural area population

(e) URBPOP: urban area population includes that of Eup“ with more than 20 thousand

(t) POPU: nation-wide population

(g) URBR: annual rate of growth of cities over 100,000 divided by the average rate of growth
in the national population

(h) URB: rate of growth in cities over 100,000

E l' WI 15 . l l
(a) TWOMIL: the two million house program dummy variable
(b) DPLSTABL: the political stability dummy variable

(c) PARK: Park’s Regime dummy variable

(d) ELECTION: the election dummy Variable

EL . Q l" H 'H
(a) NTHH: total number of households

(b) NSIZE: number of persons per household

L.. ”.15.”,15

(a) RATDEF: defense spending over total expenditures

(b) DEFR: defense spending rate of growth

(c) DEFPG : defense spending as a percentage of GNP

(d) DEFPB : defense spending as a percentage of budget

(e) SWSP: social welfare spending in mil. won

(t) SWR: social welfare spending rate of growth

(g) SWPG: social welfare spending as percentage of GNP

(h) SWPB: social welfare spending as percentage of the budget
(i) ECRTE: economic growth rate

0) PDEN: population density

E . I71 . Eli
(a) TRBALO: trade balance in million US dollars

(b) EXCHANG: exchange rate of won to the U.S. dollar
(c) INTR: interest rate

 

 

‘3 These variables include the growth rate of the total population and the ratio between the rural and urban
population.

“ Eup refers to township

‘5 These variables are related to the role of government action in constructing more housing.

115

3. Patterns of Housing Investment and Fluctuations

3-1. The Level of Housing Investment and Fluctuations

The level of housing investment which is represented in 1970 constant prices
appears in Table 27 and Figure 8. Korea became highly unstable regarding housing
investment after 1975. Housing investment in the later stage is high compare to other
Asian countries. In order to diagnose the level of housing investment, the housing
investment per capita is presented in Table 28 and Figure 9. We can see that the
housing investment per capita has increased continuously. The pattern of housing
investment per capita income is quite similar to the level of housing investment
patterns. Table 27 presents housing ﬂuctuations in terms of the percentage of change in
the level of housing investment from the previous year. Figure 8 shows rather
stationary patterns all through 1953 to 1993. The growth rates of housing investment
are very high, 10.1 % on average. The 19.07 percent of standard deviation of growth
rates represents the relatively high volatile growth rate. The periods of the Second
Economic Plan and Sixth Plan have a much higher growth rate (18. 3 percent and 18.5

percent) than those of other plan periods.

116

Table 27 The Level of Housing Investment

 

 

 

 

 

 

Real level of Housmg Investment ‘ Growth Rate Of Real Heumng Investment
Mean so”: Mat Mm on” Mean SD” Max Min. on”

13. 5 2. 98 18.7 9. 9 0.22 2. 8 2.5 49 39. 87 -37 13 9.03

22.1 6.10 32.16 17.41 0.28 12.6 10.43 27.63 -1.46 0.83

70.1 21.37 93.9 42.7 0.30 18.3 14.35 30.9 1.3 0.78

166.9 54.95 218.9 87.1 0.33 11.1 22.55 34.8 -l3.3 2.04

. » - ~ . 410.6 86.00 483.8 279.6 0.21 8.6 28.12 39.3 31.9 3.26
(1977-81):};-

Fifth Plan [2 542.1 62.94 609.9 440.0 0.12 9.08 11.49 22.8 -3.61 1.26
(198286);

' ‘ ‘“ 1225.7 539.05 1823.3 668.7 0.44 18.5 14.89 41.7 2.7 0.81

380.49 517.160 1823.35 9.87 1.36 10.1 19.07 41.7 -37.1 1.88

 

 

 

 

Note: Units: Billion won, 1970 constant prices
1) SD. means Standard Deviation
2) CD. means Coefﬁcient of Deviation

 

117

Figure 8 The Level of Real Housing Investment

 

 

Bllllon won

 

iiiiiggiiii

 

I- Level of l-bus'ng hvestment + growth rate of investment

 

 

 

 

 

 

 

Table 28 Housing Investment Per Capita

118

 

 

 

 

 

 

“mod ’7 - i f 7181389," ' 59D? 0 Max, 5 Min: c1137?)
Postwar transition‘(195361)._-, ' 58.3 N 11.26 H 76.22 145.79 0.19 '1
‘Pirstplan‘(l96266)ﬁ= 78.6 18.88 110.27 63.84 0.24
second Planf(1967'—1971) , J 222.2 62.28 285.4 141.7 0.28
ThirdPlan(1972~76)' ‘ ' 478.4 149.67 620.4 259.9 0.31
FourthPlan (1977481) - 1092.3 223.79 1269.1 767.8 0.20
Tilthiplan (1982—86) ~ . : J 1345.0 138.53 1480.9 1118.9 0.10
Sixth Plan (1987-91) ' 2874.9 1224.04 4214.1 1608.4 0.43
Research-Targett1953—93) ; _; 866.28 1104.94 4214.1 45.79 1.28

 

 

 

 

Note: Unit: won, 1970 constant prices
1) SD. means Standard Deviation
2) CD. means Coefﬁcient of Deviation

Figure 9 Homing Investment Per Capita

 

 

 

 

 

 

 

1 95319561 9591 9621 9651 9681 #119741 9771 9801 9831 9861 9891 992

 

—e-— Housing Investment per cap}

 

 

 

 

 

 

 

119

3-2. The Share of Housing Investment and Fluctuations

The share of housing investment as a percentage of GNP appears in Table 29
and Figure 10. Examining the share of housing investment overall for the 1953 - 1993
period, Korea has an average of 3.61 percent. The Sixth Plan period (1987—1991) has
recorded a 6.0 percent. The pattern of the share of housing investment increased from
1.7 percent in the First Plan to 4.9 percent in Fifth Plan, then peaked to 6.0 percent in
Sixth Plan period (1987-91).

Housing ﬂuctuations in terms of the percentage change in the share of housing
investment from the previous year appear in Table 29 and Figure 10. The average
growth rate of share of housing investment and its variance are high. It increases from
-1.4 percent in the Fifth Plan to 9.7 percent in the Sixth Plan stage. Overall, there is a

1.5 percent growth rate overall from 1953-1992.

120

Table 29 The Share of Housing Investment as a Percentage of GNP

 

 

 

 

 

 

Share of Housing Growth Rate of Share of Housing
Investment Investment
Period Mean so." Max. Min. c.D.”Mean 3.1). 1’ Max. 17111 on."
Postwartransition 1.8 0.37 2.53 1.36 0.21 -2.6 28.19 39.22 -41.68 -10.78
(19521—61)
FirstPlan 1.7 0.24 2.04 1.42 0.15 3.3 13.56 17.24 -17.70 4.14
(1962-66)
SecondPlan 3.0 0.43 3.5 2.5 0.14 8.3 17.54 22.6 ~16.5 2.11
(1967-1971)
ThirdPlan 3.8 0.83 4.9 2.9 0.22 -3.4 26.08 27.1 -34.9 -7.76
(1972—76)
FourthPlan 5.0 0.80 5.9 4.0 0.16 3.3 24.81 29.0 -33.6 7.50
(1977-81)
FifthPlan 4.9 0.67 5.9 4.3 0.14 -1.4 15.58 16.0 -18.6 -1o.75
(1982-86)
SixthPlan 6.0 1.87 8.2 41 0.31 9.7 17.05 35.3 -7.0 1.77
(1987-91)
ResearchTarget 3.61 1.82 8.2 1.36 0.50 1.50 20.63 39.2 -41.68 13.78

 

 

 

 

Note: Units: %

1) SD. means Standard Deviation

2) CD. means Coefﬁcient of Deviation

 

121

Figure 9 The Share of Housing Investment as the Percentage of GNP

 

 

 

 

 

 

 

 

 

 

 

 

      

 

 

 

9 40
8 J 30
t
7 - i ’0 , g 20
6 _- 10
5 - 0 3‘
a! D
4 -. -10
4
3 _. -20
2 t. J -30
1 JJ J ‘ -40
l l
0 ‘ J -50
(‘0 d3 (0 no
In In (O CO
e 92 92 2
!— Share of housing investment —.—grow th rate of share of housing investment I

Table 30 New Housing Construction

 

122

 

 

 

 

 

 

 

 

"Floor Area: 3:
Period.'.,~_*-;._~_._._ Mean. ' {standard 9' Maximtim“-Minimum* , Coefﬁcient
_ . 1 , . _ J .‘fDeviation ‘ ' 'J   ., v .4 J- ’ 5 » orpeviationi
Postwar transition (1953—61) 402.5667 175.7513 734.8 262.5 0.436577
First Plan(l962-66) 945.16 496.025 1731 390.6 0.524805
Second Plan (1967-1971) 4098.6 1641.185 5885 1977 0.400426
Third Plan (1972-76) 8773 2727.126 11623 4524 0.310854
Fourth Plan (1977-81) 14036.2 2661.378 17516 10308 0.189608
Fifth Plan (1982-86) 20406.8 2253.452 22518 16651 0.110427
Sixth Plan (1987-91) 45859.6 31631.17 70927 21639 0.689739
Research Target (1953-92) 13223.32 17383.74 70927 262.5 1.314627
(Unit: FA 1000 m‘)
DwellinsUniii; T
:Periodtjfj.‘ . ' ' ' ' _ Mean. ; . Standard. ._ '_Maximum ‘ Minimum' Coefﬁcient._;géz
_ -: If . - , w ' abeviatioo . w .aneviamf’f
Postwartransition (195361) 8372.333 ' 3290.638 ' 14727 i i 85154 0.393037 '
First Plan (1962-66) 14205.2 6354.965 23353 6990 0.447369
Second Plan (1967-1971) 58371.2 21374.52 80956 28812 0.366183
Third Plan (1972-76) 91388 18724.81 109347 60832 0.204894
Fourth Plan (1977-81) 94984 30208.98 124789 51644 0.318043
Fifth Plan (1982-86) 70219.4 11312.66 89255 61252 0.161105
Sixth Plan (1987-91) 94304 19359.95 122679 69110 0.205293
56553.68 39449.03 124789 5154 0.69755

Research Target (1953-92)

 

(Unit: DU)

123

Figure 11 New Housing Construction Building Permit

 

 

UnnuﬂoorAnn

 

 

140000 -.

1201100- _____

100000- .....

 

.. 140011)

#120011)

 

 

Unlt: DU

 

_______________ —-—-..20000

 

 

+hbw rbus'ng Construction Sliding Farm'ts by Floor Area
*Mw i-bus'ng Construction Blid'ng Famis by Duality Uits

 

 

 

 

 

 

 

124

New housing construction is usually measured by housing statistics of dwelling
units and ﬂoor area. In the case of Korea we can replace new housing construction
permit data for real new housing construction although there is a reliability problem
involved in using building permits to trace the recent evolution of the housing stock.
For a consistency check between the building permits data and actual construction data,
we note that the slight difference of 3 percent exists. It may be due to a limited amount
of illegal construction across all urban areas.

We see that basically new housing construction patterns are consistent with the
previous level and share of housing investment patterns (Table 30). Figure 11
demonstrates that housing quality in terms of ﬂoor area per new construction unit has
improved in Korea.

A recent estimate of new housing stock demand runs as high as 5.9 million units
by the year 2001. The estimation is based on the assumption that a close relationship
exists between per capita income and housing investment. (Kim, 1993:336)

Housing ﬂuctuations in terms of the percentage of change in new housing
construction from the previous year appear in Table 31 and Figure 12 Korea had

negative growth in the Postwar transition period and in the Fourth Plan period.

125

Table 31 The Growth Rate of New Housing Construction

 

 

 

 

 

 

 

 

 

FlOOrw‘A’rea;-ﬂ _

Period. 3' ' - 7 _' , Mean - Standard, Maximmn :_ Minimum _. Coefﬁcient .
.p 2‘;Deviation'.,lg " i. _' -iofiDeviation--
Postwartransition(195461) -2.3 ' 37.64 J ' 40.72 H -7547 ' ‘ ' ' ' -16.49 ' '
First Plan (1962-66) 24.5 18.83 47.75 5.15 0.77
Second Plan (1967-1971) 19.3 17.98 38.7 -5.2 0.93

Third Plan (1972-76) 6.3 29.02 42.9 —23.7 4.64
Fourth Plan (1977-81) -1.8 29.91 26.5 43.0 -16.48
Fifth Plan (1982-86) 12.9 17.78 38.1 -5.6 1.38

Sixth Plan (1987-91) 14.7 23.21 36.5 -20.1 1.58
Research Target (195492) 9.51 27.02 47.8 -7547 2.84

(Unit: 1000 m‘)

Dwelling-Unit. V ,

Period ' ~ """ Mean j Standard. Maximo Minimum ; coefﬁcient , .
.1 . , ; -. .. . . .1 Devan” ] i ’1 J J. ' '7 “Deviation
Postwar transition (1953-61) -5.7 40.08 41.90 -80.36 .707

First Plan (1962-66) 17.2 27.86 44.81 -19.57 1.62
Second Plan (1967-1971) 19.8 17.00 40.1 -5.5 0.86

Third Plan (1972-76) 0.0 24.88 37.0 -26.1 -1334.82
Fourth Plan (1977—81) -14.7 30.12 25.0 -50.4 -2.05

Fiﬁh Plan (1982—86) 1.6 24.97 27.8 -35.7 15.94

Sixth Plan (1987-91) 5.2 22.73 22.4 -34.1 4.36
Research Target (195492) 2.63 28.87 44.8 -80.36 10.99

 

(Unit: DU unit)

126

Figme 1 The Growth Pate of New Housing Permits

 

 

 

 

60

 

-40,

-60.

 

 

 

 

 

 

 

 

 

 

 

 

 

)-
b

 

 

h
<1-
..

1 11111111
I 7" I

'1'-

1
T

TIH VII IWYSUIIT I 111
NWO!- [x
BN5 § §

888880.883 s

FFFFI—FFFFFP“
Year

 

 

+ Growth Rate of New l-busing Construction Buid'ng Emits by Floor Area
_._Growth Rate of 'Mw l'bus'ng Construction Building Ernie by Dwel'ng this

I

 

 

 

CHAPTER SIX

TESTING MODELS OF HOUSING INVESTMENT IN KOREA

1 . Introduction

This chapter is an empirical interpretation of explanatory analysis concerning
several housing investment models. First, it presents the overall data description and
correlation relationship for various housing investment models. Second, the chapter
illuminates determinants of housing investment by using factor analysis. The chapter
factor analyses veriﬁes the implications of empirical ﬁndings of the various housing
investment models. Third, it tests the speciﬁcations of the housing investment models.

Three analytical methods were used : (1) Pearson correlation analysis, (2) factor
analysis and (3) multiple regression analysis. The analysis began with selection of
dependent variables. Out of 75 variables, 11 important variables were selected for
regression analysis. There were initially 36 independent variables which were grouped
into 6 categories according to similar characteristics. Regression equations were then
derived for the most important categories of housing investment. Qualitative variables

were treated as dummy variables.

127

128

2. Data Description and Data Collection

The data of income variables are collected at 1970 constant prices. The results
are quite different from the current prices. It is important to use constant prices because
these are more appropriate in our analysis by representing dramatic changes of trends.
Observations of population growth are based on census data, as are the estimates of
mid-year population to calculate the growth rate from the previous year. The long-term
mortgage rate is considered as the interest rate in our analysis.

Per capita GNP can be termed as the level of income, and ratio of housing
investment over the total output can be termed as share of housing investment. As stated
in the literature review, most of the studies used per capita GNP, and percent of
housing investment over the total output as dependent variables. The data description in
the housing investment models is presented in Table 32.

In order to identify important dependent variables to perform regression
analysis, initially 10 dependent variables were introduced to Show the different aspects
and diversity of housing investment. But due to the limited scope of this research, a
reduced number of categories were needed. This was done by using correlation analysis

and factor analysis.

Table 32 Descriptive Statistics

129

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Valuables PERGNPOJ ECRTE J j TRoJ TEOJ . - MsoJ DEFLAJ EXCHANGE
Np, ; 41 41 41 41 41 41 41

Min, 29.815 -37 66.533 53.6 40426.667 2.2 18

Max. , 583.447 13.8 5282.15 6291.58 288739312 122.4 88890.2

Mean 182.463 7. 563 1440.271 1535.173 735363. 538 40.029 439.444

Var. 29178.249 15.551 251786061 275633539 582552E+12 1245.73 73888.975

,J 1

3.1)}? . 170.816 3.944 1586.777 1660.221 763251.183 35.295 271.825

‘ Medi.” ‘ 110.562 7.6 587.609 851.53 561769.231 27.2 398.9

Variables CPI TRBALO PPI J DOMSAV J HSGEXPIN J INTR DPOP
N 41 41 41 40 33 40 40
Min.) 7.5 4078.667 1.1 1.358 27.16 8 -0347
. Max. 1005.8 1590.081 108.6 35.877 486.031 19.5 4.403
Mean 1 307.354 -768.109 42.015 19.736 202.869 12.726 1.768
Var. . : 100608.413 871441.39 1625.545 148.977 15068.821 8.224 0.95
s, D}? 317.188 933.51 40.318 12.206 122.755 2.868 0.974
Medi! ' 130 -781 17.91 20.345 168.772 13.87 1.574
Variables URB J HIHSH J URBRTE J HGLOPC J PHF PDEN J IRUPOP‘
N, _ g 40 33 40 27 19 41 39
Min _ 45.78 3.568 37.2 0.18 11.3 205.1 -0.467
Max. _ 22.131 155.93 83.7 301.366 2711.219 443.6 9.957
Mean ‘ ' 4.461 46.32 56.374 57.28 762.53 335.751 3.733
var, , 65.087 1872.505 248.968 6813.685 787120.71 5785.103 3.918

* . 8

S. D." ~ 8.068 43.272 15.779 82.545 8887.198 76.06 1.979
Mani. 4.919 35.497 52.8 2791.08 17.231 345.3 3.423
Variables NTHH J NSIZE J HRTE J RATDEF J NHF J swsp J SWPG
N 33 34 35 40 19 33 33
Min. ,; ' 4358 3.7 62.9 6.247 24.6 104 2.2
Max. 11907 5.87 82.5 62.189 3414.266 108708 4.8
Mean 7 7127.818 4.948 75.423 27.334 950.068 23152.242 3.752
var, J~ 488589890 0.531 25.421 169.186 13695906 996388000 0.439
s, D," ; 2210.407 0.729 5.042 13.007 1170.295 31565.619 0.662
Medi. ~ 6702 5.225 76.9 4.586 414.546 5806 3.9
Note: 8. D means Standard Deviation

 

 

 

 

130

2-1. Correlation Analysis

Pearson correlation technique is applied to investigate the correlation patterns.
Two types of correlation matrices are developed: (1) correlation among independent
variables, and (2) correlation between selected dependent variables and independent

variables.

2-1-1. The Correlation Relationship among Dependent Variables

Because of the different meanings of these measurements of housing investment,
the relationships among these patterns are to be examined. Table 33 shows the
correlation matrix of these measurements. We can see very high correlation coefﬁcients
(over 0.8) among these measurements. Since new housing construction is the major
component in the level of housing investment, it is important to ﬁnd that this variable is
statistically signiﬁcant. However, the coefﬁcient of PBBCDW and other variables are
quite low, so the PBBCDW variable is eliminated from dependent variables. Figures
13 and 14 show the relationship among these measurements which are transformed by
normalized plots.46

The original units of the series have no effect on the plot; each series has the

same amount of variation. Normalized plots (Z = (ll—2‘2) are helpful in studying the
S .D x.

 

‘6 Normalized plots (so-called 2 Scores) are normalized the series by dividing the differences of the series and its
mean by its standard deviation.

131

joint movements of two or more series with different units of measurement. We can see
quite a coincidence among these patterns of three measurement in general, although a
slightly different pattern appears between the level and share of housing investment

from 1953 to 1968.

Table 33 Correlation Matrix among Dependent Variables

 

CONNP , HIHSH ‘ PHAREA ‘ REso PBBCDW

 

~, CONNP];I] 1 1.00

; HIHSH? I 0.916 1.00
PHAREA. 0.861 0.980 1.00
RBso 5 0.860 0.984 0.982 1.00

‘Pnnch: 0.774 0.602 0.573 0.496 1.00

 

 

Note: RESOzLevel of housing investment
CONNP:Share of housing investment
PHAREA: Per capita new housing construction area
HIHSH: Per household level of housing investment
PBBCDW:New housing construction by dwelling unit

132

Figure 13 The Relationships among Housing Investment Patterns by Normalized
Plots

 

 

 

Normalized Z

 

 

 

 

 

J111
f1

11
TTTIV'TTW

,3;

.J

1953
1956 l-
1959
1962 ._
1965 I.
1968 .t
1971
1974
1977

1 1 1 1 L L
V I T V T T
a- v- s-

1983

+ Level of I-bushg tweatrrant + Shae of I-bushg hveatn‘mt
+New Buidhg ConstructbnErrrtaby FloorArea +New Buldhg ConstructionEmisbydweIrgunls

 

 

 

 

 

 

 

 

133

Table 34 and Figure 14 show the relationship among the housing
growth/ﬂuctuation patterns. We can see that highly coincident growth patterns exist.
The growth pattern between the level and share of housing investment is still coincident
with each other. The correlation coefﬁcient among housing investment growth patterns
has statistical meaning; however, it is much lower than that of the housing investment
pattern itself (over 0.5). We may ﬁnd that the ﬂoor area is more correlate than that of
the dwelling unit approach. It can be said that the ﬂoor area approach is a more

accurate representation.

134

Figure 14 The Relationships among Housing Investment Growth Patterns by
Normalized Plots

 

 

 

 

 

 

 

 

 

 

 

 

q
1----
.l
|
011
N . J i
N
= -1- --
i
o
z
-2‘l- —————
-3. ---------------------------------------
-4 1 111 1 1111111 111111111111
TIIIIIITTTTIIIIIIIlVTITﬁﬁITTHIIIIII
E N O ('0 03 N ID (I) ‘- 3 Ix O
In 8 8 8 N N N co no a:
O) O) O) O) O) a: O) O) O)
F 1- ‘- 1- v- 1- v- 1- a- s- 1- V- ‘-
Year

 

+ Growth Rate of Real Housing hvestment

+ Growth Rate of Share of Housing hvestment
_._ Growth Rate of New Construction Floor Area
+ Growth Rate of New Construction dw ell'ng units

 

 

 

 

 

 

 

 

 

135

Table 34 Correlation Matrix among Housing Investment Growth Patterns

 

 

. ' IRHIV IRCONNP   GRPABCFLO
IRCONNP 0.973
GRPABCFLO ' 0.601 0.555
GRPBBCDW 0.562 0.515 0.914

 

 

Note: IRHIV: Growth rate of the level of housing investment
IRCONNP: Growth rate of the share of housing investment
GRPABCFLO: Growth rate of the new housing construction by ﬂoor area
GRPBBCDW: Growth rate of the new housing consn'uction by dwelling unit

2-1-2. The Correlation Relationship among Independent Variables

The correlation matrix of independent variables in housing investment model is
presented in Table 35. The level of housing investment is highly correlated with the per
capita GNP(0.826). The coefﬁcient of TRO, TEO, M80, and SWSP to the Per Capita
income shows very high (over 0.97). This may dominate the relations over other
variables’ relationship. Therefore, these variables are also eliminated for regression
analysis. The results of correlation analysis are described below.

The three population variables show that they are moderately correlated.
Population density is negatively correlated with population growth (DPOP). Several
income measuring variables [per capita GNP, total revenue (TRO), total expenditure
(TEO), and money supply (MSO)] are strongly correlated to each other. Among the
four housing related variables, most of them are moderately correlated with each other.
Considering the correlation with housing invesnnent, three variables, such as, people in

housing units (NSIZE), number of households (NTHH), number of housing units (NHO)

136

are highly correlated with each other variables. However, housing supply ratio (HRTE) are

negatively correlated with all other housing variables.

137

X: .6.
2.6.6.
6NN6.
M6666.
2N6.
36.6
«66.6
mmm6-
6: .6
6x. .6-
2 N6
N6N6-
cemd.
3&6-
666.—

664 6.95 MHI7= gag Nana-lead 0.4493

3.6-

www.cr
mcn6.
«ﬁnd.
66m6.
06— .6
666.6-
m666
>696
66N6

V696

mmmd-
NN66.
m6m6-
66nd
666.—

2.96.
63.6-
658.6.
E.— .6-
36.6.
mend
9&6-
3N6
9K6
666.6

93.6

666.6-
v.36-
no.6-
6mm6

:66

666.—

m66

666.6
36.6
6696
6666
Se— .6-
.366
on: .6
N966-
66N6-
mm66-
$.66
56 6
$66
93.6.
6696-
~86-
666.—

mmm6
36.6
66m6
63 .6
6S6
m2 .6
m6n6
6v: .6
63:6-
66— .6.
wmc6-
”3.6.6
636

:66

m :6-
mom6-
3.5.6-
636
666.—

>26
N666
306
wcmd
6:6
6:6.
mend
6-.6-
6R6-
62.6.
658.6.
M: m6
65.6
634.6
.3 .6.
End.
mm“?
xxm6
63.6
666.—

£66
$6.6
”$66
63.6

366.6
3&6-
35.6
_m_ .6
666.6-
3%.6.
mm66.
3.6.6
5666

666.6

mVN6-
m6n6-
«.696-
E66
63. .6
$36
666. _

66m6
X36
666.6
:66
6.3.6
mmm6-
mm: .6
M166
6mm6-
33.6.
V3.6-
$.66
665.6

5.66

xmn6-
666.6-
_cv.6-
666.6
6mm6
N6m6
666.6
666. _

Eu maﬁa

3.66.6
6mx6
36.6
6. >6
3.66
N. _ m6-
65.6
52.6
m. _ x6-
626-
".666-
6666
N666
M666
3N6-
66.6-
~36-
«K66
_6w6
66m6
5666
94.56
666.—

 

666.6
66m6
3.56
69.6
666.6
6m. .6-
:66
M»: .6
2.6.6-
I: .6-
m6w6-
6666
m666
C666
66m6-
3496.
6x66.
:66
6666
36.6
6666
$3.6
mN66
666.—

096 g aggmamum

N666
«16.6
6666
69.6
666.6
66: .6-
3.6.6
66. .6
2.6.6-
.6— .6.
3.6.6-
3.6.6
3.6.6
xm66
6626.6-
63.6-
5966.
5666
:66
:m6
3.66
6.3.6
6m66
«£66
666.—

N666
mmw6
69.6.6
N666
T666
3.4.36-
6366
366.6
3.66-
3‘6-
6m66-
3.6.6
2.6.6
$16.6
mﬁm6-
66m.6-
_:..6-
6x66
m>c6
65.6
Nm66
:6 6
666 6
6m66
E66
666.—

$66
N066
36.6-
21.6-
N"; 6
2.6.6
2.6.6
5&6-
x666-
we. .6
«ec— .6-
cm. .6
6K; .6
Mm.— .6
93.6-
mnm6-
mmm.6-
we. 6
v~c6
66.66
666.6
6m66-
m6— .6
m6~6
:4— .6
2...»..6
666.—

£66
:66
666.6
3V6
.9666
v2 .6-
:66
6x. .6
3.6.6-
N: .6-
666.6.
9.6.6
:66
6666
3&6-
c _ m .6.
696.6-
6m666
6666
v~m6
2.6.6
:66
666.6
666.6
mx66
346.6
63.6
666.—

26.6
6wx6
62.6
$3.6
6666
N3 .6-
3.6.6
66. .6
6966-
66— .6-
-6.6-
$66
xm66
:66
ch6.
63.6.
N666-
3.66
N226
6666
6.66
mcm .6
$66
5.66
6666
N256
_m.~.6
6x66
666.—

Min.

"6:2

En—

063m
6320.
666:“me
620E
ZC_._.Um._m
mmOhé
mum:
mN_mZ
22,—.2
ZQO
6.5—mm:
mm:

6066
66.2—
2.668sz
><mEOQ
0.266...
EU
mOZ<=UXm
643756
OmE

06,—.

61,—.
6.5.06
mgzommm
0620me

waists) unaccounte- mcoE< EEoEoou ecu—«.230 mm o3:

138

000,—

356 33.6
666.— 3.0.6
666.—

mnm mzz

9.6

C36
x5x6
666.—

5.66.6
~86
66w6
onnd
666.—

5 Guam

mm: .6-
65_ .6-
wc~.6-
55_ .6.
56nd-
666.—

mwa Sum-Ed £2636“ gum-E Mada a: amid

3x6
6~x6
03:6
66N6
5cw6
6m. .6-
666.—

~: 6
mx: 6
$66
666.
cm. .6
_5~6
3.6.6.
666.—

www.6-
:66.
536.
2.3.6-
«.5w6.
me. .6
3.5.6-
6266
666.—

c: .6-
5_ m6.
66m...-
n36-
3&6-
6.5-6
3...:-
9.: .6
mm~6
666.—

3E6-
5 _ x6-
5xx6-
$.06-
_m6gﬁ
$3.6

~656-
056.6-
mw56

54.6

26.6 5mm6 636.6 $366 “.11
x666 6x56 N6x6 ~6x 6 LIZ
«.366 nm66 5m66 55.66 7:
~36 mN56 6m56 65.56 063m
.256 26.6 636.6 6V66 mmgw
wm~.6- M_N.6- 6:6- x_~...6- (Pm-6&0
:w6 3.5.6 NC56 NC56 4:2035
mg .6 5_ _ .6 mm. .6 mm. .6 O-FUmqm
56x6- 356. M656- p.656- two-569‘
5.26- 55m6- 6.3.6- x676- mhm:
666. _ 366- £666- 6666- mN_mZ
~66.— 6m.6.6 5m66 EEZ

666.— m666 ZMQL

666.— ugh—mm:

6%:

6069

~652—

ZEXmOmZ

><MECQ

041:5:-

EU

mOZ<ZUXm

_<-EmQ

Om:

Cm:-

Cy:-

MERLE

mOA—ZOmm:

2206mm

alum—mg

69.2006 mm min...

139

3. Factor Analysis and Factor Matrix

Factor analysis is a technique for reducing large sets of collinear data to a few
synthetic variables according to given criteria. This analysis is a scientiﬁc way of
screening redundant variables. It is primarily the choice of variables which determine
the results, i.e., the factors that can be identiﬁed. Therefore, the selection of variables
should be guided primarily by theoretical background i.e., by their expected relevance
to the basic determinants underlying the housing investment. A principal component
analysis is performed for using the principal component scores in a multiple regression.
It is essential to perform this analysis to estimate the effect of a particular explanatory
variable on the dependent variable. This principal component analysis has advantages
of escaping from problems associated with multicollinearity and degrees of freedom.
In view of ﬁndings of factorial determinants on housing investment, we look for
variables associated with the following characteristics: (1) socio-economic status; (2)
institutional setting; (3) demographic and urbanization; (4) housing conditions; (5)
policy effect; and (6) foreign affairs effect. The independent variables enter the analysis
and result in thirty six principal components, six of them with eignevalues of one or
above. The number of factors extracted equals the number of eignevalues exceeding
one. The 36 variables are selected from the period of 1953 to 1993, thus there is a 40 x

36 variables matrix. Table 36 presents this factor matrix.

140

Table 36 Summary of Major Factors and Variablas

 

 

 

 

Factor 3 ' VanableName IDern-ittonandUt-its ,
Sodo—EcOnomic ' PERGNPO Per capita GNP (constant) 1000won
‘ Status 3 if C . PERGNPO2 Per capita GNP2 1000 won

’ ‘ . . TRO Total Revemte (constant) Bil. won

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

. ‘ TEO Total Expenditure (constant) Bil. won
‘ .‘ ~ 3 M80 Money Supply (constant) Bil. won
, DEFLATI Deﬂator Index
' ' HSGEXPIN Average expenditure per household in the national
. budget 1000 won
‘ : CPI Consumer Price Index ( 1970 base year)
» ' . PPI Producer Price Index
, ~: ’ , ,; DOMSAV Ratio ofDomestic Savings toGNP (1954—1992) %
3 Institutional: , ECRTE Economic Growth Rate %
j Setting ‘ f ' RATDEF Share of Defense Spending over Total Expenditure %
. ' ‘ ‘ ' ‘ SWSP Social Welfare Spendiéig in Mil. won (1962-93)
SWPG Social Welfare Spending in Percentage of GNP (1962-
93)%
VPOIicy Effect , * PHF Public Housing Fund by KHB
. NHF NationalHousingFundbyKHB
. RNHNP Share of number of public housing over total number of
housing construction
. V HPOL Strength of anti-speculation measures
f HGLOPC Per capita housing loan from Korea Housing Bank 1000
, . ' won
_ j. 1 j GRMS GrowthRate ofMoney Supply %
Housing . = . *_ NTHH Total Number of Householdle
_'.Conditions.. . NSIZE Person Per Each Households
. L " HHSGRO Growth Rate of Households %
_ g, STOGRO Growth Rate of Housing Units, %
Demographicandl URBR annual rate of growth of cities over 100,000 divided by
«UrbanizatiOn i ' ’ average rate of growth in national population %
. = DPOP Rate ofGrowth %
~ PDEN Population Density Per Km2
’ IRUPOP Increase rate of urban population
= T URB Rates of Growth in cities over 100,000
g .. URBPOP Urban Population
: ’ y ‘ URBRTE Urbanization Rate
. Foreign Affairs 3 ii; INTR Interest Rate%
i Effect DCPI Inﬂation Rate %
‘ . . . ' : EXCHAN Exchange Rate of won to U.S. Dollar, won
TRBALO Trade Balance in million US dollars (constant) Mil. $

 

 

141

Communality is deﬁned as the percentage of the variance of each variable which
can be explained by a regression on the remaining variables. The six principal factors
extracted explain respectively 59.595 %, 10.884%, 7.958%, 5.505%, 4.460%, and
3.276% of the total variance - a total of almost 91.678%. The principal factors are not
unique but can be rotated to simplify the structure and improve the interpretability of
the results. The following is a discussion of the rotation according to the Varimax

criteria, which yield the best results for our research interests.

3-1. Orthogonal Rotation with the Varimax Criterion

The Varimax criterion maximizes the interpretability of factors by rotating the
axes so that each factor has a minimum number of variables with large loading on it.
The ﬁrst four of the six factors obtained by this rotation explain most of the variance of
the variables - 52.418%, 9.983%, 6.773%, and 13.066%. The rotation factor matrix is
presented in Table 37. The ﬁfth factor contributes only 4.707%.

Interpretation of the factors from the factor loadings is straightforward. The
ﬁrst factor is socio-economic status such as per capita GNP, Total Revenue, Total
Expenditure, Deﬂator, etc. The signiﬁcant relationships exist with several variables
associated with housing conditions. These include number of total households, and size
of dwelling units. The second factor is called institutional setting; it has high loadings
on the variable associated with economic growth rate. The factor is correlated with

variables associated with socio-eoonomic status such as social welfare expenditure. The

142

third factor, called policy effects, has high loadings on the variables associated with the
national housing fund variable and the housing policy variable. The fourth factor is
termed housing condition; it has no high loadings on the variables associated with it
and is thus somewhat ambiguous in its interpretation. The factor is correlated with
variables associated with policy effect. The ﬁfth factor is called demographic and
urbanization, has its largest loadings on the two variables related to urbanization; the
percentage of urban population, URB and URBR. The last factor is termed foreign
aﬁ'air eﬂ'ect. It has largest loadings on the inﬂation variable. In addition, there are high

loadings on variables associated with the interest rate and exchange rate.

143

Table 37 Factor Matrix: Component Loadings

 

 

 

 

 

PERGNPO . 0.962 0.172 0.019 0.137 0.009 0.073
f-‘PERGNP02 0.941 0.214 0.127 0.180 0.023 0.079
;ncarnp ,gt   0.212 0.690 0.401 0.241 0.141 0.163
TRO _ 0.981 0.057 0.059 0.001 0.015 0.017
f T50] ‘ ‘ - 0.985 0.053 0.088 0.119 0.002 0.005
M80 ‘ - 0.956 0.220 -0.036 0.151 0.050 0.010
; GRMS 0.211 0.350 0.091 0.034 0.710 0.434
DEFLAI 0.980 0.150 0.062 0.036 0.053 0.045
EXCHANG 0.666 0.574 0.363 0.199 0.054 0.029
cm; ., 0.967 0.214 0.057 0.013 0.049 0.085
TRBALOT 0.473 0.134 0.497 0.395 0.258 0.062
i-PPI :; 0.866 0.422 0.156 0.054 0.085 0.129
DOMSAV 0.655 0.516 0.217 0.027 0.183 0.390
‘jTHSGBXPIN 0.983 0.039 0.017 0.069 0.063 0.098
-~INTR ' 0.683 0.390 0.230 0.345 0.018 0.112
{DCPI 0.538 0.407 0.205 0.433 0.355 0.300
uDPOP 0.616 0.172 0.130 0.519 0.330 0.075
URB ~ _ . 0.167 0.191 0.716 0.466 -0.098 0.137
_..URBPoP-- .. g 0.985 0.057 0.083 0.031 0.081 0.029
.V’UIRVBRTE ~ 0.976 0.137 0.080 0.020 0.082 0.022
URBR 0.006 0.129 0.544 0.756 0.220 «0.062
PDEN 0.976 0.119 0.138 0.021 0.048 0.031
IRUPOP 0.452 0.541 0.068 0.117 0.290 0.395
1 {ii'a-f: 0.978 0.072 0.078 0.100 0.039 0.117
NSIZE 0.964 0.115 0.054 0.021 0.031 0.117
RATDEF . , 0.834 0.027 0.146 0.015 0.068 0.052
AEXPEN ’ 0.985 0.045 0.085 0.121 0.006 0.009
.sroouo . _v _ 0.532 0.010 0.612 0.204 0.082 0.123
_;I-IHSGR054 0.174 0.395 0.109 0.204 0.658 0.464
:iswsp 0.983 0.051 0.132 0.105 0.003 0.008
7' " 0.666 0.580 0.041 0.052 0.114 0.230
0.851 0.103 0.416 0.110 0.062 0.052

0.905 0.132 0.319 0.182 0.033 0.085

0.953 0.130 0.192 0.160 0.030 0.033

0.202 0.669 0.365 0.246 0.100 0.027

0. 086 0. 410 0. 705 0. 046 0. 254 0. 382

 

{Note Varlance explained by components. factor 1 (21 .,454) factor 2(3. 918), factor 3 (2. 865),

factor 4 (1.982), factor 5 (l. 606) and factor 6 (1.18)

144

Table 38 Factor Matrix: Rotated Loadings

 

Momma” Factors Factoraractorsracm ,—

 

 

var
389N992

'{EP
fieNTHI-I

PBRGNPO
MSO _' = ,:
[HSGEXPIN .

CNHF

juniper»
TRO

-.'DEFLA1,

URBRTE, f ’

T, CPI _' g f p j
NSIZE

RATDEF »
PPI :

DOMSAV
’WSP-ﬁ} ' L

TRBALO : a

= URBR

DCPI z *
EXCHANGE
HHSGRO'
GRMS . i“

:L’DPJOP‘. " ~ . ‘ ,

 

0.978
0.975
0.971
0.970
0.966
0.966
0.953
0.950
0.945
0.931
0.927
0.909
0.903
0.902
0.894
0.878
-0. 877
0.868
-0.802
0.735
0.618
0.575
0.563
-0.502
0.076
0.113
0.213
-0.075
-0. 134
-0.270
0.449
0.134
-0.372
0. 100
-0.226
0.437

0.132
0.120
0.128
0.121
0.013
0.217
0.103
0.248
0.250
0.143
0.016
0.141
0.237
0.067
0.091
0.017
-0.178
0.148
-0.152
-0.093
0.665
0.007
-0.322
-0.711
0.856
0.780
0.626
0.015
-0.176
-0.072
-0.025
0.165
0.073
0.316
0.090
-0.435

-0.039
-0.038
0.004
0.008
-0. 101
0.022
-0.026
0.065
0.082
0.055
-0.241
0.058
0.032
0.042
0.041
0.023
0.018
0.059
0. 131
0.038
0. 104
-0. 598
—0.040
0.054
0. 163
-0.077
0.097
-0.932
-0.855
0. 129
0.073
0.000
0.221
0. 166
0.374
0.262

0.053

0.166
0.181

0. 187

-0.011
0.005
0.241

0. 103

0.092
0.276
-0.063
0.360
0.302
0.411

0.407
0.466
-0.379
0.451

-0. 183
0.642
-0. 130
-0.043
0.643

-0.254
-0.067
-0.399
0.513

0.129
-0.066
—0.857
0.831

-0.021
-0.431
0.159
0.414
-0.360

—0.019
0.007
0.019
0.022
0.084
-0.016
-0.119
0.013
0.034
0. 132
0.014
0.085
0.070
0.044
0.058
0.050
0.094
0.057
0.055
0.056
0.082
-0. 121
0.076
-0.022
0.094
0.293
-0.050
0. 177
-0. 180
0.070
-0.112
0.916
0.582
-0.094
-0.129
-0.453

0. 105
0.063
0.059
0.056
0. 169
0.079
-0.031
0.023
0. 133
0.047
0.073
-0.054
0.020
-0.047
-0.082
-0.037
0.084
-0.064
—0.069
-0. 121
-0.265
0. 128
-0.097
-0. 1 12
0.089
0.269
0.047
-0. 1 19
0.077
-0.258
-0. 193
-0.060
0. 195
0.834
-0.705
0. 161

Note: Variance explained by components: factor 1 (18.871), factor 2 (3.594), factor 3 (2.438),

factor 4 (4.704), factor 5 (1.695) and factor 6 (1.703)

145

Table 39 Factor 1 ,2 and Determinants

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

EFaCtor 1 Factor 2
Rank lVariables Rotated Determinants Variables Rotated Determinants
3 Loadings 7. . Loadings '
1 HGLOPC 0.978 Policy Effect SWSP 0.975 Institutional
Setting
2 ' TEO 0.971 Socio-Economic ECRTE 0.856 Institutional
, Status Setting
3 ' AEXPEN 0. 970 Socio-Economic HPOL 0.780 Policy Effect
Status
4 ' PERGNPO2 0.966 Socio-Economic INTR -0.711 Foreign Affairs
Status Effect
5 PERGNPO 0.950 Socio-Economic DOMSA 0.665 Socio-Economic
Status V Status
6 M80 0. 945 Socio-Economic
Status
7 HSGEXPIN 0.931 Socio-Economic
f Status
8 , NHF 0.927 Policy Effect
9 TRO 0.903 Socio-Economic
Status
10 DEFLAI 0.902 Socio-Economic
. Status
11 ‘ . CPI 0.878 Socio-Economic
Status
12 » PPI 0.735 Socio-Economic

 

 

 

Status

 

 

 

146

Table 40 Factor 3, 4 and Determinants

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

{Factor _ 3 Factor _ 4

, Rank iVariables j. Rotated Determinants Variables Rotated Determinants ‘

1 PHF 0.966 Policy Effect NTHI-I 0.953 Housing
. Conditions

2 URBR -0.932 Demographic NSIZE —0.877 Housing
[ & Urban Conditions

3 W URB -0.855 Demographic DCPI -0.857 Foreign Affairs
' & Urban Effect

4 : URBPOP 0.909 Demographic EXCHANGE 0626 Foreign Affairs
. & Urban Effect

5‘ - URBRTE 0.894 Demographic SWPG 0.643 Institutional
& Urban Settmg' ‘

6 ’ RNHNP -0.705 Policy Effect

7 STOGRO 0598 Housing

. Conditions

Table 41 Factor 5, 6 and Determinants
5 Factor . i 5 Factor 6 ,

"Rank Variables . _ .‘Rotated ,Determimnts Variables . Rotated _ Determinants

' fLoadings', 7 ’ '8 Loadings 7

l . HHSGRO 0.916 Housing GRMS 0.834 Policy Effect I

Conditions

2 . PDEN 0.868 Demographic RATDEF 0802 Foreign Affairs
&Urban Effect

3 ' IRUPOP 0.582 Demographic TRABALO 0.047 Foreign Affairs
1 &Urban Effect

4 DPOP 0.453 Demographic

' &Urban

147

Principal component scores are saved, then utilize in a multiple regression
analysis.

All regression results support accepting the null hypothesis -- Korea’s economic
and political variables affect the production of urban housing. The R2 is 0.792, and
adjusted R2 is 0.743. Standard Error of Estimate is 0.886 and the F-ratio is 15.907.
This regression is signiﬁcant at 0.001 level. Table 42 presents the variables in the

regression equation.

Table 42 The Level of Housing Investment Explained by Principal Component
Scores

 

 

. Variable; ft * Coefficient srn Error STDCOEF.- . 1" m2 TAIL)
comm." 4.062 0.157 x 0.000 ’ 25.928 0.000 '
* FACTORO) ' 0.585 0.159 0.335 3.679 0.001
FACTOR(2) -1 1.016 0.159 0.581 6.382 0.000
FACTORS) ' 0.332 0.159 0.190 2.085 0.047
FACTOR(4) 0.830 0.159 0.475 5.212 0.000
FACTOR(5) 0.126 0.159 0.072 0.794 0.435
,FACTOR(6) ’ 0.478 0.159 0.274 3.006 0.006

 

 

As shown in Table 42, domestic political and economic variables and
institutional factors exercise a signiﬁcant effect on the share of housing in the total
output in Korea. The problems of multicollinearity and the number of degrees of
freedom are solved. There is no correlation between the six factors: the problem of

auto correlation is eliminated since the Durbin—Watson is 1.194 suggesting there is very

148

small autocorrelation (ﬁrst order autocorrelation 0.391). This fact supports our
hypotheses of socioeconomic effects on the level of housing investment.

Table 43 and Figure 15 demonstrate the results of factor scores by time-series.

Table 43 Factor Scores by Time-Series

149

 

 

 

 

 

 

 

 

 

Facmr ”p.71 §Factor 2 Factor '3 Factor 4 a Factor 5 Factor 6
Year Scot-es Rankiscom Rank Scores Rank Scares Rank Soot-us Rank Scores Rank
”1953 -6.0783 36 2.3846 18 0.9145 16 -1.7803 36 4.5885 2 1.2703 9
1954 -6.0469 35 3.3900 16 1.4934 11 -1.1962 31 0.5978 1 1.2844 8
1955 -5.8245 34 -9.3484 41 0.9328 15 4.2977 33 -1.5885 2 0.3065 24
1956 -5.7688 33 2.0889 19 0.4891 31 -l.6093 35 0.5978 1 1.2673 10
1957 -5.7357 32 3.2043 17 0.8568 17 2.4984 41 0.5978 1 0.9434 12
1958 -5.6934 31 3.5794 15 0.9002 32 -O.8187 30 0.5978 1 0.5224 18
*1959 -5.6567 30 3.7668 14 1.4220 12 -1.2761 32 0.5978 1 0.5789 15
1960 2.5790 41 -8.0338 40 1.5773 10 -1.7381 38 0.5978 1 1.3049 7
1961' 4.2346 40 4.5318 35 4.3256 33 2.1138 39 0.5978 1 2.3682 1
1962) 45.9719 39 -3.3312 33 0.7608 19 4.3882 34 0.5978 1 1.3244 6
1963 -6.5289 38 -5.3872 37 0.3410 30 4.7734 37 0.5978 1 0.5771 16
1964 63944 37 -6.8675 39 0.5608 22 0.3940 26 4.5885 2 1.3406 5
1965 -5.4966 28 -6.0333 38 4.8908 35 0.2959 25 0.5978 1 2.0716 3
1966 53024 27 4.6761 36 1.1294 13 0.7750 29 0.5978 1 1.0504 11
1967 -5.1967 25 4.4521 34 0.6790 21 0.5569 18 0.5978 1 0.1746 26
1968 -5.5810 29 -3.0663 32 2.3730 4 0.5117 28 0.5978 1 0.5453 28
1969 -5.1716 23 -2.7647 29 2.3390 36 0.6046 16 0.5978 1 0.3436 23
1970 4.1821 21 2.0378 26 2.1258 7 1.0063 9 0.5978 1 2.4270 40
(1971, -52714 26 2.1784 27 -5.1802 40 0.2832 24 0.5978 1 0.6362 14
1972: -5.1938 24 4.8174 25 0.1321 29 0.0699 22 4.6567 4 0.5205 20
f 1973. 4.6236 22 2.9552 31 2.4762 37 0.4014 27 0.5978 1 0.4728 22
1974 4.1696 20 2.5819 28 2.4564 3 1.8526 3 4.6567 4 0.0697 27
19751 -3.2226 19 0.2096 24 1.0562 14 0.5949 17 4.5885 2 0.5403 17
1976 2.6767 18 0.2245 23 4.9263 39 0.0359 21 0.5978 1 -5.2461 41
1977 4.6664 17 0.0587 22 0.0422 26 1.3829 6 0.5978 1 -1.0069 31
1978 0.7824 16 0.3105 21 2.4709 2 0.9689 11 0.5978 1 4.1504 32
1979 0.5736 15 1.1455 20 0.3498 23 1.1632 7 0.5978 1 0.2246 25
1980 0.1737 14 5.2681 13 0.0505 28 0.4607 20 0.5978 1 0.6383 30
1981? 0.4086 13 5.3469 11 0.7236 20 2.1161 40 0.5978 1 0.4931 21
1982’ 1.3349 12 6.6112 2 0.3358 24 0.2739 23 0.5978 1 4.7517 36
19837- 3.0598 11 6.3643 3 0.2762 25 0.4910 19 0.5978 1 0.6801 13
1984 3.8412 10 5.8117 6 4.3061 38 1.6191 5 0.5978 1 4.7732 37
1985 5.0263 9 5.5451 10 0.7880 18 0.8508 13 0.5978 1 1.4361 4
271986 6.0330 8 5.8305 4 0.2570 26 0.6137 15 -1.5885 2 4.6301 34
#1987 7.4428 7 5.5565 9 2.3442 5 1.8984 2 0.5978 1 0.5022 28
41988118644 5 5.2756 12 4.4903 34 1.9358 1 0.5978 1 4.9995 39
[1989 14.1405 4 5.8230 5 1.5877 9 1.7143 4 0.5978 1 4.4583 33
71990 16.7658 3 7.5066 1 -6.4065 41 0.6924 14 2.4703 3 4.9807 38
31991 18.5926 2 5.5909 7 2.5318 1 1.1322 8 0.5978 1 0.5215 19
«1992 20.0110 1 5.5801 8 2.2951 6 0.9981 10 0.5978 1 4.7318 35
1993 11.3524 6 2.8018 30 1.6412 8 0.9393 12 0.5978 s 1 2.1479 2

 

150

Figure 15 Factors Scores of Housing Investment by Time-Series Analysis

 

 

 

 

15

 

10

 

Factor Scores

 

 

 

 

 

 

 

 

 

 

 

 

 

+ Factor 1
-0— Factor 2
_x — Factor 3
-0— Factor 4
._...... Factor 5
* Factor 6

    
   
  
  
   

 

 

 

 

 

151

4. Empirical Results and Interpretation

Multiple regression analysis is the technique which was ﬁnally applied in order to
statistically express housing investment as a function of economic and political
characteristics. Regression equations were derived for the four dependent variables chosen
in the previous chapter. According to their characteristics, we deﬁne models as following.
1) Housing Investment Share Model (using CONNP as dependent variable) 2) Housing
Construction Area Model (using PHAREA as dependent variable) and 3) Housing
Investment Household Model (using HIHSH as dependent variable) (shown in Table 44).

For the question of how much does each independent variable contribute to an
explanation of the share of housing investment over total output, eleven socioeconomic and
political variables, also chosen in the previous chapter, were used as independent variables.

The stepwise method of variable selection process was used to derive regression
equation. Three measures, elasticities and standardized regression coefﬁcient (beta) and the

coefﬁcient of determination (R2), will be applied in interpretation of equations.

Table 44 Classification of Housing Investment Models

 

 

 

 

 

 

 

‘~ 1 .w ; _.‘ Housing » THousing‘u" ' 7 Housing'
1 § Dependent 3 InvestmentShm Commiction i, . , 7 ' ~ Investment ‘ ~ I
V'Variablesj ~- .' ’  ﬁ 'AreaModel-j j~quusehold Q
- : .7 _ . q Model f:
Vj'. Open Model Open Model Open Model
' ‘ National National National
_ ,~ g . _ * Allocation Allocation Allocation
‘ Independent L} Policy Effect Policy Effect Policy Effect
Variables x f *1 Model Model Model
' ~ ‘ _ V Financial Model Financial Model Financial Model
Comprehensive Comprehensive Comprehensive
Model Model Model

 

 

 

 

 

152

For interpretation, we consider regression coefﬁcients meeting the 90 percent
conﬁdence criterion as statistically signiﬁcant. A t-test is used for measuring the
signiﬁcance of individual coefﬁcients; and F-test is used to measure the joint statistical
signiﬁcance of all variables in an equation. With thirty-eight degrees of freedom in
each equation, there is 10 percent of the true difference is not zero if the t value is
1.684 or greater; at a higher level of conﬁdence, with a ﬁve percent chance of a true
non-zero value, t=2.021; and, at the one percent level, the value of t must equal or
exceed 2.704.

We specify ﬁve groups of models classiﬁed by each main model which follows

dependent variables.

4-1. Housing Investment Share Models

To determine the relationship between share of housing investment and the per
capita GNP, we assume that the decisions in the housing sector are strongly associated
with political and institutional settings, economic policies, and urban activities and their
interrelationships. First, the level of income tests the effect of economic capacity on
housing investment. This variable generates estimates for income elasticities. The rate
of increase in the total population is used to gauge the impact of population growth.
The rate of increase in urban population is used to analyze how the pressures of

urbanization might inﬂuence decisions on housing investments.

153

According to the regression model speciﬁed earlier Ch. Five Equation (4), the
estimated regression model found for housing investment, including all variables, is

CONNP = 18.911 + .002 PERGNPO -. 056 PERGNPOZ + .073 URBRTE +

.518 DPOP -.160 DPOP2 -.001 1R0 + .024 HVTR + .113 DOMSAV +

.048 AEXPEN - .189 DENGNP — .031 SWPG + .038 CPI —.I93 PPI -

.004 NHO -I.896 NSIE + .001 NIHH + .234 TWOMH. -.365

DPLSTABL + .018 ELECHON -.001 IRBALO -.007 EXCHANG +2. 792

R2 = .97

This complete equation explains over 97 percent of the variance in the
dependent variable, i.e., housing investment as a share of GNP, and is signiﬁcant at the
.001 level. However, not all variables should be included in the equation in order to
make a signiﬁcantly large contribution to the explanatory power of the equation.
Consequently, stepwise regression procedures are utilized in order to sort out the lesser

contributing variables in terms of beta coefﬁcient.

4-1-1. Stepwise Selection of Variables for Models

Although stepwise regression do not solve to ﬁnd best ﬁtiing model, this
technique can imply plausible models. Table 45 shows the results of housing
investment share model by this stepwise selection. However, we have to understand

that stepwise analysis is usually of no theoretical signiﬁcance.

154

Table 45 The Housing Investment Share Model by Stepwise Selection

 

 

0.015 0.009 0.326 0.070 1.556 0.131
0.063 0.043 0.108 0. 575 1.482 0.150

 

 

 

The results of the estimation of Equation (1), explaining the socioeconomic
effects on the housing investment are found in Column 1 of Table 46. They are as
expected: the share of housing investment is a positive function of per capita GNP and
interest rate, and a negative function of the trade balance and military spending. The t-

statistics of all other variables indicate statistical signiﬁcant.

155

Table 46 Regression Results: Housing Investment Share Model

 

dependent with ﬁnancial with policy CONVP CONN'PTVitu

 

variable consideration effect with saving households
(CONNP) consideration rates variables
Constant 0.122 10.126 -3.851 7.438 -3.368
(0.196) (3,610)" (3.169)" (2.879)” (4.922)”
PERGNPO 0.014 0.021 0.018 0.014
(9.245)” (6.012)” (3.428)1 (2.359)”
PERGNPOZ 0.013 0.03 0.009
(2.569)” (-5. 194) ‘ (4.174)
TRBALO 0.001 0.001 0.001 0.001 0.001
(3.661)” (8.056)” (-7.56)' (6.47)” (6533)”
HRTE 0.126 0.087
(3.78)” (2.913) 1’
INTR 0.123 -0.018
(2.425)” (4.984)2
IRUPOP 0.077
(1.645)
RATDEF 0.03 -0 013

(2.838)2 (4.175)
ELECTION 0.313

 

(4 .225)
GRMS 0.011
(1.399)
DOMSAV 0.035 0.084
(1.475) (5.375) 1’
DENGNP 0.115 0.368
(4.226) (3.108) 1’
HPOL 0.227
(2. 183) 2
NTHl-I 0.001 0.000
(0.994) (4.006) ‘1
TWOMIL 4.07 1.382 0.868
(4.777) (2.231)2 (2.022) 2’
R2 0.888 0.945 0.937 0.938 0.934
F 34.998 53.717 64.939 70.172 76.865
g 0.672 0.475 0.486 0.492 0.488

 

Notes: t-statistics in parenthesis, (1) signiﬁcant at 0.001 level, (2) signiﬁcant at 0.01,
and (3) signiﬁcant at 0.1 level.

156

ZesLResults

Hypotheses 1

Null Hypotheses (Ho) : one of the most important determinants of the level of housing

investment is not socio-economic factors in Korea.

If any variable from domain 1 has a p-value S 0.05 then reject Ho:

Table 47 Testing: Relationships Between the Independent Variables and
Socioeconomic Variables

 

 

 

Variable CEIT'icient STD Error STD COEF Tolerance T P (2

TAIL)
Constant -8.771 4.126 0.00 - -2.126 0.048
PERGNPO 0.014 0.014 1.280 0.002 0.965 0.347
PERGNPOZ -0.064 0.031 -3.363 0.001 -0.036 0.972
TRO -0.002 0.001 -2.083 0.008 -3.248 0.004
TEO —0.000 0.003 -0.080 0.001 -0.036 0.972
MSO 0.000 0.00 0.430 0.006 0.579 0.569
DEFLAl 0.009 0.017 0.166 0.031 0.517 0.611
HSGEXPIN -0.004 0.013 0289 0.004 -0.318 0.754
EXCHANGE -0.010 0.003 -1.360 0.019 —3.324 0.004
CPI -0.009 0.017 -1.469 0.000 -0.488 0.632
PPI 0.061 0.083 1.359 0.001 0.738 0.470
PDEN 0.043 0.016 1.310 0.013 2.615 0.018
DOMSAV 0.103 0.041 0.633 0.051 2.530 0.021
SWSP 0.000 0.000 4.365 0.001 2.995 0.008
Note: 1. F- latio 22.870

 

2. Durbin-Watson D Statistic 2.483
3. First Order Autocorrelation -.264

157

The p values of TRO, Exchange, PDEN, DOMSAV, and SWSP s 0.05.
Therefore we can reject Ho. The socioeconomic factors can predict the level of housing
investment in Korea.

Hypotheses 2

Null Hypotheses (H0) : housing investment as a share of total output are not explained

by the demographic and urbanization factors in Korea.

If any variable from domain 1 has a p-value s 0.05 then reject H0:

Table 48 Testing: Relationships Between the Dependent Variables and
Demographic and Urbanization Variables

 

 

 

 

 

Note "1".rluatto 28.587

2. Durbin-Watson D Statistic 1.095

3. First Order Autocorrelation .439

The p value of URBRTE s 0.05. Therefore we can reject Ho. The demographic

and urbanization factors can predict the level of housing investment in Korea.

Hypotheses 3

Null Hypotheses (H0) : the level of housing investment is not determined by the
policy effects in Korea.

If any variable from domain 1 has a p-value s 0.05 then reject H0:

158

Table 49 Testing: Relationships Between the Dependent Variables and Policy
Effect

 

 

 

Variable Coetiieient STD Error STD coﬁ Tolerance T P (2
T5113)...

Constant 2.463 0.349 0.000 - 7.062 0.000
ELECTION 1.199 0.519 0.315 0.660 2.312 0.028
TWOMIL 3.420 0.631 0.623 0.928 5.421 0.000
REALPLYl 2.208 1.181 0.265 0.610 4.870 0.071
REALPLY2 1.706 1.060 0.205 0.756 1.609 0.118
REALPLY3 3.025 0.950 0.363 0.942 3.184 0.003
GRMS 0.008 0.019 0.053 0.822 0.437 0.665
PARK 1.541 0.593 0.339 0.717 2.596 0.014
DPLSTABL 4.181 0.890 0.171 0.734 4.327 0.194
Note: 1. F-Ratio 6.339

 

2. Durbin-Watson D Statistic 1.323
3. First Order Autocorrelation .188

The p value of ELECTION, TWOMIL, RELAPLYl, REALPLY2, REALPLY3, and
PARK S 0.05. Therefore we can reject Ho. The policy effect factors can predict the
level of housing investment in Korea.

Hypotheses 4

Null Hypotheses (HO) : the level of housing investment is not determined by
housing conditions in Korea.

If any variable from domain 1 has a p-value s 0.05 then reject Ho:

159

Table 50 Testing: Relationships Between the Dependent Variables and Housing
Conditiom

 

 

 

 

Wm“

NTHH 0.002 0.001 2.593 0.008 2.639 0.014
1181213 . ’ . 1.438 0.750 0.566 0.088 1.917 0.067
N110 7 0.002 0.002 4.223 0.007 4.210 0.238
' 11136er ~ . - * 0.162 0.169 0.092 0.835 0.961 0.346
*' HHSGRo ‘ i ' .. 0.062 0.052 0.143 0.533 1.192 0.244
. srooRo __; f ;; 0.005 0.087 -0.006 0.650 0.052 0.959

 

Note: 1. F—Ratio 17.619
2. Durbin-Watson D Statistic 1.120
3. First Order Autocorrelation .437
The p value of NTHH s 0.05. Therefore we can reject Ho. The housing
conditions factors can predict the level of housing investment in Korea.
Hypotheses 5
Null Hypotheses (Ho): among the most important determinants of the level of housing

investment are not governmental and institutional factors in Korea. The budget

constraints factors do not have affected the level of housing investment in Korea.

If any variable from domain 2 has a p-value s 0.05 then reject H0:

160

Table 51 Testing: Relationships Between the Independent Variables and
Imtitutional Settings Variables

 

 

Variable ~ Coefﬁcient STD Error STDCOEF Tolerance 330‘ 9‘ P12

, a , _. ‘ _ .  _; . -g;:.TAIL)_L,
Constant 2. 208 0 918 0.000 - 2.405 0. 023
ECRTE 0.047 0.057 0.105 0.892 0.828 0.415
. RATDEF 0.022 0.027 0.102 0.897 0.805 0.428
3 swPG 0. 000 0.000 0.761 0.975 6.262 0.000

 

 

Note: 1. F-Ratio 13.834
2. Durbin-Watson D Statistic .610
3. First Order Autocorrelation .654
The p value of SWPG S 0.05. Therefore we can reject Ho. The institutional

setting factors can predict the level of housing investment in Korea.

Hypotheses 6

Null Hypotheses (H0) : housing investment as a share of total output is not
previously determined by global and foreign affairs affect in Korea.

If any variable from domain 1 has a p-value s 0.05 then reject Ho:

Table 52 Testing: Relationships Between the Dependent Variables and Global and
Foreign Affairs Effect

 

 

' Variable CoefﬁcientSTDFa'mr STDCOEF Tolerance i, P (2 -. .j
: Constant  ; 3.202 1. 263 0. 000 - 2. 536 0.016
INTR . -  -  f 0.110 0.105 0.171 0.775 1.044 0.303
TRBALO 0. 001 0.000 0.536 0.775 3.266 0.002

 

 

Note. 1. F Ratio 5. 499

2. Durbin-Watson D Statistic .532

3. First Order Autocorrelation .714

The p value of TRBALO _<. 0.05. Therefore we can reject Ho. The global and

foreign affairs effect factors can predict the level of housing investment in Korea.

161

Hypotheses 7

Null Hypotheses (Ho) : some combination of main effect variables from six
different domains will not generate a predictive model of housing investment
that is a better predictor than the main effect variables selected from any one
domain, based upon predictor variables containing a p-values s 0. 05 and an
overall regression p-value $0. 05.

If any combined set of variables from the six different domains has a p-value s
0.05 and an adjusted R2 value that is higher than any single main effect predictor, then

reject Ho:

Table 53 Testing: Results from Stepwise Regression Analysis of Combined Main
Effect Variables, illustrating the best equation

 

 

 

 

“SIZE
1:1
STOGR
URBRTE

0.113
0.000
0.229
0.011
0.027
0.036
0.110

 

 

 

-0.160 0.088 -0.304

 

 

2. Durbin-Watson D Statistic 2.548
3. First Order Autocorrelation -.308

The p values of ELECTION (policy #5), TRBALO (foreign affairs #6), NSIZE
(housing conditions #4), STOGRO (housing conditions #4), PPI (socioeconomic #1),
SWPG (institutional settings #2), and URBRTE( demographic and urbanization #3) p

value 5 0. 05 . Therefore we can reject Ho. Some combination of main effect variables

162

from six different domains will generate a predictive model of housing investment that

is a better predict than the main effect variables selected from any one domain.

SI'EIEEIZJ'RIER'II

Table 54 demonstrates that the results of seven hypotheses testing support our

alternative hypotheses.

Table 54 Synthesis of the Hypotheses Testing Results of Regression Analysis

 

 

Test Results H0, Null Hypotheses Ha. Alternative Hypotheses
Hypothesis 1 reject accept
Hypothesis 2 reject accept
Hypothesis 3 reject accept
Hypothesis 4 reject accept
Hypothesis 5 reject accept
Hypothesis 6 reject accept
Hypothesis 7 reject accept

 

 

4-1-2. Global Open Model (Foreign Affairs and Domestic Housing Investment

Approach)

AW

Economic criteria such as per capita GNP was examined in order to explain

variation in the level of housing investment, compared to demographic criteria, such as

163

population growth and urbanization rate. We assume that the major factors explaining
changes in share of housing investment over GNP are the socioeconomic variables rather
than the demographic variables.

As the importance of per capita income to housing investment is considered in
conventional studies, the relationship between the per capita GNP and housing
investment was proved to be signiﬁcant. For testing of this assumption, we try to four
approaches - linear, quadratic, reciprocal, and log from of per capita GNP. The linear
or the quadratic form of per capita GNP is important for housing investment. This
economic development level is the key variable in housing investment. Although
standard economic development theory as well as the results of statistical analysis
generally supports the importance of the economic development level in housing
investment, it is not clear why the share of housing investment tends to decline at high
levels of economic development. Furthermore, we cannot deﬁne what speciﬁc level of

economic development will housing investment drop.

164

Table 55 Nonlinear Relationship between the Share of Housing Investment
(CONNP) and GNP Per Capita

 

 

 

 

EQ. 1 (CONNP) EQ. 2 (CONNP) EQ. 3 EQ.4
(LCONNP) (LCONNP)
Constant 1.37 2.284 0.298 -2 .72
(4.922)1 (8.469)” (9.427)“ (6184)"
PERGNPO 0.018 0.009 0.001 0.000
(6.122)1 (8.882)” (8.666)“ (4 .808)”
PERGNP02 -0.016
(3.085)2
RGNP(1/GNP) 3.317
(0.072)”
LPERGNPO 0.655
(6.870)”
R2 0.795 0.765 0.664 0.852
F 73.764 61.872 75.107 106.802
SEE 0.858 0.919 0.135 0.091
Note: Equation 1 : Y = a +bx + ex"
Equation2: Y= a-b/xz
Equation3 : b1Y= a-bx
Equation4 : b1Y= a +blnx

t-statistics in parenthesis, (1) signiﬁcant at 0.001 level, (2) signiﬁcant at 0.01, and (3)
signiﬁcant at 0.1 level.

The logarithmic transformation shows the relationship between the percentage

of change in per capita GNP and in housing investment, i.e., the per capita GNP

elasticity of housing investment. The estimated values of t are over 2.0 for the linear

and squared per capita GNP terms except Equation 3 RGNP (l/GNP). This RGNP is

not statistically signiﬁcant. Furthermore, the statistical results (Equation 1) show that

per capita income alone can explain about 79.5 percent of the housing investment

variation. In Korea, quadratic form of per capita GNP ﬁts the ﬂuctuations of

socioeconomic transformation.

165

BD 1' till . . Eli

We can assume that of the two demographic measures, urbanization is a far more

important determinant of the level of housing investment than population growth at the

national level.

Table 56 Relationship between the Share of Housing Investment (CONNP) and

Demographic and Urbanization Variables

 

" E01

 

 

 

SEEN-9.”?

 

5.245
(9.920)”

0.926
(3.524)”

0.246
12.417
1.599

1.491
(2.671)”
0.018
(5.057)”
-0.016
(2.565)”
-0.068
(0.163)
0.007
(0.064)

0.783
31.604
0. 894

0. 040
(0.132)

2 '249
(4.760)”

6. 595
( 2.034)”
—0.006
(1.314)

0.104
(12865)”

0.287
(2.201)”

0.002
(4.458)

0.00
(12769)”
0.811
163.054
0. 801

0.824
56.098
0. 794

0.813
165.519
0. 796

Note: t-statistics in parenthesis, (1) signiﬁcant at 0. 001 level, (2) signiﬁcant at 0. 01, and (3)
signiﬁcant at 0.1 level.

As you see results of Equation 4, urbanization variable (URBRTE) has more

statistical signiﬁcance than population growth variables. However, as you see Equation

1, value for DPOP is ~3.524 which is signiﬁcant at the marginal level of ten percent.

166

:EL . Z 1" E11

Housing conditions effects can be represented by the impacts of housing supply
ratio, size of households, and number of households. However, we consider that the
housing supply ratio (HRTE) as a pre-condition for housing investment. In somehow,
HRTE already implies the level of existing housing investment. HRTE is the one of
existing housing stock variable as measured by the ratio of total dwelling units to total
households. Because HRTE is considered to be the theoretical precondition for housing
investment, it is not included in these housing investment models. The housing stock
variable, however, is examined before building the model. Equations 1 and 2 show that
HRTE is statistically signiﬁcant. This is one of reasons why HRTE is called “pre-
condition”. In any event, we also tests HRTE variable in the models as a comparison.

Equation 1 and 2 show that the t-statistics of the HRTE indicate statistical signiﬁcant.

167

Table 57 Relationship between the Share of Housing Investment (CONNP) and
Housing Conditions Variables

 

 

EQ. 1 EQ. 2 no. 3 EQ. 4 EQ.5 EQ.6
Constant 23.287 11.625 13.586 -3532 0.891 2.747
(7.186)” (4.018)” (10297)” (0.770) (4 .693)” (4 .679)”
PERGNPO 0.007 0.022 -0.006
(6.244)“ (3.831)” (4.197)
PERGNPO2 0.017))
(2.555)-
HRTE 0.256 0.119
(6.961)“ (3.271)”
NSIZE 4 .921 0.858
(4.281)” (1.114)
NTHH 0.001 0.001
(9.733)” (3.050)”
R2 0.518 0.783 0.624 0.755 0.753 0.765
F 35.531 57.708 53.011 30.796 94.734 48.746
SEE 1.260 0.859 1.105 0.921 0.883 0.877

 

 

Note: t-Statistics in parenthesis, (1) Signiﬁcant at 0.001 level, (2) Signiﬁcant at 0.01, and (3)
signiﬁcant at 0.1 level.

DE'lﬂﬁ' [1118! ET!

As global and foreign affairs variables, we may see the effects of interest rate
and trade balance at Table 46 column 1. The equation proves that share of housing
investment is signiﬁcantly related to the global and foreign affairs effect, showing

strong statistical signiﬁcance of effects of interest rate and trade balance.
4-1-3. National Allocation Models

Institutional setting effects and military spending effects are major national

allocation components of national allocation models.

168

|L"IS'E$

In order to prove effects of institutional settings, we try to ﬁnd the effects of
economic growth rate, military Spending, and social welfare spending. Among them,

only military Spending has a statistical signiﬁcance.

While housing subsidies encourage the production of housing services,
increasing military expenditures decreases housing investments. This implies that it is
up to the government to consider military expenditures and the transfer of housing
investments because public polices affect housing production and consumption. Excess

military spending leaves little for the production of housing.

169

Table 58 Relationship between the Share of Housing Investment (CONNP) and
Military Spending Variables

 

 

_. no 1 no. 2 . E0 3
Fit-179.1711 1. .473 , (LCONNP)
Constant 5.146 0. 265 1.060
(8.209” (4.390)” (1.115)
.3 PERGNPo 0.003
**** (7.067)”
' PERGNPez 0.003
(4.537) 1’
RATDEF -0.056 0.002
(2.709)” (4.728)”
DENGNP 0.173
‘ (0.940)
f};  R? 0.162 0.807 0.624
F g 7.339 50.089 53.011
8153—; 1.686 0.105 1.105

 

 

'Note: t-statistics in parenthesis, (1) signiﬁcant at 0.001 level, (2) signiﬁcant at 0.01, and (3)
signiﬁcant at 0.1 level.

As shown Table 58, share of nation defense over total expenditure (RATDEF)
has more statistical meanings than defense spending as percentage of GNP (DENGNP).

The regression results support our assumption and proved to be statistical signiﬁcant.

4-1-4. Policy Impact Model

In order to ﬁnd out whether the political setting plays any roles in affecting
housing investment behavior, we evaluate the effects of direct policy variables. The
effects of direct government intervention, such as Two-Million Houses Program,
election ejfect, real estate regulation policy effect, and Park ’s regime, and political

stability dummy variables are analyzed.

170

| 13 ”if in E , Emil!

Table 59 Relationship between the Share of Housing Investment (CONNP) and

Policy Effect Variables

 

EQ.I

(LCONNP)

'EQ.2

EQ3

.fEQ-‘i

 

 

 

 

SEE

.....................................

 

0. 255
(7.734)”

-0.270

(2777)”

0.722
48.545
0.125

1.870

(8.539)”
0.009
(8.94)”

0.601
(1.838)”

0.765
61.783
0. 920

2 551 , . ..

(8.659)”

3.518
(5.418)”
0.936

(2.086)”

2.978
(3.028)”
1.196
(2.202)”
0.142
3.145
1.757

1.708
(8.305)”
0.009
(12.06)”

1.462
(2.408)”
1.018
(3.091)“
0.755
30. 796
0. 921

 

signiﬁcant at 0. l level.

As shown Table 59, domestic policies affect housing production in Korea. Two-
Million Houses Program is more sensitive than the general housing policy effect and
political stability variables. Election dummy variable appeared strong in housing

investment as well as Park’s regime, and real estate deregulation measures.

171

4-1-5. Financial Model

Table 60 represents that how important ﬁnance is to housing investment?
Interest rate, producer price index, consumer price index, and deﬂator have affect the level
of housing investment. For the more detailed analysis such as sensitivity analysis is not
included in this study, however, we try to ﬁnd whether housing ﬁnance increases housing

investment.

Table 60 Relationship between the Share of Housing Investment (CONNP) and
Finance Variables

 

0.955 1.280 0.307 1.659
(2.378)” (5. 249)” (0.435) (6.019)”
0.019 0. 004 0.016 0.012
(6.362)” (3.003)” (5.358)” (2.266)”

0.017 0.01

(0.241) (1.915)”

.. 0-018

- ' * s__.;; :-:‘:‘::.;;,; (1416):)

 : DOMSAV-:7?» 0.086

‘ ‘ (4.515)”

INTR 0.138

: : s .. ._ ‘ (2.621)”

f ' 81817211: - : 0.059
(2.164)”

‘ CPI -::«  0.009
‘ {4: it , 2:: :.ii (45?

R2551; 0.806 0.832 0.823 0.784

F 51.147 91. 367 55. 706 44.836

_:g.9§é,;?;.%§:;;;:;é_:::-,‘::'.::; 0 847 0. 775 0. 806 0. 893
Note: t-statistics in parenthesis, (1) signiﬁcant at 0. 001 level, (2) signiﬁcant at 0. 01, and (3)

signiﬁcant at 0.1 level.

 

 

 

 

 

172

4-1-6. Comprehensive Housing Investment Share Models

Table 61 Summary of Comprehensive Housing Investment Share Models

 

 

EQ. 1 BO. 2 EQ.3 EQ.4 EQ.5 EQ.6
Constant 1.388 3.62 4.758 1.475 2.058 0.438
(4.451)1 (6.688)‘ (2.549) (2.314) (0.863)
PERGNPO 0.018 0.018 0.015
(5.609)1 (4.888)l (3.131)2
PERGNP02 -0.016 0.016 0.012
(-2.726)3 (-2496) (—1.595)
RGNP (l/GNP) -2094 1.773
(0.866) (0.624)
DPOP 0.213 -0.018 0.157
(0 286) (0.039) (0 318)
DPOP2 0.191 0.005 0.007
(-1 023) (0.045) (0 059)
URB 0.009 —0.063 0.008 -0.006
(0.379) (-1.401) (0.342) (0.252)
URBZ 0 0.003 0 0
(0.232) (1.026) (0.214) (0.008)
DOMSAV 0.155
(8.636)1
RDOMSAV 1.403
(1.302)
R2 0.784 0.052 0.267 0.784 0.789 0.803
F 31.692 1.012 6.739 19.955 17.082 49.032
SEE 0.893 0.052 1.598 0.919 0.922 0.849

 

 

Notes: t-statistics in parenthesis, (1) significant at 0.001 level, (2) signiﬁcant at 0.01, and (3)
signiﬁcant at 0.1 level.

We focus on estimating elasticities of the economic variables believed to be the

most important determinants by theorists and previous empirical studies.

Table 62 Elasticities and Beta Coefﬁcients, Investment Analysis

 

173

R l . E E l D .
Elasticities are weights in non-dimensional units which measure the relative
responsiveness of the dependent variable to changes in values of the independent

variables. Point elastcities for the individual variables are calculated as follows. (Burns

and Grebler, 1976:42)

Y’=a+le+b2X2+....

the partial derivative of Y’ with respect to X is taken to yield

§§= b1 +2b2X

At the means of X and Y’, denoted X and Y, this becomes

M = (5010(1). + 21,, 2?)

6 (X / X)
Beta Coefﬁcient47
This coefﬁcient indicates better measure of the relative importance of several

independent variables.

 

‘7 Beta Coefficiem measure the statistical contribution of a variable to the explantory power to the equation.

174

4-2. Housing Construction Area Models

4-2-1. Stepwise Selection of Variables for Models

Table 63 The Housing Construction Area Model by Stepwise Selection

 

 

 

 

 

 

* Variable 11 c _ ' ' 2- STDError r - STD-coat; - ; ; Tolerance e g T I; Tm: g;
Constant 227.409 97.474 ' i 0.00“” - ' 2.333 if .038"
DEFPG w . -l88.744 32.757 0.214 0.083 -5.762 0.000
, HI-ISGRO 5.012 2.017 0.033 0.645 2.485 0.029
HPOL * 75.051 9.009 0.121 0.539 8.331 0.000
pm ‘ 029.751 44.009 0.092 0.117 -2.948 0.012
92116111302 8.272 1.388 1.037 0.004 5.960 0.000
names * 24.576 4.500 0.224 0.068 5.462 0.000
TWOMIL? 224.164 45.663 0.125 0.176 4.909 0.000
popUM‘; v 0.027 0.003 0.195 0.311 10.177 0.000
DP]? 34181.9 025.432 31.595 0.052 0.672 -3970 0.002
nPOBCRr , i 0.635 0.777 0.030 0.560 .2104 0.057
‘ 1101.0; 4  0.406 0.023 1.411 0.018 17.504 0.000
12mm 0.067 0.006 0.777 0.004 00.365 0.000
1] . l l l . EElllREI

The use of CONNP as the only indicator of housing investment is not enough to
substantiate the argument and check on the speciﬁcation of the independent variables.
Share of housing as a percent of per capita GNP does not capture the real total worth of
newly produced dwellings. Using stepwise selection, we can ﬁnd the empirical results
of Table 63. Military spending and political stability, housing loan has affected the

determination of housing investment. For detail of effect of the government allocation of

175

bank loans, the total volume of construction loans at constant prices show the statistical

signiﬁcance of housing investment.

Table 64 Basic Regression Equation: The Relation Between PHAREA and Other

 

 

Variables
,3: Egt 2 x ‘ ; - . 12131214 ’Egs L“? " Suﬁ-L T-Li BQ7 :1 *i if;
Constant 2.252 -24.168 1118.737 31.617 147.889 036.953
(0.042) (0.424) (4.048)1 (0.304) (0.929) (.2505)3
PERGNPO 2.527 2.509 2.607 2.065
(4.293)1 (4.251)1 (3.982)1 (2.39)3
PERGNP02 3.234 3.303 3.147 3.884
(2.959)2 (3.014)2 (2.73)3 (2.806)2
RGNP 417.248 1035.093
(l/GNP)
(0.964) (0.732)
DPOP 083.573 -91.746 019.367
(0.635) (0.1) (0.352)
DPOP2 44.2 22.67 25.134
(0.61) (1.06) (1.166)
URB 2.591 -22.708 4.19 4.618
(0.628) (0.329) (0.949) (1.039)
URB2 0.171 1.428 0.16 0.093
(0.601) (1.191) (0.552) (0.314)
R2 0.948 0.951 0.052 0.246 0.953 0.954 0.664
F 338.123 169.11 1.01 6.029 110.444 94.597 23.08
SEE 162.822 163.029 696.12 620.833 164.866 165.042 422.182

 

 

Notes: t-statistics in parenthesis, (1) signiﬁcant at 0.001 level, (2) signiﬁcant at 0.01 , and (3)
signiﬁcant at 0.1 level.

176

Table 65 Housing Construction Model

 

., ~r "s-L‘siQt};.L”);L:§:i::.-gz:3; Invcsﬂnem
.5,; 171115121 ”m3

PERGNPO
PERGNP02
TRBALO
HRTE
DEFLAI
RATDEF
ELECTION
TWOMIL
R2

F
SEE

 

 

1760.82" ‘

(2.103)”
1.488
(1.41)
4.79
(3.103)”
0.115
(2.197)”
-22757
(0.312)”

0.958
165.695
155.253

IAMwnh

2323527 *’

(2.679)”
2. 134
(2.011)”
5.057
(3.280)”
0.114
(0.258)”
-27.696
(2755)”
4.939
(0.785)3
-5074
(0.642)

0.962
100.314
156.28

IAMmth 118er111

P0110?

(2.516)”
2.663
(2.639)”
4.686
(3.26)”
0. 124
(2.637)”
24.092
(0.552)”
-5.865
(.2261)3
-5.401
(0.884) 3
043.796
(-2.336) 3

0.970
120.725
138.355

20437290

foreign

1193.258
(2.013)”
4.089
(6.9329)”

0.151
(-3.614) 1’
05.621
(-2.165) ”
-3.268
(0.368)
4.465
(0.656)
-94.536
(0.584)
486.881
(4.028) ”
0.974
139.746
0128.857

 

Notes: t-statistics in parenthesis, (1) signiﬁcant at 0.001 level, (2) signiﬁcant at 0.01, and (3)

signiﬁcant at 0.1 level.

Trade balance has explained the determination of housing investment. Political
effects such as election, and two million houses programs show the statistical

signiﬁcance. Other institutional settings like military spending also exert a serious

impact on PHAREA.

177

4-3. Housing Investment Household Model

4-3-1. Stepwise Selection of Variables for Models

 

 

 

 

. '0-165
EXCHANGE 0.011
GRMS 0-269
""" 04.110
0.501
18.645
~5.498
0.021
0.035
0.031

 

 

0.156
0.021
0.129
8.620
0.082
6.236
3.415
0.003
0.012
0.042

  

0.122
0.000
0.007
0.476

 

The set of equations, presented in Table 67, conﬁrms the theoretical model and
the results previously presented. Equation 1 shows that per capita income is a decisive

element in the households’ housing investment as presented by the t-ststistic.

178

Table 67 Basic Regression Equation: The Relation Between HIHSH and Other
Variables

 

‘ 1302203304 Hos ; 206507
CW '353-

1 468.546 8604.287 16025.99 10.598 971.353 -

7 10703.953

. 51? (0.411) (0.499) (3.92)1 (3.628)1 (0.006) (0.36) (0.343)
30.952 28.097 27.77 23.287

, ‘ j}: (3.377)2 (2.89)2 (2.533)2 (1.592)

'j-PERGNP‘OZ 66.116 74.554 74.706 80.796

i j : j: (4.064)1 (4.129)1 (3.869)1 (3.447)2

RGNP ~ - 22391.11

(l/GNP) 3447.722 2

< (0.471) (0.823)

458.384 086.612 0747.37

 

DPOP???“ , -
. ; 2259.369
, r ' ' (0.489) (0.328) (0.459) (0.406)
-DPOP2 ; 032.042 87.309 107.746 339.262
‘- > , (0.719) (0.244) (0.295) (0.334)
URB ’ '- 65.121 -338.504 72.683 76.221 121.187
' (0.958) (0.242) (0.983) (1.013) (0.583)

012132 . 0.068 18.816 0.166 0.712
‘ 2  4: (0.227) (0.984) (0.241) (0.34)

_L?-DOMSAV: 933.396

i: I. ;, j, (4.488)1

" RDOMSAV 14869.19
' “““ (1.54)

R2 .’ i *  : 0.952 0.947 0.042 0.235 0.947 0.948 0.628
F , . 378.072 156.222 0.82 5.678 98.606 82.557 7.474
send , 3; 2655.11 2686.182 11100.65 9922.967 2760.941 2794.098 7504.677

#Notes5t-statistics in parenthesis, (1) signiﬁcant at 0.001 level, (2) signiﬁcant at 0.01, and (3)
signiﬁcant at 0.1 level.

 

179

Table 68 Housing Investment Household Model

 

 

‘ .lnvesmlént .
‘ . 5116mm , 6184-
: j} i. (HIHSH) ' '
110.513
(1.917)”
~ 0.501

 

(3.452) 1’
0.193
(0.283)

‘ 0.018
f (0.219) ”
0.182
(2.287)”
0.254

(1.662)
-1.244
(-1.678)

‘ 0.977
A. . 1‘ 187.196
7.221

IHM With 3
110399

effect ‘
100.058
(1.732)”
0.522
(3.583) 1’
0.299
(0.808) ”
0.018
(0.259) 1’
0.088
(2.099)”

-l.l61
(4.582)

16.155
(1.781)”
0.978
189.93
7.17

* 1”,“ With».
'insﬁmuonai *

91.508
(1.591)
0.522
(3.626) 1’
0.285

(0 .734) ”
0.018
(0.395) 1’
-0.983
(0.893)”

0.148
(0.581)
4.09
(0.263)
13.065
(1.406)’
0.979
166.746
7.089

 

Foreign affairs, domestic politics, and institutional settings affect housing
production in Korea as shown Table 68. Trade Balance, and per capita GNP, Two
Million Houses Program exert extensive inﬂuence on housing production. However,
changes in population do not have a signiﬁcant effect on HIHSH. Urbanization does

affect HIHSH.

CHAPTER SEVEN

CONCLUSION AND FUTURE DIRECTIONS

1 . Introduction

This concluding chapter discusses the overall result of the study. Throughout
this study, we have found some limitations of the explanatory models. On the other
hand, we suggest some recommendations for future studies for the behavior of housing
investment. The policy recommendations, and the conclusion of this study are

addressed in this ﬁnal chapter.

2. Summary of ﬁndings

Korea’s macroeconomic plans are linked to housing conditions and development
that has been organized through a series of ﬁve-year plans. Central organizing
structures of development have had a strong impact on private sector investment in
housing through strong linkages with the macroeconomy. These circumstances indicate
a signiﬁcant impact on the general components of the Korean housing market.

A general review of housing theory discloses that more and better housing was
used as a solution to societal stress, and that state intervention was justiﬁed in the
provision of housing. While existing models for housing investment related to overall

allocation of resources in a country, they failed to consider adequately macroeconomic

180

181

factors. To resolve that omission, a new model is proposed. The Burns and Grebler
model serves as the base of departure because it has conﬁrmed the effects of the
socioeconomic factors and urbanization. With the addition of new macroeconomic and
international factors we contribute to analysis of housing in developing countries.

However, in our review of previous studies of housing investment, we point out
that they suffered from several major shortcomings. (1) They focus only on socio-
economic variables, not policy effect. (2) They focus only on domestic affairs without
considering factors related to the global setting and foreign affairs. (3) They do not
consider national allocation effects within budget constraints. (4) They do not factor in
the effects of institutional settings and arrangements.

We have overcome these weaknesses by creating a new approach based on the
data from Korea from 1953 to 1993. The time series data demonstrate historical
sequences in which macroeconomic issues are apparently signiﬁcant. Our model then
includes both the housing data and macro-economic factors. This comprehensive
approach determines statistically what is otherwise only an apparent relationship
between macroeconomic factors and housing related policy experiments. For example,
the Two-Million Houses Program and several anti-real estate regulation enactments
have been tested. The result is that state policy can no longer ignore the housing sector
or treat it as a separate domain of policy when calculating the effect of macroeconomic
issues on general budgetary allocations.

Our modeling of the housing investment in Korea has been guided by what we

expect to be the most important determinants of residential investment. By ﬁrst

182

deﬁning a broad spectrum of determinants in six categories of factors, we allow for a
generous and broadly deﬁned test. By testing for signiﬁcance and variance the
determents are narrowed to about forty-seven variables divided into the six categories,
which are Socio-Economic Status, Demographic and Urbanization, Policy Effect,
Housing Condition, Institutional Setting, and Foreign Affairs. Of these exchange rates,
balance of payment, trade balance, and interest rates are considered macroeconomic,
which in ﬁnal analysis allow us to concentrate on how the macroeconomic sector
impinges upon the dynamics of the housing investment; and the inﬂuence of various
aspects of global and foreign affairs and policy. In addition, the domestic factors allow
us to correlate these macro-economic issues with the structure of institutional
arrangement and how political factors affect housing investment.

The empirical results of the new model reveal some important aspects of
housing investment in Korea. Several interesting observations emerged from the study.
Global and foreign factors are extremely inﬂuential determinants of the general
economy of Korea and the housing investment within that context.

Analysis reveals a statistically signiﬁcant relationship between historical
patterns of investment in the housing sector and the structure of Korea’s total
investment. Based on our theoretical framework, factor analysis, Pearson Correlation
Analysis, and stepwise regression analysis are used to select variables. Regression
models using time-series data are estimated to establish the empirical relationships.
Results show that housing investment in Korea is statistically explained by levels of

income, military spending, political stability, housing policy, foreign affairs, and

183

global context of ﬁnance. Demographic pressure did not show any signiﬁcant
relationship on the level of housing investment. Urbanization, as measured by the ratio
of urban population over total population, does correlate with levels of housing
investment. These results are made more meaningful by adding international,
domestic, and institutional variables to the context in which housing policy was
formulated.

This dissertation will help to clarify factors that inﬂuence the level of housing
investment in Korea. It also helps to emphasize the relationship between housing and
macro-economic factors, thus elevating housing to higher level of policy consideration.
It provides housing researchers with new insights into the nature of decision making as
affected by macroeconomic and political contexts, while including a social-historical
context. The conceptual and empirical models, though based on Korea’s example,
could be used to test for generalization for other developing countries, where housing is

used as a tool for economic development.

3. Limitations of the Study

Because the socio-economic and historical context of Korea includes prominent
government intervention to alleviate serious housing problems, this research has strong
implications concerning policy for government intervention. Yet in terms of the

explanation of housing investment, there are several major limitations in our models.

184

The ﬁrst limitation is that this study relies on the validity and accuracy of the
two data sources, ofﬁcial housing production statistics and census data published by the
government. With respect to reliance on government data, there is a question of its
dependability. Because the research of almost all of our sources is built on data which
are always questionable, we keep in mind that our sources are valid only insofar as
these data are valid.

The second limitation of our models is that several variables, including
households (NTHH), dwelling size (NSIZE), total revenue (TR), and total expenditures
(TE) cannot be tested successfully in our models because of multicollinearity problems.
It might be caused by the high correlation between the share of housing investment
over the GNP and these variables. It might be possible to develop a more sophisticated
model for explaining the relationship between variables and the behavior of housing
investment.

Finally, a general limitation is an inadequate description of the private housing
sector, which can hinder the full understanding of housing policy formulation; for
example, who dominates the private housing sector, Chaebol, investment companies, or
corporations? This information of inside structure of the construction industry will
enhance the reliance of analysis. In addition, the cycle of the construction business
assumes the strong correlation relationships with the general housing investment.
These issues affect the impact of government regulation on the housing construction
sector and the general housing investment, which present an area for future research

and study.

185

4. Recommendations for Further Research

The limitations of the study discussed above should be taken into consideration
for further research in this subject.

First, the relationship between housing stock and housing investment in Korea
can be tested further. One might analyze more intensively whether the maturity of the
housing stock condition (representing over 1 housing supply rate) is the major
determinant of the stabilized behavior of housing investment. We may further
investigate this preliminary ﬁnding by a cross-national study which includes the
conditions of maturity and non-maturity of housing stock.

Second, further research will ﬁnd new approaches to deal with several
multicollinearity problems. One method is to select new indicators of economic
development, urbanization, and p0pulation growth. Another is to ﬁnd different
analytical techniques. In addition, an adaptation of the model tested here is needed to
be able to apply this Korean case to other developing countries. The adaptation has to
deal with cross-section pooled time series analysis.

Third, if further study focuses on how government can intervene to alleviate
serious housing problems facing developing countries, one must overcome the
limitations of dummy variables that evaluate the effect of structural changes. One must
maintain the assumption of homoscedasticity. For example, one might consider a

hazard model and path analysis as new alternative methodologies. A hazard model can

186

examine each phase of capital formation and each ruling regime separately. A path
analytic technique permits the testing of direct and indirect casual effects in the

determination of a dependent variable.

5. The Application of the Theoretical Basis to The Key Housing Problems

Throughout this study, we assess the implications for public policy. Generally
speaking, there are two components of concepts of housing. First, housing is used as a
solution to social problems and, second, housing is considered as only one of several
government economic policies. In the ﬁrst case, Strassman indicates that changes in
certain government policies may lead to increases or decreases in the production of
dwelling units. Government involvement through rent control, for example,
discourages development and diverts funds from the housing market.

One of the important tests of a housing system is the equity in the distribution of
housing welfare. Housing problems in Korea may be identiﬁed in terms of the
following variables: urbanization, formation of nuclear households, income
distribution, dwelling price increases, land development policy, and ﬁnancial
repression.

With respect to housing demand, some of the measures could be proposed for
the expansion of supply such as the development of housing ﬁnance, low cost of land
and dwelling, and a cut in the size of dwelling units will surely increase the housing
demand across the board. The rationale for government intervention in housing has

been justiﬁed, and the role of government is important in dealing with relationships

187

between economic and political structures and the established housing system. The
Korean government should keep this view in mind: government has role in housing
because of its social responsibility and housing should not be left entirely to free

market forces.

6. Policy Reorientation and Policy Recommendation

Korea's housing problems manifest themselves in terms of absolute housing
shortages, relatively poor quality, over crowding, lessening affordability, and skewed
distribution of housing welfare. The most obvious reason for these housing problems
is the rapid increase in housing demand caused by urbanization, accelerated formation
of nuclear households, and past demolition of a great number of dwellings. Korea's
housing shortage has been due essentially to three variables: skewed income
distribution, skewed wealth distribution, and rapidly rising dwelling prices.

The results indicate that government intervention has a signiﬁcant impact on
housing, yet the housing sector is also coupled with macro-economic factors that are
outside the view normally taken by domestic planners. Therefore, Korea and other
countries with an apparent need for signiﬁcant government involvement should re-think
housing investment problems and type of government involvement, based simply on
uncertainty of macro-economic affects as well as on domestic issues, such as slum
clearance, land pricing systems, and public housing that have been shown to distort the

market.

188

The whole question of whether housing is an infrastructure like water supply,
sewage and road facilities should be examined, especially with respect to those
individuals such as the elderly and handicapped, who have limited market relations
because of ﬁxed income and/or state dependency. Safer government involvement
would address market imperfections which are in the context of a larger global market,
therefore respecting the increasing alignment between domestic and international
factors. Deregulation and incentives, for example, can promote development of an
efﬁcient private sector within a domestic environment, while accommodating exchange
and interest rates which are internationally determined. Korea's new housing policy
should be formulated in light of these principles. Finally, for any housing policy to

succeed, the public should be invited to participate in the planning process.

BIBLIOGRAPHY

189

Abrams, Charles. Man ’s Struggle for Shelter in an Urbanizing World, Cambridge,
Massachusetts: MIT Press, 1964.

Abrams, Charles. “Squatting and Squatters.” In Ihird Word Urbanization, edited by
Jane Abu-Lughod and Richard Hay, Jr., 293-299. Chicago: Maaroufa Press. 1977.

Achen, Christopher H. Interpreting and Using Regression. Beverly Hills: Sage
Publications, 1982.

Adelman, Irma and Robinson, Sherman. A Non-linear, Dynamic, Micro-economic
model of Korea: Factors Affecting the Distribution of Income in the Short Run.
Research Program in Economic Development, Woodrow Wilson School, Princeton
University. Discussion Paper No. 36. 1973.

Alonso, William. “Five Bell Shapes in Development.” Papers of the Regional Science
Association, 45:(1980).

Annez, Philippe and Wheaton, William C. “Economic Development and Housing
Sector: A Cross-National Model.” Economic Development and Cultural Change,
32:(4):749-766(l984).

Annez, Philippe. Housing Investment: A Cross Country Comparison. Ph. D.
Dissertation, Massachusetts Institute of Technology, 1981.

Apgar, William C., Masnick George S., and McArdle, Nancy. Housing in America:
1970-2000: The Nation ’s Housing Needs for the Balance of the 20th Century, Joint
Center for Housing Studies, Harvard University, 1991.

Baer, William C. “Filtering and Third World Housing Policy.” Third World Planning
Review. 13(1):69-82(1991).

Baer, W. C. and Koo. C. “Housing Turnover in Developing Nations: The Case of
Seoul.” Journal of Planning Education and Research, 13:(1994).

Bahl, Roy W., and Linn, Johannes, F ., Urban Public Finance in Developing
Countries, New York: Oxford University Press for the World Bank, 1992.

Balassa, Bela. “The Lessons of East Asian Development: An Overview.” Economic
Development and Cultural Change, 36(3), (April 1988).

Ball, Michael. Housing Policy and Economic Power, New York: Methen Co., 1983.

190

Barlow, James and Duncan, Simon. Success and Failure in Housing Provision:
European Systems Compared. Pergamon, Elsevier Science Ltd. 1994

Beyer, Glenn H. Housing and Society, New York: The Macmillan Company, 1965.

Beyer, Glenn H. Housing: A Factual Analysis, New York: The Macmillan Company,
1958.

Black, John and Stanford, David. Housing Policy and Finance, New York: Routledge,
1988.

Blalock, Jr. Hubert M. Social Statistics: Second Edition. New York: McGraw—I-lill
Book Company, 1972

Bourne, L. S. The Geography of Housing, London, Edward Arnold, 1981.

Bratt, Rachel G. et a1. Critical Perspectives on Housing, Philadelphia: Temple
University Press, 1986.

Brown, Gilber T. Korean Pricing Policies and Economic Development in the I960s.
Baltimore, Johns Hopkins University Press, 1973.

Browning, E. , and Browning, J. Public Finance and the Price System, N.Y.:
Macmillan, 1979.

Buckley, Robert M. Housing Finance in Developing Countries: A Transaction Cost
Approach World Bank, Policy, Planning and Research, Working Paper no. 347,
Washington D.C.: World Bank, December 1989.

Buckley, Robert, and Stephen Mayo. “Housing Policy in Developing Countries:
Evaluating the Macroeconomics Impacts.“ Review of Urban and Regional
Development Studies 2:27-47(1989)

Buckley, Robert M. and Madhusudhan, Ranjana G. The Macroeconomics of Housing 's
Role in the Economy: An International Analysis, Metropolitan Studies Program:
The Maxwell School of Citizenship and Public Affairs, Syracuse University
(Occasional Paper No. 78), 1984.

Buckley, Robert. The Measurement, Control and Targeting of Housing Finance
Subsidies, Washington D. C.: World Bank Report, June 1988.

Bulos, Marjorie and Teymur, Necdet. Housing: Design, Research, Education.
Brookﬁeld: Avebury, 1993.

‘ 191

Burns, Leland S. and Grebler, Leo. "Resource Allocation to Housing Investment: A
Comparative International Study. " Economic Development and Cultural Change,
25(1):95-121(October 1976).

Burns, Leland S. and Grebler, Leo. The Future of Housing Markets, New York, N.
Y.: Plenum Press, 1986.

Burns, Leland S., and Grebler, Leo. The Housing of Nations: Analysis and Policy in a
Comparative Framework, New York: John Wiley & Sons, 1977.

Casetells, Manuel, Lee Goh, and Reginald Y.W. Kwok with Toh Lap Kee. Economic
Development and Housing Policy in the Asian Paciﬁc Rim: A Comparative Study of
Hong Kong, Singapore, and Shenzhen Special Economic Zone. Monograpy no. 37.
Berkeley: Institute of Urban and Regional Development, June 1988.

Chan, Steve. “Catching Up and Keeping Up: Explaining Capitalist East Asia’s
Industrial Competitiveness.” The Journal of East Asian Aﬁairs 79-103, Winter
1991

Chang, Chin-Oh, Explanation and Forecasting of National Housing Investment: A
Comparative Study in Taiwan (ROC), Korea (South), Japan, and the USA.
University of Pennsylvania. Ph. D. Dissertation. 1986.

Chang, C. and Linneman.P. “Forecasting Housing Investment in Developing
Countries,” Growth and Change, 59-71, Winter 1990.

Chant, Sylvia, and Peter Ward.” Family Structure and Low-Income Housing Policy.”
Third World Planning Review 9(1):5-19(1987).

Child, Dennis. The Essentials of Factor Analysis Second Edition, Cassell, 1990.

Christian, James. Housing Finance for Developing Countries. International Union of
Building Societies and Savings Associations, 1980.

Chung, Joseph Hee-Soo. ”Housing Policies in Korea: Search for New Orientation. " In
Dynamic Transformation: Korea, NICs and Beyond edited by Lim, G. C., and
Chang, Wook, Seoul: Myung-Bo Publishing, Co., 1990.

Chung, Joseph Hee-Soo, eds. A Study on Housing Problems and Policy Developments
in Korea, Seoul: KRIHS, 1985.

Clayton Keith N. An Introduction to Statistics for Psychology and Education,
Columbus, Ohio: A Bell & Howell Company, 1984.

192

Cole, David C. and Park, Yung Chul. Financial Development in Korea, I 945-1 978.
Cambridge: MA, Harvard University Press, 1983.

Daniel, Wayne and Terrell, James. Business Statistics: Basic Concepts and
Methodology: Third Edition, Boston: Houghton Mifﬂin Company, 1983.

Donnison, D. and Ungerson, C. Housing Policy, Harmondsworth, Penguin, 1982.
Donnison, D. V. The Government of Housing, Baltimore, Penguin, 1967.

Dunteman, George H. Principal Components Analysis. Newbury Park: Sage
Publication. 1989.

Drakakis-Smith, D. W. Urbanization, Housing and the Development Process. London:
Croom Helm, 1981.

Dwyer, D. J. People and Housing in Third World Cities: Perspectives on the Problem
of Spontaneous Settlements, New York: Longman, 1979.

Economic Planning Board, the Republic of Korea. Korea ’s ﬁve-year Plan for the New
Economy, July, 1993.

Edgan, John J. et a1. Housing and Public Policy: A Role for Mediating Structures,
Cambridge, Massachusetts: Ballinger Publishing Company, 1981.

El-Kholei, Ahmed Osman. The Role of Government in Housing in Developing
Countries: The Case of Egypt. University of Illinois at Urbana-Champaign. Ph. D.
Dissertation. 1992.

El Sadek, Samir. Comprehensive Planning for Housing in Developing Countries, Ann
Arbor: University of Michigan, Monograph, 1980.

Ellickson, Bryan. Economic Analysis of Urban Housing Markets: A New Approach,
Santa Monica, CA: Rand, 1977.

Fallis, George. Housing Economics. Toronto: Butterworths, 1988.

Florida, Richard L.(eds.) Housing and the New Financial Markets, New Brunswick,
New Jersey: The Center for Urban Policy Research of Rutgers University, 1986.

Follain, J ., and Jimenez, E. “Estimating The Demand for Housing Characteristics: A
Survey and Critique.” Regional Science and Urban Economics, North-Holland
15:77-107(l985).

193

Follain, J ., and Jimenez, E. “The Demand for Housing Characteristics in Developing
Countries.” Urban Studies, 22(56):421-432(1985).

Follain, J ., Lim, G. C. , and Renaud, B., "Demand for Housing in Developing
Countries," Journal of Urban Economic, 7 :315-36(1985).

Fonseca, Jose, eds. Housing Finance in Developing Countries, Paper Presented to the
30th World Congress of IF HP, May 1970.

Foote, Nelson N. et a]. Housing Choices and Housing Constraints, New York:
McGraw-Hill Book, 1960.

Forrest, Ray and Murie, Alan. Selling the Welfare State: The Privatization of Public
Housing, New York: Routledge, 1988.

Friedman, Joseph, Emmanuel Jimenes, and Stephen K. Mayo. “The Demand for
Tenure Security in Developing Countries.” Journal of Development Economics
29: 185-198(1988).

Friedman, J. and Weinberg, D. “Housing Consumption under a Constrained Income
Transfer,” Journal of Urban Economics,11, 1982.

Friedrichs, Jurgen. Aﬁordable Housing and the Homeless, Berlin, Germany: de
Gruyter, 1988.

Government of the Republic of Korea. The 6th ﬁve-year Economic Social Development
Plan, 1988.

Grebler, Leo and Mittlebach, Frank G. The Inﬂation of House Prices: Its Extent,
Causes, and Consequences, Lexington, Massachusetts, Lexington Book, 1977.

Grigsby, William G. and Rosenburg, Louis. Urban Housing Policy, New Brunswick,
New Jersey: The Center for Urban Policy Research of Rutgers University, 1975.

Grigsby, William, Baratz, Morton and Maclennan, Duncan. Shelter Subsidies for Low
Income Households, Research Report Series, Department of City and Regional
Planning, University of Pennsylvania, 1983.

Grimes, Orville F. Housing for Low-Income Urban Families: Economics and Policy in
the Developing Countries. World Bank Research Publication, Baltimore: The Johns
Hopkins University Press, 1976.

194

Ha, Seong-Kyu and Merrett, S. “Assessing Housing Consumption Requirements, The
Case of Seoul.” The Third World Planning Review, 6(4): 1984.

Ha, Seong—Kyu. "Urban Redevelopment, and Social Inequality in Korea.” In Dynamic
Transformation: Korea, NICs and Beyond edited by Lim, G. C., and Chang,
Wook, Seoul: Myung-Bo Publishing, Co., 1990.

Ha, Seong-Kyu, eds. Housing Policy and Practice in Asia, NY : Croom Helm, 1987.
Harman, H. Harry. Modern Factor Analysis. The University of Chicago Press. 1960.

Harsman, Bjorn, and John M. Quigley, eds. Housing Markets and Housing
Institutions: An International Comparison. Boston: Kluwer Academic Publishers,
1991.

Hegedus, J. “Reconsidering the Roles of the State and the Market in Socialist Housing
Systems.” International Journal of Urban and Regional Research 11( 1): 79-
97(March 1987).

Hartman, Chester W. Housing and Social Policy, Englewood Cliffs, New Jersey:
Prentice-Hall, inc., 1975 .

Head, John. G. Public Goods and Public Welfare. Durham, N. C. : Duck University
Press, 1974.

Headey, B. Housing Policy and the Developed Economy, London, Croom Helm, 1977.

Heilbrun, James, Urban Economics and Public Policy, Second Edition, NY: St.
Martin's Press, 1981.

Hirshman, A. D. The Strategy of Economic Development, Yale University Press, 1955.

Ho, Lee-Jay and Kim, Yon Hyung (eds.). Economic Development in the Republic of
Korea:A Policy Perspective, Honolulu, Hawaii: University of Hawaii Press, 1991.

Hoffman, Michael, Barbara Haupt, and Raymond J. Struyk. International Housing
Markets: What We Know; What We Need To Know. Prepared for Fannie Mae,
Ofﬁce of Housing Policy Research, Draft. Washington, D. C.: Urban Institute,
August 1991.

Howenstine, E. Jay. Attacking Housing Costs: Foreign Policies and Strategies, New
Brunswick, New Jersey: The Center for Urban Policy Research of Rutgers
University, 1983.

195

Howenstine, E. Jay. “Appraising the Role of Housing in Economic Development.”
International Labor Review, 75:26 (January 1957).

Hughes, J. W. and Sternlieb, George. The Dynamics of America ’s Housing, New
Brunswick, New Jersey: The Center for Urban Policy Research of Rutgers
University, 1991.

Hula, Richard, C. Market ~Based Public Policy. NY. :St. Martin’s Press, 1988.

Jimenez, Emmanuel “Tenure Security and Urban Squatting. ” Review of Economics and
Statistics 66(4):556-67(1984).

J in, Young Hwan. An Analytical Model of Residential Location: With Application T o
The Housing Market In Seoul, Korea. Graduate Field of City and Regional
Planning, Cornell University, 1985.

Jones, Gareth and Peter M. Ward. (eds.) Methodology for Land and Housing Market
Analysis edited by Gareth Jones and Peter M. Ward. Cambridge, MA: Lincoln
Institute of Land Policy, 1994.

Kaplan, Abraham. The Conduct of Inquiry: Methodology for Behavioral Science.
Scranton: Pennsylvania, Chandler Publishing Company. 1964.

Keare, Douglas H. , and Parris, Scott. Evaluation of Shelter Programs for the Urban
Poor: Principal Findings, Washington DC: The World Bank, 1982.

Kihl, Young Whan. “Sourth Korea in 1989: Slow Progress Toward Democracy.”
Asian Survey. 8(1):67-73(1990).

Kim, Eun Mee. “The Industrial Organization and Growth of the Korean Chaebol:
Integrating Development and Organizational Theories” Gary Hamliton (ed.),
Business Networks and Economic Development in East and Southeast Asia. Hong
Kong: Hong Kong University Press. April 1990.

Kim, Eun Mee. “Chaebol: The Victor and Villain of Korean Development. ” Joint
Korea- U. S. Academic Symposium. Impact of Recent Economic Developments on
U. S.-Korean Relations & Paciﬁc Basin. (1: 134-150) 1991.

Kim, Eun Mee. “Contradictions and Limits of A Developmental State: With
Illustrations From the South Korean Case.” Unpublished Paper. Department of
Sociology, University of Southern California. January 1992.

 

196

Kim, Hyun-Sik. Housing Tenure and Residential Mobility: Multi-Choice Logit
Approach. Ph. D dissertation. School of Architecture and Urban Planning,
Princeton University, 1981.

Kim, Jeong-Ho “Residential Mobility: Issues and Policy Implications-- A Case in
Seoul, Korea.” Paper presented at the International Seminar on Metropolitan
Management, Seoul City University, October 1987.

Kim, Jeong-Ho. “The Study of Housing Demand and Rental Housing.” In The
Direction of Korean Housing Policy, Korea National Housing Corporation, 1983.

Kim, Jeong-Ho. ”Changing Perspectives for Korea Housing Policy." In Dynamic
Transformation of Societies edited by Lim, G. C, and Lee, M. H. Seoul: Nanahm
Publishing House, 1993.

Kim, Jeong-Ho. Residential Mobility and Consumption Adjustment: Case in Seoul,
Unpublished Ph. D. Dissertation, The University of Michigan, 1987.

Kim, Joon 800. “The Analysis of Housing Supply and Demand Equilibrium and its
Policy Implication.” Housing Finance Bimonthly Review, The Korea Housing
Bank, 17(4), August 1984.

Kim, Joon Soo. “Time-Series Analysis of Determinants of Housing Supply and
Demand in Korea.” The Study of Korea Development, Korea Development
Institute, 5(4), 1983.

Kim, Kyung-Hwan. An Analysis of Ianﬁciency of An Urban Housing Market: The Case
of Seoul, Korea. Department of Economics, Princeton University, Ph. D.
Dissertation. 1987.

Kim, Kyung-Hwan. Housing Prices, Aﬁ'ordability, and Government Policy in Korea.
World Bank, Internal Discussion Paper, Report No. IDP 99, July 1991.

Kim, Kwan-Young. “Determinants of Housing Supply-Demand and Housing Finance
Sector in Korea.” Housing Finance Bimonthly Review, The Korea Housing Bank,
21(2), April 1988.

Knight, Malcom, Loayza, Norman, and Villanueva, Delano. The Peace Dividend:
Military Spending Cuts and Economic Growth. International Monetary Fund, IMF
Working Paper, 1995.

Korea National Housing Corporation. Housing Policy Development Research. August
1983.

197

Kuzenet, Paul, W. “An East Asian Model of Economic Development: Japan, Taiwan,
and South Korea.” Economic Development and Cultural Change, 36(3), April
1988.

Kuznets, Paul W. Economic Growth and Structure in the Republic of Korea, New
Haven: Yale University Press, 1977.

Kuzenets, 8. ”Economic Growth and Income in Equality. " American Economic Review,
45, 1965.

Kuzenets, 8., “Quantitative Aspects of The Economic Growth of Nations, Part V.

Long Term Trends in Capital Formation Proportions.” Economic Development and
Cultural Change, Part II, July 1961.

Kwon, Joo An. Economic Analysis of the Greenbelt Amenity in Seoul, Korea.
Department of Economics, Michigan State University, Ph. D. Dissertation. 1995.

Lansley, S. Housing and Public Policy, London: CroomHelm, 1979.

Lamb, Geoffrey and Valexiana Kallab. Military Expenditure and Economic
Development. A Symposium on Research Issues, 185. World Bank Discussion
Papers. World Bank, Washington D. C. 1995.

Laquian, Aprodicio A. Basic Housing: Policies for Urban Sites, Services, Shelter in
Developing Countries, Ottawa, Canada: International Development Research
Centre, 1983.

Lawley, D. N. and Maxwell, A. E. Factor Analysis As a Statistical Method. London:
Butterworths. 1963.

Lee, K., and Sohn, K. A Study on the Optimum Allocation of the Housing Finance,
Korea Research Institute for Human Settlement, February, 1989.

Leeuw, Frank de and Struyk, Raymond J. The Web of Urban Housing: Analyzing
Policy with a Market Simulation Model, Washington D. C.: Urban Institute Press,
1975.

Leeuw, Frank de. “The Demand For Housing: A Review of Cross-Section Evidence.”
The Review of Economics and Statistics, February, 1971.

Lewis-Beck, Michael 8. Applied Regression: An Introduction, Beverly Hills: Sage
Publications, 1984.

198

Lim, Gill-Chin. "Housing Policies for the Urban Poor in Developing Countries,"
Journal of American Planning Association, 53(2): 176-185(Spring 1987).

Lim, Gill-Chin. Political Economy of Housing In Soviet Union: Comparative Synthetic
Approach, Michigan State University, October, 1991.

Lim, Gill-Chin, and Lee, Man-Hyung. “Housing Consumption in Urban China.”
Journal of Real Estate Finance and Economics, 6: 89-102, 1993.

Lim, Gill-Chin and Lee, Man-Hyung. "Political Ideology and Housing Policy in
Modern China.” Journal of Environment and Planning C, 8, 1990.

Lim, Gill-Chin. and James Follain. “Determinants of Home Ownership in a
Developing Economy: The Case of Korea.” Urban Studies 17: 13-23(1980).

Lipton, Michael, Why Poor People Stay Poor: Urban Bias in World Development ? ,
Cambridge, Massachusetts: Harvard University Press, 1980.

Lowry, Ira S. “Filtering and Housing Standards: A Conceptual Approach.” Land
Economics. 36(4):362-370(November 1960).

Machlup, Fritz. Methodology of Economics and Other Social Sciences, New York:
Academic Press, 1978.

Maclennan, Duncan. Housing Economics London: Longman, 1982.
Malpass, P, and Murie. A, Housing Policy and Practice, London, Macmillan, 1982.

Malpezzi, Stephen and Gwendolyn Ball. Rent Control in Developing Countries. World
Bank Discussion Papers No: 129. Washington, D. C.: The World Bank, 1991.

Malpezzi, Stephen. Urban Housing and Financial Markets: Some International
Comparisons.” Urban Studies. 27(6):971-1022(1990).

Malpezzi, Stephen, and Stephen K. Mayo with David J. Gross. Housing Demand in
Developing Countries. World Bank Staff Working Paper no. 733. Washington, D.C.:
World Bank, 1985.

Malpezzi, Stephen, and Stephen K. Mayo. “User Cost and Housing Tenure in Developing
Countries.” Journal of Development Economics 25: 197—220(1987).

Malpezzi, Stephen, and Stephen K. Mayo. “The Demand for Housing in Developing
Countries: Empirical Estimates from Household Data.” Economic Development and
Cultural Change 35(4):687-721(July 1987).

199

Mayo, Stephen, and Stephen Sheppard. “Housing Supply and the Effects of Stochastic
Development Control.” Urban Development Division. Washington, D.C.: The
World Bank, January 1990.

Mayo, Stephen K. “Household Preferences and Expenditure.” In Shelter, Settlement &
Development, edited by Lloyd Rodwin, 60-72. Boston: Allen & Unwin, 1987.

Mayo, Stephen K., Stephen Malpezzi, and David J. Gross. “Shelter Strategies for the
Urban Poor in Developing Countries.” Research Observer 1(2): 183-203(July 1986).

McCallum, Douglas, and Stan Benjamin. “Low-Income Urban Housing in the Third
World: Broadening the Economic Perspective.” Urban Studies 22:277-287(1985).

Marcussen, Lars, Third World Housing in Social and Spatial Development: The Case
of Jakarta, Sydney: Avebury, 1990.

Marris, Peter, The Meaning of Slums and Patter of Change, Monograph, School of
Architecture and Urban Planning, University of California, Los Angels.

Mason, Edward S. et al. The Economic and Social Modernization of the Republic of
Korea: Studies in the Republic of Korea: I 945-1 975 , Cambridge, Massachusetts:
Harvard University Press, 1980.

Mathey, Kosta, eds., Housing Policies in the Socialist Third World, London, England:
Mansell Publishing Limited, 1990.

Matlack, William F ., Statistics for Public Policy and Management, Boston,
Massachusetts: Duxbury Press, 1980.

McKean, Roland. “The Unseen Hand in Government.” American Economic Review,
496-505 (January 1965).

Meier, Kenneth J ., and Brudney, Jeffrey L., Applied Statistics for Public
Administration, Third Edition, Belmont, California: Wadsworth Publishing
Company, 1993.

Merrett, Stephen, State Housing in Britain, London: Routledge & Kegan Paul, 1979.

Ministry of Construction, ‘94 Comprehensive Housing Construction Plan, 1994. 2.

Ministry of Construction, Evaluation of Public Sector Housing Program, Korea
Research Institute for Human Settlement, February, 1989.

200

Ministry of Construction, Housing Construction Plan in the 6th ﬁve-year Economic
Social Development Plan, 1988. 8.

Ministry of Construction, The Study of Future Direction of Housing Policies, 1985.

Miron, John R. Housing Aﬁordability and Willingness to Pay, Research Paper, Centre
for Urban and Community Studies, University of Toronto, 1984.

Mittelbach, Frank eds. , Current and Future Directions of Real Estate and Urban
Development in the United States, University of California, Los Angeles, 197 8.

Mohan, Rakesh, Urban Economic and Planing Models: Assessing the Potential for
Cities in Developing Countries, World Bank Staff Occasional Papers, Baltimore:
The Johns Hopkins University Press, 1979.

Moore, Terry. “What Allow Planners to Do What They DO? A Justiﬁcation from
Economic Theory.” Journal of American Planning Association, October 197 8.

Muth, Richard F. Cities and Housing: The Spatial Pattern of Urban Residential Land
Use. Chicago: The University of Chicago Press, 1969.

Murie, A., Niner P., and Watson, C., Housing Policy and the Housing System,
London, George Allen & Unwin, 1976.

Murison, H. S., and Lea, J. P. eds. , Housing in Third World Countries: Perspectives
on Policy and Practice, New York, N. Y.: St. Martin’s Press, Inc., 1979.

Nesslein, Thomas S. “Housing: The Market versus the Welfare State Revisited.”
Urban Studies 25:95—108(1988).

Olsen, Edgar. “A Comparative Theory of the Housing Market.” American Economic
Review, 1969.

Olson, Mancur. The Logic of Collective Action: Public Goods and the Theory of
Groups. Cambridge, MA: Harvard University Press, 1965.

Payne, Geoffrey K. Urban Housing in the Third World, Boston: Routledge & Kegan
Paul, 1977.

Pamuk, Ayse. Housing Production and Transaction in Developing Countries: Do Formal
Rule Systems Matter? Ph. D. Dissertation. University of California at Berkeley. 1994.

Peattie, Lisa R. “Housing Policy in the Developing Countries: Two Puzzles.” World
Development 7: 1017-22(1979).

201

Perlman, Janice. “Misconceptions about the Urban Poor and the Dynamics of Housing
Policy.” Journal of Planning Education and Research 6(3): 187-196(Spring 1987).

Phares, Donald, eds. A Decent Home and Environment: Housing Urban America,
Cambridge, Massachusetts: Ballinger Publishing Company, 1977.

Polinsky, Mitchell, A. “The Demand for Housing: A Study in Speciﬁcation and
Grouping,” Econometrics, Vol. 45, No. 2, March, 1977.

Pollakowski, Henry 0. Urban Housing Markets and Residential Location, Lexington,
Massachusetts: Lexington Book, 1982.

Pozdena, Randall Johnston. The Modern Economics of Housing: A Guide to Theory and
Policy for Finance and Real Estate Professionals New York: Quorum Books,
1988.

Pugh, Cedric, Housing and Urbanization:A Study of India, London: Sage
Publications, 1990.

Quigley, John M. “What Have We Learned About Urban Housing Market?” In
Current Issues in Urban Economics edited by Peter Mieszkowski, and Mahlon
Straszheim, 391-429. Baltimore: Johns Hopkins University Press, 1979.

Renaud, Bertland. National Urbanization Policy in Developing Countries. New York:
Oxford University Press, 1981.

Renaud, Bertland. Housing and Financial Institutions in Developing Countries: An
Overview. World Bank Staff Working Papers no. 658. Washington, DC. , World Bank,
1984.

Renaud, Bertland. “Compounding Financial Repression with Rigid Urban Regulations:
Lessons of the Korean Housing Market.” Review of Urban and Regional Development
Studies 1:3-22(1989).

Renaud, Bertland. “Another Look at Housing Finance in Developing Countries.” Cities,
4(1):28-34(February 1987).

Renaud, Bertland. ”Resource Allocation to Housing Investment: Comments and Future
Results. " Economic Development and Cultural Change, 28(2): 389—399(January
1980).

Renaud, Bertland. “Confronting A Distorted Housing Market: Can Korean Policies
Break With The Past?” Paper presented at the Korean -US Symposium on Korean
Social Issues, Graduate School of International Relations and Paciﬁc Studies,
University of California at San Diego, June 26-27, 1992.

202

Rhee, Ha-Hyong. A Game Theoretic Analysis on the Housing Problems and Housing
Policies. Department of Public Administration, Graduate School of Korea
University, Ph D. Dissertation, 1983.

Rho, Y. H. and Ha, S. K. The Evolution of Housing Policies in I980s and its Future
Task, Korea Research Institute for Human Settlements, 1987.

Rouse, John E. Urban Housing: Public and Private - A Guide to Information Sources,
Detroit, Michigan: Gale Research Company, 1978.

Rostow, W.W. Theorist of Economic Growth from David Hume to the Present: with a
Perspective on the Next Century, Oxford: Oxford University Press, 1990.

Samuelson, P. E., “The Pure Theory of Public Expenditure.” Review of Economics
and Statistics. 386-369 November 1954.

Rostow, W.W. The Stage of Economic Growth. New York: Cambridge University
Press. 1971.

Sayegh, Kamal S. Housing: A Canadian Perspective, Ottawa: Academy Book, 1987.

Schussheim, Morton J. Toward a New Housing Policy: the Legacy of the Sixties, New
York: NY: Committee for Economic Development, 1969.

Shefer, Daniel. “The Demand for Housing, and Permanent Income, in Indonesia.”
Urban Studies 27(2): 259-272(1990).

Shidlo, Gil, eds. Housing Policy in Developing Countries, London, England:
Routledge, 1990.

Shih, Ko Ching. American Housing A Macro View. Seattle: WA Vanguard Press, Inc.,
1990.

Smart, A. “Invisible Real Estate: Investigations into the Squatter Property Market.”
International Journal of Urban and Regional Research 10(1):29-45(1986).

Smith, Lawrence B., Kenneth T. Rosen, and George Fallis. “Recent Developments in
Economic Models of Housing Development. ” Journal of Economic Literature
26:29-64(March 1988).

Smith, Wallace F ., Housing: The Social and Economic Elements, Berkeley: University
of California Press, 1971.

203

Smith, Wallace F., Urban Development: The Process and the Problems, Berkeley:
University of California Press, 1971.

Song, B. N. and Struyk, R. J. Korean Housing: Economic Appraisal and Policy
Alternatives, Korea Development Institute, 1976.

Stafford, D. C. The Economics of Housing Policy, London, Croom Helm, 1978.

Stegman, Michael A. Housing Investment in the Inner City, Cambridge, Massachusetts:
MIT Press. 1972.

Steve Chan. ”Catching Up and Keeping Up: Explaining Capitalist East Asia's
Industrial Competitiveness. " The Journal of East Asian Affairs, Winter 1991.

Strassman, Paul W.“ The Construction Sector in Economic Development” Scottish
Journal of Political Economy, 17:393-402, 1970.

Strassman, Paul W. “Measuring the Employment Effects of Housing Policies in
Developing Countries.” Economic Development and Cultural Change 24(3):623-
87(1982).

Strassman, Paul W. “Third World Housing.” In The Encyclopedia of Housing,
Garland, 1993.

Strassmann, Paul W. The Transformation of Urban Housing, World Bank Research
Publication, Baltimore: The Johns Hopkins University Press, 1982.

Strassmann, Paul W. “Housing Market Intervention and Mobility.” Urban Studies,
28(5): 759-771 1991.

Strassmann, W. P. and Blunt, Alistair, Land, Income, Mobility and Housing: The Case
of Metro Manila, Unpublished Paper, Michigan State University, 1992.

Strassmann, Paul W. Third World Housing, Unpublished Paper, Michigan State
University, 1993.

Straszheim, Mahlon. An Economic Analysis of the Urban Housing Market, New York,
National Bureau of Economic Research, Columbia University Press, 1975.

Struyk, Raymond J and Turner, Margery A. Guidelines For Creating A Housing
Finance Strategy in A Developing Country, Washington D. C.: Urban Institute
Press, 1987.

204

Struyk, Raymond J. Assessing Housing Needs and Policy Alternatives in Developing
Countries, Washington D. C.: Urban Institute Press, 1988.

Struyk, Raymond J ., and Turner, Margery Austin. Finance and Housing Quality in
Two Developing Countries: Korea and The Philippines, Washington D. C.: The
Urban Institute Press, 1986.

Struyk, Raymond J and Margery A. Turner. “Econometric Analysis: Measuring the
Impacts of Rent Controls in Urban Housing Markets.” In Housing the Poor in the
Developing World: Methods of Analysis, Case Studies and Policy edited by A. Graham
Tipple, and Kenneth G. Willis, 169-188. London: Routledge, 1991.

Struyk, Raymond J. “The Housing Needs Assessment Model.” Journal of American
Planning Association 53(2):227-234(Spring 1987).

Struyk, Raymond J. , and Margery Austin Turner. Finance and Housing Quality in TWO
Developing Countries: Korea and the Philippines. Washington D. C.: The Urban
Institute Press, 1986.

Sub, S. H. Macroeconometric Analysis of Korean Real Estate Market. Seoul: Korea. Hong
Mun $3. 1994.

The Planning and Coordination, The Ofﬁce of Prime Minister. The Evaluation of
Economic Development Plan (Housing Sector), every year.

Tipple, A. Graham and Kenneth G. Willis, eds. Housing the Poor in the Developing
World: Methods of Analysis, Case Studies and Policy London: Routledge, 1991.

Turner, John F. C. and F ichter, Robert. Freedom to Build: Dweller Control of the
Housing Process, New York: The Macmillan Company, 1972.

Turner, John F. C. Housing by People: Towards Autonomy in Building Environments.
NY: Pantheon Books, 1977.

U.S. Department of Housing and Urban Development. National Analysis of Housing
Aﬁordability: A Framework for Local Housing Strategies. Washington, D. C.:
H.U.D., November, 1993.

Vilet, Willem van, Harvey Choldin, William Michelson, and David Popenoe, eds.
Housing and Neighborhoods: Theoretical and Empirical Contributions. New York:
Greenwood Press, 1987.

Weinberg, Daniel H. “The Determinants of Intra-Urban Household Mobility.”
Regional Science & Urban Economics, 9(2/3): 219-246 (May/August 1979).

205

Whittick, Arnold et.al. Encyclopedia of Urban Planning, 1974.

World Bank. Housing: Enabling Markets to Work-- A World Bank Policy Paper.
Washington D.C.: Urban Development Division, The World Bank, 1993.

World Resources Institute. Environmental Almanac, 1992.
Worldwatch Institute. State of the World, 1990, 1991, 1992, 1993.

Yeh, SH, and Laquian, A. Housing Asia ’s Million ’s Problems, Policies, and
Prospects for Low—Cost Housing in South Asia, Ottawa, Canada: International
Development Research Centre. 1979.

Yong, Phang Sock. Housing Markets and Urban Transportation: Economic Theory,
Econometrics and Policy Analysis for Singapore, San Francisco: CA, McGraw-
Hill Book Co., 1992.

Yoon, Il-Seong. Owner-Occupied Housing in South Korea: An Analysis of the Structure
of Provision. Ph. D. Dissertation, Department of Sociology, University of Essex.
1994.

Yum, Jae Sun. A Study on the Factors Affecting a Housing (Product)- Mix Decisions
with reference to Apartment Site Development, Ph. D. dissertation, Dongguk
University, 1989.

APPENDICES

206

APPENDIX 1 . List of Variables

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

c:Pr1

 

Consumer Price Index ( 1953 base year)

 

Lubel Name Contents f. if .................... " * Units ‘
{GNPU 97; Current GNP Bil. won
:IGNPD ‘ V Constant GNP in US dollar at 1990 10 Mil. $
GNPO L_ . f Constant GNP (W) Bil. won
EGDPU‘ ' ’ Current GDP Bil. won
{GDPO l i _x I ; Constant GDP Bil. won
rQNII- ff ‘ ‘ '7 National Income Bil. won
7NIOC , National Income (constant) Bil. won
j NDI : '7 National Disposable Income Bil. won
nNDIO ., 7. _‘ National Disposable Income (constant) Bil. won
PGFC'FU , . Current ﬁxed capital formation Bil. won
j GFCFO ‘ . . ~ Constant ﬁxed capital formation Bil. won
GDCFU Current Gross Domestic Capital Formation Bil. won
j-GDCFO' ~ j Constant Gross Domestic Capital Formation Bil. won
IRFCE ~ Q Increase Rate of Final Consumption Expenditure %
E'IRGFCF ~- Z, Increase Rate of Gross Fixed Capital Formation %

: IRGFCFI ‘3 Increase Rate of Gross Fixed Capital Formation (constant) %
jCONSU: Construction Bil. won
{CONSO ' f 7, Construction (constant) Bil. won
; RESU, 5} ~ . ' Current Capital Formation in Resid. Bldg Bil. won
'RESO; ' . . Constant Capital Formation in Resid. Bldg Bil. won
IRHINVE 1 f Growth Rate of Level of Housing Investment %
IRHIV 1 Growth Rate of Level of Housing Investment (constant Reso %

r ., . . j based)

CONPCT Constant Housing Investment as % of Constant GDP %
{GPCPOG Constant GDP per Capita 1000won
ir-GPCPOZQ Q} (GDP/Constant Capita)z 1000won
CONNP * f. *' f Constant Housing Investment as of Constant GNP %
ﬁPERGNP Per capita GNP lOOOwon
:PERGNPOE“; Per capita GNP (constant) Mi1.won
FRGNP .: - If? l/GNP -
EGNPIND: GNP Index -
ECRTE .. . Economic Growth Rate %

TR ,. j Total Revemre Bil. won
TRO _ _ <  “ Total Revenue (constant) Bil. won
TE 6 “ Total Eyenditure Bil. won
{TEO ~ ,9. Total Expenditure (constant) Bil. won
EMSTWQ l. ~ ‘“ MoneLSupplL Bil. won
MSO if Money Supply (constant) Bil. won
f GRMS f ' f Growth Rate of Money Supply %
DEFLATOR Deﬂator Index -
{'DEFLAI ' Deﬂator Index -
:‘(GRDEFA ’f Gowth Rate of Deﬂator %
EXCHAN; .. ’ Exchange Rate of won to U.S. Dollar won.

 

 

207

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

‘ CPI - V ‘ Consumer Price Index ( 1970 base year) -
'[ TRBAL 1 ' * Trade Balance in million US dollars Mil. $
‘ TRBALO 3 Trade Balance in million US dollars (constant) Mil. $
IGRTRBAL Growth Rate of Trade Balance %
'ijPIj , ~ : _ y , Producer Price Index -
{SAVf , f ~ _ Gross Saving Bil. won
CSAVR - f? f Ratio of Private Savingg disposable Income %
TDOMSAV 1 7 , Ratio of Domestic Savigngs to GNP (1954-1992) %
i RDOMSAV l/DOMSAV -
INIPC " i Z" Per Captia Natioml Income 1000 won
PERGNPO2 - Per capita GNPz 1000 won
“HSGEXPIN j Average expenditure per household in the national budget (1962-92) 1000 won
[INTR ' '3 3 Interest Rate %
E'DCPI ' 1' Growth Rate at Consumer Price Index %
'jPRSAV. . : Private Saving Bil. won
CPERGNPD 1 _ Per Capita GNP in US dollar 3
RTPERGNP . Growth Rate at Per Captia GNP %
HIPC I (If g 7 Housing Investment Per Capita won
. IRCONNP . . Growth Rate at Constant Housing Investment as of Constant GNP %
j'POPU ‘ i * Population 1000
POPO ‘ V - Pop. Incities over 100,000 1000
J DPOP _: w j ~ Rate of Growth %
3DPOP2 " [3 (Rate of Growth)2 -
fURB . 1q_ l Rate of Growth in cities over 100,000 %
URB2 I . 0 (Rate oﬂgrowm in cities over 100,000)2 -
[POPUM ‘ i__ Pop. in millions 1000
EPOPOM j . f *- ,{ Pop in cities over 100,000 (in millions) Mil.
Q URBPOP} j Urban population 1000
CRURPOP __ Rural Population 1000
URBRTE. f Urbanization Rate %
QURBRTEZ ‘ Urbanization Ratez -
URBR . ' ' . annual rate of growth of cities over 100,000 divided by average %
* f. = 'p L rate of growth in national population
URBR2 ‘ 7: ~. .. URBR2 -
PDEN “ , “ . Population Density Per sz
IRUPOP ; ‘ Increase rate of urban population %
'ZPABC. . 1 j Permit Authorized for building construction: Total, Number of dwelling unit
‘ building
[PABC FLO ' " Permit authorized for building comtruction: total, ﬂoor area 1000 m2
:ZPABCFLOD- ' 1 Permit authorized for New Housing Construction by dwelling use 1000 at2
:W‘: _ 7“, 1 .5“ ,3 by ﬂoor area
[PBBCDW * V Permit authorized for New Housing Construction by dwelling use dwelling unit
j} j I * I .1 ' .g by du
NTHH } Total Number of Households 1000
NSIZE ' i = Person Per Each Households -
ﬁNHO' ? if »' } Number of Housing Units 1000
“ GRPABCFL [. Growth Rate of the New Housing Construction by ﬂoor area %
:‘GRPBBCD j : Growth Rate of the New Housing Construction by dwelling unit %
W I’ '1 C 7
[HRTE '_ .. Housing Supply Rate %

 

 

208

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

THNC ‘, . .i Number of Housing Constructed. 1000
,NDEX f . " National Defense Expenditure Bil. won
'NDEXO ” ’5 National Defense (constant) Bil. won
RATDEF ' . j Share of National Defense over Total Expenditure %
DEFSPI’ . i * Defense spending in Mil. won Mil.W.
DEFR,’ ; ‘ . ._ Defense spendigate of growth %
DEFPG ' I - ' . Defense spending as percentagﬁe of GNP %
- STCF * V ' Share of Fixed Capital Formation %
[*SHTOGD. 3, Share of GDP over Residential Building %
{YPCNI . ; ' x Share of housing ﬂestment over national income %
VLGPCPC ‘ log form of constant GDP per caﬂa -

.LNIPC . ' log form of per capita national income -
”LPERGNP log form of per capita GNP -

LCONPCT. . logform of constant housing investment as % of constant GDP -

LYPCNI V ' log form of share of housing investment over national income -

f’ LCONNP, ‘ log form of share of housing investment over GNP -

PHAREA if I Per Capita new housig construction area 0.001 m2
j LPHAREA' log form of PHAREA -
[AEXPEN . I per capita average expenditure in the national budget 1000 won
'LAEXPEN ’ log form of AEXPEN -

JPARK . , , The Park’s Regimezdummy variable -
ELECTION The Election Dummy Variable -

TWOMIL I , The Two Million Housing Construction Plan Dummy variable -
DPLSTABL, , The political stability dummy variable -

STOGRO . , ‘ growth rate of housing units (NHO) %
'HHSGRO ’ . rowth rate of households (NTHH) %

; DENGNP ‘ ‘ Defense spending as percentage of GNP %
TBR TaxBurdenRate %
: UMM. H j Unregualted interest on money markets %
SWSP . , Social Welfare Spending in Mil. won (1962-93) Mil . won
1 m " ' Growth Rate of Social Welfare Spending (1962-93) %
: SWPG 1 f i - Social Welfare Spending in Percentag of GNP (1962-93) %
SWPB- : . f I , Social Welfare Spendipgis Percentage of Budget (1962-93) %

HIHSH ] ‘~ ’ 1 . 1 Per Household Housing Investment 1000 won
HGLOPC -‘ f I Per capita housinQan from Korea Housigrg Bank 1000 won
_fHGLO . i f :, Total Housing loan from Korea HousingBank Billion won
NHF. . ‘ 'f 1 National Housing Fund by KHB Billion won
TPHF‘ . ' - f » ,' I Public Housing Fund blKHB Billion won
" HPOL 1 ’ ,,_ 7 f Strength of anti-speculation measures
~ NHGUN . * . Total number of Housinngonstruction dwelling units
ijHGUNP- } Total number of public housing construction dwellng units
LNHGUNE} Total number of private housing construction dwelling units
HCFUN ‘ ~ {C Total Housing Construction Fund Billion won
QRINTj , f Real Interest Rate %

TIWPI f L ' f Wholesale price index (1970=100) %
{MSGNPH j V» 4 M2 /GNP %
EPDSGNP- ' - j ~ I: Domestic Savings as percentgge of GNP %
ijSGNP 5 * Foreign Savings as Erecntage of GNP %
{EXGNP f Exports as percentaggf GNP %

 

 

209

APPENDIX H. Site of New Towns Around Seoul

(”x/"l
o n r )
san .f._ / .
F \j a \l
I Sangkye .)
’\ prAfV .\- .r'l ./ ./
' ./ /. Seoul L. J "
Inchon r'\.-\ i? J _r"'
.1 )Mokdong ‘_ I /
1311;533:111}- 1th ' K/
\} Kwangmyong ,f'u' I'J'
/° \-/"/ 6 gr .
. . ‘ Kwachon 0
Bundang
. 0
/ Pyongchon
OAnsan
751'
Suwon
a - 4.06m

 

210

APPENDIX HI. List of Sample Nations In Model

 

 

BG modelescoumms) “

 

 

 

 

West Germany

Australia Australia
Belgium Belgium
Bolivia Bolivia
Canada Canada
Chile Chile
Colombia Colombia
Costa Rica Costa Rica
Denmark Denmark

El Salvador El Salvador
Finland Finland
France France
Greece Greece
Honduras Honduras
Iceland Iceland

Iraq Iraq

Ireland Ireland
Israel Israel

Italy Italy
Jamaica Jamaica
Japan Japan
Kenya Kenya
Luxembourg Malta

Malta Netherlands
Netherlands Norway
Norway Panama
Panama Philippines
Philippines Portugal
Portugal Puerto Rico
Puerto Rico South Africa
South Africa South Korea
South Korea Spain

Spain Sweden
Sweden Switzerland
Switzerland Thailand
Thailand Turkey
Turkey U.S.

U.S. United Kingdom
United Kingdom West Germany

 

 

211

APPENDIX IV. Glossary

Chaebol

Chonsei

housing services

Sal-Wol-Sei
Elasticities
housing program

pyung

conglomerate dominating the Korean economy, as a group of
companies under a single and centralized control whose level of
total sales is one of the highest throughout all industries. For
more detail see Kim, Eun Mee. 1990, 1991, 1992.

the tenant pays a large cash deposit - its size is usually a third to
a half of the price of the housing - to the landlord at the beginning
of the tenancy instead of monthly rental payments. The deposit is
refunded at the end of the lease period. The landlord takes the
interest on the deposit as rent.

a phrase that deﬁnes the utility a resident gets by living in a
dwelling units. It refers to the sum of all services, inclusive of
neighborhood attribute provided by a housing unit during some
period of time such as space, privacy, availability and
dependability of utilities, and other features of the unit which
provide comfort and pleasure.

substantial deposit with monthly rent

weights in nondimensional units that assess the response of the
dependent variable to changes in the explanatory variable.

components of a wide array of public policies with multiple
goals.

the standard measurement unit for housing is the pyung; it is
equal to 3.3 In2

212

APPENDIX Variables For Time-Series Analaysis
ll: IIC'IIEI'-S'!l'

(a) CONNP: national housing construction investment as a percentage of the GNP
(GNPO) (1953-1993) Unit: %
CONNP is GNPO divided by RESO.
Source: Bank of Korea, Economic Statistics Yearbook, 1965, 1970, 1974, 1980,
1981, 1986, 1989, 1990, 1991, 1993, 1994.

(b) PHAREA: housing construction area per person (1962-1992)
PHAREA is PABCFLO(the national new housing construction area) divided
by the total population size(POPU). Unit: 0.001 m2
Source: Bank of Korea, Economic Statistics Yearbook, 1965, 1970, 1974, 1980,
1981, 1986, 1989, 1990, 1991, 1993, 1994.
(c) HIHSH: housing construction investment household (1953-1993)
HIHSH is RESO(the national housing construction investment) divided by the
total number of households (NTHH)
Source: Economic Planning Board, Housing Census, 1990. 1985. 1980.

Economic Planning Board, Economic Statistical Yearbook, 1965, 1970, 1974,
1980, 1981, 1986, 1989, 1990, 1991, 1993, 1994.

E11 llC'llﬁI'-S'!l°

Independent variables are classiﬁed into six categories: socio-economic status,
institutional setting, urbanization and demographic change, housing conditions, policy

effect, and global and foreign affairs effect.

|S°-E '5} 11'”

In order to investigate the relationship between housing investments and key

variables in the national economy, we focus on estimating the elasticity of housing

213

investments with respect to income. In addition to income and price variables, we can
include several other independent variables which may affect housing investment.

(a) PERGNPO: per capita constant gross national product (1953-1993)
PERGNPO is GNP divided by the total population (POPU).
Unit: 1000 won
Source: Economic Planning Board, Economic Statistical Yearbook, 1965 , 1970,
1974, 1980, 1981, 1986, 1989, 1990, 1991, 1993, 1994.

(b) TRO: constant total revenue; Unit: Billion won
SourcezEconomic Planning Board, Economic Statistical Yearbook, 1965, 1970,
1974, 1980, 1981, 1986, 1989, 1990, 1991, 1993, 1994.

(c) TEO: constant Total Expenditure; Unit: Billion won
SourcezEconomic Planning Board, Economic Statistical Yearbook, 1965, 1970,
1974, 1980, 1981, 1986, 1989, 1990, 1991, 1993, 1994.

((1) M80: constant money supply; Unit: Billion won
SourcezEconomic Planning Board, Economic Statistical Yearbook, 1965, 1970,
1974, 1980, 1981, 1986, 1989, 1990, 1991, 1993, 1994.

(e) Deﬂator: deﬂator index based on 1970
SourcezEconomic Planning Board, Economic Statistical Yearbook, 1965, 1970,
1974, 1980, 1981, 1986, 1989, 1990, 1991, 1993, 1994.
Bank of Korea, National Accounts, 1987.

(f) INTR: interest rate; Unit (%)
Source: Economic Planning Board, Economic Statistical Yearbook, 1965 , 1970,
1974, 1980, 1981, 1986, 1989, 1990, 1991, 1993, 1994.

(g) DOMSAV: ratio of domestic savings to GNP (1954-1992)
SourcezEconomic Planning Board, Economic Statistical Yearbook, 1965, 1970,
1974, 1980, 1981, 1986, 1989, 1990, 1991, 1993, 1994.
Bank of Korea, National Accounts, 1987.

(h) UMM: unregulated interest on money markets
SourcezEconomic Planning Board, Economic Statistical Yearbook, 1965, 1970,
1974, 1980, 1981, 1986, 1989, 1990, 1991, 1993, 1994.
Bank of Korea, National Accounts, 1987.

(i) DCPI: growth rate at consumer price index
SourcezEconomic Planning Board, Economic Statistical Yearbook, 1965, 1970,
1974, 1980, 1981, 1986, 1989, 1990, 1991, 1993, 1994.

214

Bank of Korea, National Accounts, 1987.

(j) SAVR: ratio of private savings to disposable income; Unit: %
Source:Economic Planning Board, Economic Statistical Yearbook, 1965, 1970,
1974, 1980, 1981, 1986, 1989, 1990, 1991, 1993, 1994.
Bank of Korea, National Accounts, 1987.

(k) AEXPEN: average expenditure per person in the national budget (1962-1992)
AEXPEN is the national budget divided by the total population size (TOTPOP).
Source: Economic Planning Board, Economic Statistical Yearbook, 1965 , 1970,
1974, 1980, 1981, 1986, 1989, 1990, 1991, 1993, 1994.

(l) HSGEXPIN: average expenditure per household in the national budget; Unit :
1000 won
Source:Economic Planning Board, Economic Statistical Yearbook, 1965, 1970,
1974, 1980, 1981, 1986, 1989, 1990, 1991, 1993, 1994.

(rn) CPI: Consumer Price Index based on 1970
Source:Economic Planning Board, Economic Statistical Yearbook, 1965, 1970,
1974, 1980, 1981, 1986, 1989, 1990, 1991, 1993, 1994.

(n) PPI: Producer Price Index based on 1970
Source: The Bank of Korea, Price Statistics Summary, 1994.

BE" ”.15.”

These variables are related to the role of government action in constructing

more housing.

(a) DEFSP: defense spending in mil. won ( 1961 - 1993)
Source: National Statistics Ofﬁce, Korea Economic Indicator, 1965, 1970,
1975, 1980,1985, 1990.

(b) DEFR: defense spending rate of growth (1961 - 1993)
Source: National Statistics Ofﬁce, Korea Economic Indicator, 1965 , 1970,
1975, 1980,1985, 1990.

(c) DEFPG : defense spending as a percentage of GNP (1961-1993)
Source: National Statistics Ofﬁce, Korea Economic Indicator, 1965, 1970,
1975, 1980,1985, 1990.

215

(d) DEFPB : defense spending as a percentage of budget (1961-1993)
Source: National Statistics Ofﬁce, Korea Economic Indicator, 1965, 1970,
1975, 1980,1985, 1990.

(e) SWSP: social welfare spending in mil. won (1961 - 1993)
Source: National Statistics Ofﬁce, Korea Economic Indicator, 1965, 1970,
1975, 1980,1985, 1990.

(f) SWR: social welfare spending rate of growth (1961 - 1993)
Source: National Statistics Ofﬁce, Korea Economic Indicator, 1965, 1970,
1975, 1980,1985, 1990.

(g) SWPG: social welfare spending as percentage of GNP ( 1961 - 1993)
Source: National Statistics Ofﬁce, Korea Economic Indicator, 1965, 1970,
1975, 1980,1985, 1990.

(h) SWPB: social welfare spending as percentage of the budget ( 1961 - 1993)
Source: National Statistics Ofﬁce, Korea Economic Indicator, 1965, 1970,
1975, 1980,1985, 1990.

(i) PDEN: population density; Unit: Per Km2
Source:Economic Planning Board, Population and Housing Census, 1985.

(j) ECRTE: economic growth rate; Unit: %
Source:Economic Planning Board, Economic Statistical Yearbook, 1965, 1970,
1974, 1980, 1981, 1986, 1989, 1990, 1991, 1993, 1994.

(K) RARDEF: share of defense spending over total expenditure Unit: %

Source:Economic Planning Board, Economic Statistical Yearbook, 1965, 1970,
1974, 1980, 1981, 1986, 1989, 1990, 1991, 1993, 1994.

:12 1’ III! "13'21

These variables include the growth rate of the total population and the ratio
between the rural and urban population.

(a) DPOP: annual rate of increase in the total population (1953-1993)
DPOP is the rate of increase in the total population size (POPU).
Source:Economic Planning Board, Population and Housing Census, 1985.

(b) IRUPOP: rate of increase in the urban population (1953-1993)

216

IRUPOP is based on the aggregate population of cities which includes cities and
Eup (population of over 20,000).

Source:Economic Planning Board, Population and Housing Census, 1985.

(c) URBRTE: rate of urbanization

Source: Ministry of Home Affairs, Municipal Yearbook of Korea, 1970, 1980,
1990.

Economic Planning Board, Population and Housing Census, 1985.

(d) RURPOP: rural area population

Source: Ministry of Home Affairs, Municipal Yearbook of Korea, 1970, 1980,
1990.

(e) URBPOP: Urban area population includes that of Eup with more than 20 thousand
residents; Unit: Thousand Person.

Source: Economic Planning Board, Population and Housing Census, 1985.

(f) POPU: nation-wide population
Source:Economic Planning Board, Population and Housing Census, 1985.

(g) URBR: annual rate of growth of cities over 100,000 divided by the average rate of
growth in the national population; Unit: %
URBR is IRPOPU divided by DPOP.
Source:Economic Planning Board, Population and Housing Census, 1985.

(h) URB: rate of growth in cities over 100,000

URB is based on the aggregate population of cities which includes cities over
100,000 (POPO).
Source:Economic Planning Board, Population and Housing Census, 1985.

DEL . 31'°II'11

(a) HRTE: Housing Supply Rate
Source: Ministry of Construction, Yearbook of Construction Statistics, 1993.

(b) NHO: Number of Constructed Housing Units: 1000 units

Source:Economic Planning Board, Major Statistics of Korean Economy, 1977,
1982.
Ministry of Construction, Yearbook of Construction Statistics, 1993.

(c) NTHH: Total Number of Households

Source:Economic Planning Board, Major Statistics of Korean Economy, 1977,
1982.
Ministry of Construction, Yearbook of Construction Statistics, 1993.

217

(d) NSIZE: number of persons per household (1953-1993)
NSIZE is the total population size (POPU) divided by the total number of
households (NT HH).
Source: Economic Planning Board, Major Statistics of Korean Economy, 1977,
1982.
Ministry of Construction, Yearbook of Construction Statistics, 1993.

(e) ROWNHO: Ownership Occupied Rate
Source:Ministry of Construction, Yearbook of Construction Statistics, 1993.

(f) PABC: Number of buildings
Source: Economic Planning Board, Economic Statistics Yearbook,1994.
Economic Planning Board, Korea Statistical Handbook, 1978, 1979, 1980,
1985, 1986, 1987, 1988

(g) PABCFLO: Permits authorized for building construction
Source: Economic Planning Board, Economic Statistics Yearbook,1994.
Economic Planning Board, Korea Statistical Handbook, 1978, 1979, 1980,
1985, 1986, 1987, 1988

(h) PABCFLODW: dwelling by ﬂoor area
Source: Economic Planning Board, Economic Statistics Yearbook,1994.
Economic Planning Board, Korea Statistical Handbook, 1978, 1979, 1980,
1985, 1986, 1987, 1988

(i) PBBCDW: dwelling by use
Source: Economic Statistics Yearbook,1994.
Economic Planning Board, Korea Statistical Handbook, 1978, 1979, 1980,
1985, 1986, 1987, 1988

(i) HNC: amount of Housing Constructed. Unit: 1000 Houses
Source: Economic Statistics Yearbook 1994.
Economic Planning Board, Korea Statistical Handbook, 1978, 1979, 1980,
1985, 1986, 1987, 1988

EEI' E55 ll'll

(a) TWOMIL: the two million house program dummy variable (1953-1993)
1 = 1987-1992, 0 = other years

(b) DPLSTABL: the political stability dummy variable (1953-1993)
1 = 1979, 1980, 1981, 1988, 0 = other years

218

(c) PARK: Park’s Regime dummy variable

((1) ELECTION: the election dummy Variable

Eli . m. Eh

(a) TRBALO: trade balance in million US dollars
Source: Economic Planning Board, Major Statistics of Korean Economy, 1977,
1982.

(b) EXCHANG: exchange rate of won to the U.S. dollar; Unit: won Economic
Source: Planning Board, Korea Statistical Handbook, 1978, 1979, 1980, 1985,
1986, 1987, 1988
Economic Planning Board, Major Statistics of Korean Economy, 1977, 1982.

(C) INTR: interest rate: unit: %
Source: Economic Planning Board, Korea Statistical Handbook, 1978, 1979,
1980, 1985, 1986, 1987, 1988

DataSnurces

Economic Planning Board, Housing Census, 1990. 1985. 1980.

Economic Planning Board, Population and Housing Census, 1985.

Bank of Korea, National Accounts, 1987.

Korea Housing Bank, The Statistical Yearbook of Banking Services, 1987.

Economic Planning Board, Major Statistics of Korean Economy, 1977, 1982.

Bank of Korea, Economic Statistics Yearbook, 1965, 1970, 1974, 1980, 1981,

1986, 1989, 1990, 1991, 1993, 1994.

7. Government of The Republic of Korea, The Second Comprehensive National
Physical Development Plan: 1982-1991. 1982.

8. National Bureau of Statistics, Economic Planning Board, Korea Statistical
Yearbook, 1981,

9. Bank of Korea, National Income Statistics Yearbook, 1953-1967.

10. Economic Planning Board, Korea Statistical Handbook, 1978, 1979, 1980, 1985,
1986, 1987, 1988

11. Economic Planning Board, Korea Stastistical Yearbook, 1970, 1992.

12. Ministry of Reconstruction, Korea, Development of the Korean Economy, 195 8.

13. Korea Housing Bank, Housing Economic Statistical Yearbook, 1994.

14. The Korea National Housing Corporation, Housing Handbook, 1994.

15. Ministry of Construction, Yearbook of Canstmctian Statistics, 1993.

16. Ministry of Finance, Financial Savings Statistic, 1990, 1991, 1992, 1993.

17. Natonal Statistics Ofﬁce, Monthly Statistics of Korea, every month from December,
1992 to March, 1994.

18. Natonal Statistics Ofﬁce, Papulatin and Housing Census, 1985, 1990.

19. Natonal Statistics Ofﬁce, ‘90 Population & Housing Census, 1980, 1985, 1990.

99:59.“?

219

20. Natonal Statistics Ofﬁce, Annual Report an the Family Income and Expenditure
Survey, 1985, 1986, 1987, 1988, 1989, 1990, 1991, 1992, 1993.
21. Ministy of Home Affairs, Municipal Yearbook of Korea, 1992, 1993.

22. Ministry of Construction, Land Price Statistics, 1983, 1984, 1985, 1986, 1987,
1988, 1989, 1990, 1991, 1992, 1993.

23. Natonal Statistics Ofﬁce, Major Statistics afFareign Economy, 1993.

220

APPENDIX VI. Exchange Rates, Weight, and Measures

WM

W011 -
W011
W011
WOIl

W011

won
won =
won = 1
Source: IMF, IFS. 1977.1
Economic Planning Board, Major Statistics of Korean Economy, 1977.
The Bank of Korea, Economic Statistics Yearbook, 1991, 1992, 1993, 1994.

 

2._Measures

1 pyung equals 3.3 m2

1 kilometer = 1000 meters = 0.5397 miles = 39,370 inches = 3,281 feet
1 kilometer” = 1.55 *109 inches” = 1.076*107 feet” = 0.3862 mile”

"1111111111111